Petroleum Biotechnology - Duhalt - (NAFTI - IR) PDF

Studies in Surface Science and Catalysis 151
PETROLEUM BIOTECHNOLOGY
Developments and Perspectives
This page is intentionally left blank
Studies in Surface Science and Catalysis
Advisory Editors: B. Delmon and J.T. Yates
Series Editor: G. Centi
Vol. 151
PETROLEUM BIOTECHNOLOGY
Developments and Perspectives
Edited by
Rafael Vazquez-Duhalt
Institute of Biotechnology
National University of Mexico
Morelos, Mexico
and
Rodolfo Quintero-Ramirez
Mexican Petroleum Institute
Colonia San Bartolo
Atephehuacan, Mexico
ELSEVIER
Amsterdam - Boston - Heidelberg - London - New York - Oxford - Paris - San Diego
San Francisco - Singapore - Sydney - Tokyo
ELSEVIER B.V. ELSEVIER Inc. ELSEVIER Ltd ELSEVIER Ltd
Sara Burgerhartstraat 25 525 B Street, Suite 1900 The Boulevard, Langford Lane 84 Theobalds Road
P.O. Box 211, 1000 AE Amsterdam San Diego, CA 92101-4495 Kidlington, Oxford OX5 1GB London WC1X 8RR
The Netherlands USA UK UK
© 2004 Elsevier B.V. All rights reserved.
This work is protected under copyright by Elsevier B.V., and the following terms and conditions apply to its use:
Photocopying
Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the
Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for
advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational
institutions that wish to make photocopies for non-profit educational classroom use.
Permissions may be sought directly from Elsevier's Rights Department in Oxford, UK: phone (+44) 1865 843830, fax (+44)
1865 853333, e-mail: permissions@elsevier.com. Requests may also be completed on-line via the Elsevier homepage (http://
www.elsevier.com/locate/permissions).
In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive,
Danvers, MA 01923, USA; phone: (+1) (978) 7508400, fax: (+1) (978) 7504744, and in the UK through the Copyright Licensing
Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 20 7631 5555;
fax: (+44) 20 7631 5500. Other countries may have a local reprographic rights agency for payments.
Derivative Works
Tables of contents may be reproduced for internal circulation, but permission of the Publisher is required for external resale or
distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and
translations.
Electronic Storage or Usage
Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter
or part of a chapter.
Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any
means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher.
Address permissions requests to: Elsevier's Rights Department, at the fax and e-mail addresses noted above.
Notice
No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material
herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages
should be made.
First edition 2004

Library of Congress Cataloging in Publication Data
A catalog record is available from the Library of Congress.
British Library Cataloguing in Publication Data
A catalogue record is available from the British Library.
ISBN: 0 444 51699 9

ISSN: 0167 2991 (Series)
@ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).
Printed in The Netherlands.
V
PREFACE
Without a doubt, historians will describe 20th and 21st centuries as the oil-based society. One
hundred years ago oil exploitation began, first as a source of energy and later to include oil as
a source of raw material. In addition to the 1 trillion barrels that have already been harvested,
recent estimations shows that about 3 trillion barrels of oil remain to be recovered worldwide,
half from proven reserves and half from undeveloped or undiscovered sources. Oil production
is expected to peak sometime between 2010 and 2020, and then fall inexorably until the end
of this century. After the production peak, the more expensive fuel sources will come into
production. These include hard-to-extract oil deposits, tarry sands, and Synfuels from coal
that requires alternative or complementary to conventional oil refining technologies.
Our society has an inexorable challenge: to increase the production of goods and
services for people, using new process technology that should be energetically efficient and
environmental friendly. This also will be the case for the petroleum industry. Improvements
in conventional oil refining processes such as cracking, hydrogenation. isomerization,
alkylation. polymerization, and hydrodesulfurization, certainly will occur. Nevertheless, non-
conventional biotechnological processes could be implemented. In contrast to the available
processes, biological processing may offer less severe process conditions and higher
selectivity for specific reactions. Biochemical processes are expected to be low demand
energy processes and certainly environmentally compatible.
The primary target of the petroleum industry is to enhance and maintain a continuous
oil production. Preconceived ideas and misconceptions about biotechnology continue to limit
the applications of biological processes in the chemical industry. Nevertheless, there are
biotechnological processes that have been demonstrated to be industrially successful and that
are shown to be sufficiently stable, productive and economic for commercial applications.
Even if wastewater treatment and soil bioremediation are common biotechnological
applications in the oil industry, petroleum biotechnology is still in its infancy. Doubtless,
though, biotechnology will play an increasingly important role in future industrial processes.
In this book, experts from 11 countries critically discuss the developments and perspectives of
biotechnological processes for the petroleum industry.
An integrated approach into the possibility of using petroleum biotechnology
throughout the value chain of an oil company is presented. The authors discuss the evaluation
of biotechnology as a general toolbox for solving some of the technology problems of today
and future possibilities to implement new refinery processes. Petroleum refining could be
enhanced by biochemical reactions in which the specificity exceeds by far these of chemical
reactions. The selective removal of sulfur, nitrogen, and metals from petroleum by
biochemical reactions performed by microorganisms and/or enzymes is discussed. Increasing
supply of heavy crude oils and bitumens has increased the interest in the conversion of the
high-molecular weight fractions of these materials into refined fuels and petrochemicals. This
upgrading has typically been accomplished either with high-temperature and expensive
processes thermal conversion (cracking or coking) or by catalytic hydroconversion. In
contrast to the available processes, biological processing may offer less severe process
conditions and higher selectivity to specific reactions. Enzymatic transformations of
asphaltenes in non- conventional media, and biological upgrading to improve the quality of
certain crude oils and liquid fuels could be envisaged, using biocatalysts to decrease
aromaticity and sensitize aromatic heterocycles to subsequent heteroatom removal.
Bioprocessing would complement conventional refining technologies and result in improved
fuel quality at lower capital and operating costs and with reduced environmental impact.
vi
Innovative new processes could be explored, such as methanol production from

methane. Methane monooxygenases are unique among known catalytic systems in their
ability to convert methane to methanol under ambient conditions using dioxygen as the
oxidant. The unusual reactivity and broad substrate profiles of methane monooxygenases
suggest many possible applications in the petrochemical industry. In addition, the ability of
anaerobic bacteria to convert petroleum into methane and thereby generate useful energy is a
very interesting alternative. On the other hand, biological production of hydrocarbons by
bacteria is revisited and its potential is explored, not only as an environmentally-friendly fuel
supply, but also as a renewable source for basic petrochemicals.
Microbial colonization of metal surfaces drastically changes the classical concept of
the electrical interface commonly used in inorganic corrosion. Corrosion is a leading cause for
pipe failure, and is a main component of the operating and maintenance costs of gas and oil
industry pipelines. The cost of corrosion to the gas and oil industries was estimated in 2001 to
be about $13.4 billion/yr and of this as much as $2 billion/yr may be due to microbially-
induced corrosion. In order to moderate the economic importance of corrosion in the oil
industry, molecular tools are used to study its microbial complexity.
The current knowledge of the indigenous deep subsurface microbial community in
petroleum reservoirs shows an enormous physiological diversity and constitutes a complex
ecosystem with an active biogeochemical cycling of carbon and minerals. "Souring" of oil
reservoirs by the formation of hydrogen sulfide has been a problem since the beginning of
commercial oil production. Sulfate-reducing bacteria are the culprits that produce this noxious
gas, leading to souring. This microbial process in wastewaters and oil field waters can be
controlled by another group of microbes, known as nitrate-reducing bacteria. The use of
nitrate to control microbially-produced sulfide in oil fields is a proven biotechnology that is
grossly under-used by the petroleum industry. Its effectiveness has been demonstrated in
many laboratory investigations and in some field studies. Nitrate has replaced biocides in
some of the oil fields in the North Sea, and the results have been very positive. It is now very
clear that land-based oil field operators should seriously consider using this proven
biotechnology to control, and possibly eliminate, microbially-induced souring and the
problems associated with H2S formation.
Environmentally-related biotechnological processes were pioneered in the petroleum
industry. Oil spill bioremediation technologies epitomize modern environmental techniques,
working with natural processes to remove spilled oil from the environment while minimizing
undesirable environmental impacts. The application of biological wastewater treatment in the
frame of a process integration treatment technology will hopefully close the water cycle
allowing "zero discharge" in the petroleum industry. Nowadays, water should be considered
as one of the main raw materials of the petroleum industry and its treatment and reuse with
advanced treatment technology should be applied. On the other hand, phytoremediation is an
emerging technology that is based on sound ecological engineering principles, and that has
developed into a more acceptable technology for the remediation of soils and groundwater
polluted with residual concentrations of petroleum hydrocarbons. The advantages of using
phytoremediation include cost effectiveness, aesthetic advantages, and long-term
applicability. Finally, biological air treatment systems are among the established technologies
that can be applied to control volatile organic compounds and odor emissions, and they are
applicable for a wide range of volatile pollutants found in the petroleum industry. Biological
treatment of polluted air emissions results from the competence of active microorganisms,
including bacteria, yeast, and fungi, to transform certain organic and inorganic pollutants into
compounds with lower health and environmental impact. Their applications are growing
continually based on scientific and technological developments.
vii
The powerful tools of molecular biochemistry can be used to improve the enzyme
stability and efficiency. These techniques may be applied to the particular needs of the
petroleum industry. In addition, the enzymes isolated from extremophilic microorganisms are
extremely thermostable and generally resistant to non-conventional conditions such as organic
solvents and extreme pH. Thus, many enzymes and enzymatic proteins are still to be
discovered.
Rafael Vazquez-Duhalt
The only way to discover the limits of the possible is to go beyond them into the impossible.
(Arthur C. Clarke).
ix
Table of Contents
Preface v
List of Contributors xiii
Chapter 1
Use of Petroleum Biotechnology throughout the value chain of an oil company:
An integrated approach.
H.Kr. Kotlar, O.G. Brakstad, S. Markussen and A. Winnberg
Statoil ASA. Trondheim, Norway 1
Chapter 2
Petroleum biorefining: the selective removal of sulfur, nitrogen, and metals
J.J. Kilbane II" and S. Le Borgneb
a
Gas Technology Institute, Illinois U.S.A.
b
Instituto Mexicano del Petroleo, Mexico 29
Chapter 3
Enzymatic catalysis on petroleum products
M. Ayala" and R. Vazquez-Duhaltb
a
lnstituto Mexicano del Petroleo. Mexico
b
Instituto de Biotecnologia, UNAM, Mexico 67
Chapter 4
Prospects for biological upgrading of heavy oils and asphaltenes
K.M. Kirkwood, J.M. Foght, and M.R. Gray
University of Alberta, Canada I 13
Chapter 5
Whole-cell bio-processing of aromatic compounds in crude oil and fuels
J.M. Foght
University of Alberta, Canada 145
Chapter 6
Biocatalysis by methane monooxygenase and its implications for the petroleum
industry
T.J. Smith" and H. Dalton 3
a
University of Warwick, United Kingdom
Sheffield Hal lam University, United Kingdom 177
X
Chapter 7
Biocorrosion
H.A. Videla" and L.K. Herrerah
a
University of La Plata, Argentina
University of Antioquia, Colombia, 193
Chapter 8
Molecular tools in microbial corrosion
X. Zhu and J.J. Kilbane II
Gas Technology Institute, Illinois U.SA. 219
Chapter 9
Potential applications of bioemulsifiers in the oil industry
H. Bach" and D.L. Gutnick1'
b
Tel-Aviv University, Tel-Aviv, 69978, Israel
a
Taro Pharmaceuticals New York, U.S.A. 233
Chapter 10
Anaerobic hydrocarbon biodegradation and the prospects for microbial
enhanced energy production
J.M. Suflita", I.A. Davidova\ L.M. Gieg", M. Nanny" and R.C. Prince1'
"University of Oklahoma, U.S.A.
b
ExxonMobil Research and Engineering Co., U.S.A. 283
Chapter 11
Using nitrate to control microbially-produced hydrogen sulfide in oil field waters
R.E. Eckford and P.M. Fedorak
University of Alberta, Edmonton, Canada 307
Chapter 12
Regulation of toluene catabolic pathways and toluene efflux pump expression
in bacteria of the genus Pseudomonas
J.L. Ramos, E. Duque, M.T. Gallegos, A. Segura and S. Marques
Estacion Experimental del Zaidin, CSIC, Granada, Spain 341
Chapter 13
Bacterial hydrocarbon biosynthesis revisited
B. Valderrama
Instituto de Biotecnologia, UNAM. Mexico 373
Chapter 14
The microbial diversity of deep subsurface oil reservoirs
N.-K. Birkeland
University of Bergen, Norway 385
xi
Chapter 15
Biotechnological approach for development of microbial enhanced oil recovery
technique
K. Fujiwara11, Y. Sugai1', N. Yazawa1', K. Ohno\ C.X. Hong" and H. Enomoto1
a
Chugai Technos Co. Ltd., Japan
Akita University, Japan
c
Japan National Oil Corporation, Japan
PetroChina Company Limited, China
c
Tohoku University, Japan 405
Chapter 16
Phytoremediation of hydrocarbon-contaminated soils: principles and applications
R. Kamath, J. A. Rentz, J. L. Schnoor and P. J. J. Alvarez
University of Iowa, U.S.A. 447
Chapter 17
Biological treatment of polluted air emissions
S. Revah* and R. Auria"
a
Universidad Autonoma Metropolitana-lztapalapa, Mexico.
b
Universite de Provence, France 479
Chapter 18
Bioremediation of marine oil spills
R. C. Prince and J. R. Clark
ExxonMobil Research & Engineering Co. 495
Chapter 19
Biotreatment of water pollutants from the petroleum industry
E. Razo-Flores, P. Olguin-Lora, S. Alcantara and M. Morales-Ibarria
Institute Mexicano del Petroleo, Mexico 513
xiii
List of Contributors
S. Alcantara
Institute Mexicano del Petroleo
Eje Central Lazaro Cardenas 152. C.P. 07730, Mexico D.F.
P. J. J. Alvarez
Department of Civil and Environmental Engineering, Seamans Center
University of Iowa, Iowa City, Iowa, U.S.A. - 52242
R. Auria
Laboratoire 1RD de Microbiologie, Universite de Provence
CESB/ESIL, Case 925, 163 Avenue de Luminy 13288, Marseille Cedex 9 France
M. Ayala
Institute Mexicano del Petroleo.
Eje Central Lazaro Cardenas 152, San Bartolo Atepehuacan 07730 Mexico DF, Mexico
H. Bach
Department of Molecular Microbiology and Biotechnology, Tel-Aviv University
Tel-Aviv, 69978, Israel
N.-K. Birkeland
Department of Biology, University of Bergen, Box 7800, N-5020 Bergen, Norway
O.G. Brakstad
Sintef Materials and Chemistry, Trondheim, Norway
.1. R. Clark
ExxonMobil Research & Engineering Co.
Annandale, NJ 08801
H. Dalton
Department of Biological Sciences, University of Warwick
Coventry CV4 7AL, United Kingdom
I.A. Davidova
Institute for Energy and the Environment and Department of Botany and Microbiology,
University of Oklahoma, Norman, OK 73019, USA.
E. Duque
Estacion Experimental del Zaidin. CS1C
C / Profesor Albareda 1, 18008 Granada, Spain
xiv
R.E. Eckford
Department of Biological Sciences, University of Alberta
Edmonton, Alberta, Canada T6G 2E9
H. Enomoto
Department of Geoscience and Technology, Graduate School of Environmental Studies,
Tohoku University, Aramaki, Aoba-ku, Sendai 980-0845, Japan
P.M. Fedorak
Edmonton, Alberta, Canada T6G 2E9
J. M. Foght
Edmonton, Alberta Canada T6G 2E9
K. Fujiwara
Chugai Technos Co. Ltd.
9-20 Yokogawa-Shinmachi Nisi-ku Hiroshima City 733-0013, Japan
M.T. Gallegos
Estacion Experimental del Zaidin, CSIC
L.M. Gieg
M.R. Gray
Department of Chemical and Materials Engineering, University of Alberta
Edmonton, Alberta, Canada T6G 2G6
D.L. Gutnick
Present address, Biotechnology Research Laboratories. Taro Pharmaceuticals U.S.A.,
3 Skyline Drive, Hawthorne, New York, 10532, U.S.A.
L.K. Herrerab
Faculty of Engineering, University of Antioquia, Medellin, Colombia
C.X. Hong
PetroChina Company Limited, Jilin Oilfield Company
Jilin province, China
R. Kamath
XV
J.J. Kilbanell
Gas Technology Institute, 1700 S. Mt. Prospect Rd.. Des Plaines 1L 60018
K.M. Kirk wood

Department of Chemical and Materials Engineering, University of Alberta
H.Kr. Kotlar
Statoil ASA, R & D Center, Postuttak, N-7005 Trondheim, Norway
S. Le Borgne 1
Institute Mexicano del Petroleo, Eje Central Lazaro Cardenas 152, Col. San Bartolo
Atepehuacan, 07730 Mexico D.F., Mexico
S. Markussen
Department of Marine Environmental Technology, Trondheim, Norway
S. Marques
M. Morales-lbarria
Instituto Mexicano del Petroleo
Eje Central Lazaro Cardenas 152, C.P. 07730, Mexico D.F.
M. Nanny
Institute for Energy and School of Civil Engineering and Environmental Science,
K. Ohno
Technology Research Center, Japan National Oil Corporation
1-2-2 Hamada, Mihama-ku, Chiba 261-0025, Japan
P. Olguin-Lora
Instituto Mexicano del Petroleo
Eje Central Lazaro Cardenas 152, C.P. 07730, Mexico D.F.
R. C. Prince
Annandale. NJ 08801
J.L. Ramos
xvi
E. Razo-Flores,
Institute Potosino de Investigation Cienti'fica y Tecnologica
Camino a la Presa San Jose 2055,. C.P. 78216, San Luis Potosi, SLP, Mexico.
.1. A. Rentz
University of Iowa, Iowa City, Iowa. U.S.A. - 52242
S. Revah
Department of Process Engineering, Universidad Autonoma Metropolitana-Iztapalapa
(UAM-I). Apdo. Postal 55-534, 09340 Mexico D.F., Mexico
J. L. Schnoor
A. Segura
T.J. Smith
Biomedical Research Centre, Sheffield Hallam University
Howard Street, Sheffield SI 1WB, United Kingdom
.I.M. Suflita
University of Oklahoma, Norman. OK 73019, USA.
Y. Sugai
Akita University Venture Business Laboratory
1-1 Tegatagakuen-cho Akita City ,010-8502, Japan
B. Valderrama
Departamento de Ingenieria Celular y Biocatalisis, Universidad Nacional Autonoma de
Mexico. AP 510-3. Cuernavaca, Morelos, 62250, Mexico.
R. Vazquez-Duhalt
Instituto de Biotecnologia, UNAM.
Apartado Postal 510-3 Cuernavaca, Morelos 62250 Mexico
H.A. Videla
Department of Chemistry. College of Pure Sciences, IN1FTA, University of
La Plata, Argentina
A. Winnberg
Department of Biotechnology, N7465 Trondheim, Norway
xvii
N. Yazawa
Technology Research Center, Japan National Oil Corporation
1-2-2 Hamada. Mihama-ku, Chiba 261-0025, Japan
X. Zhu
Gas Technology Institute, 1700 S. Mt. Prospect Rd., Des Plaines 1L 60018
R. Vazquez-Duhalt and R. Quintero-Ramirez (Editors)
© 2004 Elsevier B.V. All rights reserved. l
Chapter 1
Use of Petroleum Biotechnology throughout the value

chain of an oil company: An integrated approach.
H.Kr. Kotlara, O.G. Brakstad", S. Markussenc and A. Winnbergc.
a
Statoil ASA, R & D Center, Postuttak, N-7005 Trondheim, Norway
Sintef Materials and Chemistry, bDept. Marine Environmental Technology,

c
Dept. Biotechnology, N7465 Trondheim, Norway
1. INTRODUCTION TO AN INTEGRATED APPROACH

The history of biotechnology goes thousands of years back in time. One of the
very first written statements of biotechnology is found in the Bible, telling that
Lot was drinking wine, made through fermentation around 2000 B.C.E. In
modern time Antoni van Leeuwenhoeck was the first to observe a micro-
organism in a primitive microscope in 1684. Louis Pasteur discovered how to
protect against diseases by vaccination, using heat-inactivated organisms,
around 1863. In 2002 the gene sequence of the human genome was completed.
Biotechnology is continuously expanding, and will play an increasingly
important role in future industrial process. Petroleum biotechnology is a very
young and exiting part of these industrial possibilities
It is well established that petroleum reservoirs contain active and diverse
populations of microorganisms. Microbial growth within oil reservoirs has
traditionally been associated with biofouling and souring. Furthermore, the
potentials for microbial improved oil recovery (MIOR) have been investigated
for many decades (see chapter 15)[1]. Recently, nitrate injection was introduced
as a method for curing reservoirs "contaminated" by sulphate-reducing
prokaryotes (see chapter 11)[2]. However, petroleum biotechnology possesses
several other opportunities besides MIOR and nitrate injection. This chapter will
focus on some of these issues.
The primary target of the petroleum industry is to enhance and maintain a
continuous oil production. In 1998/1999 Statoil initiated an R&D program
2
looking into the possibility of using petroleum biotechnology as an integrated

approach throughout the value chain of the oil company.
There were three main objectives:
1: Evaluation of biotechnology as a general toolbox for solving some of the
technology problems of today.
2: Investigate future possibilities; e.g. to start refinery processes in the reservoir
using dedicated microorganisms.
3: To generate a resource base for new genetic information achieved from the
organisms in the reservoir.
These objectives may be achieved through focusing on biotechnology as a
new business concept of interest to the company. Coverage of all aspects of
biotechnology would be an enormous task. However, the enhanced in-house
understanding of reservoir microbiology has served as a basis for the few
selected areas described below:
. New techniques in exploration and production:
Application of molecular biology techniques as new tools for specific
identification and characterization of hydrocarbon sources during exploration
and production. Samples may come from drill cuttings from exploration
wells; produced oil and formation water; sediments from sea floor seep zones;
etc.
• Biological well treatments (preventive medication):
Clogging of wells by scaling, hydrates, etc. may be prevented by applying
environmentally friendly biological produced chemicals. This may be
achieved by developing self-sustained, natural existing or bioengineered
microbial populations placed inside the reservoir. The target is to produce
biological substances that can replace traditional chemicals, and that this
remediation will increase treatment lifetime to ensure a continuous oil
production.
• Bioreactors:
Low energy biological processes for up-grading of oil to improve quality and
thereby reduce penalty pricing. Various types of bioreactors and enzyme
systems can replace traditional catalysts for certain chemical reactions, waste
handling or the production of bio-energy.
• New application of extremophiles:
New thermophilic and piezophilic enzyme system can enable new bio-
engineering processes and products for applications in the above-mentioned
areas, or give rise to entirely new products and business opportunities.
Combined approaches of microbiology, biochemistry and DNA technology

are used to obtain microorganisms with specifically designed metabolic
3
functions. Such organisms can be applied in reservoirs for the production of

various treatment products or enzymes in situ. Thermophilic enzymes may also
be employed to overcome possible fundamental problems related to the growth
characteristics of these microorganisms. Additionally, the "gene-pool" of the
indigenous microbial assemblages of the reservoir have direct implication to the
success of the product in the above suggested business areas.
Environmental aspects/public awareness: Apart from providing technical

solutions, the outcome of this program will have a great impact on meeting the
environmental challenge of the future. The Norwegian authorities consider many
of the production chemicals applied in the fields today as harmful, and in the
Norwegian sector of the North Sea there is a program for phasing out such
chemicals, replacing them with more environmentally acceptable alternatives.
Biotechnology may provide us with more environmentally friendly alternatives.
Value generation: This program will contribute to increasing and maturing

the reserve base (upstream), as well as creating business opportunities or
increasing market shares downstream. The Fig. 1 below illustrates the potential
influence of biotechnology throughout the entire value chain within an oil
company.
Fig. 1. Biotechnology throughout the value chain.

4
The main challenges are related to:

• The biological activities in a reservoir are still poorly understood. Growth
control of reservoir microbes, and the knowledge to achieve this control, will
be crucial. In bioreactor-type processes, however, this will be possible.
. There are fundamental questions related to energy pathways and reaction rates
that need to be resolved. Direct use of tailor-made enzyme system might
bypass some of these obstacles.
• In bioreactors, the main challenge is to achieve sufficient reaction rates that
are required for a commercial process. This is not a challenge from the
microbiological aspect only, but also from a chemical engineering point of
view.
Acquiring new knowledge: In order to balance the beneficial and detrimental

effects of microbial growth in the reservoir, new knowledge is required. Growth
and possible excretion of products under different reservoir conditions are not
well known. To date, various types of chemicals are injected into the reservoir in
order to maintain or restore oil production, e.g. to counteract or minimize the
influence of scaling, hydrate and asphaltene precipitation. Occasionally,
chemicals and antibiotics are injected to prevent microbial growth. Some of
these chemicals are known to serve as energy source for the microorganisms, i.e.
nitrogen, phosphor and carbon sources. [3-4]. Reservoir conditions vary
significantly, and thus, the microbial communities will respond differently
depending on this external influence. It is imperative to acquire in depth
understanding of the growth and production of microbial products under the
different reservoir conditions. In this respect modeling tools may be used to
simulate how the changes will influence on the indigenous microorganisms.
Joint efforts from internal experts and external collaborators are vital to the
success of this type of projects. Much knowledge on microbial technologies
already exists but the molecular biology approach represents a bold and
important step forward.
The nature of this research requires long-term commitment and support from
the R&D management. A thorough understanding and awareness of the ethical
implications is needed for all involved.
2. MICROBIAL DNA FINGERPRINT TECHNIQUES IN EXPLORATION AND

PRODUCTION
Several studies have documented microbial communities in hot oil reservoirs

(see chapter 14)[5-9]. Indigenous microbial communities have also been
detected in core samples and water saturated regions of reservoirs [10].
Members of indigenous reservoir communities may include strictly anaerobic
sulfate-reducing prokaryotes [5, 11-12] and methanogens [13-15], as well as
5
other microbes [9, 15]. Thus, one would expect to find genetic markers of
microbial activities both during exploration, drilling and production.
Statoil has filed a patent application for utilization of DNA technologies
as a tool for identification and characterization of hydrocarbon sources during
drilling or sampling from sea floor seep zones. Drill cuttings from exploration
wells, sediments from sea floor seep zones or other specimens could be analyzed
with a selection of specific DNA probes/markers. These specific DNA probes
are taken from microbes found to be linked to different oil producing fields in
the North Sea and other sources. The energy sources for these organisms will be
constituents of the oil, gas or others, specific for the reservoir zones and
conditions of the particular field
This genetic tool may give valuable information on possible migration
routes of the hydrocarbon from the source rock. Specific recognition patterns
might also be used in monitoring different reservoir zones during production,
and further indicate the individual contribution of the particular zone to the
overall production. Possibly, sweep efficiency pattern could be calculated.
Detection of DNA from drill cuttings, sediments, or core samples during
explorative drilling may result in defined species pattern, resulting in indications
of potential hydrocarbon bearing zones (Fig. 2).
Fig. 2. System for characterization of microbes in exploration cores by culture-dependent and

-independent approaches, based on 16S rRNA gene sequencing. The sequences are used for
the generation of DNA probes to be used for screening of cores.
6
2.1. Microbial diversity in oil reservoirs

It is essential to establish databases of the microbial ecology in petroleum
reservoirs. Genetic tools for exploration and production can then be developed.
The knowledge of the in situ microbial activities should be improved through an
interdisciplinary collaboration between specialists in petroleum exploration and
production, chemists and microbiologists. Understanding the interactions
between the biosphere and the geosphere is essential.
The microbial diversity of two North Sea reservoirs (termed reservoir A and
B) has been studied in some detail [16-17]. Both a culture collection and a 16S
rDNA library have been established for these reservoirs.
2.1.1. Culture-independent methods

Culture-independent methods have recently been used for the
characterization of microbial communities in some oil reservoirs [9-10]. In these
studies, DNA was extracted directly from reservoir samples (produced water,
core samples, drill cuttings etc.) This approach was used for the comparison of
microbial assemblages in some North Sea reservoirs with different reservoir
characteristics and production histories. In our studies microbial communities
differed significantly between the reservoirs (Fig. 3). Sequence studies of 16S
rDNA clones from reservoir A showed that 32 % of the clones aligned to the
sulfide reducing thermophile Archaeoglobus fulgidus, while bacterial clone
inserts aligned to a variety of types, including Sphingomonas, Herbaspirillum,
Nevskia, Aquabacterium, Alcanivorax, Bacillus and Acetobacterium. Clones
from reservoir B were dominated by sequences aligning to the a-proteobacteria
Erythrobacter, the sulfide-oxidizing e-proteobacteria Arcobacter, the
halotolerant y-proteobacterium Halomonas, and the thermotogales Geotoga.
Several of the microbial genes detected in our studies have been found in
produced fluids or enrichment cultures from oil reservoirs in the Pacific Ocean
or Canada [9, 18]. The differences in the assemblage compositions between oil
reservoirs and other subsurface structures may reflect the geochemical
influences on the community structures [19-20]. Biodegraded oils dominate the
world's petroleum inventory, and microbial activities play an essential role in
most oil reservoirs [21]. Recent studies have emphasized the impact of an active
potentially indigenous subsurface community [19].
2.1.2. Culture-based methods

Most studies of reservoir communities have been conducted by culture-
based methods [7-8, 22-23]. As a supplement to the culture-independent
characterization of the two North Sea oil reservoirs, culture-based methods were
used to study the diversity of the cultivable microbes in produced fluid from the
reservoirs. Enrichment media for fermentatives, methanogenes, sulfide-
oxidizers, sulphate-reducers and acetogenes were designed, and cultures from
7
the two reservoirs showed dominance of small rods, single or in short chains,
and sheathed rods (Thermotogales like). Pure isolates were obtained from only
one of the reservoirs, reservoir A. Even though the enrichments from the other
reservoir, reservoir B, showed a variety of organisms, it was not possible to
obtain any pure isolates from these. The 16S rDNA clones from these
enrichments aligned to Thermosipho japonicus, Bradyrhizobium and
Aquabacterium. 16S rDNA clones from isolates from reservoir A, showed
dominance of Archaeobglobus fulgidus, Methanococcus thermolithotrophicus,
Thermococcus sibiricus and Thermosipho japonicus. Several of the sequences
abundant in the cultures were not found in the clone library from the culture-
independent approach (2.1.1). This is in accordance with other studies [9], and
suggests that several of the predominant members of the enrichment cultures
(e.g. Thermosipho) are not the predominant member of the reservoir
communities, but show fast-growing characteristics in several of the culture
media. Other cultures included a-, P-, s- and y-Proteobacteria Sphingomonas,
Stenotrophomonas, Halomonas meridiana, and Geospirillum, and the Gram-
positive bacterium Thermoanaerobacter ethanolicus.
Fig. 3. DGGE analysis of PCR-amplified 16S rDNA sequences from two North Sea oil
reservoirs, reservoir A (1, 2) and reservoir B (3, 4, 5). Only sample 2 contained fluids with
seawater penetration.
8
Thermophilic species of Thermotogales, Archaeoglobus, Thermoanaero-

bacter, Methanococcus and Thermococcus have been reported from high-
temperature oil reservoirs [6-9, 14]. Several of these microbes are typical sulfur-
utilizers, being active in desulphurization of crude oil. These microbes may be
the predominant sources for H2S generation rather than typical sulphate-
reducing bacteria, and interestingly several of them were enriched in culture
media designed for SRB.
2.1.3. Detection of specific microbes

Monitoring of microbes in the oil reservoir has traditionally been
accomplished by culture methods, e.g. MPN methods for quantification of
viable sulphate-reducing bacteria (SRB), as recommended by the American
Petroleum Institute [24]. Some commercial techniques have also been
introduced, for instance a commercialized immunoassay for semi-quantification
of the SRB-specific enzyme APS reductase [25]. Monitoring may also include
molecular biology methods. Currently, two RNA-based methods are
investigated, fluorescence in-situ hybridization (FISH) and nucleic acid
sequence-based amplification (NASBA). By using RNA detection mainly the
metabolic active cells are assessed. The FISH methods include fluorescence-
labeled DNA probes for the targeting of specific microbes. An example is given
in Fig. 4 where bacteria, archaea, Archaeoglobus, Arcobacter and Erythobacter
are enumerated in production fluids from two reservoirs. These methods may be
further refined for offshore analysis by using field equipment, e.g. the Microcyte
fluorescence cell counter. NASBA is an isothermic alternative to PCR [26].
Real-time miniaturized lab-on-a-chips systems are currently under development
with the NASBA technology as basis [27].
2.1.4. Characterization ofmicrobial dynamics by microarrays

Nucleic acid microarrays have recently been introduced for phylogenetic
identification in microbial ecology. Basically, microarrays consist of series of
specific DNA probes (grabber probes) that are printed on glass slides. Sample
nucleic acids are extracted and labeled (e.g. by fluorescence) and incubated on
the slides, followed by recording. Labeled detector probes may be used for
detection as alternatives or supplements to labeled target DNA [28]. The
microarrays are made quantitative by employing reference DNA to normalize
variations in spot size and hybridization (29). The methods provide a powerful
tool for parallel detection of 16S rRNA genes [30-31] and may be particularly
useful for environmental studies of phylogenetically diverse groups. Although
most arrays are based on the PCR amplification of target genes prior to array
hybridization, systems have also been described where direct profiling of
extracted rRNA from environmental samples have been used [32]. Printed slides
may be brought offshore and target genes quantified directly on the platforms by
9
portable devices. Arrays have also been established for the assessment of
functional gene diversities and distribution, for instance with genes from the
nitrogen cycling [33-34]. For offshore conditions the sulphur and nitrogen
cycles may be addressed during curing of biological souring by nitrate injection.
3. BIOREACTOR: POTENTIAL USE OF BIOCATALYSTS IN CRUDE

OIL UP-GRADING AND REFINING
Until recently, research within oil biotechnology mainly focused on bio-

degradation and bioremediation in connection with clean up after oil spills, and
less on the application of microbial systems in industrial processes. However,
the interest in the latter has been growing the last years, addressing problems
like asphaltenes, high sulfur content, the poor transportability of heavy crudes
due to high viscosity, the presence of heavy metals and polyaromatic/
heterocyclic compounds (see chapters 2, 3, 4 and 5).
The aim of our activity is to use biotechnological processes in up-grading of
"problem" oils/heavy oil and refinery fractions. The overall scope is to define
microbial/biotechnological technologies along the crude oil value chain that will
give the potential highest cost-benefits, competing with or being superior to
existing methods, or even better, provide solutions where no acceptable methods
exist. In the current program there has been focused on:
Fig. 4. FISH enumeration of the total concentrations of cells (DAPI), bacteria (EUB338),
archaea (ARCH915), Arcoglobus (ARGLO605) and thermotogales (THERSI672) in produced
fluids from two North Sea reservoirs, Reservoir A and Resevoir B wl and w2.
10
Reduction of the viscosity of heavy crudes through partial degradation of

waxes and/or asphaltenes, thereby increasing the transportability.
Microbial or enzymatic ring opening of polyaromatic hydrocarbons in
refinery distillates in order to increase the fraction of aliphatic components.
Removal of heavy metals such as nickel and vanadium from crude oils
through microbial sequestering, thereby simplifying the subsequent refining
of the crude.
Although chemical means to tackle the above problems exist, they are often
relatively expensive and may lead to pollution of the environment.
Biotechnological processes may represent new and more environmentally
friendly alternatives for value enhancement of heavy oils and partially distilled
petroleum products.
3.1. Pre-refining
Up-grading of crude oils by biocatalytic processes may take place anywhere
from down-hole to the refinery; in the reservoir, at the wellhead, during tanking,
transport and storage. The pre-refining opportunity is to utilize the time slot
from the start of drainage in the reservoir to the crude reaches the refinery stage.
At any of these stages, a specially designed biocatalyst could be introduced (see
Fig. 1). Although there will be considerable differences between traditional
crude oils and the heavy crudes in physical handling as well as refinery
processes, the chemistry of the compounds that need to be bio-converted could
be close relatives within the same classes.
3.1.1. Increased transportability by biocatalytic cleavage of heavy compounds

Extraction, transportation and handling of heavy oils often represent a
problem due to high viscosity. Several classes of molecules are important in
building viscosity. These are asphaltenes, waxes and the more heavy fractions of
polyaromatics. Controlled biodegradation of asphaltenes and waxes in heavy
crudes are highly desirable, as these processes could lead to a substantial
economical gain (see chapter 4).
Wax is degraded by several bacterial species that use the degradation
products for their metabolic pathways [35-36]. Efficient methods for isolation of
wax-utilizing microorganisms with the help of selective media, bacteriophages,
and paraffin wax baiting system have been developed [37-38]. Although the
enzymology of the wax degradation is not understood, some clues have been
obtained through studies of wax biosynthesis by certain bacteria, such as
Acinetobacter spp. [39-40].
11
Biodegradation of asphaltenes seems to represent a more challenging

problem - very few publications is found on this subject. However, several
studies have shown that biodegradation of asphaltenes occurs in nature [41], and
that certain bacteria, such as Acinetobacter and Providencia, proliferate in
environments containing high amounts of asphaltenes [42]. Fungi capable of
"erosion" of hard coal due to the cleavage of asphaltenes have also been
reported [43], as well as combined steam/bacteria treatment of asphaltene
depositions [44]. In addition, biodegradation of bitumen has been observed [45],
and bacteria like Pseudomonas, Flavobacterium, Acinetobacter, and
Caulobacter growing on bitumen-contaminated surfaces have been described
[46].
Potential processes are not limited to the natural occurring microorganisms
and their native enzymes. By gene technology it is possible to improve key
enzymes by rational engineering and by use of "gene shuffling" techniques.
These methods make it possible to rapidly "adapt" a given enzyme to new
substrates, or dramatically change the enzyme's properties such as Km, pH and
temperature optimum [47]. The modified enzyme(s) may then be introduced
into the appropriate microorganism(s) and its over-production, may greatly
enhance the ability of this microbe(s) to reduce the viscosity of heavy oils.
3.1.2. Demineralization - Biosorption of heavy metals

Demineralization of heavy oils that contain considerable amounts of Ni and
V is an important issue for oil industry due to refinery stage catalyst poisoning.
Several reports describing the use of microorganisms for bioremediation of
environments polluted with heavy metals, suggest that the use of microbes for
demineralization of heavy oils is possible [48-49].
Six mechanisms for microbial resistance to heavy metals have been
described: exclusion by a permeability barrier, intra- and extra-cellular
sequestration, active transport by efflux pumps, enzymatic detoxification, and
reduction of sensitivity of cellular targets to metal ions [50]. For
demineralization of heavy oils, sequestration and enzymatic detoxification seem
to be the most relevant mechanisms to study. In our current work we have just
entered this particular field of research.
3.2. Biocatalytic refining, distillate quality improvements

Perio-refining or post-refining technologies might also be of interest.
Although some of these areas have been addressed elsewhere in this book, we
would like to convey some of our own work (see chapters 2, 3, 4 and 5).
12
3.2.1. Selective ring opening

The mechanisms, the biochemical pathways, and the genetics of degradation
and bioconversion of hydrocarbons in general, and polycyclic aromatic
hydrocarbons in particular have been extensively studied [51-53]. The research
has mainly concentrated on biodegradation and bioremediation in connection
with cleanup after oil spills etc., and less on the application of these systems in
processes. However, the interest in the latter has been growing the last years. In
the petroleum industry there is a desire for products with a larger fraction of
aliphatic components, and thus a higher H/C-ratio, and microbial/enzymatic ring
opening of aromatics may be used to achieve this (see chapter 5).
Development of biocatalysts for aromatic- and heterocyclic ring opening,
including nitrogen compounds such as the polycyclic compound carbazole is of
particular interest. Middle distillate fractions from thermochemical conversion
of heavy oils contain di- and tricyclic aromatics with low fuel value. These are
currently upgraded by expensive high pressure-high temperature chemical
hydrogenation. A Canadian research group [54-55] has suggested an alternative
to thermochemical cracking: "microbial cracking" - a two-step process where
the aromatic rings first are cleaved enzymatically by a blocked mutant under
"near ambient conditions", followed by hydrogenation of the oxygenated
product under mild chemical conditions. Our group is currently engaged in a
project, "Upgrading of crude oils and refined products" involving selective ring
opening of aromatic distillates. In this work, a blocked mutant of Sphingomonas
is used for studies of bioconversion of aromatic distillates in a bioreactor [56].
Bioconversion of aromatic compounds in a real feedstock from crude oil in a

bioreactor system. The content of polyaromatic hydrocarbons (PAH's) in the
diesel fuels contribute to low cetane numbers and particle emissions from
combustion. The present study focuses on the use of a continuous bioreactor
system for up-grading of light gas oil (LGO) feed stock from the refinery as a
potential industrial process. This is done by biocatalytic ring opening of the
PAH's to generate a more paraffmic diesel fuel.
Two different bacterial strains, Sphingomonas yanoikuyae N2 and
Pseudomonas fluorescence LP6a 21-41 (donated by Dr. Julia Foght, University
of Edmonton Canada), and a mixed blend of six different strains were compared
for biocatalysis of the PAH's in the LGO feed stock using a fed batch
reactor/semi-continuous reactor.
13
Fig. 5. Schematic outline of the procedure for making blocked mutants with an inactive
enzyme by gene disruption.
The P. fluorescence LP6a 21-41 was obtained by transposon mutagenesis

and its genetic background remains unknown (see chapter 5). The PAH
degradation pathway of S. yanoikuyae was genetically engineered in order to
obtain a recombinant strain accumulating one of the intermediates, 2-
hydroxychromene-2-carboxylate. Thus, the degradation of PAH would
terminate after the ring opening. This is important for keeping the octane
number of the hydrocarbon fraction, and this was achieved by inactivating the
gene encoding the specific hydratase-aldolase enzyme, (NahE), by gene
disruption (Fig. 5). The mixed blend consisted of six different strains obtained
from commercial culture collections and isolates from mud samples collected at
a water purification plant. The organisms in this combined blend were not
genetically modified to terminate the degradation of PAH's after the ring-
opening step.
The LGO feed stock did not have any toxic effects in concentrations up to 50
vol%. In these studies, a continuous feed of 20 vol% was used. Comparison of
the N2 and Lp6a 21-41 mutants show that the two strains have different uptake
mechanisms and different preferences for certain PAH's. The N2 strain shows
the highest conversion of the least substituted aromates (Fig. 6), while Lp6a 21 -
41 show a somewhat broader specificity range (data not shown).
14
Fig. 6. Bioconversion of light gas oil by the specially designed Sphingomonas spp. N2.
In order to apply the concept to a real industrial process, higher degrees of

conversion of the more substituted aromatic compounds are necessary. The
enzyme systems in the PAH degrading pathway of N2 were found to be too
specific. Using the mixed biocatalytic blend a broader range of substrate
conversion was observed. More than 30 % of both the di- and the tri aromatic
compounds were removed from the LGO feedstock; in addition, approximately
30 % of the sulfur containing substrates was removed (Fig. 7). As already
mentioned, the mixed blend had not been genetically modified to terminate the
degradation of PAH's after the ring-opening step. The further uses of this mixed
biocatalytic blend with respect to developing an industrial process; will demand
genetic modification of the strains
The results achieved in the fed batch reactor are now being verified in a
continuous bioreactor to mimic a potential industrial process. Figure 10 shows
the schematic outline of the continuous bioreactor.
In conclusion, microorganisms with biocatalytic pathways that will
selectively convert aromatic compounds in a crude hydrocarbon mixture without
degrading aliphatic compounds exist. Such strains have been used as model
systems for studies of bioconversion of aromatic distillates (LGO) from the
refinery in a bioreactor system.
15
Fig. 7. Efficient bioconversion by a mixed biocatalyst.
The PAH degradation pathway of Sphingomonas yanoikuyae DSM 6900

have been genetically modified in order to obtain a recombinant strain that
terminates the PAH degradation after the ring-opening.
The LGO feedstock from the refinery has been shown to have no toxic
effects on the tested organisms, S. yanoikuyae mutant N2 and Pseudomonas
fluorescence LP6a mutant 21-41, in concentrations up to 50 vol%. This is of
vital significance, because in an overall technological process it will be of
importance to keep the water volumes as low as possible.
The uptake mechanism and also the substrate specificity differ between the
two strains. The substrate specificity seems to be rather narrow for each
(both) of the strains, non- or mono substituted PAH's were the preferred
substrates.
Importantly, no C is lost by breaking the C - C chains in the blocked
mutants. The organisms are not gaining energy by the reaction. It is of value
that neither the fuel properties nor the cetan number are lost.
A broader range of substrate specificity was observed with a mixed
biocatalytic blend. More than 30% of both the di- and tri aromatic
16
compounds and approximately 30 % of the sulfur containing substrates were

removed from the LGO feedstock in a continuous bioreactor system.
In future refinery processes this might replace the energy-expensive

distillation processes. These results suggest that bioreactor systems have the
potential for up-grading of hydrocarbon refinery fractions, heavier distillates and
possibly crude oils. In the years to come governmental regulations will be very
strict on both PAH and sulfur content in the diesel fuel. These preliminary
studies are thought as initial steps in a process of making a more environmental
acceptable diesel fuel with dramatic reduction in both PAH's and sulfur content,
while still maintaining adequate fuel combustion values (Fig. 8). This will be a
bio-upgraded environmental friendly diesel.
Bioconversion for more

environmental friendly diesel fuel
Fig. 8. Bio-reactor for conversion of PAH's in a real feedstock from crude oil
17
Study of pure enzyme vs. whole cell based biocatalysts. In future investigations
this will include "the aromatic ring opening dioxygenase system". The
Sphingomonas yanoikuyae N2 will be used as a model system for comparing
enzyme and whole cell biocatalysts. In many instances it is an advantage to use
pure enzyme systems instead of whole cells as biocatalysts (see chapter 3).
Enzyme reactions are specific and easy to control, they can be carried out in
non-aquatic environments, and enzymes, as other chemical catalysts, will not
consume carbon i.e. the carbon content in the fuel will be preserved. The
opening of the aromatic ring (e.g. naphthalene, Fig. 9) is a four step enzymatic
process starting with a dioxygenase reaction, then a dehydrogenation followed
by a second dioxygenase reaction and finally an isomerization. The first
oxygenation requires NADH, but the formed NAD+ is recycled to NADH in the
dehydrogenation reaction. The challenge is to develop a system where this
multistep enzyme reaction could proceed efficiently in a cell free system.
Fig. 9: Metabolic pathway of naphthalene showing the enzymes involved See reference [57],
18
3.2.2. Bioreactors
Bioconversion of refinery fractions may take place using growing or resting
cells, "dead" cells, or immobilized cells or enzymes as biocatalysts. Aromatic
ring-opening involves a multistep metabolic pathway. Multistep enzymatic
reactions often require co-factors and/or reducing power (NAD (P) H) that has
to be regenerated or supplied for the enzymatic reaction to take place. Thus,
whole cells, rather than pure enzymes, are often required. The biocatalysts are
usually contained in the aqueous phase and the reaction take place either in this
phase or at the interface between the aqueous and the organic/oil phase. The
components in the refinery fraction that are being up-graded usually show low
water solubility, while the converted products usually are more soluble in the
aqueous phase than in the organic/oil phase. Mass transfer of substrates and
products between the water and oil phase is a major challenge. To achieve
adequate mass transfer, reactors capable of generating a large interface between
oil and water should be chosen. Various types of bioreactors have been
employed by others [58], including stirred tank reactors, airlift reactors,
emulsion phase contactors reactor and fluidized bed reactors. The current
investigation has used stirred tank reactors run in batch, fed-batch and
continuous mode with free growing or resting cells. However, immobilized cells
and enzymes are included in the next phase of studies.
Fig. 10. Schematic of a bioreactor for continuous feed of LGO.

19
Continuous processes are well suited for multiphase processes. In the

continuous bioreactor based on a stirred tank reactor in fig. 10, a continuous
stream of substrate (oil phase) is run through the reactor while the biocatalyst (in
the water phase) is recycled. Recycling of the biocatalysts reduces the amount of
water needed in the process. The overall economy of the process is also
dependent upon the lifespan of the biocatalyst and their stability in water/oil
media. In a continuous reactor it is possible to regenerate or boost the
biocatalyst. In the current studies, problems have been encountered connected to
formation of stable emulsions. The emulsion increases mass transfer, but the
stable emulsions made phase separation problematic. Currently, different
approaches are explored to solve this problem.
Enzymes or cells may be immobilized by binding or adsorption to
membrane surfaces or beads, or by entrapment in a matrix. In a continuous
reactor with an immobilized biocatalyst, it is possible to have a higher
biocatalyst concentration, little or no water in the reactor, and the product
separation is easy. Reactions with purified enzymes might be easier to control
compared to whole cell biocatalysts (see chapter 3). Whole cells may contain
different metabolic pathways and could lead to production of several unwanted
by-products. By co-immobilization of series of enzymes in the water phase of
the reactor, it might be possible to run multi step enzymatic reactions.
Realistic cost of developing new technology. New technologies are often

met with obstructive arguments. Sentences like "it cannot be done" and "it is
impossible" are customary. Such arguments are "progress killers", and within
the oil industry, new techniques will have to compete with traditional
technology that has been optimized for the last 50 years. A lesson can be learned
from the Canadians. None of their syncrude technologies for mining bitumen
would have been available today if they had listened to the "wise guys" 14 years
ago. At that time the operational cost of the technology was more than 30
US$/bbl, today the operational cost is down to around 10 US$/bbl.
The OPEX (operational expenditure) profile (Fig. 11) illustrates the cost
developments in developing new technology for mining bitumen. This curve
profile is believed to be quite universal for most new technology
implementations.
4. WELL TREATMENTS TO SECURE CONTINUOUS PRODUCTION

BY PREVENTIVE MEDICATION.
MICROBES AS SELF-GENERATING SYSTEMS
Preventive medication could be defined as intelligent treatment concepts

performed in advance during the complementation phase, before the impairment
20
in productivity occur in the well. The preventive actions are to avoid the onset of
these predicted situations.
With the advance in drilling and completion, increasing number of complex
and expensive wells are being installed, e.g. multilateral, multi-zones, sidetrack
and horizontal. The infrastructures that are in place, such as flow lines and
platforms, also enable the targeting and drainage of the additional reserves found
near the exiting fields. Very often these additional oil and/or gas are produced
via tieback and satellite facilities. Successful treatments of stimulation, scale
squeeze and tubing deposit removal in these wells can no longer rely on the
traditional method of bullheading. Special tools such as coil tubing and
inflatable plug will be needed to place the chemicals accurately down-hole.
Intervention in these wells will be prohibitory expensive due to tools hire,
personnel and extended period of deferred oil production (tools run). It is
important to realize that for certain type of completion, well re-entry is almost
impossible despite accepting the financial penalty. There is clearly a need to
develop an intervention free system for these wells that allow the flow of oil
unhindered and preferably with the chemicals pre-delivered down-hole.
Syncrude Canada OPEX
Fig. 11. OPEX profile in developments of new technology for mining bitumen. The curve
shows the measured cost until 1998, then the further projection. The bars in 99, 00 and 01 are
the actual cost. (Maurice B. Dusseault, personal communication).
21
4.1. Preventive treatment: Increased productivity by self-generating- or

more environmental friendly treatment system/processes (scale, hydrate,
asphaltenes, wax, etc.)
The generation of effective production chemicals could be achieved using a
self-sustained, natural existing or bio-engineered, microbial population. This
will protect and free the well from most other intervention treatment and could
be of great economical interest to an oil company, enhancing both well recovery
and well productivity.
This will imply the search for microbes that have the genetic machinery to
produce certain treatment chemicals (i.e. organic acids, enzymes, surfactants,
antifreeze-proteins etc). Alternatively, genetic engineering could be used to
introduce this capability to the organisms. Such organisms could be introduced
to the near well bore area by various means (i.e. squeezed with/without solid
support, immobilized, combined with nutrients, etc), to produce the treatment
chemicals.
If the organism is not fit for life under the reservoir conditions, the bacteria
can be used in bioreactors to produce the desired product.
Bypassing the problems of placements: Correct placement of the treatment
fluids is of crucial importance to the overall treatment success. Numerous
treatments have failed due to poor placement. Nonetheless, in many wells,
especially in gravel packed wells, uniform placement is difficult to achieve.
With this new technology placement should no longer be the problem.
The strategies of this new technology are illustrated in fig. 12 and include:
Placement of the treatment during the completion stage. This can be done
either by bullheading the specially designed organism together with nutrients
into the formation, or by coiled tubing (CT) deployment.
Use of porous particles soaked with the product placed inside the gravel
packs at the completion face.
Use of micro encapsulation, with the desired microorganism together with
nutrition inside the capsules. Inject far beyond the critical matrix in the well.
If successful, this concept constitutes the only possible self sustained and
lasting method by which production chemicals can be produced in situ and to
allow wells to operate free of most interventions.
22
In situ production of treatment chemicals
Fig. 12. Schematic view of in situ production of treatment chemicals.
4.2 Green treatment products

In order to prove the basic concept of the above-mentioned technology of
preventive medication for secured production, the following approaches have
been made:
A synthetic gene, coding for polyaspartate (polyAsp), has been cloned in
E.coli. In a construct with 75 basepairs, coding for 25 amino acids, with a fusion
protein included, the polyAsp polypeptide was expressed in the host cell.
Most service companies in the oil industry are supplying polyAsp as a
combined scale - and corrosion inhibitor. Recently, polyAsp has also proved to
be an efficient bridging agent, boosting the squeeze lifetime of traditional scale
inhibitor jobs. PolyAsp is classified as a green treatment product, being more
than 60 % biodegradable and non-toxic. From 2005 the Norwegian government,
through chart 12 and the Norwegian Pollution Authorities, SFT, will implement
a "zero harmful discharge" policy for the Norwegian sector of the North Sea.
This will focus the search for more environmental friendly treatment products.
On shore bioreactor: E. coli will not survive during reservoir conditions.
However, the bacteria can be used in bioreactors to produce the desired product.
Bioreactor production of PolyAsp might prove to be economically feasible.
23
Down hole: Work is in progress, introducing the corresponding synthetic

gene construct into a vector, compatible with extremophiles. This is a first step
towards down hole application
5. NEW APPLICATION OF EXTREMOPHILES IN OIL RELATED

INDUSTRY
5.1. Bioprospecting of the gene pools

Oil quality may be linked to microbial growth in oil reservoirs. This has
been substantiated in fields with biodegraded heavy oils. Although biogenic
reservoir processes seem to be slow [21] oil is utilized as carbon source and
water as a source of inorganic nutrition. The reservoir microbes, acting at high
temperature and pressure, have preferences or tolerance for these extreme
conditions. Enzymes from extremophilic microbes may be tailor-made for
industrial systems run at high temperature and pressures, i.e. systems in which
enzymes from mesophilic microbes will not function. Such enzyme systems
may be utilized inside the reservoir, in bioreactors, in waste handling or in
energy processes. DNA technology may be used to link appropriate enzyme
systems to microbes growing at relevant temperature and/or pressure conditions.
An immediate prerequisite for the utilization of microbes and enzymes from
the hot oil reservoirs will be to perform surveys of the genetic pools within the
reservoirs. The knowledge about microbial species in these environments is
constantly increasing, but the understanding of the interactions between the
microbes and their environments is still limited. It will be essential to
characterize active enzyme systems in the reservoirs. Complete genomes have
been sequenced for several microbes detected in oil reservoirs, including
Archaeoglobus fulgidus and Methanococcus jannaschi [59-60]. Recent progress
in molecular microbial ecology has revealed that traditional culturing methods
fail to represent microbial diversity in nature, since only a small proportion of
viable microorganisms in most environmental samples are recovered by
culturing techniques. Methods to investigate the full extent of microbial
genomes in nature include the use of BAC (bacterial artificial chromosome)
vectors or random shotgun sequencing techniques [61-62]. These approaches
also have potentials for characterization of the complete genomic structures in
oil reservoirs. Besides explaining microbial structure-function relationships in
the reservoirs, the genomic libraries may be excellent tools for prospecting of
novel biocatalysts [63].
5.2. Thermophilic/extremophilic enzymes

New application of extremophilic/thermophilic enzyme systems: The concept
is to investigate the commercial utilization of thermophiles. These organisms
have enzyme systems working at high temperature, and often at high pressure.
24
Such enzymes are tailor-made as catalysts in industrial processes performed at

extreme conditions. Enzymes from most mesophilic microbes will not function
as the high temperature will denaturate their proteins (e.g. the enzymes). Such
enzyme systems will work placed either inside the reservoir, in bioreactors, in
waste handling or in energy processes.
5.3. Future prospective

The petroleum biotechnology is still in its infancy and will play an
increasingly important role in the future industrial processes. Within the oil
company it will have a substantial economical impact throughout the value
chain. This will influence on the development of:
New techniques in exploration and production
Biological well treatments (Preventive medication)
Biocatalytic up-grading of oil
New application of extremophiles
Acknowledgement
The authors would like to thank Statoil for the permission to publish this
book chapter and for their support in the "Applied Biotechnology" program.
Many thanks to our special adviser, Hakon Rueslatten, for valuable help and
discussions.
REFERENCES
[1] M.J. Mclnerney and K.L. Sublette, In: C.J Hurst, G.R Knudsen, MJ Mclnerney, L.D
Stetzenbach, and M.V. Walter (eds.), Manual of Environmental Microbiology, ASM
Press, Washington, D.C., 1997, pp. 600-607
[2] M. Nemati, TJ. Mazutinec, G.E. Jenneman, G. Voordouw, J. Ind. Microbiol.
Biotechnol. 26 (2001) 350.
[3] D. Lee, D. Lowe and P. Grant, 47th Annu. Cim. Petrol. Soc. Tech. Mtg. (Calgery),
Vol.2, Pap. no. cim. 96- 09.
[4] R.S. Bryant, SPE/DOE - 35356 1 (1996) 127.
[5] J.T. Rosnes, T. Torsvik, and T. Lien, Appl. Environ. Microbiol. 57 (1991) 2302.
[6] K.O. Stetter, R. Hubert, E. Blochl, M. Kurr, R.D. Eden, M. Fielder, H. Cash and I.
Vance, NATURE 365 (1993) 743.
[7] S. L'Haridon, A.L. Reysenbach, P. Glenat, P. Prieur and C. Jeanthon, NATURE. 377
(1995)223.
[8] Y. Takahata, M. Nishijima, T. Hoaki and T. Marauyama, Appl. Environ. Microbiol. 66
(2000) 73.
[9] VJ. Orphan, L.T. Taylor, D. Hafenbradl and E.F. DeLong, Appl. Environ. Microbiol.
66 (2000) 700.
[10] I. Spark, I. Patey, B. Duncan, A. Hamilton, C. Devine and C. McGovern-Traa, Clay
Minerals 35 (2000) 5.
25
[11] C. Tardy-Jackuenod, P. Caumette, R. Matheron, C. Lanau, O. Arnauld and M. Magot,

Can J. Microbiol. 42 (1996) 259.
[12] J. Beeder, R.K. Nilsen, T. Thorstensen and T. Torsvik, Appl. Environ. Microbiol. 62
(1996)3551.
[13] T.K. NG, P.J. Weimer and L.J. Gawel, Geomicrobiol. J. 7 (1989) 185.
[14] R. Nilsen and T. Torsvik, Appl. Environ. Microbiol. 62 (1996) 728.
[15] G.S. Grassia, K.M. McLean, P. Glenat, J. Bauld and A.J. Sheehey, FEMS Microbiol.
Ecol. 21 (1996)47.
[16] O.G. Brakstad, S. Ramstad, G. Eidsaa, B.M. Hustad and H.K. Kotlar, 99th Annual
Meeting of the American Society for Microbiology, Chicago, (1999)
[17] O.G. Brakstad, K. Bonaunet and H.K. Kotlar, Proceedings to the Oil & Gas Science and
Technology Conference on Microbiology of hydrocarbons: state of the art and
perspectives, Paris, June 6-7 (2002).
[18] G. Voordouw, S.M. Armstrong, M.F. Reimer, B. Fouts, A.J. Telang, Y. Shen and D.
Gevertz, Appl. Environ. Microbiol. 62 (1996) 1623.
[19] V.J. Orphan, S.K. Goffredi, E.F. Delonga and J.R. Boles, Geomicrobiol. J. 20 (2003)
295.
[20] K. Takai, M.R. Mormile, J.P. McKinley, F.J. Brockman, W.E. Holben, W.P. Kovacik
and J.K. Fredrickson, Environ. Microbiol. 5 (2003) 309.
[21] I. M. Head, D.M. Jones and S.R.Larter, NATURE 426 (2003) 344.
[22] M. Magot, B. Ollivier and B. Patel, Antonie van Leeuwenhoek 77 (2000) 103.
[23] R.K. Nilsen, J. Beeder, T. Thorstenson and T. Torsvik, Appl. Environment. Microbiol.
62 (5) (1996) 1793.
[24] API, (1975)
[25] Gawel et al, Eur. Pat. Appl., No. 0272916 Al (1987)
[26] M. Uyttendaele, R. Schukking, B. Vangemen and J.Debevere, J. Appl. Bacteriol. 77
(1994) 694.
[27] A. Gulliksen, L. Solli, F. Karlsen, H. Rogne, E. Hovig, T. Nordstrom and R.Sirevag,
Anal. Chem. 76 (2004) 9.
[28] A. Spiro, M. Lowe and D. Brown, Appl. Envir. Microbiol. 66 (2000) 4258.
[29] J.C. Cho and J.M. Tiedje, Appl. Environ. Microbiol. 68 (2002) 1425.
[30] D.Y. Guschin, B.K. Mobarry, D. Proudnikov, D.A. Stahl, B.E. Rittmann and A.D.
Mirzabekov, Appl. Envir. Microbiol. 63 (1997) 2397.
[31] W.T. Liu, A.D. Mirzabekov and D.A. Stahl, Environ. Microbiol. 3 (2001) 619.
[32] S. El Fantroussi, H. Urakawa, A.E. Bernhard, J.J. Kelly, P.A. Noble, H. Smidt, G.M.
Yershov and D.A. Stahl, Appl. Envir. Microbiol. 69 (2003) 2377.
[33] L.Wu, D.K. Thompson, G. Li, R.A. Hurt, J.M. Tiedje and J. Zhou, Appl. Envir.
Microbiol. 67 (2001) 5780
[34] G. Taroncher-Oldenburg, E.M. Griner, C.A. Francis and B.B. Ward, Appl. Envir.
Microbiol. 69(2003)1159.
[35] H.M. Alvarez, O.H. Pucci and A. Steinbuchel, Appl. Microbiol. Biotechnol. 47 (1997)
132.
[36] F. Kawai, M. Shibata, S. Yokoyama, S. Maeda, K. Tada and S. Hayashi,
Macromolecular Symposia 144 (1999) 73.
[37] A.R. Massengale, R.A. Ollar, S.J. Giordano, M.S. Felder and S.C. Aronoff, Diagn.
Microbiol. Infect. Dis. 35 (1999) 177.
[38] F. McKenna, K.A. El-Tarabily, S. Petrie, C. Chen and B. Dell, Lett. Appl. Microbiol.
35 (2002) 107.
[39] S. Reiser and C. Somerville, J. Bacteriol. 179 (1997) 2969.
26
[40] T. Ishige, A. Tani, Y. Sakai and N. Kato, Appl. Environ. Microbiol. 66 (2000) 3481.
[41] Z. Wang, M. Fingas, E.H. Owens, L. Sigouin and C.E. Brown, J. Chromatogr. A. 926
(2001) 275.
[42] D.R. Kadavy, B. Plantz, C.A. Shaw, J. Myatt, T.A. Kokjohn and K.W. Nickerson, Appl.
Environ. Microbiol. 65 (1999) 1477.
[43] M. Hofrichter, F. Bublitz and W. Fritsche, Fuel. Proc. Technol. 52 (1997) 43.
[44] A.Y. Zekri and R. El-Mehaideb, J. Petrol. Sci. Eng. 37 (2003) 123.
[45] T.L. Potter and B. Duval, Environ. Sci. Technol. 35(2001)76.
[46] W. Dott and D. Schoenen, Zentralbl. Bakteriol. Mikrobiol. Hyg. 180 (1985) 436.
[47] K.A. Powell, S.W. Ramer, S.B. Del Cardayre, W.P. Stemmer, M.B. Tobin, P.F.
Longchamp and G.W. Huisman, Angew. Chem. Int. Ed. Engl. 40 (2001) 3948.
[48] D.L. Gutnick and H. Bach, Appl. Microbiol. Biotechnol. 54 (2000) 451.
[49] A.C. Greene, B.K. Patel and A.J. Sheehy, Int. J. Syst. Bacteriol. 47 (1997) 505.
[50] M.R. Bruins, S. Kapil and F.W. Oehme, Ecotoxicol. Environ. Saf. 45 (2000) 198.
[51] T. Bugg, J.M. Foght, M.A. Pickard, M.R. Gray, Appl. Environm. Microbiol. 66 (12)
(2000) 5387.
[52] D.T. Gibson and R.E. Parals, Curr. Opinion in Biotechnol. 11 (2000) 236.
[53] D.R. Boyd, N.D. Sharma and C.C.R. Allen, Curr. Opinion in Biotechnol. 12 (2001) 564.
[54] P.M. Fedorak, M.A. Pickard, M.R. Gray and J.M. Foght, Prepr. Symp. Am. Chem. Soc.
Div. Fuel Chem. 43 (3) (1998) 515.
[55] J.M. Foght, P.M. Fedorak, M.A. Pickard and M.R. Gray, 48. Ann. Tech. Meet. Petr.
Soc. (Calgary), Paper 97-13:1-9. (1997)
[56] H.K. Kotlar, K. Rasmussen, K. Grande, M. Ramstad, S. Markussen, A. Winnberg, S.
Zotchev and M. Gimmestad, In proceedings of 225th ACS National Meeting, New
Orleans, LA, March 23-27 2003.
[57] The University of Minnesota Biocatalysis/Biodegradation Database
http://umbbd.ahc.umn.edu/naph/naph_image_map.html
[58] B.L. McFarland, DJ. Boron, W. Deever, J.A. Meyer, A.R. Johnson and R.M. Atlas,
Crit. Rev. Microbiol. 24 (2) (1998) 99.
[59] H.P. Klenk, R.A. Clayton, J.F. Tomb, O. White, K.E. Nelson, K.A. Ketchum, RJ.
Dodson, M. Gwinn, E.K. Hickey, J.D. Peterson, D.L. Richardson, A.R. Kerlavage, D.E.
Graham, N.C. Kyrpides, R.D. Fleischmann, J. Quackenbush, N.H. Lee, G.G. Sutton, S.
Gill, E.F. Kirkness, B.A. Dougherty, K. McKenney, M.D. Adams, B. Loftus, S.
Peterson, C.I. Reich, L.K. McNeil, J.H. Badger, A. Glodek, L.X. Zhou, R. Overbeek,
J.D. Gocayne, J.F. Weidman, L. McDonald, T. Utterback, M.D. Cotton, T. Spriggs, P.
Artiach, B.P. Kaine, S.M. Sykes, P.W. Sadow, K.P. DAndrea, C. Bowman, C. Fujii,
S.A. Garland, T.M. Mason, G.J. Olsen, CM. Fraser, H.O. Smith, C.R. Woese and J.C.
Venter, NATURE 390 (1997) 364.
[60] C.J. Bult, O. White, G.J. Olsen, L.X. Zhou, R.D. Fleischmann, G.G. Sutton, J.A. Blake,
L.M. FitzGerald, R.A. Clayton, J.D. Gocayne, A.R. Kerlavage, B.A. Dougherty, J.F.
Tomb, M.D. Adams, C.I. Reich, R. Overbeek, E.F. Kirkness, K.G. Weinstock, J.M.
Merrick, A. Glodek, J.L. Scott, N.S.M. Geoghagen, J.F. Weidman, J.L. Fuhrmann, D.
Nguyen, T.R. Utterback, J.M. Kelley, J.D. Peterson, P.W. Sadow, M.C. Hanna, M.D.
Cotton, K.M. Roberts, M.A. Hurst, B.P. Kaine, M. Borodovsky, H.P. Klenk, CM.
Fraser, H.O. Smith, C.R. Woese and J.C. Venter, SCIENCE 273 (1996) 1058.
[61] M.R. Rondon, P.R. August, A.D. Bettermann, S.F. Brady, T.H. Grossman, M.R. Liles,
K.A. Loiacono, B.A. Lynch, LA. MacNeil, C. Minor, CL. Tiong, M. Gilman, M.S.
Osburne, J. Clardy, J. Handelsman and R.M. Goodman, Appl. Envir. Microbiol. 66
(2000)2541.
27
[62] G.W. Tyson, J. Chapman, P. Hugenholz, E.E. Allen, RJ. Ram., P.M. Richardson, V.V.
Solovyev, E.M. Rubin, D.S. Rokshar and J.F. Banfield, NATURE 428 (2004) 37.
[63] S. Voget, C. Leggewie, A. Uesbeck, C. Raasch, K.-E. Jaeger and W.R. Streit, Appl.
Envir. Microbiol. 69 (2003) 6235.
©2004 Published by ElsevierB.V. 29
Chapter 2
Petroleum biorefining: the selective removal of sulfur,

nitrogen, and metals
J.J. Kilbane IIa and S. Le Borgne"
a
Gas Technology Institute, 1700 S. Mt. Prospect Rd., Des Plaines, Illinois 60018
b
Instituto Mexicano del Petroleo, Eje Central Lazaro Cardenas 152, Col. San
Bartolo Atepehuacan, 07730 Mexico D.F., Mexico
1. INTRODUCTION
The quality of petroleum is progressively deteriorating as the highest quality

petroleum deposits are preferentially produced. Consequently the concern about
the concentrations of compounds/contaminants such as sulfur, nitrogen, and
metals in petroleum will intensify. These contaminants not only contribute to
environmental pollution resulting from the combustion of petroleum, but also
interfere with the processing of petroleum by poisoning catalysts and
contributing to corrosion. The selective removal of contaminants from
petroleum while retaining the fuel energetic value is a difficult technical
challenge. New processes are needed and bioprocesses are an option. Existing
thermo-chemical processes, such as hydrodesulfurization, can efficiently remove
much of the sulfur from petroleum but the selective removal of sulfur from
compounds such as dibenzothiophene, the removal of organically bound
nitrogen, and the removal of metals cannot be efficiently accomplished using
currently available technologies. The specificity of biochemical reactions far
exceeds that of chemical reactions. The selective removal of sulfur, nitrogen,
and metals from petroleum by biochemical reactions performed by
microorganisms and/or enzymes has been demonstrated. However, further
research is needed before biorefining technology can be commercialized. This
chapter reviews the status of biorefining and discusses topics requiring further
research.
30
The geochemical conversion of organic matter into petroleum is a slow

and inefficient process. It is estimated that 23.5 tonnes of plant material/biomass
are required to form a single liter of petroleum during geological periods of time
[1]. Moreover, the current rate of energy consumption is 400 times greater than
the capacity of the planet to produce biomass. It behooves us to utilize our fossil
fuel legacy as efficiently as possible while avoiding environmental damage.
Environmental regulations limit the amount of sulfur oxides emitted to the
atmosphere by the combustion of fossil fuels by regulating the concentration of
sulfur in these fuels. In particular, transportation fuels are severely regulated.
For example, the permissible concentration of sulfur in diesel has been
progressively decreased over the past decade from 500 ppm to 10 to 15 ppm [2].
Environmental regulations do not specify concentration limits for nitrogen and
metals in transportation fuels, but such regulations may be forthcoming as these
compounds inhibit the catalytic converters used to cleanse exhaust gases from
vehicles. The future use of petroleum products to power fuel cells may provide a
further impetus to decrease the sulfur, nitrogen, and metal content of petroleum
derived fuels as reforming catalysts and fuel cell electrodes are sensitive to
impurities at ppm levels [3].
Sulfur, nitrogen, and heavy metals in petroleum not only contribute to
environmental pollution when oil is burned, they also decrease the efficiency of
catalytic cracking and hydrotreating processes in oil refineries by poisoning the
involved catalysts [4-6]. Although the focus on biologically upgrading
petroleum has mainly been on sulfur, the ability of certain microorganisms to
metabolize organonitrogen compounds may be particularly important because
organonitrogen compounds are associated with the majority of metals in
petroleum [4, 7]. Thus, by metabolizing nitrogen-containing compounds, it may
be possible to simultaneously achieve the selective removal of nitrogen and
heavy metals, mainly nickel and vanadium, from petroleum. Furthermore,
nitrogen compounds contribute to the instability of petroleum byproducts [7-9].
The selective removal of sulfur, nitrogen and heavy metals from oil would be
highly desirable, but the existing physicochemical processes are not completely
effective and moreover they are not environmentally friendly.
1.1 Composition of crude oils

The sulfur content of crude oil can vary from 0.03 to 7.89 wt% [2, 4, 7,
10]. Sulfur is present in crude oil almost exclusively as organic sulfur. While
there are multiple types of organosulfur compounds such as mercaptans,
sulfides, disulfides and thiophenes, the most abundant form of sulfur in
petroleum is usually thiophenic [7, 10]. Thiophenic sulfur often comprises 50%
to 95% of the sulfur in crude oil and derived fractions, and alkylated derivatives
31
of dibenzothiophene are the most common organosulfur compounds typically

found in crude oil and fractions used to produce diesel. Alkylated derivatives of
benzothiophene are the most abundant organosulfur compounds in gasoline [11].
The sulfur, nitrogen, and metal content is preferentially associated with the
higher molecular weight components of crude oils and, consequently, heavy
crude oils typically have higher sulfur, nitrogen, and metals content than light
crude oils. Similarly, when crude oil is refined, the sulfur, nitrogen, and metals
concentrate into the high molecular weight fractions.
Nitrogen compounds typically found in crude oil consist of heterocycles
such as quinoline and carbazole, which are examples of basic and non-basic
organonitrogen compounds, respectively [4, 8, 9]. The total nitrogen content of
crude oil is typically about 0.3%, but it can be as high as 5%. Basic
organonitrogen compounds in petroleum usually comprise 25 to 30% of the total
nitrogen and include compounds such as quinoline and pyridine [12]. The
majority of basic organonitrogen compounds in crude oils are actually alkylated
derivatives of quinoline and pyridine. Quinolines and related nitrogen
heterocycles have relatively high solubility in water and can be significant
environmental contaminants. Quinoline and related compounds have been
shown to be hepatocarginogens in rodents, and mutagens in toxicity tests [4, 13-
19]. Non-basic organonitrogen compounds typically comprise 70 to 75% of the
total nitrogen in crude oils and alkylated derivatives of carbazole are the most
typically found non-basic organonitrogen compounds. Basic compounds like
quinoline are generally more reactive in the inactivation of catalysts than non-
basic compounds. However, non-basic compounds can potentially be converted
to basic compounds during the refining/catalytic cracking process. The
inactivation of catalytic cracking and hydrotreating catalysts decreases the
efficiency of operation of a refinery and results in lower yields of transportation
fuels [5]. The heavy metals found in greatest abundance in crude oil are nickel
and vanadium, which are both potent inhibitors of refining catalysts. These
heavy metals are typically associated with nitrogen compounds [20].
1.2 The need for new technologies to upgrade crude is intensifying

The standard practices used within the petroleum industry over decades
are not capable of treating heavy oils and residuum, so the need for new
technologies is intensifying (see chapter 4). The sulfur, nitrogen, and heavy
metal content as well as the average molecular weight of available crude oils in
the U. S. and in the world has increased significantly in recent years and will
continue to increase due to the progressive depletion of light crude oils
reservoirs [2, 6, 11, 21-24]. Multiple factors including physics, chemistry,
environmental concerns and market forces contribute to this trend of increasing
the heteroatom and metal content as well as the molecular weight of presently
32
available crude oils. However, the bottom line is that light crude oil is more
readily recovered, and more readily processed/refined, than heavy oil.
Consequently, deposits of light, low heteroatom content oil are preferentially
brought into full production while known deposits of heavy/high heteroatom
content petroleum are produced at less than full capacity or may even be idle.
Moreover, as a deposit of light oil is harvested, the lighter fractions are
preferentially removed such that after primary or secondary production mainly
heavy oil remains. Because of this irreversible trend, the time when available
crude oil is predominantly or exclusively heavy with a high heteroatom and
heavy metals content is not far off.
The chief concern is for the sulfur content of petroleum, but the nitrogen
and metal content of petroleum is also of concern due to environmental,
processing and corrosion concerns [23]. In North America, over 3 trillion barrels
of known petroleum reserves are largely untapped or underutilized because of
their high sulfur content and viscosity [22]. It is well known in petroleum
chemistry that sulfur and heavy metals are preferentially associated with the
higher molecular weight fractions of oil [4, 7]. So, not only is light oil easier to
produce because of its physical properties, but it also contains significantly less
undesirable impurities in comparison with heavy oils. Sulfur, nitrogen and heavy
metal impurities are of great environmental concern since they originate acid
rain as a consequence of sulfur and nitrogen oxides emissions from the
combustion of petroleum derived fuels, and potential health effects due to high
concentrations of heavy metals on combustion ashes [2, 10, 24, 25]. Some sulfur
and nitrogen heterocycles are suspected carcinogens [8, 12] and sulfur
compounds in oil have been implicated in the corrosion of pipelines and refinery
equipment [7, 24, 26]. Heavy metals content, mainly nickel and vanadium,
contributes to the poisoning of catalysts used in hydrodesulfurization or in
catalytic cracking [5, 20, 23]. In addition to catalysts poisoning by heavy metals,
sulfur and nitrogen in heterocyclic compounds are capable of poisoning catalysts
by causing electronic modifications in Pd, Pt, Ni, and Ru compounds. The
poisoning of catalysts exasperates the problems associated with the processing
of heavy oils and residuum by interfering with the methods employed to reduce
the heteroatom content and molecular weight, i.e. hydro treatment and cracking.
The quantity of heavy oils to be processed is increasing not only due to
the depletion of light oils but also to the increasing demand for cleaner
transportation fuels and other low molecular weight products, so the importance
of technologies capable of dealing with heavy oils and residuum has increased
[11, 27]. This increased demand for lower molecular weight petroleum products
seems incompatible with the use of heavy oils as primary feedstocks because of
their metals, nitrogen and sulfur content that increases the production of coke
and gas and accelerates catalyst deactivation [6]. But at the same time, the need
to obtain greater quantities of gasoline, diesel and aviation fuels from each
33
barrel of oil demands increased attempts to further process residuum. Therefore,

the petroleum industry is clearly forced to process increasing quantities of heavy
crude oils and heavy residuum which have increased levels of environmental
contaminants and are not efficiently treated by existing technologies.
In the years 1972 through 1985, the U.S. petroleum industry spent
approximately $1.4 billion in capital and operating expenditures for dealing with
pollution abatement. The National Petroleum Refineries Association estimated
that meeting Clean Air Act regulations, i.e. achieving a sulfur content of 0.05%
for diesel fuel in 1994, had cost about $3.3 billion in capital expenditures and
$1.2 billion in annual operating costs [22, 24, 28]. Similar estimations are not
available for the desulfurization, denitrogenation and/or demetallation of heavy
crude oils but considering that diesel fuel is far easier to desulfurize and handle
than heavy oils and residuum, it is possible to predict that the costs associated
with the upgrading of heavy oils and residuum would be correspondingly higher.
Because of the quantity of petroleum consumed in the U.S., the price differential
between high and low sulfur fuels and the opening or reopening of markets for
high sulfur petroleum reserves, it is estimated that the value of an alternative
desulfurization technology as biodesulfurization, which is capable of upgrading
existing production, is in excess of $10 billion annually in the U.S [22]. Clearly
then, there is a need for alternative technologies to deal with heavy oils and
residuum and there is ample economic incentive for the development of
appropriate technologies that are immune from the technical problems that limit
existing technologies. Biorefining, which is defined here as the application of
biotechnology to the upgrading of petroleum, is one such technology.
2. BIODESULFURIZATION
Effective technologies for the treatment of heavy crude oils have been and
continue to be a topic of keen interest. Hundreds of processes related to the
desulfurization of heavy oils have been described in the patent literature, and the
interest in such processes has steadily increased [2, 29]. Hydrodesulfurization
can be used to desulfurize heavy oils and residuum, but does not lead to a
significant decrease in molecular weight. The predominant method for
upgrading heavy oils to decrease molecular weight and increase the yield of
transportation fuels is the use of fluid catalytic cracking (FCC) [4, 30].
However, FCC cannot achieve desulfurization. Moreover hydrodesulfurization
and FCC catalysts are poisoned in the process of treating heavy oils because of
the presence of sulfur and nitrogen heterocycles and heavy metal contaminants
[5, 31]. Therefore, a combination of technologies is needed to address both the
removal of heteroatoms and the decrease in molecular weight in order to
mitigate environmental problems and to get the greatest yield of value added
34
products. Of all of the chemical forms of sulfur in crude oil, the most recalcitrant
to hydrodesulfurization is the thiophenic sulfur in thiophene and
dibenzothiophene derivatives. Because of the abundance of alkylated
dibenzothiophenes in crude oil and the recalcitrance of these compounds to
hydrodesulfurization, there is a high level of interest in technologies that can
effectively desulfurize dibenzothiophenes [2].
Researchers have been examining the possibility of biodesulfurization of
petroleum or other fossil fuels for over 4 decades [32]. Presently, there is no
commercial operation for biodesulfurization of fossil fuels, however several
economic studies indicate a favorable prospect of developing such a technology
[10, 24, 25, 27, 33]. Numerous microorganisms have been described in the
literature that are capable of utilizing dibenzothiophene (DBT) as sole source of
carbon, energy and sulfur. However, the complete degradation of organosulfur
compounds is not beneficial for upgrading crude oils and derived fuels. The
selective cleavage of carbon-sulfur bonds in DBT and derivatives is preferred,
this way sulfur is selectively removed and the calorific value of the treated fuel
remains intact.
The first microorganism that was shown to be capable of selectively
cleaving carbon-sulfur bonds in crude oil, coal, and a wide range of model
compounds, resulting in the selective removal of sulfur and the retention of
carbon and calorific value, is Rhodococcus erythropolis IGTS8 (ATCC 53968)
[34]. Subsequently, numerous other bacteria capable of selectively cleaving
carbon-sulfur bonds in DBT were isolated and characterized. The biochemical
pathway used by these aerobic microorganisms to desulfurize DBT compounds
has been termed the 4S pathway due to the progressive oxidation of sulfur that
occurs through 4 steps [2, 35]. The selective removal of sulfur from DBT and
from crude oil by anaerobic bacteria has also been reported. Sulfate-reducing
bacteria such as Desulfovibrio desulfuricans have been shown to metabolize
DBT to H2S and biphenyl [36]. The desulfurization of oil under anaerobic
conditions avoids costs associated with aeration, and has the advantage of
liberating sulfur as a gas. However, an anaerobic biodesulfurization process has
not been developed due to low reaction rates, safety and cost concerns, and the
lack of identification of specific enzymes and genes responsible for anaerobic
desulfurization. Consequently, aerobic biodesulfurization has been the focus of
the majority of research [2, 32].
2.1. Substrate range of desulfurization

A desulfurization competent moderate thermophile was recently isolated,
Mycobacterium phlei GTIS10, that metabolizes DBT by the same pathway as R.
erythropolis IGTS8 [37]. A comparison of the capabilities of these two
microorganisms that have optimum growth temperatures of 30°C and 50°C
35
respectively, reveals a great deal about biodesulfurization. The range of

substrates used as sulfur sources by M. phlei GTIS10, as shown in Table 1, is
quite broad and essentially the same as reported for R. erythropolis IGTS8 [34].
The majority of bacterial cultures isolated based on their ability to
metabolize DBT are reported to be unable to metabolize benzothiophene and/or
thiophene [38-40]. While R. erythropolis IGTS8 has been reported to be unable
to utilize thiophene and/or benzothiophene [38, 40], we find that M. phlei
GTIS10, as well as R. erythropolis IGTS8, grew well with benzothiophene or
thiophene as sole sulfur sources [37]. The utilization of benzothiophene as a
substrate for the desulfurization enzymes of R. erythropolis IGTS8 has also been
reported by others [41], so that substrate utilization of desulfurization competent
cultures is somewhat controversial. This is probably due to the fact that
adaptation by repeated subculturing in media containing thiophene,
benzothiophene, or mixtures of thiophene or benzothiophene plus DBT are often
required to establish good growth for cultures originally isolated based on their
ability to metabolize DBT.
The end product of DBT metabolism by M. phlei GTIS10 is 2-HBP, which
has been reported to be bactericidal, bacteriostatic, and an inhibitor of the
desulfurization enzymes [37]. Concentrations of 200 uM of 2-HBP are reported
to be inhibitory and 400 uM of 2-HBP has been reported to completely prevent
growth and desulfurization activity of Corynebacterium, Rhodococcus and
Gordona cultures [27, 42-44]. It has been stated that bacterial strains with
increased tolerance for 2-HBP are needed for a viable petroleum
biodesulfurization process [19, 27]. M. phlei GTIS10 appears to be more
resistant to 2-HBP than other previously reported desulfurizing bacteria.
The 4S pathway for the desulfurization of DBT is shown in Fig. 1 [35, 45-
47]. The dszC gene encodes the dibenzothiophene monooxygenase that
catalyzes the conversion of DBT into DBT sulfone (DBTSO2). The dszA gene
encodes the dibenzothiophene-5,5-dioxide monooxygenase that catalyzes the
conversion of DBTSO2 into 2-hydroxybiphenyl-2-sulfinate (HBPSi). The dszB
gene encodes 2-hydroxybiphenyl-2-sulfinate sulfinolyase that catalyzes the
conversion of HBPSi into 2-hydroxybiphenyl (2-HBP) and sulfite [43, 48]. The
dszABC genes are transcribed as an operon found on a large plasmid, pSOX [49,
50]. An unlinked fourth gene, the dszD gene encoding a NADH-FMN
oxidoreductase, is an accessory component of the desulfurization pathway and
allows the regeneration of the cofactors needed for the monooxygenase reactions
catalyzed by DszC and DszA [51, 52]. The enzymology of the 4S
desulfurization pathway has been firmly established using purified enzymes
from several desulfurization competent bacterial species and from the results of
genetic analyses.
36
Table 1.
Range of organosulfur substrates used as sole source of sulfur for growth
byMp«ezGTIS10
37
An alternative desulfurization pathway has been found in Gordona sp.

strain 213E [38] that converts benzothiophene to 2-(2'-hydroxyphenyl) ethan-1-
al but is unable to metabolize DBT. The limited substrate range of this pathway
and a comparative lack of biochemical and genetic information have resulted in
the majority of research concerning biodesulfurization being focused on the 4S
pathway [2].
Fig. 1. The 4S metabolic pathway for DBT desulfurization. DszC is the DBT
monooxygenase, DszA the DBT sulfone monooxygenase, DszB the HPBSi desulfmase and
DszD is a flavin reductase. I, DBT; II, DBT sulfoxide; III, DBT sulfone; IV,
hydroxyphenylbenzenesulfmate; V, 2-hydroxybiphenyl.
38
Table 2.
Concentrations of metabolites of dibenzothiophene produced by M. phlei GTIS10 and R.
erythropolis IGTS8
1
= All concentrations of metabolites are expressed as |j.g/ml of the ethyl acetate extract (1 ml
total) derived from each culture grown with 1,440 (ig DBT as the sole sulfur source.
Dihydroxybiphenyl #1 and #2 have identical molecular formulas, but could not be assigned
specific molecular structures based on data available.
2.2. Range of metabolites produced by R. erythropolis IGTS8 versus M.

phlei GTIS10 from DBT
Because both R. erythropolis IGTS8 and M. phlei GTIS10 metabolize DBT
via the 4S pathway, but have optimum temperatures of 30°C and 50°C
respectively, the metabolites produced from DBT by these two cultures were
compared. Of all of the DBT metabolized by R. erythropolis IGTS8, 95% was
converted into 2-HBP; whereas, for M. phlei GTIS10 only 65% of the DBT
metabolized was converted to 2-HBP [37]. The lack of quantitative conversion
of DBT to 2-HBP has been noted in reports concerning other desulfurization
cultures where as little as 54% mole of DBT metabolized could be accounted for
as 2-HBP [53]. M. phlei GTIS10 produced greater levels of dibenzothiophene
sulfoxide and dihydroxybiphenyls than did R. erythropolis IGTS8. A significant
difference between data for M. phlei GTIS10 and R. erythropolis IGTS8 are the
results obtained for dihydroxybiphenyls (Table 2). About 3.3% of the DBT
metabolized by M. phlei GTIS10 was converted to dihydroxybiphenyls, while
no dihydroxybiphenyls were observed in the metabolites produced by R.
erythropolis IGTS8. While in the experiment reported in Table 2 no
dihydroxybiphenyls were observed as metabolites produced by R. erythropolis
39
IGTS8, occasionally trace amounts of dihydroxybiphenyls were observed as

metabolites of DBT produced by R. erythropolis IGTS8 [35].
This difference in the relative abundance of DBT metabolites produced by R.
erythropolis IGTS8 versus M. phlei GTISIO may be due to differences in the
thermotolerance of the desulfurization enzymes in these two cultures. To further
explore the thermostability of the desulfurization enzymes, the R. erythropolis
IGTS8 dszA, dszB and dszC genes were cloned individually and overexpressed
in E. coli to determine the thermostability of corresponding enzymes [37]. Each
desulfurization gene was cloned into the E. coli expression vector pQE80 and
the DszA, B, and C proteins containing polyhistidine residues at their N-termini
were produced and purified by affinity chromatography. Polyacrylamide-SDS
gel electrophoresis was performed to confirm the presence of proteins of the
correct molecular weight in the cell lysates.
Samples containing the DszA, DszB and DszC proteins were pre-incubated at
30°, 37°, 45°, 52°, 60°, 65°, and 72°C for 30, 60, and 120 minutes. The enzyme
solutions were assayed by adding the appropriate substrate and/or cofactors,
incubated at 30°C for an hour and then the amount of product formed was
determined by HPLC analysis. The results of these analyses indicate that DszB
activity (conversion of HBPSi into 2-HBP) is not inhibited by pre-incubation at
30° or 37°C, but little or no activity is seen when samples are pre-incubated at
temperatures of 45°C or higher even with exposure times as brief as 30 minutes.
The half-life of dibenzothiophene monooxygenase, DszC, was 1 hour at 45°C
and the half-life of DszA was 1 hour at 60°C. We examined the M. phlei
GTISIO FMN oxidoreductase (DszD) for thermal inactivation and it was found
to function with little inactivation up to 45 °C but was progressively inactivated
by exposure to higher temperatures [37].
Data obtained here regarding the thermostability of desulfurization enzymes
derived from M. phlei GTISIO are in agreement with results obtained for the
purified desulfurization enzymes isolated from Rhodococcus cultures [43, 48,
54-56]. The DszC enzyme from R. erythropolis Dl was inactivated by pre-
incubation at 45°C for 30 minutes and DszA purified from R. erythropolis Dl
was inactivated by pre-incubation at 60°C for 30 min. Since DszB catalyzes the
last step in the desulfurization pathway and is responsible for the release of
sulfur from DBT it must be functional in order to allow cultures to grow and
utilize DBT as sole sulfur source and it is required to allow the production of 2-
HBP. Therefore, it is surprising that the DszB enzyme appears to be thermally
inactivated in vitro by exposure to temperatures as low as 45°C yet some
activity is detected in whole cells at temperatures of 57°C [37] . While the DszB
enzyme may be the most thermolabile enzyme in the desulfurization pathway,
thermal inactivation is not instantaneous and it apparently has enough residual
activity even at 57°C to allow the accumulation of 2-HBP to be detected. The
more rapid thermal inactivation of purified desulfurization enzymes as
40
compared with desulfurization activity detected in whole cells requires further

investigation. Table 2 shows that dihydroxybiphenyls were observed in the
supernatant of M. phlei GTIS10 cultures grown at 45°C. Oldfield et al [46]
describes a minor desulfurization pathway in which the dibenzothiophene-5,5-
dioxide monooxygenase (dszA product) produces dihydroxybiphenyls from
biphenylene sulfone. Biphenylene sulfone can be formed from non-enzymatic
oxidation of 2-hydroxybiphenyl-2-sulfmate (HBPSi). Perhaps at higher
temperatures, the DszB enzyme is inactive and the minor desulfurization
pathway catalyzed by DszA is active to provide sulfur to the cells.
The dibenzothiophene monooxygenase encoded by dszC is responsible
for the conversion of DBT into dibenzothiophene sulfoxide and then to
dibenzothiophene sulfone. The elevated levels of dibenzothiophene sulfoxide
detected in M. phlei GTIS10, as compared with R. erythropolis IGTS8 (Table 2),
might suggest that DszC is temperature sensitive and is less active at 45 °C than
at 30°C. Also the faster growth of M. phlei GTIS10 with DBTSO2 versus DBT
suggests that the DszC activity may be thermolabile. However, the results of in
vitro experiments do not support this conclusion as DszC exhibits activity at
45 °C while DszB is the most thermolabile desulfurization enzyme [37]. The
DszC enzyme requires other cofactors/substrates for proper functioning: FMNH2
and oxygen. The FMN oxidoreductase encoded by dszD is responsible for
providing FMNH2 to the reactions catalyzed by DszC [51]. The accumulation of
DBTSO in the supernatant of the M. phlei GTIS 10 45°C growing cell
experiment might indicate that the FMN oxidoreductase DszD was significantly
inactivated in vivo at 45 °C and that the DszC catalyzed reaction did not go to
completion due to a lack of cofactors. However, this explanation is inconsistent
with the fact that DszD activity in M. phlei GTIS 10 extracts showed very little
inactivation at 45°C. Moreover, the half-life of the DszD enzyme from R.
erythropolis Dl was determined to be about 17 minutes at 50°C, slight activity
was detectable even after incubation at 72°C for 60 minutes [51].
These data demonstrate that the thermostability of enzymes determined in
vitro is not necessarily a good predictor of the functional range of an enzyme in
vivo, and that the same operon can yield metabolic pathways with different rate
limiting steps and different yields of metabolites in different hosts.
2.3. Comparison of the desulfurization activity of R. erythropolis IGTS8

and Paenibadllus sp. All-2
The isolation of the first desulfurization-competent thermophilic bacteria,
Paenibadllus sp. All-2, was reported in 1997 [57]. However, the specific
desulfurization activity of this culture (0.008 uM 2-HBP/hr/g dry cell weight
{DCW}) was low in comparison to previously characterized mesophilic cultures
(0.083 to 1.23 uM 2-HBP/hr/g DCW) [11, 27, 39, 52, 57-62].
41
The entire thermophilic desulfurization operon (tdsABC) of Paenibacillus

sp. All-2, which shows substantial homology to the dsz ABC operon of R.
erythropolis IGTS8, was cloned into E. coli and expressed with a maximum
activity of 0.155 uM 2-HBP/min/g DCW [61, 63]. The tds genes showed 61% to
73% homology to the corresponding dsz genes and the Tds enzymes showed
51.5% to 64.5% homology to the corresponding Dsz enzymes [61]. Methyl,
ethyl and propyl DBT derivatives were metabolized differently by Paenibacillus
Al 1-2 compared to R. erythropolis KA2-5-1 (which is essentially identical with
R. erythropolis IGTS8) [64]. These asymmetric substrates can yield two
desulfurized products depending upon which C-S bond is cleaved first by
DszA/TdsA. The ratios of the isomers produced by each strain were influenced
both by the position and size of the alkyl groups, but the two strains generally
showed opposite preferences for the reaction pathway. Desulfurization activity
generally decreased with increasing size of the alkyl substitutent in both
Paenibacillus and Rhodococcus. However, the substrate range of the two
desulfurization pathways differ [64]. Thiophenic sulfur compounds that are
recalcitrant to hydrodesulfurization include 4,6-dialkyl DBTs and 7-alkyl
benzothiophenes. The Tds enzymes showed more tolerance for substituted
DBTs than the Dsz enzymes and the Dsz enzymes did not metabolize 7-alkyl
BTs whereas the Tds enzymes did. The substrate range of cell-free
systems/lysates was broader than for the whole cell biodesulfurization catalysts,
which suggests that the transport or the bioavailability of organosulfur
compounds was limiting in both desulfurization catalysts [65]. While the
enzymatic rates of the Tds enzymes were substantially less than with the Dsz
enzymes, the different substrate range and thermotolerance of Tds versus Dsz
enzymes make them valuable resources for possible future directed evolution
experiments aimed at developing improved desulfurization enzymes.
2.4. Role of the desulfurization trait in nature

The cleavage of carbon-sulfur bonds in molecules such as DBT liberates
sulfur making it available as a nutrient to support the growth of bacteria. The
widespread occurrence of desulfurization-competent bacteria in samples
obtained from diverse environments and geographic locations indicates that the
ability to obtain sulfur from organic substrates is an important and fairly
common survival strategy for some bacterial species [2, 66]. Apparently,
microenvironments exist, even in soils where inorganic sulfur is relatively
abundant, where sulfur is a growth limiting nutrient. Clearly then, there is a
selective advantage in many natural environments for bacteria that can utilize
organic compounds such as DBT to obtain sulfur. The importance of the 4S
pathway to the survival of some bacteria in nature is also illustrated by the fact
that the dsz genes are found on conjugal plasmids and located in the proximity
of insertion sequences [49, 67]. While laboratory data demonstrating the
42
conjugal transfer of plasmids containing the dsz genes or the transposition of dsz
genes is sparse, the distribution of dsz genes in bacterial cultures strongly
support the hypothesis that these genes are commonly subjected to horizontal
transfer in nature. Indeed the DNA sequences of dsz genes from numerous
bacterial cultures isolated in geographically distinct locations have been found to
be nearly identical. Various Rhodococcus [68], Mycobacterium [37], Gordona
[59], Corynebacterium [69], Arthrobacter [70], Enterobacter [39],
Stenotrophomonas [39], Klebsiella [39], Bacillus [71], and Nocardia [72]
species have been isolated that possess dsz gene sequences that are identical or
highly homologous to the DNA sequence of the dsz gene of R. erythropolis
IGTS8. However, some variation in the sequences of the dsz genes has been
observed. The dsz genes of the moderate thermophile Paenibacillus sp. A 11-2
and Nocardia asteroides are only 52-65% and 89% homologous to R.
erythropolis IGTS8, respectively [61, 73]. Moreover, PCR amplification of dsz
genes from soil samples revealed relatively few variations in dsz gene
sequences, with the majority of variations found in dszA, and even then
homology to the R. erythropolis IGTS8 dszA sequence was 95% or more [66].
It is interesting to note that while several bacterial genera apparently
participate in horizontal transfer of dsz genes in nature, and laboratory studies
demonstrate that dsz genes and enzymes function well in Pseudomonas and E.
coli strains, a naturally occurring desulfurization-competent Pseudomonas sp. is
rarely encountered [74] and a desulfurization competent E. coli isolate has never
been reported [2]. The reasons for the restricted range of distribution of dsz
genes in nature are currently unknown, but one factor may be the ability of
bacterial species to withstand exposure to substrates such as petroleum.
Laboratory studies indicated that Pseudomonas sp. containing dsz genes could
efficiently metabolize DBT in aqueous culture or DBT added in a solvent such
as hexadecane. However, the ability of these same Pseudomonas cultures to
metabolize DBT in diesel oil or other petroleum product is much reduced [75,
76]. Naturally occurring desulfurization-competent bacterial cultures are almost
exclusively gram positive or gram variable and it may be that the cell
wall/membrane structure of gram negative bacterial species is less able to
tolerate exposure to petroleum compounds and solvents. There are many
thousands of gram positive and gram variable bacterial species, yet the observed
occurrence of dsz genes in naturally occurring desulfurization-competent
bacterial isolates is restricted to only a few species. Clearly, more remains to be
learned about the role dsz genes play in microbial ecology and the functioning of
desulfurization enzymes in different bacterial hosts [77].
Other topics regarding biodesulfurization that are not well understood are
the access of desulfurization enzymes to insoluble and high molecular weight
substrates, and the mechanism by which the sulfur liberated from organosulfur
substrates by the desulfurization enzymes is subsequently incorporated into
43
biomass. When R. erythropolis IGTS8 was first isolated, it was obtained from an
enrichment culture growing in a defined mineral salts medium devoid of
inorganic sulfur [78, 79]. All essential nutrients were present in abundance with
the exception of sulfur, which was supplied in the form of coal or DBT creating
an environment where any bacterial species that could utilize organically bound
sulfur had a strong selective advantage. The mixed culture that grew with DBT
as the sole source of sulfur was streaked onto a variety of agar plates allowing
pure cultures to be obtained from each of the types of colonies present. Then
each pure culture was tested individually to determine if it could utilize DBT as
a sole source of sulfur for growth.
It soon became clear that none of the pure cultures most readily isolated
from the desulfurization-competent mixed culture were capable of utilizing DBT
as a sole sulfur source. Perseverance in investigating this desulfurization-
competent mixed culture eventually led to the isolation of a relatively slow
growing pure culture that was demonstrated to utilize DBT as a sole source of
sulfur, and this culture was subsequently identified as R. erythropolis IGTS8
[34]. R. erythropolis IGTS8 was present at low abundance in the original
desulfurization-competent mixed culture and even when pure cultures of R.
erythropolis IGTS8 and a desulfurization-deficient, but faster growing, bacterial
culture such as Enterobacter cloacae, were combined in various ratios and used
to inoculate growth experiments in sulfur-limited media where DBT was the
sole source of sulfur, R. erythropolis IGTS8 invariably emerged as the least
abundant species in the resulting culture [34].
These results cannot be explained if DBT is taken up into the cytoplasm
of/?, erythropolis IGTS8 and only then is DBT converted to 2-HBP and sulfite,
unless it is also hypothesized that sulfite is then excreted. While intracellular
metabolism of DBT by R. erythropolis IGTS8 is stated to occur [46] there is no
evidence for DBT transport/uptake in desulfurization competent Rhodococcus
cultures [25], nor is there evidence for mass transfer limitations in DBT
metabolism [25, 80]. If sulfur is liberated intracellularly within R. erythropolis
IGTS8 and sulfur is the growth limiting nutrient, it seems unlikely that sulfite
would be excreted extracellularly unless intracellular oxidation of sulfite to
sulfate were not possible. Sulfate has been demonstrated to be the form of
inorganic sulfur that is utilized by R. erythropolis IGTS8 [81] while sulfite has
been demonstrated to be the form of sulfur obtained as a product of the
desulfurization of DBT [43]. Further research into sulfite and sulfate metabolism
by desulfurization competent cultures is warranted.
One hypothesis that is consistent with the observation that faster growing
desulfurization-deficient bacterial species can dominate mixed cultures when R.
erythropolis IGTS8 is the only desulfurization competent culture present, is that
desulfurization of DBT occurs in association with the external surface of R.
erythropolis IGTS8 cells. The Dsz proteins are known to have membrane-
44
spanning domains [47, 82] so that the desulfurization pathway may function in
association with the cell membrane such that extracellular substrates and
intracellular cofactors can both be accessed. A further consideration regarding
the localization of the Dsz enzymes is the size of some of the substrates that can
be metabolized. Solid coal particles and high molecular weight coal derived
polymers can be effectively desulfurized [83-85], yet there has never been a
report documenting the intracellular uptake of substrates such as coal by any
bacterial species. Moreover, the size of coal particles vastly exceeds the size of
bacterial cells in experiments where biodesulfurization has been demonstrated to
remove 72% of organic sulfur without otherwise altering the composition of the
coal [85]. There is no evidence whatsoever that desulfurization enzymes are
excreted from R. erythropolis IGTS8 cells, but the size of substrates metabolized
and the ability of other bacterial species to successfully compete for sulfur
liberated from organosulfur substrates by R. erythropolis IGTS8 make it likely
that desulfurization does not occur intracellularly, but in association with the
external surface of cells.
A consequence of the fact that desulfurization-deficient bacterial species
can successfully compete for sulfur liberated from organosulfur substrates by R.
erythropolis IGTS8 is that a selective pressure favoring the evolution of a high
specific activity for desulfurization enzymes is created. In a mixed culture
environment where sulfur is the growth limiting nutrient and R. erythropolis
IGTS8 is the only desulfurization-competent culture, this bacterium must
liberate many times more sulfur than it needs to meet its own nutritional
requirements because competition from other bacteria leaves only a fraction of
the utilizable sulfur actually available for use by R. erythropolis IGTS8 [34]. If
this dynamic typified the natural environment for R. erythropolis IGTS8 and
other desulfurization competent cultures it would be reasonable to expect that a
high level of desulfurization activity would have evolved in such cultures.
However, that is not the case and even when grown as pure cultures, all
naturally occurring desulfurization competent cultures have levels/activities of
desulfurization enzymes that are growth limiting rather than capable of
supplying sulfur in excess of the needs of the culture [2]. This further illustrates
that we have much to learn about the role of Dsz enzymes in nature and the
characterization of the microenvironment occupied in nature by R. erythropolis
IGTS8 and other desulfurization-competent bacteria. Nevertheless, it is worth
considering that enrichment cultures and directed evolution experiments
designed to obtain cultures with higher levels of desulfurization activity may
benefit from the intentional use of mixed cultures.
2.5. Influence of the bacterial host on biodesulfurization

M. phlei GTIS10 appears to be highly similar to R. erythropolis IGTS8 as
regards to biodesulfurization capability except that the maximum growth
45
temperature for R. erythropolis IGTS8 is about 33°C whereas M. phlei GTIS10

can grow up to temperatures of 52°C [37]. Figure 2 illustrates that R.
erythropolis IGTS8 showed maximal activity at 30°C and a progressive loss of
activity at 37°C and 45°C, while no activity is observed at temperatures of 52°C
or above. On the other hand, M. phlei GTIS10 exhibits activity over the
temperature range of 25°C to 57°C with maximal activity at 45°C to 50°C, and
no activity at 62°C.
Since these two cultures show such different temperature ranges for
growth it was anticipated that the sequences of the dsz genes in M. phlei GTIS10
versus R. erythropolis IGTS8 would be different or contain mutations conferring
thermostability; however, both cultures were found to contain the pSOX plasmid
encoding dszABC genes having identical DNA sequences [37]. It is likely then
that M. phlei GTIS 10 acquired the dsz genes by the conjugal transfer of the
pSOX plasmid from R. erythropolis IGTS8 or other desulfurization-competent
bacteria.
Taken altogether the results shown in Fig. 2, plus the knowledge that
DNA sequence analysis showed that both cultures contain identical dsz genes,
indicate that the ability of the dsz enzymes to function is greatly influenced by
the bacterial host strain. However, enzymes other than, or in addition to, those
encoded by the dsz operon may contribute to the desulfurization activity and
range of metabolites of DBT produced by M. phlei GTIS 10. Both M. phlei
GTIS 10 and R. erythropolis IGTS8 exhibit maximal desulfurization activity
corresponding to the optimum growth temperature of each culture, 50° and 30°C
respectively, and then desulfurization activity declines in concert with
decreasing cell viability at higher temperatures. The desulfurization pathway
requires NADH, FMNH2, and oxygen in order to complete the conversion of
DBT to 2-HBP. The host must supply these factors so that the functional
temperature range of the desulfurization pathway is seen to be different in two
different bacterial hosts possibly reflecting the ability of each bacterial species
to provide cofactors and reaction substrates at various temperatures. Transport
of substrates and products may also contribute to desulfurization activity. The
observation that the dsz operon had two apparent temperature maximums in two
different bacterial hosts perhaps suggests that if the dsz operon could be
expressed in a thermophilic bacterial host the desulfurization enzymes may
function at even higher temperatures.
Other researchers have also reported obtaining desulfurization competent
cultures that have identical dsz gene sequences, but exhibit different phenotypes.
Several research groups reported that even though multiple desulfurization
competent cultures isolated and examined were Rhodococcus species they
exhibited different specific activities for DBT, yields of 2-HBP, activity with
4,6-dimethyl DBT, and sensitivity to hexadecane [86, 87]. The reasons for these
phenotypic differences among highly similar bacterial cultures that possess
46
identical desulfurization genes are unknown. However, it is clear that the host
contributes to the functioning of the desulfurization pathway in yet
uncharacterized ways so that the manipulation of the dsz (or tds) genes alone
may be insufficient to yield bacterial cultures with substantially higher
desulfurization activity, such as would be required for a commercial
biodesulfurization process.
2.6 Desulfurization activity of various cultures

The maximum specific desulfurization activity for M. phlei GTIS10 was
1.1 ±0.07 umole 2-HBP/min/g DCW [37]. The maximum specific
desulfurization activity for R. erythropolis IGTS8 observed at the Gas
Technology Institute (1.2 ± 0.08 umole 2-HBP/min/g DCW) is higher than
previous studies where other researchers reported specific desulfurization
activity values ranging from approximately 0.6 to 5.8 umole 2-HBP/min/g
DCW [46, 57]. The reason why our culture of R. erythropolis IGTS8 showed
higher desulfurization specific activity than previously reported may be due to
the continuous culturing (> 10 years) of this bacteria in our laboratory under
conditions where DBT, or other organosulfur compounds, served as sole sulfur
source for growth. The highest desulfurization specific activity reported for the
thermophilic Paenibacillus sp. strain A l l - 2 was approximately 0.08 umole 2-
HBP/min/g DCW [57]. Bacterial cultures containing cloned desulfurization
genes from Paenibacillus (tdsABC) were reported to have a maximum
desulfurization specific activity of 0.16 umole 2-HBP/min/g DCW [61], while
strains containing cloned Rhodococcus desulfurization genes (dszABC) were
reported to have a maximum desulfurization specific activity of 4.7 umole 2-
HBP/min/g DCW [52, 58]. The moderate thermophile Mycobacterium phlei
WU-F1 was described as having greater desulfurization activity than
Paenibacillus sp. strain Al 1-2, but no specific activity data was reported [53].
2.7. Genetic modifications to increase desulfurization activity

Largely due to the interest in biodesulfurization and the high percentage
of desulfurization competent bacteria that are Rhodococcus species as well as
other potential applications of this genus, there has been a lot of research on the
genetics of Rhodococci [88, 89]. Multiple cloning and shuttle vectors are
available and genetic manipulation of Rhodococcus can now be conveniently
and reliably performed [90-92]. However, one area of genetic research that has
received comparatively little attention is gene expression in Rhodococcus.
Overexpression of genes in Rhodococcus has been reported [89, 90, 92, 93];
however, an array of gene expression vectors and a knowledge of the consensus
sequences of transcriptional promoters in Rhodococcus is lacking.
47
The coordinated expression of the desulfurization genes has been shown to be

important in obtaining maximum desulfurization activity as well as the yields of
the pathway intermediates. The flavin oxidoreductase that supplies the cofactors
needed by DszC and DszA/(TdsC and TdsA) has been the subject of several
investigations. The DszA, B, and C proteins are active in E. coli, but the level of
flavin oxidoreductase is low in comparison with Rhodococcus [52, 94, 95], so a
flavin oxidoreductase gene from Vibrio harveyi was cloned and expressed in an
E. coli strain containing the dszABC genes [52]. This resulted in increasing FMN
oxidoreductase levels from 0.03 umol/min/mg protein to 1.1 umol/min/mg
protein. Co-expression of the dsz and flavin genes resulted in the highest rates of
DBT transformation (51 mg/hr/g DCW), but accumulation of intermediates,
mainly DBTSO2, rather than full conversion to 2-HBP, was observed suggesting
that DszB is the rate-limiting enzyme in the desulfurization pathway. Other
studies of the 4S pathway confirmed that DszB was limiting the global
desulfurizing activity [11, 27].
Fig. 2. Resting cells of M. phlei GTIS10 exhibit specific desulfurization activity at higher
temperatures than resting cells of R. etythropolis IGTS8. The amount of 2-HBP produced by
the conversion of DBT by each culture after incubation for 24 hours at various temperatures
was quantified by HPLC analysis. Rate of change in 2-HBP concentration was calculated
from the linear portion of the curve, generally the first 4 hours of the incubation. The specific
desulfurization activity values recorded are averages of three replicate samples from three
separate experiments for a total of nine data points. Standard deviation was less than 10 %. O,
M. phlei GTIS10; • , R. erythropolis IGTS8.
48
The highest rate of 2-HBP formation was obtained without the cloned
FMN oxidoreductase, perhaps because the accumulation of intermediates
inhibited the Dsz enzymes. The maximum amount of 2-HBP produced was 0.2
mM regardless of the amount of DBT added or the incubation time, which
suggests that inhibition by 2-HBP is also a key factor limiting biodesulfurization
efficiency in E. coli, and probably in other bacterial hosts as well [52]. Oshiro et
al. [94] screened 80 bacterial and 20 yeast cell extracts to find flavin reductases
with the best ability to support DszC and DszA activity. A flavin oxidoreductase
from Paenibacillus polymyxa was found to be the best allowing 3.5 to 5-fold
better activity of DszC and DszA as compared with the Rhodococcus flavin
oxidoreductase.
Rhodococcus strains containing increased copies of dszABC genes on
plasmids or integrated into the chromosome have resulted in higher DBT
conversion rates, but also in the accumulation of pathway intermediates as
DBTSO and DBTSO2 [58, 96]. When the copy number of the dszD gene was
increased in Rhodococcus erythropolis KA 2-5-1 cultures containing their
natural complement of dszABCD genes, then DBTSO and DBTSO2
accumulation occurred. However, if the copy number of all of the dszABCD
genes was increased then the accumulation of intermediates was avoided, but
only when the correct balance between dsz genes was achieved [58]. Derivatives
of R. erythropolis KA 2-5-1 originally had a specific desulfurization activity of
0.05 mmol/g DCW/hr while derivative cultures that, in addition to their natural
complement of dszABCD genes, contained a plasmid with one copy of the
dszABC genes had a specific desulfurization activity of 0.14 mmol/g DCW/hr;
0.19 mmol/g DCW/hr with dszABCD genes on a plasmid, and 0.28 mmol/g
DCW/hr with two copies of dszABC genes and one dszD gene on a plasmid.
Derivative cultures that contained additional copies of the dszABC operon or the
dszD gene did not yield cultures with higher enzymatic activity. Similar results
were obtained for the expression of various combinations of dsz and tds genes in
Rhodococcus, E. coli, and Pseudomonas hosts [96, 97], demonstrating that in
order to obtain bacterial cultures with the highest possible desulfurization
activities it is necessary to obtain the proper ratio of desulfurization enzymes
and cofactors.
The results of experiments in which different copy numbers,
combinations of dsz/tds genes and promoters were used also revealed that a limit
in desulfurization activity is reached that can not be overcome by increasing the
amount of the Dsz/Tds enzymes in cells [24, 76, 93, 95, 98]. There are other
factors affecting desulfurization activity and/or the intrinsic properties of the
Dsz/Tds enzymes needs to be improved if higher desulfurization activity is to be
achieved. A way of improving the intrinsic properties of enzymes is directed
evolution.
49
Rational protein engineering studies have been a powerful tool, enabling

the modification of some enzymes to increase their thermostability, shift their
pH optima and alter their substrate specificity [2, 73]. However, in order to do
this, a fair amount of information is required such as the amino acid sequence,
the three dimensional structure and the location of the active site of the protein.
Even when all of this information is available, protein engineering is an
uncertain undertaking that can be expensive and time consuming. It is more
straightforward to use site specific mutagenesis to increase the hydrophobicity
of proteins, however even this requires uncertain theoretical predictions. The
amino acid sequences of the three desulfurization enzymes can be inferred from
their DNA sequences; however, detailed knowledge of their active sites or three
dimensional structures of these proteins is currently unavailable [43, 48, 51, 54].
To obtain this information would take a considerable amount of time and
resources and would not be sufficient to guarantee success in developing
enzymes with improved activity or thermostability. In addition, while there are
some general rules/trends that have emerged regarding the thermostabilization
of proteins (such as increasing the hydrophobicity of proteins by placing proline
residues at beta-turns in proteins and adding disulfide bonds), these are very
general rules that often do not hold true and are difficult to implement. Other
methods to increase the thermotolerance of enzymes such as immobilization or
post-production modification techniques are not available since
biodesulfurization requires the intervention of three enzymes and their
associated cofactors, requiring the need to use intact cells rather than
immobilized enzyme systems.
The thermostabilization of the three-enzyme desulfurization pathway (and
the fourth enzyme which supplies cofactors) would be a daunting if not an
overwhelming task to accomplish using protein engineering given the current
state of knowledge. Fortunately, this information is not necessary in order to
employ directed evolution to obtain thermostable desulfurization enzymes.
Through the use of mutagenesis combined with natural selection, or screening
for enzymatic activity, a directed evolution process can be employed to obtain
thermostable derivatives of the desulfurization enzymes. In this method, no prior
knowledge of the enzymes three-dimensional structure or even complete amino
acid sequence is required. The method mimics nature's own protein engineering
system: evolution. All that is needed is a powerful screen or selection so that the
desired enzyme traits can be identified. The process of natural selection can then
be accelerated in the laboratory to evolve the desired traits. In a short period of
time, researchers have modified many enzyme traits such as thermostability, pH
optima, substrate specificity and organic solvent tolerance [99-103].
Evolution is accomplished through the combined action of mutation,
recombination and selection. Both general and site-specific mutagenesis of
targeted genes can be used. In a typical experiment, part or all of a gene or
50
operon is subjected to a general mutagenesis method such as error-prone PCR

[100]. Gene shuffling techniques can also be used to increase recombination
frequencies and speed the evolutionary process. The gene is then transferred into
the host organism and the selection or screen is applied. Candidate mutants with
the desired phenotype are then identified and analyzed. These mutants can then
be used for further rounds of mutagenesis so that a stepwise approach can be
used to gradually evolve the trait of interest.
Such a process has been used to improve the desulfurization activity of
the DszC enzyme [73]. The dszC genes of R. erythropolis IGTS8 and Nocardia
asteroides A3H1 are 89% homologous. More specifically, these two variants of
the dszC gene contain alterations at 127 locations resulting in proteins that differ
at 38 amino acid positions. This diversity of dszC genes was used in the DNA
shuffling/RACHITT technology developed by Enchira Corp. (formerly Energy
Biosystems Corporation) to yield a DszC derivative with a 70% higher
enzymatic activity [73]. Similar experiments to obtain improved derivatives of
other desulfurization enzymes have not been reported. The results obtained with
improved derivatives of DszC are promising, but improved derivatives of all of
the desulfurization genes are required in order to achieve high levels of
improvement in desulfurization activity for the complete 4S pathway. However,
since there are other, yet to be identified, cellular components that contribute to
the functioning of the 4S pathway, improvements of desulfurization genes only
may not be sufficient to achieve very high desulfurization activity compatible
with a commercial application [24].
2.8. Development of processes for the biodesulfurization of diesel and crude

oil
A cell-free biodesulfurization process would be impractical. The
requirement of a three-enzyme pathway along with cofactors will prohibit the
use of purified enzyme systems for a practical biodesulfurization process. A
commercial biodesulfurization process will have to employ intact bacterial cells
as biocatalysts [24, 104, 105]. Because of this, the use of immobilization or
other post-production modification techniques to enhance the thermotolerance of
enzymes would not be practical for use in an industrial process. It has been
proposed however, that individual enzymes that do not require cofactors may be
used. Chloroperoxidase and cytochrome C could be used to catalyze the
oxidation of DBT to DBTSO2, as well as oxidize petroporphyrins resulting in
the release of metals (see chapter 3)[2, 20]. It is necessary to add hydrogen
peroxide and water to these peroxidase reactions, but high rates of conversion of
DBT to DBTSO2 have been reported. Since DBTSO2 has a higher boiling point
than DBT, it has been proposed that distillation can be used to obtain sulfur-free
petroleum fractions following peroxidase treatment [2]. This approach has not
51
yet been tested in pilot scale experiments so the costs and efficiency of such a
process are not yet known.
Energy BioSystems Corporation (EBC) conducted a comprehensive
evaluation, particularly as regards crude oil and fractions, of the
biodesulfurization technology originally developed by the Institute of Gas
Technology (IGT) [now known as the Gas Technology Institute (GTI)] under a
program funded by the U.S. Department of Energy. Encouraged by their
experimental results and feedback from the petroleum industry, EBC licensed
the technology, and assembled a team of executives, engineers, and scientists
from the petroleum industry, committed to the commercialization of
biodesulfurization technology.
The development of bioprocesses for biodesulfurization of petroleum
have been almost exclusively focused on the use of biocatalysts that are
derivatives of, or related to, R. erythropolis IGTS8, and diesel has been the
target for the development of the first biodesulfurization processes [24, 25].
EBC was the first organization to seriously attempt the development of a
commercial biodesulfurization process. They chose diesel fuel desulfurization as
the target for initial process development efforts because environmental
regulations mandating a reduction of the maximum permissible concentration of
sulfur in diesel to 50 ppm had been proposed and existing refinery processes
were not able to efficiently and economically meet this requirement [25]. The
most abundant organosulfur compounds in diesel includes DBT and its
derivatives which are recalcitrant to traditional hydrodesulfurization but are
good substrates for biodesulfurization [2].
The application of any technology, chemical or biochemical, to the
treatment of petroleum requires a highly efficient process as the resulting
products are low priced commodities [105]. Moreover, the volume of petroleum
processed, even at a small refinery, dwarfs the scale of bioprocesses typically
used in the pharmaceutical and biotechnology industries. To address these
process concerns, EBC claimed to have achieved a 200-fold improvement in the
specific activity of the R. erythropolis IGTS8 biocatalyst using a combination of
medium improvement, reaction conditions and genetic engineering [24].
Moreover, process engineering research increased the volumetric reaction rate
(oil/water ratio), biocatalyst life and solved separations issues. Specific details
about EBC's biodesulfurization process and the results achieved were not
published, but a desulfurization rate of 20 umole DBT/min/g DCW was stated
as a target for a commercially successful process [25]. The literature contains a
large amount of information regarding the use of genetic engineering to achieve
higher desulfurization rates as previously discussed in this chapter. It has been
shown that R. erythropolis IGTS8 biocatalysts are capable of functioning at 9-
to-1 oil-to-water ratios [106], and maximum cell yields in fed batch fermentation
52
were reported to be 92 g DCW/liter [107]. Maintaining high cell densities and

catalytic rates was accomplished by EBC by employing a cell recycling and
regeneration step in their process [24]. The R. erythropolis IGTS8 cells were not
used in a process mode in which they were required to grow in the presence of
diesel that served as a sole source of sulfur. Instead, separation techniques,
typically a hydrocyclone oil/water separator, were used to recover cells so that
they could be cleansed of reaction products, regenerated by aeration and
nutrients, and recycled in a concentrated form back to the biodesulfurization
process. An air lift reactor was used in the EBC process to minimize energy
costs in the process, and cell suspensions were found to be more efficient than
immobilized cell preparations [24, 25].
A schematic representation of the diesel biodesulfurization process
initially envisioned by EBC is shown in Fig. 3 [24]. Initially crude oil would be
distilled to yield products that include diesel. Hydrodesulfurization (HDS)
would first be used to reduce the sulfur content of the diesel, followed by
biodesulfurization (BDS) to further reduce the sulfur content, ultimately yielding
a product having 50 ppm sulfur or less. This process scenario takes advantage of
the fact that hydrodesulfurization and biodesulfurization complement each other.
Hydrodesulfurization can be operated more efficiently and at lower cost by
treating only those compounds that are most reactive. Similarly, the
biodesulfurization process operates more efficiently by receiving a feed that has
a lower total sulfur content and contains almost exclusively only those
organosulfur compounds that can be treated most effectively by
biodesulfurization. The biodesulfurization process operates at lower
temperatures and pressures than hydrodesulfurization. This allows hydro-
desulfurization to operate at lower temperatures and pressures than if
hydrodesulfurization was the only desulfurization treatment step. This scenario
has advantages of 70-80% lower CO2 emissions and energy consumption, and
safer operating conditions as compared with hydrodesulfurization alone [108].
Apparently even these improved operating parameters were not sufficient
for a commercially viable process for the biodesulfurization of diesel. As
discussed previously, the rate limiting step in the 4S pathway is the cleavage of
the second/final carbon-sulfur bond catalyzed by DszB. Rather than, or in
addition to, obtaining further improvements in the catalytic rate of DszB, an
alternative biodesulfurization process scheme is illustrated in Fig. 4 [24]. The
oxygenated sulfinate byproduct could be recovered as a value added byproduct,
since it has surfactant properties, that would then improve the economics of the
biodesulfurization process. However, even this version of a biodesulfurization
process for diesel was not commercially successful and EBC went out of
business.
53
Fig. 3. Overview of an integrated hydrodesulfurization and biodesulfurization process for

diesel oil.
The key reasons that EBC did not succeed in developing a commercially
viable biodesulfurization process included changes in the environmental
regulations and improvements in hydrodesulfurization technologies. When EBC
began process engineering efforts to develop a commercial process for the
biodesulfurization of diesel, the environmental regulations specified a maximum
total sulfur content of 50 ppm and the existing hydrodesulfurization processes
could not efficiently achieve that goal. However, while EBC was involved in the
challenging task of implementing the first bioprocess in the petroleum industry
(other than waste remediation), stricter environmental regulations were proposed
decreasing the maximum permissible sulfur content in diesel to 10 to 15 ppm.
Additionally, during this same time frame, improvements were made in
hydrodesulfurization technology that allowed these lower sulfur levels to be
reached [109].
Integrating a biodesulfurization process into a refinery is the only way to
treat a product such as diesel, but this requires a substantial modification of
current operations in a refinery and requires that the biodesulfurization process
operate at the same speed and reliability as other refinery processes so as not to
disrupt normal refining operations. It is very challenging for any new technology
to be embraced by a conservative industry such as the petroleum industry so that
employing biodesulfurization as a component of refinery operations met with
understandable opposition. However, alternative ways of implementing a
biodesulfurization process exist (see chapter 4)[104].
54
Fig. 4. Overview of the Energy Biosystems Corporation process for the simultaneous
biodesulfurization diesel oil and the production of a sulfinate/surfactant byproduct.
Biorefining can complement existing technologies by specifically addressing

compounds/contaminants refractory to current petroleum refinery processes.
Heteroatoms such as nitrogen, metals and sulfur can poison the catalysts used in
catalytic cracking and hydrotreating processes [5, 109]. Existing refineries are
not capable of operating efficiently with heavy crude oils and residuum that
have high heteroatom content [110]. Bioprocesses could be used to pre-treat oil
reducing the heteroatom content allowing the use of heavy crude oils that could
not otherwise be treated with existing refinery processes [104]. Biorefining
processes can also be used in conjunction with existing processes to meet the
increasingly stringent environmental requirements for contaminant reduction.
The development of a bioprocess to selectively remove sulfur, nitrogen, and
associated metals, from crude oil and residuum will allow existing refineries to
process lower quality oils that they could not otherwise accept. The reduction of
sulfur, nitrogen and metals in petroleum would allow refineries to operate more
efficiently, decrease costs and protect the environment.
55
A potentially attractive means of implementing a petroleum

biodesulfurization process could be to treat heavy oil on site in the production
field prior to the initial separation of petroleum from produced water [104].
Since biodesulfurization would require a water wash/separation step to remove
the biocatalysts and liberate the sulfur from treated petroleum and since a water
separation step is a normal component of petroleum production, performing
biodesulfurization on produced oil would minimize the required processing
steps. The produced water could be re-injected on site for secondary
recovery/water flooding operations and could therefore eliminate the need for
wastewater treatment [4]. The reinjection of sulfur-laden water into a petroleum
field generates concerns about the potential souring of wells, but strategies for
the prevention and control of H2S formation are routinely used in existing
petroleum production operations and could be employed to deal with water
resulting from biodesulfurization treatments. A schematic illustration of a
biorefining process used to treat crude oil in association with production
operations is shown in Fig. 5.
This scenario could greatly improve the economics of a biodesulfurization
process and could fit into existing petroleum production procedures with
minimal modifications. This approach would require biodesulfurization reactors
to be present at petroleum production sites; however, it is possible that
conventional petroleum storage tanks can be modified for use in
biodesulfurization treatments. The main concern about this biodesulfurization
approach would be the temperature of the produced oil and the thermal tolerance
of biocatalysts. Desalting and dewatering processes for petroleum are normally
performed at temperatures of from 60°C to 100°C, and the use of elevated
temperatures will be increasingly important in the treatment of heavy crude oils
and residuum in order to deal with the viscosity of these heavy oils [4, 23, 110].
If thermophilic cultures could be used to desulfurize heavy oils in
conjunction with desalting and dewatering processes, then both the viscosity and
sulfur content of heavy oils could be simultaneously reduced allowing upgraded
lighter oils to be sent to refineries for subsequent processing. The
hydrodesulfurization and FCC of these upgraded oils would be easier [30, 110].
Thus, biodesulfurization of heavy crude oils would fit well within current
practices of the petroleum industry, but the biodesulfurization should be
performed at high temperatures (60°C to 100°C).
Microbial cultures that can selectively desulfurize petroleum have already
been identified, but cultures that will efficiently desulfurize petroleum at
thermophilic temperatures are not yet available [37]. Performing biorefining
processes at higher temperatures is not only more compatible with existing
industry practices, but would also result in higher catalytic rates and the reduced
viscosity of petroleum at higher temperatures would allow lower processing
56
costs. Thermophilic microorganisms have not been well studied and no

systematic examination of thermophilic cultures for possible use in biorefining
has been reported [111, 112].
Biotechnology may one day solve many problems confronting the
petroleum industry today, but a biorefining process will have to operate on a far
greater scale and at less cost than any current biotechnology process. For any
process to be viable in the petroleum industry, it must not only be capable of
treating the complex mixture of chemicals that comprise petroleum but it must
also treat very large volumes in a cost effective way. Many enzymes catalyze
reactions relevant to biorefining goals but they must be improved in numerous
ways before practical and economical bioprocesses can be developed. Metabolic
engineering and directed evolution approaches are possible means for the
development of bioprocesses relevant to the petroleum industry. Specific
development needs include: developing cultures capable of performing
biodesulfurization at thermophilic temperatures (60° to 100°C), developing
cultures with higher levels of desulfurization activity, and process development
research for the biodesulfurization of heavy oils.
Fig. 5. Overview of a process for the biodesulfurization of crude oil.

57
3. BIODENITROGENATION OF PETROLEUM
The removal of organically bound nitrogen from crude oil, without the loss of
significant calorific value, requires the selective cleavage of carbon-nitrogen
bonds. The selective cleavage of carbon-sulfur bonds in crude oil using
biocatalysts has been demonstrated [2, 24, 113] and it may be possible to
selectively cleave carbon-nitrogen bonds using biocatalysts developed from
microorganisms capable of metabolizing compounds such as quinoline and
carbazole [114]. The cleavage of carbon-nitrogen bonds resulting in the
conversion of quinoline to 8-hydroxycoumarin and ammonia has been
demonstrated [115], and the genes that encode the enzymes participating in the
quinoline degradation pathway have been identified and sequenced [116]. The
removal of nitrogen from crude oil by a quinoline degrading culture,
Pseudomonas ayucida IGTN9m, has also been demonstrated [115]. However,
the abundance of quinoline relative to other organonitrogen compounds in crude
oil is low and existing quinoline degradation enzymes have a narrow substrate
range. Consequently, even though removal of 68% of quinoline from crude oil
was demonstrated the total nitrogen content was reduced by only 5%. An
appropriate topic for future research is the development of cultures that express
higher levels of quinoline degrading enzymes that have wider substrate ranges,
but it is also important to develop biocatalysts that can remove nitrogen from
other compounds typically found in petroleum such as carbazole.
Fig. 6. Carbazole Degradation Pathways. The top pathway illustrates the existing carbazole
degradation pathway that results in overall degradation, whereas the bottom pathway
illustrates a potential pathway for the selective removal of nitrogen from carbazole that could
be developed using metabolic engineering.
58
Carbazole is a good model compound that is representative of the

nitrogen-containing compounds present in the greatest abundance in many
petroleum fractions [12, 117]. For developing a biological process for the
removal of nitrogen from petroleum, none of the presently known carbazole-
degrading cultures are particularly appropriate because nitrogen is only removed
in the course of complete degradation [114, 118].
A variety of carbazole-degrading microorganisms have been reported in
the literature including Sphingomonas, Pseudomonas, Mycobacterium,
Ralstonia and Xanthomonas species [114, 119-125]. Insofar as biodegradation
pathways have been investigated, these differing species of carbazole degraders
follow a similar carbazole degradation pathway that begins with the oxidative
cleavage of the hetorocyclic nitrogen ring of carbazole to form 2'-
aminobiphenyl-2,3-diol. This compound is then oxidized through meta cleavage
yielding 2-hydroxy-6-oxo-6-hexa-2e,4z-dienoate. The next metabolic steps
result in the degradation of one of the aromatic rings releasing carbon dioxide.
In existing pathways nitrogen is released from carbazole only after substantial
carbon degradation. Figure 6 illustrates the carbazole degradation pathway
employed by currently known carbazole utilizing cultures as well as a potential
pathway for selective removal of nitrogen from carbazole that could be created
using metabolic engineering by combining the CarA enzyme from carbazole
degraders such as Sphingomonas sp. GTIN11 with a suitable deaminase.
Some carbazole-degrading cultures, like Sphingomonas sp. CB3 [124],
have been found to contain carbazole dioxygenases that are related to biphenyl
oxidases while other cultures, like Pseudomonas resinovorans CAIO [121],
contain carbazole dioxygenases that show no close relationship to other
characterized oxidases. CARDO consists of three components: a dioxygenase,
ferredoxin, and ferredoxin reductase. In some carbazole-degrading cultures, like
Pseudomonas resinovorans CAIO [121], the dioxygenase is a single protein, but
in other cultures, like Sphingomonas sp. CB3 [124], the dioxygenase is
comprised of two subunits. The arrangement of genes encoding CARDO also
differs significantly as the four genes in Sphingomonas sp. CB3 (carAa, carAb,
car Ac and car Ad) are contiguous and arranged in that order [124], while the
genes encoding CARDO in Pseudomonas resinovorans CAIO are not
contiguous [126]. Additionally, the carbazole dioxygenase enzyme of
Sphingomonas sp. CB3 has a rather narrow substrate range and does not
metabolize naphthalene, dibenzothiophene, phenanthrene, or fluorene unlike
Pseudomonas resinovorans CAIO [121]. While the arrangement of genes
encoding the enzymes involved in carbazole degradation in Sphingomonas sp.
GTIN11 is similar to the order found in Pseudomonas resinovorans CAIO, the
carA, carB, and carC genes of Sphingomonas sp. GTIN11 do not show
significant homology to the car genes present in either Pseudomonas
resinovorans CAIO or Sphingomonas sp. CB3. Moreover, while several
59
carbazole-degrading microbial cultures are known, the ability of these cultures

to selectively remove nitrogen from crude oil has only been tested for
Sphingomonas sp. GTIN11 [118].
Therefore several bacterial cultures are known that can utilize carbazole
as a sole nitrogen source, but no culture is known that can selectively cleave
both C-N bonds in carbazole while leaving the rest of the molecule intact.
Sphingomonas sp GTIN11 [118] was demonstrated to metabolize carbazole, and
to a lesser extent Cl and C2 derivatives of carbazole, from petroleum. As much
as 95% of the carbazole present in crude oil was removed as a consequence of
biotreatment by Sphingomonas sp. GTIN11. However, the reduction in the total
nitrogen content of the crude oil was relatively modest. This highlights the need
for developing of improved cultures that contain appropriate biochemical
pathways for the cleavage of carbon-nitrogen bonds in organonitrogen
compounds and that have broad substrate ranges that encompass the diverse
mixture of organonitrogen compounds typically found in crude oil.
The genes encoding the carbazole degradation pathway of Sphingomonas
sp. GTIN11 have been cloned and sequenced [118]. The reaction catalyzed by
CarA converts carbazole to 2'-aminobiphenyl-2,3-diol accomplishing the
cleavage of the first C-N bond in carbazole. There are no known deaminases or
amidases that can metabolize 2'-aminobiphenyl-2,3-diol and accomplish the
cleavage of the final C-N bond [114, 118]. If a deaminase that will recognize 2'-
aminobiphenyl-2,3-diol as a substrate could be found then the gene encoding
this enzyme could be combined with the carA genes (carAa, carAc, and carAd
encoding for the carbazole dioxygenase, ferredoxin and ferredoxin reductase
respectively) from Sphingomonas sp. GTIN11 and thereby construct a synthetic
operon for the selective removal of nitrogen from carbazole. A preferred
bacterial strain would lack the carB and carC genes so that complete
biodegradation of carbazole would be avoided and the final product would be
2',2,3-trihydroxybiphenyl (or a similar compound).
There are reports of mixed cultures of thermophilic bacteria that
selectively reduce the nitrogen content of crude oil by as much as 45% [112].
However, other investigators have not verified these results, and the identity of
the microbial species responsible for carbon-nitrogen bond cleavage is
unknown. Consequently, nothing is known about the biochemistry or the
genetics of this nitrogen removal phenomenon, so that there is no
straightforward means of furthering and improving this process.
While the selective removal of nitrogen from crude oil has been
demonstrated and information about the biochemistry and genetics of quinoline
and carbazole degradation is available, it is clear that much remains to be done
before a commercially viable process for the biodenitrogenation of petroleum
can be considered. The chief need is for enzymes with broader substrate ranges.
60
4. BIOPROCESSES FOR THE REMOVAL OF METALS FROM

PETROLEUM
The use of biotechnology to reduce the concentration of metals in petroleum is

the least studied topic in biorefining research [2]. Chloroperoxidase and
cytochrome C peroxidase could be used to catalyze the oxidation of
petroporphyrins resulting in the release of metals (see chapter 3). The removal
of 53% nickel and 27% vanadium from crude oil has been reported using
chloroperoxidase [20]. Peroxidase reactions can be accomplished using enzymes
rather than whole cells because a single enzyme, rather than a multi-step
pathway, is involved and no cofactors are required. Water and hydrogen
peroxide must be provided in order to promote this reaction, but a practical
bioprocess for the removal of metals from petroleum may be possible. Since
most metals in petroleum are associated with organonitrogen compounds it may
be possible that improved biodenitrogenation processes will simultaneously
reduce the nitrogen and the metal content, thereby avoiding the need for a
separate metal removal process.
5. CONCLUSIONS AND FUTURE RESEARCH PRIORITIES
Heavy crude oils and residuum constitute a significant, and constantly

increasing portion of world reserves. These heavy oils possess high calorific
content yet have comparatively low market values mainly because of high sulfur
and metals content and high viscosity/molecular weight. The sulfur and nitrogen
content is of environmental concern due to potential sulfurous and nitrous
emissions from petroleum combustion. Metals, and to a lesser extent sulfur and
nitrogen, present in heavy crude oils can poison catalysts used in
hydrodesulfurization therefore limiting the effectiveness of current technologies
to remove sulfur and nitrogen from these oils. Furthermore, the catalytic
cracking process used to convert crude oil to lower molecular weight products is
negatively affected by the sulfur, nitrogen and metal content. The high
viscosity/high molecular weight of these oils limits the amount of higher value
petroleum byproducts such as gasoline, aviation fuel, and diesel fuel that can be
obtained as well as causes increased operating costs. These problems associated
with heavy oils have prompted the preferential utilization of light crude oils. As
light crude oils are consumed at a disproportional/high rate the amount of heavy
oil as a percentage of remaining world petroleum reserves continues to increase.
New technologies capable of dealing with heavy oils to mitigate
environmental concerns and increase byproduct yields in a cost effective manner
are needed: biorefining may provide an answer.
61
Perhaps the best way of implementing a biorefining process is to integrate

it into existing industry practices to the greatest degree possible and in such a
way as to complement existing industry practices. It is believed that biorefining
can be particularly useful in the treatment of heavy crude oils as the technology
can remove sulfur, nitrogen and metals and simultaneously reduce the
viscosity/molecular weight of oil as a consequence of carbon-sulfur and carbon-
nitrogen bond cleavage, and can tolerate heavy metal and salt concentrations
typically found in heavy oils and produced/formation waters. If biorefining can
be used in conjunction with desalting and dewatering steps in oil production
operations and reduce the sulfur, nitrogen, and metal content of crude oil prior to
the oil being sent to refineries then existing refining technologies could be used
with an expanded range of low quality oils that could not otherwise be treated.
While the potential of biorefining has been demonstrated, more development is
needed before the technology can be successfully commercialized for the
treatment of heavy oils. Specific development needs include: developing
cultures capable of performing biorefining at thermophilic temperatures (60° to
100°C), cultures with higher levels of enzymatic activity, new biochemical
pathways for the selective cleavage of carbon-nitrogen bonds, new enzymes
with broader substrate ranges, and process development research for the
biorefining of heavy oils and residuum.
REFERENCES
[I] B. Mason, Nature October 29, (2003), www.nature.com/nsu/031027/031027-

031023.html.
[2] S. Le Borgne, and Quintero R., Fuel Processing Technol. 1641 (2003), 1.
[3] H. Kim, J. M. Vohs, and R. J. Gorte, Chem. Commun. (Camb) (2001) 2334.
[4] L. J. Drew, Kirk-Othmer Encyclopedia of Chemical Technology (1996) (Kroschwitz, J.
I., and Howe-Grant, M., Eds.), pp. 342-476.
[5] L. L. Hegedus, and McCabe, R. W., Catalyst Rev. 23 (1981) 377.
[6] S. Reeson, Energy World 235 (1996) 9.
[7] J. G. Speight, The Chemistry and Technology of Petroleum, Marcel Dekker Inc, New
York 1980.
[8] C. S. Hsu, K. Qian, and W. K. Robbins, J. of High Resolution Chromatography 17
(1994)271.
[9] C. S. Creaser, F. Krokos, K. E. O'Neill, M. J. C. Smith and P. G. McDowell, J. Am. Soc.
Mass Spectrometry 4 (1993), 322-326.
[10] Shennan-J-L, J. Chem. Technol. Biotechnol. 67 (1996) 109.
II1] B. L. McFarland, D. J. Boron, W. Deever, J. A. Meyer, A. R. Johnson, and R. M. Atlas,
Crit. Rev. Microbiol. 24 (1998) 99.
[12] G. W. Mushrush, E. J. Beal, D. R. Hardy, and J. M. Hughes, Fuel Processing
Technology 61 (1999) 197.
[13] A. Donetti, E. Cereda, A. Ezhaya, and R. Micheletti, J. Med. Chem. 32 (1989) 957.
[14] E. V. Brown, and R. Isbrandt, J Med Chem 14 (1971) 84.
62
[15] J. Jacob, A. Schmoldt, C. Augustin, G. Raab, and G. Grimmer, Toxicology 68 (1991)

181.
[16] K. G. Kropp, and P. M. Fedorak, Can. J. Microbiol. 44 (1998) 605.
[17] F. A. Leighton, Fundam. Appl. Toxicol. 12 (1989) 787.
[18] T. Maruyama, L. L. Wotring, and L. B. Townsend, J. Med. Chem. 26 (1983) 25.
[19] T. McFall, G. M. Booth, M. L. Lee, Y. Tominaga, R. Pratap, M. Tedjamulia, and R. N.
Castle, Mutat. Res. 135 (1984) 97.
[20] L. Mogollon, R. Rodriguez, W. Larrota, C. Ortiz, and R. Torres, Appl. Biochem.
Biotecchnol. 70-72 (1998) 765.
[21] Monticello-D-J, and W. R. Finnerty, Ann. Rev. Microbiol. 39 (1985) 371.
[22] P. Kassler, Energy exploration and exploitation (1996) (Jenkins, G., Ed.), pp. 229-242
Multi-Science Publishing Co, Berkshire, United Kingdom.
[23] P. O'Connor, L. A. Gerritsen, J. R. Pearch, P. H. Desai, S. Yanik, and A. Humphries,
Hydrogen Processing 11 (1991) 76.
[24] M. A. Pacheco, E. A. Lange, P. T. Pienkos, L-Q. Y. M. P. Rouse, Q. Lin, and L. K.
Linguist, National Petrochemical & Refiners Association (1999), pp. AM-99-27, San
Antonia, Texas.
[25] D. J. Monticello, Curr. Opin. Biotechnol. 11 (2000) 540.
[26] Thomas-J-A, G. Ganapathi, and E. L. Stover, Res. J. Water Pollution Control Fed. 63
(1991)475.
[27] B. L. McFarland, Curr. Opin. Microbiol. 2 (1999) 257.
[28] Hart's Diesel Fuel News (1998).
[29] Kilbane-J-J, Trends in Biotechnology 7 (1989) 97.
[30] J. R. Harris, Hydrocarbon Processing 75 (1996) 63.
[31] T. Isoda, S. Nagao, X. Ma, Y. Korai, and I. Mochida, Enegy & Fuels 10 (1996) 1078.
[32] J. J. Kilbane, Trends in Biotechnology 7 (1989) 97.
[33] Sandhya-S, Indian Journal of Microbiology 36 (1996) 1.
[34] K. J. Kayser, B. A. Bielaga-Jones, K. Jackowski, O. Odusan, and J. J. Kilbane, J. Gen.
Microbiol 139 (1993) 3123.
[35] J. R. Gallagher, E. S. Olson, and D. C. Stanley, FEMS Microbiol. Lett. 107 (1993) 31.
[36] Kim-T-S, H. Y. Kim, and B. H. Kim, Biotechnol. Lett. 12 (1990) 757.
[37] K. J. L. C. Kayser, H.-S. Park, J.-H. Kwak, A. Kolhatkar, and J. J. Kilbane II, Appl.
Microbiol. Biotechnol. 59 (2002) 737.
[38] S. C. Gilbert, J. Morton, S. Buchanan, C. Oldfield, and A. McRoberts, Microbiology
144(1998)2545.
[39] M. Kobayashi, T. Onaka, Y. Ishii, J. Konishi, M. Takaki, H. Okada, Y. Ohta, K.
Koizumi, and M. Suzuki, FEMS Microbiol. Lett. 187 (2000) 123.
[40] C. Oldfield, N. T. Wood, S. C. Gilbert, F. D. Murray, and F. R. Faure, Antonie Van
Leeuwenhoek 74 (1998) 119.
[41] T. Matsui, T. Onaka, Y. Tanaka, T. Tezuka, M. Suzuki, and R. Kurane, Biosci.
Biotechnol. Biochem. 64 (2000) 596.
[42] J. H. Chang, Y.J. Kim, B.H. Lee, K.S. Cho, H.W. Rhu, Y.K. Chang, and H.N. Chang,
Biotechnology Progress 17 (2001) 876.
[43] N. Nakayama, T. Matsubara, T. Oshiro, Y. Moroto, Y. Kawata, K. Koizumi, Y.
Hirakawa, M. Suzuki, K. Muruhashi, Y. Izumi, and R. Kurane, Biochem. Biophys. Acta
1598 (2002) 122.
[44] S. Nekodzuka, T. Nakajima-Kambe, N. Nomura, J. Lu, and T. Nakahara, Biocatalysis
Biotrans. 15 (1997) 17.
63
[45] K. A. Gray, O. S. Pogrebinsky, G. T. Mrachko, L. Xi, D. J. Monticello, and C. H.

Squires, Nat. Biotechnol. 14 (1996) 1705.
[46] C. Oldfield, O. Pogrebinsky, J. Simmonds, E. S. Olson, and C. F. Kulpa, Microbiology
143(1997)2961.
[47] C. S. Piddington, B. R. Kovacevich, and J. Rambosek, Appl. Environ. Microbiol. 61
(1995)468.
[48] L. M. Watkins, R. Rodriguez, D. Schneider, R. Broderick, M. Cruz, R. Chambers, E.
Ruckman, M. Cody, and G. T. Mrachko, Arch. Biochem. Biophys. 415 (2003) 14.
[49] C. Denis-Larose, D. Labbe, H. Bergeron, A. M. Jones, C. W. Greer, J. al-Hawari, M. J.
Grossman, B. M. Sankey, and P. C. Lau, Appl. Environ. Microbiol. 63 (1997) 2915.
[50] C. Denis-Larose, H. Bergeron, D. Labbe, C. W. Greer, J. Hawari, M. J. Grossman, B.
M. Sankey, and P. C. Lau, Appl. Environ. Microbiol. 64 (1998) 4363.
[51] T. Matsubara, T. Oshiro, Y. Nishina, and Y. Izumi, Appl. Environ. Microbiol. 67 (2001)
1179.
[52] D. S. Reichmuth, H. H., H. W. Blanch, and J. D. Keasling, Biotechnol. Bioeng. 67
(2000) 72.
[53] T. Furuya, K. Kirimura, K. Kino, and S. Usami, FEMS Microbiol Lett 204 (2001) 129.
[54] B. Lei, and S. C. Tu, J. Bacteriol. 178 (1996) 5699.
[55] T. Ohshiro, K. Suzuki, and Y. Izumi, J. Ferment. Bioeng. 83 (1997) 233.
[56] T. Ohshiro, T. Koshima, K. Torii, H. Kawasoe, and Y. Izumi, J. Biosci. Bioeng. 88
(1999)610.
[57] J. Konishi, Y. Ishii, T. Onaka, K. Okumura, and M. Suzuki, Appl. Environ. Microbiol.
63(1997)3164.
[58] K. Hirasawa, Y. Ishii, M. Kobayashi, K. Koizumi, and K. Maruhashi, Biosci.
[59] S. K. Rhee, J. H. Chang, Y. K. Chang, and H. N. Chang, Appl. Environ. Microbiol. 64
(1998)2327.
[60] M. Kobayashi, K. Horiuchi, O. Yoshikawa, K. Hirasawa, Y. Ishii, K. Fujino, H.
Sugiyama, and K. Maruhashi, Biosci. Biotechnol. Biochem. 65 (2001) 298.
[61] Y. Ishii, J. Konishi, H. Okada, K. Hirasawa, T. Onaka, and M. Suzuki, Biochem.
Biophys. Res. Commun. 270 (2000) 81.
[62] J. Konishi, T. Onaka, Y. Ishii, and M. Suzuki, FEMS Microbiol. Lett. 187 (2000) 151.
[63] Y. Ishii, J. Konoshi, M. Suzuki, and K. Maruhashi, J. Biosci. Bioeng. 90 (2000) 591.
[64] T. Onaka, J. Konishi, Y. Ishii, and K. Maruhashi, J. Biosci. Bioeng. 92 (2001) 193.
[65] J. Konishi, H. Okada, K. Hirasawa, Y. Ishii, and K. Maruhashi, Biotechnol. Lett. 24
(2002) 1863.
[66] G. F. Duarte, A. S. Rosado, L. Seldin, W. de Araujo, and J. D. van Elsas, Appl. Environ.
Microbiol. 67 (2001) 1052.
[67] S. A. Denome, and K. D. Young, Gene 161 (1995) 33.
[68] T. Matsui, K-I. Noda, Y. Tanaka, K. Maruhashi, and R. Kurane, Curr. Microbiol. 45
(2002) 240.
[69] T. Omori, L. Monna, Y. Saiki, and T. Kodama, Appl. Environ. Microbiol. 58 (1992)
911.
[70] L. Serbolisca, F. de Ferra, and I. Margarit, Appl. Microbiol. Biotechnol. 52 (1999) 122.
[71] K. Kirimura, T. Furuya, Y. Nishii, Y. Ishii, K. Kino, and S. Usami, J. Biosci. Bioeng. 91
(2001) 262.
[72] J. H. Chang, S. K. Rhee, Y. K. Chang, and H. N. Chang, Biotechnol. Prog. 14 (1998)
851.
64
[73] W. M. Coco, W. E. Levinson, M. J. Crist, H. J. Hektor, A. Darzins, P. T. Pienkos, C. H.

Squires, and D. J. Monticello, Nat. Biotechnol. 19 (2001) 354.
[74] J. D. Van Hamme, and O. P. Ward, Appl. Environ. Microbiol. 67 (2001) 4874.
[75] J. J Arensdorf, A.K. Loomis, P.M. DiGrazia, D.J. Monticello, and P. T. Pienkos, Appl.
[76] K. Watanabe, K. Noda, Y. Ohta, and K. Maruhashi, Biotechnol. Lett. 24 (2002) 897.
[77] S. Krawiec, Develop. Ind. Microbiol. 31 (1990) 103.
[78] J. J. Kilbane II, Resource Conservation & Recycling 3 (1990) 69.
[79] J. J. Kilbane, Trends in Biotechnology 7 (1989) 97.
[80] E. N. Kaufman, J. B. Harkins, and A. P. Borole, Appl. Biochem. Biotechnol. 73 (1998)
127.
[81] L. Setti, P. Farinelli, S. D. Martino, S. Frassinetti, G. Lanzarini, and P. G. Pifferi, Appl.
Microbiol. Biotechnol. 52 (1999) 111.
[82] S. A. Denome, C. Oldfield, L. J. Nash, and K. D. Young, J. Bacteriol. 176 (1994) 6707.
[83] D. L. Stoner, J. E. Wey, K. B. Barrett, J. G. Jolley, R. B. Wright, and P. R. Dugan, Appl.
[84] J. J. Kilbane, and K. Jackowski, Biotechnol. Bioeng. 40 (1992) 1107.
[85] G. P. Huffman, N. Shah, F. E. Huggins, L. M. Stock, K. Chatterjee, J. J. Kilbane, and
M. M. Chou, Fuel 74 (1995) 549.
[86] S. Abbad-Andaloussi, C. Lagnel, M. Warzywoda, and F. Monot, Microbial Technology
32 (2003) 446.
[87] G. Castorena, C. Suarez, I. Valdez, G. Amador, L. Fernandez, and S. Le Borgne, FEMS
Microbiol. Lett. 215 (2002) 157.
[88] E. R. Dabbs, Antonie-Leeuwenhoek-Journal of Microbiology 74 (1998) 155.
[89] M. J. Larkin, R. DeMot, L. A. Kulakov, and I. Nagy, Antonie-Leeuwenhoek-Journal of
Microbiology 74 (1998) 133.
[90] M. Vesely, M. Patek, J. Nesvera, A. Cejkova, J. Masak, and V. Jirku, Appl. Microbiol.
Biotechnol. 61 (2003) 523.
[91] R. DeMot, I. Nagy, A. DeSchrijver, P. Pattanapipitpaisal, G. Schoofs, and J.
Vanderleyden, Microbiology 143 (1997) 3137.
[92] M. Arenskotter, D. Baumeister, R. Kalscheuer, and A. Steinbuchel, Appl.
Environ.Microbiol. 69 (2003) 4971.
[93] E. Franchi, F. Rodriguez, L. Serbolisca, and F. de Ferra, Oil & Gas Science & Technol.
Rev. IFP 58 (2003) 515.
[94] T. Ohshiro, Y. Aoi, K. Torii< and Y. Izumi, App. Microbiol. Biotechnol. 59 (2002) 649.
[95] P. Galan, E. Diaz, and J. L. Garcia, Environ. Microbiol. 2 (2000) 687.
[96] T. Matsui, K. Hirasawa, K. I. Koizumi, K. Maruhashi, and R. Kurane, Biotechnol. Lett.
23(2001)1715.
[97] Y. Ishii, T. Ohshiro, Y. Aoi, M. Suzuki, and Y. Izumi, J. Biosci. Bioeng. 90 (2000) 220.
[98] K.-I. Noda, K. Watanabe, and K. Maruhashi, J. Biosci. Bioeng. 95 (2003) 504.
[99] J. C. Moore, and F. H. Arnold, Nat. Biotechnol. 14 (1996) 458.
[100]F. H. Arnold, L. Giver, A. Gershenson, H. Zhao, and K. Miyazaki, Ann. N Y Acad. Sci.
870 (1999) 400.
[101] J. Hoseki, T. Yano, Y. Koyama, S. Kuramitsu, and H. Kagamiyama, J. Biochem.
(Tokyo) 126 (1999) 951.
[102]T. Yano, S. Oue, and H. Kagamiyama, PNAS 95 (1998) 5511.
[103]T. Yano, and H. Kagamiyama, PNAS 98 (2001) 903.
[104]R. G. Shong, Division of Fuel Chemistry, American Chemical Society 44 (1999), 1-4.
[105]J. L. Shennan, J. Chem. Technol. Biotechnol. 67 (1996) 109.
65
[106] S. Patel, J. J. Kilbane, and D. A. Webster, J. Chem. Technol. Biotechnol. 69 (1997) 100.
[107]O. Yoshikawa, Y. Ishii, K-I. Koizumi, T. Ohshiro, Y. Izumi, and K. Marahashi, J.
Biosci. Bioeng. 94 (2002) 447.
[108]L. Linguist, and M. Pacheco, Oil & Gas Journal (1999), pp. 45-48.
[109] S. T. Oyama, and Y-K. Lee, American Chemical Society, Fuel Chemistry Division 48
(2003), 173-174.
[110]S. Reeson, Energy World 235 (1996) 9.
[111]M. S. Lin, T. Premuzic, J. H. Yablon, and W. M. Zhou, Appl. Biochem. Biotechnol. 57-
58(1996)659.
[112]E. T. Premuzic, M. S. Lin, M. Bohenek, and W. M. Zhou, Enegy & Fuels 13 (1999)
297.
[113] J. J. Kilbane II, Final Report, Energy BioSystems Project No. 40308-02 (1992).
[114]M. J. Benedik, P. R. Gibbs, R. R. Riddle, and R. C. Wilson, Trends in Biotechnology 16
(1998)390.
[115] J. J. Kilbane, R. Ranganathan, K. J. Kayser, L. Cleveland, C. Ribiero and M. M.
Linhares, Appl. Environ. Microbiol. 66 (2000) 688.
[116]U. Frerichs-Deeken, B. Goldenstedt, R. Gahl-Janben, R. Kappl, J. Huttermann, and S.
Fretzner, European J. Biochem. 270 (2003) 1567.
[117]S. Mitra-Kirtley, O. C. Mullins, J. van Elp, S. J. George, J. Chen, and S. P. Cramer, J.
Am. Chem. Soc. 115 (1993) 252.
[118] J. J. Kilbane II, A. Daram, J. Abbasian, and K. J. Kayser, Biochem. Biophys. Res.
Comm. 297 (2002) 242.
[119]H. Habe, Y. Ashikawa, Y. Saiki, T. Yoshida, H. Nojiri, and T. Omori, FEMS Microbiol.
Lett. 211(2002), 43.
[120]K. Kirimura, H. Nakagawa, K. Tsuji, K. Matsuda, R. Kurane, and S. Usami, Biosci.
[121]H. Nojiri, J. W. Nam, M. Kosaka, K. I. Morii, T. Takemura, K. Furihata, H. Yamane,
and T. Omori, J. Bacteriol. 181 (1999) 3105.
[122]N. Oichiyama, T. Omori, and T. Kodama, Biosci. Biotech. Biochem. 57 (1993) 455.
[123] J. Schneider, R. J. Grosser, K. Jayasimhulu, W. Xue, B. Kinkle, and D. Warshawsky,
Can. J. Microbiol. 46 (2000) 269.
[124]J. M. Shepherd, and G. Lloyd-Jones, Biochem. Biophys. Res. Comm. 247 (1998) 129.
[125]R. R. Riddle, P. R. Gibbs, R. C. Wilson, and M. J. Benedik, J. Ind. Microbiol.
Biotechnol. 30 (2003) 6.
[126]S. I. Sato, J. W. Nam, K. Kasuga, H. Nojiri, H. Yamane, and T. Omori, J. Bacteriol. 179
(1997)4850.
© 2004 Elsevier B.V. All rights reserved. 67
Chapter 3
Enzymatic catalysis on petroleum products

M. Ayala" and R. Vazquez-Duhaltb
a
Instituto Mexicano del Petroleo. Eje Central Lazaro Cardenas 152, San Bartolo
Atepehuacan 07730 Mexico DF, Mexico
b
Instituto de Biotecnologia, UNAM. Apartado Postal 510-3 Cuernavaca,
Morelos 62250 Mexico
1. INTRODUCTION
The contemporary society is highly dependent on oil supply for energy,

transportation, food production, and in general, industrial production. A century
ago the oil exploitation began, first as a source of energy and now as a source
both of energy and raw material. Thus, history will describe our time as the oil
based society. Nature took 500 millions years to accumulate the world's oil;
nevertheless, the world's petroleum could be consumed in two centuries [1]. The
inexorable production peak is estimated to occur sometime between 2010 and
2020 and then the oil resources will be drastically reduced at the end of this
century (Fig. 1) [2]. When the world's oil reserves become scarce, the more
expensive fuel sources as hard-to-extract oil deposits, tarry sands, and synfuels
from coal will be brought to the front of production.
Fig. 1. Petroleum availability estimation from World Resource Institute (1996)

68
New technologies should improve the refining efficiency in terms of the

consumed energy and the environmental impact of the processes. Developments
and implementation of new technologies for conventional processes, such as
cracking, hydrogenation, isomerization, alkylation, polymerization, and
hydrodesulfurization could be expected. Nevertheless, the introduction of non-
conventional technologies, representing potential substitutes or complementary
processes to traditional oil refining, may happen. Significant progress has been
made in the last decades in technologies such as membrane separation [3],
supercritical extractions [4-6], and many others. Biotechnology is among the
new fields that might be introduced to the oil refining industry. The potential
application of biochemical catalysis in the petroleum refining industry has been
recently reviewed in a prospective analysis of the available data on the microbial
and enzymatic modification of oil products [7]. The proposed biotechnological
processes should be considered either alternative or complementary to
conventional oil refining technologies. The introduction of such novel non-
conventional techniques in the petroleum industry may improve its energetic
efficiency and reduce its environmental impact.
The first biotechnological processes applied in the oil industry were
environmental processes, such as wastewater treatment and soil bioremediation.
However, there are other potential uses for biotechnological processes in the oil
industry. It is important to point out that so far no enzymatic or biochemical
processes exist in the oil refining industry, thus this chapter shows a prospective
analysis of data on enzymatic transformations of petroleum products and their
derivatives, in order to evaluate the possible introduction of biotechnological
processes in the petroleum refining industry.
Preconceived ideas and misconceptions about enzymes continue to limit
the applications of enzymes on biotransformations in the chemical industry.
Nevertheless, the fact is that there are industrially-successful examples of
biocatalytic process that show enzymes to be sufficiently stable, productive and
economic for commercial applications. Enzymes have huge breadth of scope in
the types of reactions that may be catalyzed and the chemical nature of
compounds that may be transformed. Biocatalysis is no panacea, but still it is in
its infancy and significant progress may be expected.
Enzymes are far more efficient than chemical catalysts: high specificity,
low substrate concentration, and mild reaction conditions are the most
interesting properties of enzymes in this regard. Enzyme-catalyzed reactions
usually display characteristically high turn-over numbers, with rate accelerations
approaching or exceeding 108. In terms of productivity and if we consider single
step enzymatic transformations, productivities of tens or even hundreds of grams
69
of product per liter per hour have been achieved. Biocatalytic reactions that have
been successfully applied in the industry, at large scale, include: production of
high fructose corn syrup, fatty acids and triglyceride oils, aspartame, acrylamide,
antibiotic precursors, amino acids, S-2-chloropropionic acid, polylactic acid and
cyclodextrins [8].
At first glance, enzymes seem to be more expensive than chemical
catalyst. Enzyme prices ranges from $ 100 per kilogram, as for crude
preparations of amylase, to $ 100,000 per kilogram, as for lactic dehydrogenase.
However, the key cost to consider in biocatalysis should be not the cost of the
enzyme itself, but rather the cost-contribution of the final product. This cost
contribution could be as low as $ 0.10 per kilogram as in the case of aspartase in
the L-aspartic acid production. When compared with the cost of other catalyst,
especially those with similar selectivity, the prices of enzymes are not very
different (Table 1).
Still the industrial use of enzymatic catalysts is limited by their instability
under harsh conditions, which are usually found in large-scale processes.
Nevertheless, chemical and genetic modifications of enzymes to improve both
activity and stability, together with solvent engineering and new catalytic
activities from extremophiles microorganisms will provide better biocatalysts
for the specific needs of the petroleum industry. Non-conventional uses of
enzymatic transformation are still in their infancy. Non-aqueous systems, high
temperatures and hydrophobic substrates are the three main characteristics of oil
industry that represent the most important challenges for the enzymatic catalysis
to be applied in the petroleum refining industry. The success of biocatalysis in
the petroleum industry depends on the development of biocatalysts able to
perform transformations of oil products in non-aqueous systems and stable
under the conditions usually found in the refineries.
Table 1
Bulk enzyme and chemical catalyst pricesa.
a
Adapted from Rozzell [8].
70
Large amounts of water are incompatible with the refining processes, as

demonstrated in the case of microbial desulfurization process (see chapter 2) [9].
Since petroleum is a hydrophobic material, it is suitable to speculate that new
enzymatic processes for the oil industry should be carried out in non-aqueous
systems. The use of reaction mixtures containing organic solvents reduces mass
transfer limitations, promoting the establishment of productive interactions
between the enzyme and the hydrophobic substrates (oil-derived compounds).
Fortunately, it is possible to have enzymatic activity in non-aqueous systems
with very low water content, almost anhydrous. Biocatalysis in non-aqueous
media has increased significantly the range of practical applications of enzymes
[10]. The abundant information on enzymatic activity in hydrophobic solvents,
such as hexane, toluene and many other organic solvents has been extensively
reviewed [10, 11], so it is reasonable to expect enzymatic activity in petroleum
or petroleum fractions. In addition, a biocatalyst placed in a non-aqueous
medium shows interesting properties, such as improved thermostability, higher
substrate accessibility, adjustable selectivity, and high storage stability [12, 13].
Enzymatic reactions can be performed at more than 120°C in organic solvents,
even if the enzyme is not thermostable in aqueous media [14].
Any enhancement of the thermal stability of an enzyme would confer
significant operational advantages such as higher reaction rates, increased
substrate solubility in lower viscosity media, productive shifts in thermodynamic
equilibrium and reduced risks of microbial contamination. Genetic approaches
have yielded significant results in obtaining more thermostable enzymes, with
higher temperature optima through directed evolution [15-17]. A combination of
rational and random mutagenesis has been used to obtain a fungal peroxidase
174-times more thermostable than the wild type protein [18]. However, although
directed evolution seems to be a powerful tool to enhance biocatalytic
performance, its main drawback is that the knowledge of the properties gained
during site directed mutagenesis or evolution processes can not be used as a
general method to be applied to other proteins. On the other hand, chemical
modification seems to be a more general method to improve intrinsic properties
of proteins such as stability and activity without a deeper knowledge of the gene
or protein structure. Several chemical methods have been employed to obtain
more stable protein derivatives, including plastic conjugates [19, 20],
crosslinked enzyme crystals (CLECs) [21-24], attachment to polysaccharides
[25, 26], and chemical modification with amphiphilic polymers [27-30].
Catalytically active and stable enzymes at temperatures higher that 80°C
are considered hyperthermophilic enzymes. Only three enzymatic preparations
have been shown to be active and stable at temperatures higher than 100°C in
aqueous systems, two hydrolytic enzymes: a pegylated trypsin [27] and a
dextran-glycosylated amylase [26], and a pegylated cytochrome c performing
peroxidase-like catalysis [30].
71
This work is a prospective analysis of data on enzymatic transformations

of petroleum products and their derivatives, in order to evaluate the possible
impact of biotechnological processes on the petroleum refining. This prospective
analysis emphasizes enzymatic biodesulfurization and enzymatic asphaltene
upgrading, but promising carbon-carbon bond enzymatic activation is also
discussed.
2. DESULFURIZATION
Petroleum contains sulfur compounds, which structure and concentration varies

depending on the crude oil source. The combustion of these compounds
produces sulfur oxides that have a negative environmental impact. Sulfur oxides
combined with water in the atmosphere are the principal source of acid rain;
moreover, these oxides poison catalytic converters in cars, leading to increased
hydrocarbons emissions to the air. In order to reduce the environmental damage
caused by these oxides many countries have regulated its release, for example
by lowering the sulfur content in fuels [31]. Currently, sulfur content is reduced
via hydrodesulfurization, a chemical process in which organic sulfur is
converted to hydrogen sulfide in the presence of an inorganic catalyst. During
this process, crude oil is reacted with hydrogen at high pressures (150 to 3000
psi) and high temperatures (290° to 455°C).
While the sulfur in thiols, sulfides and thiophenes present in the lighter
fractions of crude oil are readily removed by hydrodesulfurization, in the heavier
fractions a significant amount of sulfur is present within polynuclear aromatic
molecules such as benzothiophenes (BT) and dibenzothiophenes (DBT). These
larger and more complex molecules, particularly those with alkyl substitutions
near the sulfur atom, are not easily hydrodesulfurized [9]. Regulatory agencies
require today around 500 ppm of sulfur in fuels, a specification that can be
accomplished with the current desulfurization technology. However, by the end
of this decade the required sulfur content in fuels is expected to be less than 15
ppm [31]. As the availability of light oil decreases, there is a need to process
heavier oil. Energy and capital costs of hydrotreating would substantially
increase in order to achieve a deep level of desulfurization. Therefore, there is a
significant interest in low-cost desulfurization technologies that might
complement hydrotreating. The fact that some developing technologies, such as
oxidation or acid treatment, promote secondary reactions affecting other
components in crude oil is a drawback to its use as a desulfurization strategy
[32]. Amongst the more promising alternatives is biodesulfurization, due to its
low-energy requirements and high specificity [9, 33].
72
2.1 Microbial desulfurization

Microbial desulfurization is a process based on the removal of organic
sulfur from petroleum mainly by bacterial cell systems. After three decades of
efforts focused on the use of metabolically active bacterial cells for fuel
desulfurization, and even pilot plant trails, the process shows to be limited by
several factors: large amounts of water needed for the process, mass transfer
limitations in the two phase reactions, and large time of batch reaction [7, 9].
This subject is extensively reviewed in chapter 2.
In order to develop a commercially feasible technology, any designed
biocatalyst must perform under the conditions found in refinery processes. A
major limitation for large-scale application of available biocatalysts is their low
activity and stability. The complexity of the microbial process for
desulfurization, involving several enzymes and cofactors, seems to make the use
of whole cells the only choice. However, the use of whole cells for
desulfurization implies some bottlenecks such as mass transfer problems,
product inhibition and metabolic repression. Some of these problems have been
addressed by genetic manipulation of the system [34]. Hydrodesulfurization
process removes from 70 to 3000 mg of sulfur (g catalyst)"1 h'1 [35], so that
enzymes that are both more active and recognize a larger set of substrates are
needed. Novel genetic strategies such as gene shuffling might provide such
enzymes [36]. Furthermore, nearly all described microbial desulfurization
processes take place by mixing a cell suspension with oil [37,38]. It is desirable
to avoid the formation of a stable water-in-oil emulsion in order to facilitate oil
recovery, so that the oil/water volumetric ratio should be carefully adjusted.
Otherwise, the large-scale energetic cost of separating the multiphase system
could strongly impact the economics of the whole process.
Recently an intermittent process with immobilized cells was described by
the Petroleum Energy Center of Japan [39]. By entrapping cells of Rhodococcus
erythropolis KA2-5-1, it was possible to devise a two-phase system
(immobilized cells and oil) to desulfurize a model oil containing 100 ppm of
DBT. According to this report, the biocatalyst was easily recovered, reactivated
and reused during 900 h. However, the desulfurizing activity in the two-phase
system was lower than in the three-phase system, probably due to interference of
the support with the diffusion of substrates and products.
2.2 Enzymatic desulfurization

Besides the microbial option, enzymatic desulfurization represents a
promising alternative for biotechnological processes applied to the oil industry.
The advantages and disadvantages of enzymatic desulfurization when compared
with metabolic desulfurization are shown in Table 2. On one hand, the use of
microbes requires the maintenance of the entire cellular machinery in order to
regenerate the cofactors and carry out desulfurization. Furthermore microbes
73
need an aqueous phase to accomplish sulfur removal, while enzymes are able to
function in media containing very low water content. Thermodynamic water
activity (aw) influences both enzyme activity and stability, as water acts as a
lubricant altering the flexibility of enzyme molecules. Protein mobility and
therefore protein unfolding is restrained in a low water content medium.
However, a certain amount of protein-bound water is essential to allow enough
molecules flexibility to execute catalysis [40]. Thus it is possible to optimize
enzyme performance in hydrophobic media by controlling aw [10, 41-42]. In
addition, it has been shown that in certain organic media enzymes are active and
more thermostable than in aqueous media [12, 13, 43], and it is possible to
perform enzymatic transformations at temperatures higher than 100°C.
Although sulfur elimination might not be achieved by a single enzymatic
step, the enzyme-mediated transformation of sulfur-containing compounds may
facilitate its removal. An enzymatic procedure to reduce the sulfur content from
straight-run diesel has been described [44]. A fungal chloroperoxidase from
Caldariomyces fumago was able to oxidize the sulfur-containing fraction of
untreated diesel containing 1.6% sulfur, in the presence of low concentrations of
hydrogen peroxide.
Figure 2 shows gas chromatograms with both Flame Ionization (FID,
general) and Flame Photometric (FPD, sulfur selective) detectors. The
distribution of compounds in straight-run diesel fuel before and after oxidation
with chloroperoxidase, are shown in panel a and b, respectively. The oxidation
is clearly detected by the increase of boiling point (retention time) of these
compounds on the gas chromatogram monitored by the sulfur selective detector
(FPD). The higher boiling point of the oxidized compounds allowed its removal
by a distillation step. Microdistillation of both chloroperoxidase-oxidized and
untreated diesel fuels monitored by FID and FPD (Fig. 3) shows that the
hydrocarbon distillation profile changes slightly after enzymatic treatment. In
contrast, the sulfur selective detector (FPD) shows a significant change of the
distillation profile, in which most of organosulfur compounds were effectively
oxidized and their boiling points increased after enzymatic treatment.
Table 2
Process characteristics of enzymatic and metabolic desulfurization.
Enzymatic desulfurization Metabolic desulfurization

Activity in low water systems Activity in aqueous phase
Activity at temperature higher than 100°C Inactivation at high temperature
Activity in toxic systems Sensitive to toxics
Activity only on organosulfur compounds Needs carbon source
Life-time depending on molecule stability Self-producing catalytic system
74
Fig. 2. Gas chromatograms of straight ran diesel fuel before (a) and after (b) chloroperoxidase
treatment [44].
Oxidized sulfur compounds can then be removed by a distillation process

(Table 3). After distillation, the sulfur content in the enzymatically oxidized
diesel fuel is only 0.27%, while for the untreated fuel is 1.27%. The distillation
of the straight-run diesel fuel (1.6% sulfur) to a final distillation point of 325°C
produced a distillate containing 66% of the total sulfur, while if the diesel fuel
was previously oxidized with chloroperoxidase, the obtained distillate contained
only 12% of the total sulfur.
Thus, by using an enzymatic oxidation with chloroperoxidase coupled
with a distillation process it is possible to obtain a diesel fuel with 6-times lower
sulfur concentration than straight-run diesel fuel. Few hydrocarbons are
transformed during the enzymatic treatment, and after distillation an additional
12% of them remain in the residue (Table 3).
75
Fig. 3. Microdistillation of both chloroperoxidase-oxidized and untreated diesel fuels

monitored by FID (general detection) and FPD (sulfur selective detection) [45].
Chloroperoxidase was able to oxidize a wide range of sulfides,

benzothiophene and dibenzothiophene [45]. However, a drawback for this
procedure is the potential modification of aromatic hydrocarbons due to the
enzymatic treatment. The presence of chlorinated hydrocarbons was detected
when individual reactions were performed with chloroperoxidase [46]. The
generation and combustion of such compounds is environmentally undesirable,
as it would deteriorate the fuel value and the air quality.
Table 3
Sulfur content of straight-run diesel fuel after enzymatic oxidation with chloro-
peroxidase from Caldariomyces fumago followed by a distillation to 325°C as final
distillation point.
Distillation Enzymatic -H distillation

TPH (%) Sulfur (%) TPH (%) Sulfur(%)
Destillate 83 1.27 71 0.27
Residue 17 3.21 29 5.51
TPH. Total petroleum hydrocarbons.
76
Nevertheless it is expected that in a complex mixture, sulfur compounds

would be preferentially oxidized due to the higher affinity and activity of
chloroperoxidase towards these substrates [46]. Other enzymes are known to
catalyze sulfoxidation, but chloroperoxidase shows higher activity and broader
specificity [47-50]. Some of the sulfur-containing substrates transformed by
chloroperoxidase are listed in Table 4.
Enzymatic desulfurization shares some of the challenges of microbial
desulfurization: the stability and activity of the biocatalyst must be appropriate
in order to develop a commercially competitive process. Several factors
influence these properties. One of them is enzyme preparation. Usually enzymes
are immobilized on an inert support in order to avoid product contamination and
simplify biocatalyst recovery. Immobilization might improve stability of the
enzyme by multiple point attachment [21]. Moreover, the chemical and physical
characteristics of the support determine the microenvironment of immobilized
enzymes, therefore altering its activity as well as the diffusion of substrates and
products [51].
Another factor influencing biocatalysts performance is water content in
nonaqueous media. As mentioned earlier, the thermodynamic water activity
influences both enzyme activity and stability in hydrophobic media. Depending
on the solvent system and on the enzyme, optimal water content must be found
in order to enhance enzyme performance [52]. It is know that some polar
organic solvents interact unfavorably with proteins, displacing water molecules
that may be essential for activity and accelerating protein unfolding [53, 54].
Hence sometimes enzyme folding is better preserved in the presence of non-
polar organic solvents and even thermostability might improve.
Finally, substrate partition between bulk solvent and the active site of the
enzyme may limit the reaction rate. It has been demonstrated that the partition of
hydrophobic substrates to the active site of an enzyme is less favorable in the
presence of organic solvents [55, 56]. The KM value offers an indication of the
unfavorable partition since it increases with the organic solvent content [57].
Though stabilization and immobilization of chloroperoxidase for use in organic
solvents have been accomplished, substrate availability needs additional study.
Thermodynamic understanding of substrate partition and the influence of active
site environment might provide the basis for solving this problem.
The most important challenge for the scale-up of the enzymatic process
using chloroperoxidase is, doubtless, the stability of the enzyme against the
peroxide inactivation. The present and future commercial uses of peroxidases
have been limited, mainly, by the low stability of peroxidases in the presence of
their natural substrate, hydrogen peroxide. All hemeproteins, including
peroxidases, are inactivated in the presence of catalytic concentrations of
hydrogen peroxide [58]. This inactivation process is specially important in the
absence of reducing substrates and its mechanism has not been fully elucidated.
77
Table 4
Sulfur-containing substrates of chloroperoxidase.
In order to overcome the described shortcomings, several tools may be

used to enhance enzyme desulfurization, such as protein engineering, solvent
engineering and immobilization procedures. Biocatalyst design should be
conducted contemplating parallel increments in activity and stability. It has been
observed that improvement of one characteristic might result detrimental for
other biocatalyst properties [18]. Thus by selecting for simultaneous
enhancements of different properties, more suitable biocatalysts might be
obtained [36, 18].
78
Thus, enzymatic biodesulfurization is a promising alternative to achieve

low sulfur levels in oil-derived streams. Some enzymes able to perform
modifications on sulfur-containing compounds that facilitate its removal have
been identified. It has been estimated that refineries investment for achieving
deep desulfurization would be considerably reduced (by 50%) if combined
hydrodesulfurization-biodesulfurization processes are implemented [9]. While
microbial desulfurization has been well studied for the last ten years and
significant advances have been achieved, enzymatic desulfurization is a less
explored field with great potential as it lacks some of the important drawbacks
of microbial desulfurization. In order to represent an economically viable
alternative, biodesulfurization processes must adequate to the conditions found
in a refinery. Specifically, both activity and stability of current biocatalyst must
be enhanced. Additional effort must be done to design an appropriate biocatalyst
considering all stages of the process. Importantly, preparation and recuperation
strategies of biocatalyst should not be neglected.
3. ENZYMATIC TRANSFORMATION OF ASPHALTENES
Asphaltenic and viscous heavy oils from bituminous deposits are a huge energy
reserve to be exploited in next decades. More than 70 countries possess
bituminous deposits. In Canada only, the oil reserve considered to be technically
recoverable is estimated to be 280-300 Gb (billion of barrels), larger than the
Saudi Arabia oil reserves estimated at 240 Gb [59]. These highly asphaltenic
resources must be rigorously treated in order to convert them into an upgraded
crude oil before them can be refined to produce gasoline and other fuels.
Asphaltene, the highest molecular weight fraction of petroleum, is a dark
amorphous solid specially rich in heteroatoms (S, O, N), and metals (Fe, Ni, V)
[60-62]. Many problems associated with either recovery, separation or
processing of heavy oils and bitumens, are related to the presence of high
concentration of asphaltenes. This fraction is thought to be largely responsible
for other adverse oil properties such as high viscosity and the propensity to form
emulsions, polymers and coke. The molecular structure of asphaltenes has been
an enigma for seven decades [62]. From numerous investigations there are
indications that asphaltenes are condensed aromatic cores containing alkyl and
alicyclic moieties. Heteroatoms, such as nitrogen, oxygen and sulfur are present
as non- and heterocyclic groups. A significant amount of porphyrins
(petroporphyrins) can be found containing mainly nickel and vanadium. A
hypothetical asphaltene molecule is shown in Fig. 4. The complexity of the
asphaltene chemical nature is evident by the difficulty of analysis of both their
molecular weight and structure.
The asphaltenic fraction is recognized as the most recalcitrant fraction of
oil. So far, there is no clear evidence that asphaltenes are degraded by microbial
79
activity (see chapters 1 and 4). Some reports on oil biodegradation claim the
degradation of asphaltenic fraction by mixed bacteria [63, 64]. However, none
of these reports described the analytical results of extractable materials
recovered from appropriate sterile controls. On the other hand, although
microorganisms have been found associated with bitumens containing high
amounts of asphaltenes [65], the asphaltenic fraction did not support bacterial
growth and no changes in asphaltene content could be found after bioconversion
of heavy oils and asphaltenes [66, 67]. Because the asphaltene content was
usually determined gravimetrically after n-alkane precipitation, the reported
changes could be attributed to the disruption of the asphaltenic matrix by the
production of surfactants during bacterial growth, liberating trapped
hydrocarbons. Therefore, most of the asphaltene losses during microbial activity
could be considered to be abiotic losses [68].
Nevertheless, a clear experimental evidence that enzymes are able to
modify asphaltene molecules has been reported [69]. Chloroperoxidase from the
fungus Caldariomyces fumago was able to transform petroporphyrins and
asphaltenes, and this modification was significantly higher in systems containing
organic solvent than in aqueous systems [69, 70]. Asphaltenes and
petroporphyrins are highly hydrophobic materials, thus mass transfer limitations
are expected in aqueous reactions. The biocatalytic oxidation of a
petroporphyrin rich-fraction of asphaltenes in the ternary solvent system and in
the presence of hydrogen peroxide was performed. Chloroperoxidase catalyzed
reaction produced notable spectral changes in the petroporphyrin rich-fraction of
asphaltenes (Fig. 5). The destruction of petroporphyrins by chloroperoxidase in
the presence of hydrogen peroxide leads to the removal of Ni and V from
asphaltene molecules, as in the case of synthetic nickel and vanadium
porphyrins (Table 5).
On the other hand, a doubly modified cytochrome c (PEG-Cyt-Met) was
able to catalyze the oxidation of a petroporphyrin rich-fraction of asphaltenes in
the ternary solvent system and in the presence of 100 mM of tert-butyl
hydroperoxide [71]. As chloroperoxidase, the PEG-Cyt-Met catalyzed reaction
produced spectral changes in the petroporphyrin rich-fraction of asphaltenes
(Fig. 5). The oxidative porphyrin ring disruption entails the simultaneously
release of metal. The biocatalytic process with PEG-Cyt-Met removed 95% of
the vanadium and 74 % of the nickel (Table 5). The destruction of the
petroporphyrin molecules is conformed by the Soret band loss and metal
removal.
80
Fig. 4. Asphaltene molecule proposed by Strausz et al. [62].
Figure 5. Absorption spectra of the petroporphyrin rich-fraction of asphaltenes after

biocatalytic treatment. Control without treatment (a); control without biocatalyst and in the
presence of hydroperoxide (b); reaction with one addition of biocatalyst (c); and reaction after
a second addition of biocatalyst (d). [69, 71].
81
Table 5
Nickel and Vanadium removal from petroporphyrin rich fractions of asphaltenes by
chloroperoxidase-mediated reaction.
Heavy metal Chloroperoxidase [69] Chloroperoxidase [70] Chemically modified

cytochrome c [71]
Nickel 20% 57% 74%
Vanadium 19% 52% 95%
Petroporphyrin-rich fractions from asphaltenes with and without

biocatalytic treatement were analyzed by Fourier transform infrared
spectroscopy (FTIR) (Fig. 6). Significant differences could be detected mainly
as an increased proportion of oxygen containing groups, such as hydroxyl (3310
cm"1), carboxyl (1300 cm"1, 1770 cm"', 1710 cm"1), aldehydes (1730 cm"'),
sulfoxides (1040 cm"1), sulfones (1130 cm"1), and sulfonates (1160 cm"1, 1260
cm"1). Sulfur is, after carbon, the most important element in asphaltene
molecules (Fig. 4), and most of it is contained in thiophenes and organic sulfides
moieties. According to the FTIR spectrum, PEG-Cyt-Met also catalyzes the
oxidation of carbon atoms from asphaltenes molecules.
Biocatalytic cracking, or biocracking, is probably the most interesting
biotechnology target for heavy oil upgrading (see chapter 4). Gel permeation
chromatography (GPC) monitored by a diode array detector of both untreated
and biocatalytically oxidized petroporphyrin are shown in Fig. 7 [71]. The
isoabsorbance contours corresponding to the Soret band (maximal absorbance at
403 nm and elution time of 10.77 min) disappeared after the biocatalytic
oxidation as expected, and in agreement with the spectra shown in Fig. 6.
Nevertheless, gel permeation chromatograms show higher molecular weight
distribution of oxidized petroporphyrin than the control distribution. However,
these results should be taken cautiously because the oxidation process, which
introduces polar groups in the molecules, may affect the asphaltenes aggregation
state. Asphaltenes are a very complex mixture and are defined only by their
solubility properties: the asphaltenic fraction is insoluble in short-chain n-
alkanes, specially pentane. As mentioned above, due to their complexity, the
order of magnitude of asphaltene molecular weight is still controversial [72, 73,
74].
The 'H NMR analysis of control and treated samples showed the following
regions (Fig. 8): Hy (y+ CH3) from 0.5 to 1.0 ppm; Hp (P+ CH2): HR from 1.0 to
1.6 ppm, HN from 1.6 to 2.0 ppm; H a (a CH2) from 2.0 to 4.0 ppm and Har (CH
aromatic) from 6.0 to 9.0 ppm. Here, HR and HN refers to P-protons in aliphatic
chains and naphthenic rings, respectively. The spectrum from untreated fraction
(control) was similar to those found in other asphaltenes samples [76]. However,
82
the spectra from both untreated and treated samples showed an important 5.29
ppm signal, probably due to protons from non aromatic (C = C) double bonds,
which are not detectable in whole asphaltenes fractions. The main differences of
enzyme treated samples when compared with the untreated fraction appeared in
the saturated hydrocarbon region: a quartet placed on 2.28 ppm, a triplet placed
on 2.49 ppm, and a singlet placed on 3.63 ppm. These signals seem to be
originated from the hydrocarbon chains of polar compounds, may be from
oxygenated compounds as 13C spectrum shows (see below). The singlet shift can
be attributed to ether or alcohol groups. The ester-amide signal (4.3-4.36 ppm)
was very important in the oxidized sample, while was minor in the control
petroporphyrins.
The 13C NMR analysis showed the presence of 58.78 ppm and 46.11 ppm
shifts in the control, which are attributed to the C-N bond (Fig. 9). These signals
disappeared in the oxidized petroporphyrins. Signals between 10 ppm and 60
ppm are usually assigned to the hydrocarbon chains. The NMR spectrum from
oxidized petroporphyrins showed a more intense terminal methyl (-CH3) signal
than in the untreated sample. The (-CH2-) / (-CH3) intensities ratio was lower in
the treated asphaltenes fraction than in the untreated ones, which could be
attributed to the presence of shorter alkyl chains or more branched chains. Thus,
this lower ratio could be the consequence of molecule cracking. The aromatic
region of the spectra (110 ppm to 160 ppm) showed significant differences. The
control showed a signal-hill between 133 ppm and 146 ppm, which include the
carbon atoms corresponding to heteroatom moieties (S, N, O), aromatic carbons
bonded to alkyl moieties, and aromatic carbons bonded to other aromatic
carbon. This signal-hill disappeared in the oxidized fraction, suggesting a loss of
heteroatoms or alkyl derivatives in the aromatic molecules. A reduction in the
number of substituted aromatic carbons and an increase of the number of
aromatic carbons bonded to hydrogen are observed.
The enzymatic treatment of asphaltenes is an interesting alternative for the
removal of heavy metals in order to reduce catalyst poisoning in hydrotreatment
and cracking processes. On the other hand, enzymatic cracking of asphaltenes
molecules should not be excluded. The enormous amount of energetic resource
found as asphaltenes-rich deposits justify the exploration of alternative
upgrading technologies.
83
Fig. 6. FTIR spectra of untreated and biocatalytically treated porphyrin-rich fractions from
asphaltenes. FTER was performed using the film-spreading technique [71].
Fig. 7. Absorbance contours from gel permeation chromatography (GPC) of untreated and
biocatalytically treated porphyrin-rich fractions from asphaltenes [71].
84
Fig. 8. H NMR analysis of control and enzymatically treated petroporphyrin-rich fraction of

asphaltenes.
Fig. 9. C NMR analysis of control and enzymatically treated petroporphyrin-rich fraction of

asphaltenes.
85
4. OXIDATION OF AROMATIC HYDROCARBONS
The ability of microorganisms (bacteria and fungi) to modify polycyclic aromatic

hydrocarbons (PAHs) by oxidation is well known, and a number of
comprehensive reviews have been written on the microbial metabolism of PAHs
[75-79]. The controlled partial oxidation of aromatic hydrocarbons by molecular
oxygen (dioxygen) is both highly desirable, from an environmental point of view,
and difficult [80]. Oxidation of organic matter to carbon dioxide and water by
dioxygen is a thermodynamically highly favourable process. Fortunately for
biological systems, the kinetic barrier is large. The main problem in biological
systems is how to promote the reaction whilst at the same time limiting the
damage caused by indiscriminate attack of dioxygen. Nature achieves this by
using metalloenzymes, many of which contains iron porphyrin groups
(hemoenzymes) as active-site responsible for activating the dioxygen. The next
section is an overview focused on the biocatalytic oxidation of PAHs in vitro by
using enzymatic and non enzymatic proteins.
4.1. Lignin peroxidase

Lignin is the most abundant renewable organic material next to cellulose
and it is an aromatic polymer. Lignin is mainly decomposed by higher
basidiomycetous fungi that cause the white-rot wood decay. Because insolubility
and complexity of this substrate, ligninolytic microorganisms have evolved to
secrete multiple lignin peroxidase enzymes. These enzymes act as non-specific,
diffusable oxidative catalysts that serve to degrade lignin. Lignin peroxidases
from Phanerochaete chrysosporium are the most extensively studied. Twelve
years ago, Sanglard et al. [81] reported the first evidence of enzymatic oxidation
of benzo(a)pyrene by lignin peroxidase. The reaction mixture contained crude
lignin peroxidase and a H2O2 generating system with glucose oxidase and
glucose. Purified enzyme is able to oxidize PAHs with ionization potential (IP)
lower than 8 eV in the presence of H2O2 [82, 83]. In addition, the specific activity
was correlated with the IP of PAHs (Fig. 10). In most of cases, oxidation products
were identified as quinones, although hydroxylated compounds were also
detected (Table 6).
Lignin peroxidases may also participate on the oxidation of aromatic
intermediates. Phenanthrene (IP = 8.03 eV) is not substrate for lignin peroxidase.
Nevertheless, its oxidized metabolite 9-phenanthrol is transformed to
phenanthrene-9,10-quinone by lignin peroxidase in vitro [86]. Lignin peroxidase
is able to catalyze other environmentally high-risk compounds, such as
polychlorinated phenols [87], sulphur and nitrogen organocompounds [85, 88],
and industrial dyes [89, 90].
86
Fig. 10. The influence of ionization potential of PAHs on the specific activity of lignin
peroxidase oxidation [82].
4.2. Manganese peroxidase

Ligninolytic microorganisms also produce extracelluar manganese-
dependent peroxidases. As in the case of lignin peroxidases, manganese
peroxidases are a family of isoenzymes produced by ligninolytic fungi [91, 92].
Manganese peroxidases are heme glycoproteins that require Mn(II) for its
activity. Mn(II) is oxidized to Mn(III), which behaves as a low-molecular-weight
mediator that diffuse to remote regions of the lignin molecule and initiate the
oxidation process. The oxidation of Mn(II) to Mn(III) is dependent on the
presence of chelating agents, such as lactate, succinate or malonate. This fungal
peroxidase is, so far, the only known enzyme system that utilizes soluble
Mn(II)/Mn(III) as an obligatory redox couple.
Manganese peroxidase is able to oxidize some xenobiotics including PAHs
[93]. Enzymatically generated Mn(III) oxidize low redox potential
methoxybencenes during lignin degradation [94]. During lipid peroxidation, this
enzyme is able to oxidize phenanthrene [95], which has an IP higher than 8 eV
and is not oxidized by lignin peroxidase [85]. The slow oxidation of phenanthrene
to 2,2'-diphenic acid supported by manganese peroxidase requires Mn(II),
oxygen, and unsaturated lipids. Fluorene, a PAH which again is not a substrate for
lignin peroxidase, is oxidized by manganese peroxidase in a lipid peroxidation-
dependent reaction [96]. The product of fluorene is 9-hydroxyfiuorene via 9-
fluorenone, and this reaction is inhibited by free-radicals scavengers. Manganese
peroxidase could also be involved in the oxidation of another PAH with high IP,
chrysene [97]. Table 7 shows the slow oxidation of PAHs in vitro with
manganese peroxidase from Phanerochaete chrysosporium during lipid
peroxidation.
87
Table 6
Products from in vitro oxidation of polycyclic aromatic hydrocarbons by lignin
peroxidase from Phanerochaete chrysosporium.
PAH Products Ref.

Anthracene 9,10- Anthraquinone 83,84
Acenaphthene 1-Acenaphthenol 83
1 - Acenaphthenone 83
1 -Methylanthracene 1 -Methyl-9,10-anthraquinone 83
2-Methylanthracene 2-Methyl-9,10-anthraquinone 83
9-Methylanthracene 9-Methyl-9,10-anthraquinone 83
9-Methyl-10-anthranone 83
9,10-Anthraquinone 83
Pyrene 1,8-Pyrenedione 83,84
1,6-Pyrenedione 82
Benzo(a)pyrene 1,6-Benzo(a)pyrenedione 87
3,6-Benzo(a)pyrenedione 87
6,12-Benzo(a)pyrenedione 87
PAH oxidation by this lipid peroxidation-mediated system with manganese

peroxidase showed to be significantly slower than the lignin peroxidase oxidation.
Nevertheless, manganese-dependent lipid peroxidation with extracelluar extracts
from Phanerochaete chrysosporium oxidize PAH in complex mixtures, such as
creosote, and the oxidation rate is correlated with the ionization potential [98]. In
addition the catalytic activity seems to be very stable and constant activity could
be found during more than 48 hours.
Extracellular crude extracts and semipurified manganese peroxidase from
Bjerkandera sp. are able to transform anthracene to anthraquinone [83, 99].
Ligninolytic fungi Phanerochaete leavis HHB-1625 which produces high levels
of manganese peroxidase and no lignin peroxidase, is able to transform PAH in
liquid culture [100]. Extracelluar extract from this strain was able to oxidize
anthracene, phenanthrene, benz(a)anthracene, and benzo(a)pyrene in the presence
of Mn(II) and hydrogen peroxide.
4.3. Versatile peroxidase

Active lignin-degrading strains of Pleurotus eryngii were shown to
produce a peroxidase, different from P. chrysosporium peroxidases, that can
efficiently oxidize Mn(II) to Mn(III), but can also carry out Mn(II)-independent
activity on aromatic substrates [101]. This novel manganese-lignin peroxidase
hybrid enzyme, now called versatile peroxidase (VP), was also described in
Bjerkandera sp. BOS55. This enzyme is able to oxidize various phenolic and
nonphenolic substrates such as 2,6-dimethoxyphenol, guaiacol, ABTS, and
veratryl alcohol, in the absence of Mn(II) [102], Similar hybrid peroxidases have
been reported in Pleurotus eryngii [103-106], Pleurotus pulmonarius [107],
88
Pleurotus ostreatus [108], as well as in Bjerkandera adusta [102, 103, 109,

110].
In the VP from B. adusta the oxidation of Mn(II) to Mn(III) proceeds
optimally at pH 4.5, while the LiP-like activity requires more acidic conditions,
showing maximum rates at pH 3.0 [104]. VP seems to have a long range
electron transfer pathway similar to that postulated for LiP [111]. The
spectroscopic characterization of VP by EPR and electronic absorption
techniques showed a protein centered radical in the presence of an excess of
hydrogen peroxide which was assumed to be a tryptophanyl radical [112]. In
addition the enzyme shows high identity with LiP (58-60%) and MnP [55%]
from Phanerochaete chrysosporium [91]. The heterologous expression of VP in
Aspergillus nidulans confirmed the ability of this hybrid enzyme to oxidize both
Mn(II) and also different aromatic compounds in the absence of manganese
[106].
The ability of the VP to oxidize PAHs was recently reported by Wang et
al [113]. Oxidation of PAHs was examined by a purified VP isoenzyme in the
presence and absence of Mn(II). PAH oxidation was reduced by the presence of
Mn(II) and the inhibition kinetics were shown to be partially noncompetitive.
The substrates were anthracene and its methyl derivatives, pyrene and
benzo(a)pyrene, with IP of 7.43 eV or lower (Table 8). The PAH metabolites of
the Mn-independent reaction were identified as the corresponding quinones. The
pH optimum of the Mn-independent oxidation was around pH 4, while for the
Mn-dependent reaction it was pH 3. The kinetic constants for the Mn-
independent oxidation of 2-methylanthracene at pH 4 were determined, and the
values we obtained were a kcat of 145 min"1, KM,apP for the aromatic substrate of
23.8 mM, and KM,app for hydrogen peroxide of 0.2 mM.
4.4. Cytochromes P450

Cytochromes P450 form a superfamily of hemoenzymes that were
originally named as a pigment having maximum absorbance at 450 nm in the
presence of CO and unknown function. Biologically occurring substrates for
cytochrome P450 include fatty acids, steroids, eicosanoids, lipid hydroperoxides,
retinoids, arginine, acetone and acetol. Interestingly, there is a very large number
of xenobiotic compounds that are substrates for cytochrome P450. These
xenobiotic substrates includes drugs, procarcinogens, antioxidants, solvents,
anesthesics, dyes, pesticides, petroleum products, alcohols, flavorants and
odorants. Ironically, in addition to its beneficial roles in metabolism, biosynthesis
and detoxification, cytochromes P450 are implicated as the activators of many
chemical carcinogens.
89
Table 7
In vitro PAHs oxidation with manganese peroxidase in lipid
peroxidation reactions [96].
. , Oxidation rate
Aromatic compound
r , ...
(nmol/h)
Fluorene 3.10
Benz(a)antrhracene 1.08
Benzo(a)pyrene 0.96
Anthracene 0.93
Dibenz(a,c)anthracene 0.60
Benzo(e)pyrene 0.31
Diphenylmethane 0.30
Benzo(c)phenanthrene 0.21
Benzo(b)fluoranthene 0.19
Fluoranthene 0.14
Phenanthrene 0.06
Table 8
Oxidation of aromatic compounds by versatil peroxidase at pH 4.0 in the absence
ofMn(II)[113].
PAH Ionization potential (eV)a Specific activity (min"1)

9-Methylanthracene 7.25 52 (±2.7) b
C
1 -Methylanthracene NA 22.4±(1.7)
2-Methylanthracene 7.37 12.4 (±1.1)
Anthracene 7.41 2.5 (+0.01)
Benzo(a)pyrene 7.41 0.32 (±0.04)
Pyrene 7.42 0.008 (±0.01)
Benzo(e)pyrene 7.43 NRd
Chrysene 7.60 NR
Carbazol 7.68 2.4 (±0.05)
1 -Methylphenanthrene 7.70 NR
Acenaphthene 7.76 NR
Phenanthrene 7.91 NR
Dibenzothiophene 7.93 NR
Fluoranthene 7.95 NR
Naphthalene 8.15 NR
a
Photoelectron spectroscopy values (http://webbook.nist.gov)
b
Values in parentheses are standard deviations.
C
NA, not available
d
NR, no reaction
90
Cytochromes P450 are widely distributed in living organisms and have

been found, for instance, in mammals, fish, yeast, bacteria, and plants [114].
Cytochromes P450 are part of multienzymatic systems called monooxygenases
which catalyze the activation of dioxygen and the transfer of one oxygen atom to
substrates, with the simultaneous consumption of NAD(P)H.
The monooxygenase cycle of cytochrome P450 has been well characterized
[49]. It has been observed that cytochromes P450 are able to carry out oxidations
with exogenous single-oxygen-atom donors like H2O2, alkyl-hydroperoxides,
iodosobenzene, amine oxides, and peracids. These oxidations are observed in
vitro with cytochrome P450 alone, without consumption of NAD(P)H. The
oxidation of benzo(a)pyrene catalyzed by cytochrome P450 in the presence or
absence of hydroperoxides yield different products. While the major products
formed in the presence of cumene hydroperoxide are quinones [117], only
phenols are formed with NAD(P)H in the absence of the hydroperoxide. After the
first measurement of benzo(a)pyrene hydrolase activity based in a direct
fluorometric method [118], several studies have been carried out in order to
determine the kinetic constants of both microsomal and purified preparations
[115, 116, 119, 120]. However, because the large number of P450 enzymes that
can be present in a single organism, and because multiple species with distinct
conformations and substrate recognition profiles coexist in a biological
membrane, data form microsomal preparations should be considered cautiously.
Interaction of these complex systems with PAHs could be resolved by using rapid
kinetic technique [121]. Data from highly purified preparations from yeast with
different single-oxygen-donors are showed in Table 9. Benzo(a)pyrene has been
extensively used as substrate, nevertheless other PAHs are oxidized by P450
enzymes in vitro, such as anthracene and alkylanthracenes [122], phenanthrene
[123], and the most potent carcinogen among all PAH, dibenzo(a,l)pyrene [124].
Table 9
Kinetic constants of purified cytochrome P448 from Saccharomyces
cerevisiea for oxidation of benzo(a)pyrene [115, 116].
System kcat (min"1) KM ( U M )

Reconstituted with NADPH 33 0.017
Cumene hydroperoxide 125 0.022
Hvdrosen oeroxide in situ 200 0.034
91
Fig. 11. Engineering cytochrome P450 BM-3 for oxidation of polycyclic aromatis hydrocarbons
[127].
In order for cytochrome P450 to be an effective catalyst, the enzyme must

efficiently bind substrates. Thermodynamic studies on the substrate binding to the
active site of rat liver cytochrome P450 by using a series of aromatic
hydrocarbons [125] showed that substrate hydrophobicity is an important driving
force that determines substrate affinity. The predominant force involved in
binding is the ability of the active site to draw the aromatic hydrocarbon from the
aqueous phase [126].
Using hydrophobic interaction analysis to design a new biocatalyst, site-

directed mutagenesis has been used to modify a bacterial cytochrome P450 [128].
This was the first report of rotational redesign of cytochrome P450 in which by
changing only one active-site residue the affinity for different substrates was
changed. The modified substrate pocket allowed tight binding of a novel
substrate, diphenylmethane. Heme domain also influences substrate affinity, as
showed when native prosthetic group is replaced by an heme dimethyl ester [129].
Esterification of the heme propionates groups removes the negative charges from
the vicinity of substrate-binding site, increasing its hydrophobic nature, and thus
increasing the substrate affinity.
Significant work has been performed on these enzymes using molecular
techniques. Laboratory evolution of cytochrome P450 from Pseudomonas putida
for peroxide-mediated hydroxylation of naphthalene has been performed [130].
The obtained mutants showed, in the absence of cofactors through the "peroxide
shunt" pathway, more than 20-fold higher activity than the native enzyme for
92
naphthalene hydroxylation. Moreover, cytochrome P450 has been enginnered into

a catalyst for the oxidation of PAHs [127, 131]. Compared with the activities of
wild type, those of the mutants improved by up to 4 orders of magnitude (Fig.
11).
4.5. Cytochrome c
Cytochromes c are part of the energy-conserving electron transport
systems. In living systems no catalytic activity of cytochrome c has been
described. The ability of cytochrome c to act as catalyst in vitro has been
reviewed [132]. Lipid peroxidation, hydroperoxide cleavage, hydroxylation of 4-
nitrophenol and oxidation of 2-keto-4-thiomethyl butyric acid in the presence of
hydrogen peroxide have been reported. Peroxidase activity of cytochrome c has
been also demonstrated by the oxidation of various electron donors including
ABTS (2-2'-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) and 4-amino-
antipyrine. In addition, cytochrome P450-like oxidative reactions such as N- and
O-demethylations, S-oxidations and olefin epoxidation are catalyzed by free and
immobilized cytochrome c in the presence of hydrogen peroxide or other organic
hydroperoxide [132, 133, 134]. It has also been observed that aromatic substrates
of cytochrome c interact with the heme group as ligand rather than as a substrate
[133, 135].
Table 10
Specific activity of yeast cytochrome c on aromatic compounds [135].
Aromatic compound Product Specific activity (min"1)

Dibenzothiophene Dibenzothiophene sulfoxide 3.2 (±0.1)
Anthracene 9,10-Anthraquinone 2.1 (±0.1)
Pyrene 1,8-Pyrenodione 1.3 (±0.3)
Benzothiophene Benzothiophene sulfoxide 1.0 (±0.2)
Carbazole Unknown 0.9 (±0.1)
Acenaphthene NR
Chrysene NR
Fluoranthrene NR
Fluorene NR
Phenanthrene NR
Triphenylene NR
NR, no reaction detected.
93
Table 11
Kinetic constants of wild-type and variants of yeast iso-1-
cytochrome c for pyrene oxidation [135].
kcat K M ,app kcat/KM,app

Variant (s'1) (mM) (s"1 M"1)
Ala86;ThrlO2 0.17 4.0 33
Phe67;ThrlO2 0.10 3.3 32
Ala72;ThrlO2 0.13 4.0 33
Ala52;ThrlO2 0.18 4.7 39
Ala73;ThrlO2 0.28 7.5 38
Ala87;ThrlO2 0.39 3.9 99
Phe82;CyslO2(WT) 0.31 9.7 32
Ala79;ThrlO2 3.28 101.8 32
Table 12
Oxidation of polycyclic aromatic hydrocarbon by unmodified- and methylated
poly(ethylene)glycol-modified-cytochrome c [137].
Specific activity (min"1)

Aromatic compound Unmodified PEG-Cyt-Met
7,12-Dimethylbenzanthracene 24.59 (±1.52) 80.33 (±3.83)
1,2:3,4-Dibenzanthracene NR 16.60 (±2.24)
Azulene 2.26 (±0.29) 14.32 (±0.57)
3-Methylcholanthrene 1.88 (±0.07) 10.96 (±0.54)
7-Methylbenzo(a)pyrene NR 7.56 (±0.42)
1,2:5,6-Dibenz anthracene NR 5.70 (±0.31)
Triphenylene NR 5.27 (±1.05)
Dibenzothiophene 0.67 (±0.06) 4.73 (±0.05)
Anthracene 0.33 (±0.06) 3.09 (±0.32)
Thianthrene 0.49 (±0.06) 1.41 (±0.08)
Pyrene 0.51 (±0.05) 0.97 (±0.03)
Fluoranthene NR 0.65 (±0.09)
Acenaphthene NR 0.40 (±0.01)
Benzo(a)pyrene 0.22 (±0.02) 0.39 (±0.06)
Fluorene NR 0.22 (±0.01)
Phenanthrene NR 0.17 (±0.02)
Chrysene NR NR
9,10-Dimethylanthracene NR NR
Naphthalene NR NR
Biphenyl NR NR
NR. No reaction detected
94
The first oxidation of an aromatic hydrocarbon with cytochrome c in the presence

of hydroperoxide was reported by Akasaka et al. [136]. Hydroxylation of benzene
was carried out in organic solvent with less than 5% of water, and with
immobilized protein. Free cytochrome c was unable to perform this reaction. The
capacity of yeast cytochrome c to perform oxidations of PAHs has also been
reported [135]. Biocatalytic activities on 11 aromatic compounds were tested in a
system containing 10% acetonitrile and 1 mM H2O2. The specific activities for the
oxidation of these compounds are shown in Table 10. Anthracene, pyrene,
dibenzothiophene, benzothiophene and carbazole were oxidized by the catalytic
activity of yeast cytochrome c. No correlation between substrate IP and specific
activity was observed when using cytochrome c as catalyst.
Site-directed mutagenesis has been performed on yeast cytochrome c; it
was observed that Phe82 substitution significantly altered the kinetic behaviour of
the protein. The Gly82:ThrlO2 variant showed 10 times more catalytic activity
and a ten-fold catalytic efficiency than the wild-type iso-1-cytochrome c [135].
For the oxidation of pyrene, the different variants of yeast cytochrome c showed
different catalytic constants (Table 11). Lysine 79 residue is placed at the edge of
the solvent access to the heme group, and its substitution by alanine produced a
protein with higher kcat but also higher KM, resulting in similar catalytic
efficiency. These results show that site-directed mutagenesis could be a tool for
the design of a better biocatalyst for PAHs oxidation.
In addition to genetic techniques, chemical modification has been
performed on horse heart cytochrome c [137]. Free amino and carboxylic groups
of horse heart cytochrome c were modified by chemical reaction with
poly(ethylene)glycol (PEG) moieties. As a consequence of the chemical
modification the heme environment in the active site was altered. Cytochrome c
with a double modification: PEG on free amino groups and methyl esters on
carboxylic groups (including propionates of heme), was able to oxidize 17
aromatic compounds from 20 tested, while the unmodified protein was only able
to oxidize 8 compounds (Table 12). Thus, chemical modification of biocatalyst
could be also a tool for the design of new biocatalyst with environmental
proposes. As mentioned above, cytochromes c are very stable proteins, and it is
possible to perform on it a large variety of chemical reaction without a negative
effect on the activity.
4.6. Hemoglobin
In the presence of hydrogen peroxide, hemoglobin has been reported to
oxidize aniline [138], lipids [139], S- and N-heterocycles [134, 140, 141] and
other organic substrates [141]. This protein could be considered as an antioxidant
in red blood cells [142]. Biocatalytic activity of hemoglobin on PAHs has been
tested with 12 compounds in the presence of hydrogen peroxide [143]. Among
the aromatic compounds tested, 6 were oxidized (Table 13). As in the case of
95
cytochrome c, and in contrast with lignin peroxidase, no correlation between the

extent of oxidation by hemoglobin and the ionization potential of the substrates
was found. Interestingly, hemoglobin is able to oxidize fluorene, while with lignin
peroxidase and cytochrome c no reaction was detected [85, 135]. Reaction
products were identified as quinones, and are the same that those obtained with
lignin peroxidase and cytochrome c. The product from carbazole oxidation was
not reported; however, this product could be a polymer such as in oxidation with
lignin peroxidase [88].
The catalytic mechanism of hemoglobin seems to be similar to that
of other hemoproteins. Early studies using ESR trapping techniques showed free
radical involvement in the oxidative reactions by organic hydroperoxides and
erythrocytes [144]. ESR studies, as for cytochrome c, detected peroxyl and
alkoxyl radicals produced by reaction of hydroperoxides and hemoglobin [145].
These experiments suggested that the formation of free radicals involves high-
valence-state iron complexes [139]; molecular oxygen could be involved, in part,
in the oxidation reactions [141].
4.7. Chloroperoxidase
Chloroperoxidase from Caldariomyces fumago (CPO) is a 42 kDa
extracellular heme glycoenzyme containing ferriprotoporphyrin IX as prosthetic
group [146]. CPO exhibits a broad spectrum of chemical reactivities; even
though in vivo it functions mainly as a peroxide-dependent chlorinating enzyme,
it also catalyzes peroxidase-, catalase- and cytochrome P450-type reactions of
dehydrogenation, H2O2 decomposition and oxygen insertion, respectively, in
vitro [147].
Table 13
Biocatalytic oxidation of aromatic compounds by hemoglobin and hydrogen
peroxide [143].
Compound Reacted substrate (%) Product

9-Hexylanthracene 100 9,10-Anthraquinone
Anthracene 91 (±4) 9,10-Anthraquinone
Carbazole 84 (±28) Unknown
Pyrene 85 (±10) 1,8-Pyrenodione
Dibenzothiophene 74 (± 1) Dibenzothiophene sulfoxide
Fluorene 49 (±30) 9-Fluorenone
Acenaphthene NR
Chrysene NR
Dibenzofuran NR
Fluoranthene NR
Phenanthrene NR
96
Table 14
Specific activity of chloroperoxidase from Caldariomyces
fumago against aromatic compounds.
Aromatic compound Specific activity (min"1)

9-Methylanthracene 758 (± 27)
Azulene 676 (+ 34)
Anthracene 134 (±14)
2-methylanthracene 107 (±8)
7,12-Dimethylbenzanthracene 87 (± 8)
Benzo[a]pyrene 84 (± 6)
7-Methylbenzo [ajpyrene 81 (±7)
Acenaphthene 65 (± 8)
Pyrene 53 (± 6)
Benzo[ghi]perylene 45 (+ 7)
Perylene 25 (± 10)
Biphenylene 10 (± 0.5)
Phenanthrene 7 (±0.1)
Fluoranthene 3 (± 0.2)
Fluorene 1.9 (±0.13)
Triphenylene 0.8 (± 0.09)
Naphtalene 0.6 (±0.01)
Biphenyl NR
Dibenzofuran NR
Anthrone NR
NR: no reaction detected.
Chloroperoxidase was able to chlorinate 17 of 20 aromatic hydrocarbons

assayed in the presence of hydrogen peroxide and chloride ions [46] (Table 14).
Reaction rates varied from 0.6 min"1 for naphthalene to 758 min"1 for 9-
methylanthracene. Mono-, di- and tri-chlorinated compounds were obtained
from the chloroperoxidase-mediated reaction on aromatic compounds.
Chlorination of aromatic hydrocarbons could be interesting for the production of
fine chemicals from petroleum products.
Figure 12 shows the correlation between IP values and specific activity
for PAHs. Because ionization potential could be defined as the energy involved
in taking out one electron from the substrate molecule, this correlation suggest a
one-electron mechanism with a free radical-mediated reaction. Only PAH's with
IP lower than 8.52 eV were halogenated (Table 14). In general, the lower the IP
of the PAH, the higher the specific activity of the chloroperoxidase for that
substrate. The IP value of 8.52 eV appears to be a threshold, as none of the
compounds tested having higher ionization potentials were transformed by
chloroperoxidase [45]. This threshold value is significantly higher than those
97
reported for other peroxidases. Lignin peroxidase is able to oxidize PAH's and
form quinones up to a PAH's IP of 8.0 eV [85] and manganese peroxidase from
P. chrysosporium shows a threshold value for PAH's substrates of 8.1 eV [98].
4.8. Laccase
Laccases (EC 1.10.3.2) are copper-containing enzymes widespread in
white rot fungi which catalyze the oxidation of a variety of aromatic phenols and
anilines, reducing oxygen to water. Their characteristics have been
comprehensively reviewed [148, 149]. While the substrate range for laccase is
normally limited to phenolic substrates, it can be extended to nonphenolic
compounds with the addition of mediating substrates such as ABTS and HBT
[150-155]. In vitro oxidation of PAH's has been demonstrated by purified fungal
laccases [156-160]. The rate of oxidation of several PAH's has been shown to be
enhanced by the addition of the cooxidant ABTS [158-161]. Purified laccase of
C. gallica transformed 7 of 10 PAHs examined in the presence of ABTS (Table
15). Benzo[a]pyrene, 9-methylanthracene, 2-methylanthracene, anthracene,
biphenylene, acenaphthene, and phenathrene were oxidized by laccase [160].
The synthetic or natural mediating substances acts as free-radical mediators.
These mediators are sbstrates for laccase and tranformed into free radicals by
one electron subtraction, and then they diffuse and oxidize the aromatic
compound prducing, as peroxidases, mainly quinones. Unlike peroxidases, no
clear relationship between the substrate ionization potential and first-order rate
constant could be detected.
Fig. 12. Influence of the PAH ionization potential on the chloroperoxidase activity.
98
Table 15
First rate constants of reactions of C. Gallica laccase with polycyclic aromatic
hydrocarbons [160].
PAH Rate constant Ionization potential

(h"1) (eV)
9-Methylanthracene 240 7.23
Benzo[a]pyrene 83 7.12
Acenaphthene 10 7.7
Anthracene 5.2 7.55
2-Methylanthracene 4.9 7.42
Biphenylene 3.8 7.58
Phenanthrene 0.8 8.03
Pyrene NR 7.72
Fluoranthene NR 7.76
Azulene NEO 7.43
NR, no reaction detected.
NEO, nonenzymatic oxidation.
The effects of mediating substances were examined by using anthracene

as the substrate (Fig. 13). The presence of 1 mM of hydroxybenzotriazine
(HBT) induced an oxidation rate of anthrecene of 2.4 h"1, while 1 mM of ABTS
showed a rate constant of 5.2 h"1, but the 1 mM ABTS plus lmM HBT
increased the oxidation rate to 45 h"1, nine fold compared with the oxidation
rate in the presence only of ABTS
Fig. 13. Effects of the free-radical mediators HBT and ABTS on the anthracene oxidation by
purified laccase from C. gallica [160].
99
5. ENZYMATIC C-C BOND ACTIVATION
Short-chain alkanes such as methane, ethane and propane are the most
abundant and cheapest hydrocarbons available. Nevertheless, they are not
used directly as raw material for production of more valuable products,
mainly because C-H bonds in alkanes are not easily activated. Thus, alkanes
are used mainly as fuels to produce energy. However, the direct alkane
activation to produce valuable petrochemicals would exploit an inexpensive
hydrocarbon feedstock. Based on commercial and process viability, some of
the most promising routes for direct alkane activation have been identified
[162]. Table 18 shows some of the potential routes to produce important
petrochemicals as well as the conventional industrial feedstocks and currently
used processes. It has been estimated that the alternative technologies could
represent cost savings of up to USD 380/metric ton over the conventional
processes [162]. Efficient production of petrochemicals by direct activation of
alkanes remains a challenge. Particularly, oxidation of alkanes into useful
products is one of the major issues in catalysis research. Yields must be kept
low when using metal-based catalysts in order to keep selectivity [163].
Besides, as products are more reactive than substrates, subsequent oxidation
of partially oxidized alkanes leads to undesirable or low-value products.
There are several reports in the literature regarding the transformation
of saturated hydrocarbons by microorganisms. There have been found
microorganisms able to mineralize or degrade Q to C44 alkanes. Table 19 lists
the enzymes identified to catalyze the most common transformation, usually
an hydroxylation, of different alkanes. This section will briefly describe
substrate specificity, activities and limitations of the most representative
enzymatic systems for alkane oxidation.
Table 18
Processes for production of basic and intermediate petrochemicals [162].
Potential Alkane Usual Conventional Process Product

Feedstock Feedstock
Methane Reforming Methanol
Methane Hydrocarbons Steam cracking Ethylene
Methanol Oxidation/dehydrogenation Formaldehyde
Ethylene Oxychlorination Vinyl chloride
Ethane Methanol Carbonylation Acetic acid
Ethylene Oxidation Acetaldehyde
Ethylene Oxidation Ethylene oxide
Propylene Ammoxidation Acrylonitrile
Propane Propylene Oxidation Acrylic acid
Propylene Chlorhydrination, epoxidation Propylene oxide
100
5.1. Methane monooxygenase

The methane monooxygenase (MMO) system is expressed in micro-
organisms able to use methane as an energy source (methanotrophs) (see
chapter 6). These gram-negative microbes are able to use methane as energy
and carbon source. However they are not able to grow on larger alkanes. Two
forms of the enzymatic system have been described: a membrane associated,
or particulate, methane monooxygenase (pMMO) and a soluble cytoplasmatic
methane monooxygenase (sMMO) [164]. While pMMO is a membrane
protein produced by all known methanotrophs, sMMO is expressed only by a
subset of them. Moreover, sMMO shows a wide range of substrate specificity,
an it is able to catalyze the oxidation of alkenes, aromatic, alicyclic and
heterocyclic compounds, whereas pMMO only mediates the oxidation of a
small group of short-chain alkanes and alkenes [164].
pMMO and sMMO are evolutionarily unrelated. It has been
demonstrated that pMMO is an iron copper protein, which is produced only
under conditions of copper sufficiency; regarding its mode of action, it has
been related to ammonia monooxygenases [165]. sMMO, on the other hand,
is an iron-containing enzyme produced only under copper-limiting conditions.
It has been suggested that sMMO may provide a competitive advantage in
copper-depleted sites, enabling the methanotrophs to colonize a wider range
of environments. sMMO is comprised by three components: an oxygenase, a
reductase and a coupling protein [166]. This system has been extensively
characterized. The NADH-dependent oxidation reaction catalyzed by sMMO
is depicted in Figure 14. So far pMMO has been poorly characterized, mainly
due to its unstability to dioxygen exposure in cell-free fractions. Purification
procedure is usually long and require strict anaerobic conditions in order to
maintain its activity [167, 168]. Furthermore, most purified pMMO
preparations show low activity; in vitro rates represent only about 1-5% of
physiological rates.
Table 19
Enzymes able to catalyze the oxidation of alkanes.
Enzyme (EC number) Paraffinic substrate Active site Ref.
p-Methane monooxygenase Methane, Ci to C5 linear Trimeric copper 164

(1.14.13.25) alkanes cluster
s-Methane monooxygenase Methane, C5 to C7 linear and Diiron cluster 164
(1.14.13.25) branched alkanes
Alkane hydroxylase C5 to C24 alkanes Diiron cluster 169-171
(1.14.15.3)
P-450 monooxygenase Cyclohexane, C5 to C§ alkanes Heme 172
(1.14.14.1)
101
Fig. 14. Steps involved in the oxidation reaction catalyzed by alkane hydroxylase and
methane monooxygenase (A) and cytochrome P450 (B).
5.2. Alkane hydroxylase

The alkane hydroxylase system (Alk) present in Burkholderia cepacia
and Pseudomonas, Acinetobacter and Rhodococcus strains is a three-
component monooxygenase, comprising an hydroxylase, a rubredoxin and a
rubredoxin reductase [173]. The hydroxylase component is membrane-bound,
while the rubredoxin and rubredoxin reductase components are soluble,
cytoplasmic proteins. This enzyme is able to oxidize medium and long-chain
linear alkanes using reducing equivalents from NADH or NADPH as shown
in Fig. 15. The most studied system is the enzyme from Pseudomonas putida
Gpol, which is able to oxidize C5 to C12 alkanes [169]. Enzymes from other
organisms are able to oxidize larger alkanes, such as the enzyme from
Acinetobacter sp. strain ADP1 (C]2 to C]8) [174], the enzyme from
Rhodococcus sp. (C]2 to C16) [175], the two enzymes from Acinetobacter sp
strain M-l (C]6 to C22 and >C22) [176] and the two enzymes from
Pseudomonas aeuroginosa PAO1 (C12 to C20 and Ci5 to C24) [171].
Apparently, microorganisms have enzymes with different specificities that are
expressed depending on the available substrate.
Even though the organisms producing both sMMO and pMMO can
catalyze the oxidation of several alkanes in addition to methane, they are
unable to grow on any of them. On the other hand, organisms producing Alk
use medium-chain alkanes as energy and carbon source. The alkane is
oxidized to alcohol by Alk, while further oxidation to aldehyde and
carboxylic acid is catalyzed by different enzymes. The carboxylic acid then
enters the fatty acid degradation pathway and is used as an energy source. In
order to make use of the Alk system for the production of oxidized
intermediates, the metabolic reactions must be interrupted, such that the
desired product accumulates.
102
Table 20
Relative activities of alkane hydroxylase for the oxidation of
medium-chain alkanes [177].
Substrate Major product R elative rate (k substrate/ k n-octane)
n-Hexane None 0
n-Heptane n-Heptanol 0. 58
n-Octane n-Octanol 1
n-Nonane n-Nonanol 0.83
n-Decane n-Decanol 0. 16
n-Undecane n-Undecanol < 0.05
n-Dodecane None 0
An example of these has been demonstrated by Bosetti and coworkers

[177]. A plasmid containing the three components of the Alk systems was
constructed and introduced to a Pseudomonas strain lacking the alcohol
dehydrogenase. This enzyme catalyzes the second step on the metabolism of
alkanes, that is, the oxidation of alcohol to aldehyde. As a result, the
reeombinant bacteria was able to oxidize alkanes to alcohols, without
oxidizing them further. As the reeombinant bacteria cannot use the alkane for
growth, a carbon source must be supplied to the microorganism. Following
this strategy, the bacteria was able to transform C7 to Cn alkanes to their
corresponding alcohols (see Table 20).
5.3. Cytochrome P450

Another type of enzymes capable of oxidizing alkanes belong to the
cytochrome P450 family. These enzymes, unlike MMO and Alk, are heme
proteins. They catalyze the oxidation of substrates in the presence of
NAD(P)H and usually the system consists of two components: an hydroxylase
and a reductase. Cytochrome P450 present in yeast are able to convert alkanes
to oxidized products. In particular, Candida sp. are able to convert >Ci2
alkanes to a,co-dicarboxylic acids that are secreted to the medium. The first
reaction is the ©-oxidation of the alkane to the corresponding alcohol and it is
catalyzed by a cytochrome P450. Further oxidation to the acid is catalyzed by
a fatty alcohol oxidase and a fatty aldehyde deshydrogenase. The fatty acid is
oxidized again by the same enzymes, to produce the diacid [178]. However,
these cytochromes P450 are usually membrane-bound and have a
multicomponent nature, which makes them difficult to produce in large
quantities using reeombinant techniques.
103
Table 21
Alkane hydroxylation by different enzymatic systems [172]
Enzymatic system Substrate Maximum rate (min"1)

P450 BM-3 139-3 Hexane 3800
P450BM-3 Hexane 182
Alk Octane 210
sMMO Methane 222
The most promising example in this family is the unspecific bacterial

cytochrome P450 BM-3 from Bacillus megaterium. This enzyme is better
suited for a potential industrial application as it has several advantages over
other enzymes of the kind, such as being a soluble (not membrane-bound)
single polypeptide chain that is readily expressed in E. coli [179]. A very
attractive characteristic of cytochrome P450 BM-3 is that, unlike MMO, Alk
or other cytochromes P450, the hydroxylase and reductase domain are
comprised in the same polypetide.
The natural activity of the enzyme is to catalyze the hydroxylation of
C12 to C18 fatty acids with the concomitant consumption of dioxygen and
NADPH. This enzyme has been engineered by Arnold and coworkers using
laboratory evolution techniques to produce mutant 139-3, which acquired the
capability of catalyzing the oxidation of medium-chain alkanes [178]. This
mutant is the fastest known enzyme for alkane hydroxylation, as it is more
than seventeen times faster than the MMO or Alk enzymatic systems already
described. Table 21 shows the activity of mutant 139-3 compared with other
systems for the catalytic oxidation of alkanes. Moreover, several mutants have
been obtained from 139-3 mutant that catalyze the regio- and enantioselective
production of alcohols from smaller alkanes. Some of the mutants are from
1.1 to 4.5 faster than 139-3, when as shown on Table 22.
6. OPORTUNITIES AND CHALLENGES
In this century all industries, including the petroleum industry, should apply
energetically efficient production processes with reduced environmental
impact: this is their main challenge. In addition to the expected improvements
of conventional processes, the use of new and non conventional techniques
for petroleum refining should be evaluated. Doubtless, biotechnology is
among the non conventional techniques to be explored. Enzymatic catalysis
with high transformation efficiency, high specificity and mild reaction
conditions offers a wide range of possibilities. The analysis of the available
data on microbial and enzymatic transformations of oil products shows
104
several opportunities for some sectors of the petroleum industry, such as deep
desulfurization and denitrogenation, and asphaltene upgrading.
The powerful tools of molecular biochemistry can be used to improve
the enzyme stability and efficiency. These techniques may be applied to the
particular needs of the petroleum industry. Protein engineering is generally
understood as the use of site-directed or random mutagenesis to alter the
properties of a protein or enzyme. In addition, the enzymes isolated from
extremophilic microorganisms are extremely thermostable and generally
resistant to non conventional conditions such as organic solvents and extreme
pH. Thus, many enzymes and enzymatic proteins are still to be discovered. In
addition, over the past two decades people have seen many examples of the
improvement of biocatalysts by chemical and genetic techniques.
Still, there is not any enzymatic process ready to by applied in the
petroleum refining industry, and three main research fields may be suggested
to obtain an enzyme catalysts to be used in the petroleum industry:
1) The search of new enzymatic activities upon petroleum products,
specially from extreme environments. New microorganisms are currently
discovered from extreme environments such as thermal vents in the ocean
deep and fossilized salt rocks. The enzymes isolated from extremophilic
microorganisms are extremely thermostable and generally resistant to organic
solvents and extreme pH. Enzymes from these microorganisms working in
non-aqueous systems at temperatures higher as 200°C (operating conditions
found in refineries) could be expected. Moreover at high temperatures, the
hydrocarbons bioavailability and solubility is increased. All these unknown
organisms are a potential source of new enzyme forms with different
physicochemical properties: the potential source of biocatalytic activity for
the oil industry could be there.
2) The improvement of the enzymatic activities in very low water
systems, in order to increase the transformation rates using petroleum
fractions without further addition of water. Since petroleum is a hydrophobic
material, it is suitable to speculate that new enzymatic processes for the oil
industry should be carried out in non-aqueous systems. The use of reaction
mixtures containing organic solvents reduces mass transfer limitations,
promoting the establishment of productive interactions between the enzyme
and the hydrophobic substrates (oil-derived compounds). In addition, a
biocatalyst placed in a non-aqueous medium shows interesting properties,
such as improved thermostability, higher substrate accessibility, adjustable
selectivity, and high storage stability. The study of the relationship between
the solvent properties and the enzyme activity seems to be essential to
understand and to improve the biocatalytic processes.
105
Table 22
Improvement of cytochrome P450 BM-3 for the catalysis of small alkanes oxidation [179]
Enzyme mutant Substrate Major product Relative rate (k mutam/ k i39.3)

Mutant 139-3 Propane n-propanol 1
Octane 2-octanol 1
Mutant J Propane n-propanol 2.5
Octane 2-octanol 1.37
Mutant 9- 10A Propane n-propanol 1.91
Mutant 9-10A-A82L Propane n-propanol 4.5
3) Finally, the enzyme design by genetic and chemical methods. The

economical and technical feasibility of a large scale enzymatic process
depends mainly on the activity and stability of the biocatalyst under the actual
conditions found in the petroleum refining industry. Molecular biochemistry
efforts should be directed to improve enzyme activity and stability in
petroleum fractions with almost no addition of water and at the temperatures
found usually in refineries.
REFERENCES
[I] R.A. Kerr, Science 281 (1998) 1128.

[2] C.J. Campbell and J.H. Laherrere, Sci. Am. 278 (1998) 78.
[3] W.C. Lai and K.J. Smith, Fuel 80 (2001) 1121.
[4] D.S. Scott, D. Radlein, J. Piskorz, P. Majerski and T.J.W. DeBruijn, Fuel 80 (2001)
1087.
[5] M.N. Dadashev and G.V. Stepanov, Chem. Technol. Fuels Oils 36 (2000) 8.
[6] S.J. Park, C.J. Kim and B.S. Rhee, Ind. Eng. Chem. Res. 39 (2000) 4897.
[7] R. Vazquez-Duhalt, E. Torres, B. Valderrama and S. Le Borgne S., Energy Fuel 16
(2002) 1239.
[8] J.D. Rozzell, Bioorg. Med. Chem. 7 (1999) 2253.
[9] B.L. McFarland, D.J. Boron, W. Deever, J.A. Meyer, A.R. Johnson and R.M. Atlas,
Crit. Rev. Microbiol. 24 (1998) 99.
[10] A. M. Klibanov, Nature 409 (2001) 241.
[II] A. Schmid, J.S. Dordick, B. Hauer, A. Kiener, M. Wubblolts and B. Witholt, Nature
409 (2001) 258.
[12] J. S. Dordick, Enzyme Microb. Technol. 11(1989) 192.
[13] J. S. Dordick, Y. L. Khmelnitsky and M. Sergeeva, Curr. Opin. Microbiol. 1 (1998)
311.
[14] A. Zaks and A.M. Klibanov, Science 224 (1984) 1249.
[15] S. Akanuma, A. Yamagishi, N. Tanaka, and T. Oshima, Protein Sci. 7 (1998) 698.
106
[16] L. Giver, A. Gershenson, P.O. Freskgard and F.H. Arnold, Proc. Natl. Acad. Sci.
USA 95 (1998) 12809.
[17] H. Zhao and F.H. Arnold, Protein Eng. 12 (1999) 47.
[18] J. R. Cherry, M.H. Lamsa, P. Schneider, J. Vind, A. Sevendsen, A. Jones and A.H.
Pedersen, Nat. Biotechnol. 17 (1999) 379.
[19] A. Aksoy, H. Tumturk and N. Hasirci, J Biotechnol. 60 (1998) 37.
[20] P. Wang, M.V. Sergeeva, L. Lim and J.S. Dordick, Nat. Biotechnol. 15 (1997) 789.
[21] C.P. Govardhan, Curr Opin. Biotechnol. 10 (1999) 331.
[22] N.L. St Clair and M.A. Navia, J. Am. Chem. Soc. 114 (1992) 7314.
[23] S.S. Wong and C.L.J. Wong, Enzyme Microb. Technol. 14 (1992) 866.
[24] M. Ayala, E. Horjales, M.A. Pickard and R. Vazquez-Duhalt, Biochem. Biophys.
Res. Comm. 295 (2002) 828.
[25] J.P. Lenders and R.R. Crichton, Biotechnol. Bioeng. 26 (1984) 1343.
[26] R.A.K. Srivastava, Enzyme Microb. Technol. 13 (1991) 164.
[27] H.F. Gaertner and A.J. Puigserver, Enzyme Microb. Technol. 14 (1992) 150.
[28] D. Garcia, F. Ortega and J.L. Marty, Biotechnol. Appl. Biochem. 27 (1998) 49.
[29] M.J. Hernaiz, J.M. Sanchez-Montero and J.V. Sinisterra, Enzyme Microb. Technol.
24(1999)181.
[30] H. Garcia-Arellano, B. Valderrama, G. Saab-Rincon and R. Vazquez-Duhalt,
Bioconjugate Chem. 13 (2000) 1336.
[31 ] EPA (2000) EPA420-F-00-057
[32] A.M. Aitani, M.F. Ali and H.H. Al-AIi, Petrol. Sci. Technol. 18 (2000) 537.
[33] L. Linguist and M. Pachecho, Oil Gas J. Feb 22 (1999) 46.
[34] M.A. Kertesz and C. Wietek, Appl. Microbiol. Biotechnol. 57 (2001) 460.
[35] J. Klein, D.E.A. Catcheside, R. Fakoussa, L. Gazso, W. Fritsche, M. Hofer, F.
Laborda, I. Margarit, H.J. Rehm, M. Reich-Walber, W. Sand, S. Schacht, H.
Schmiers, L Setti and A. Steinbuchel, Appl. Microbiol. Biotechnol. 52 (1999) 2.
[36] W.M. Coco, W.E. Levinson, M.J. Crist, H.J. Hektor, A. Darzins, P.T. Pienkos, C.H.
Squires ad D.J. Monticello, Nat. Biotechnol. 19 (2001) 354.
[37] M.J. Grossman, M.K. Lee, R.C. Prince, K.K. Garrett, G.N. George and I.J. Pickering.
Appl. Environ. Microbiol. 65 (1999) 181.
[38] M.J. Grossman, M.K. Lee, R.C. Prince, V. Minak-Bernero, G.N. George and I.J.
Pickering, Appl. Environ. Microbiol. 67 (2001) 1949.
[39] M. Naito, T. Kawamoto, K. Fujino, M. Kobayashi, K. Maruhashi and A. Tanaka,
Appl. Microbiol. Biotechnol. 55 (2001) 374.
[40] J.A. Rupley and G. Carreri, Adv. Protein Chem. 41 (1991) 37.
[41] A.M. Klibanov, Trends Biotechnol. 15 (1997) 97.
[42] P.J. Hailing, Curr. Opin. Chem. Biol. 4 (2000) 74.
[43] M. Tuena de Gomez-Poyou and A. Gomez-Poyou, Crit. Rev. Biochem. Mol. Biol. 33
(1998)53.
[44] M. Ayala, R. Tinoco, V. Hernandez, P. Bremauntz and R. Vazquez-Duhalt, Fuel
Process Technol. 57 (1998) 101.
[45] M. Ayala, N.R. Robledo, A. Lopez-Munguia and R. Vazquez-Duhalt, Environ. Sci.
Technol. 34 (2000) 2804.
[46] R. Vazquez-Duhalt, M. Ayala and F.J. Marquez-Rocha, Phytochemistry 58 (2001)
929.
[47] M.P.J. van Deurzen, F. van Rantwijk and R.A. Sheldon, Tetrahedron 53 (1997)
13183.
107
[48] S. Colonna, N. Gaggero, C. Richelmi and P. Pasta, Trends Biotechnol. 17 (1999)

163.
[49] J.H. Dawson and M. Sono, Chem. Rev. 87 (1987) 1255.
[50] F. van de Velde, F. van Rantwijk and R.A. Sheldon, Trends Biotechnol. 19 (2001)
73.
[51] W. Tischer and V. Kasche, Trends Biotechnol. 17 (1999) 326.
[52] F. Secundo, S. Spadaro, G. Carrea and P.L.A. Overbeeke, Biotechnol. Bioeng. 62
(1999) 554.
[53] R. Jaenicke, J. Biotechnol. 79 (2000) 193.
[54] Y.L. Khmelnitsky and J.O. Rich, Curr. Opin. Chem. Biol. 3 (1999) 47.
[55] J.L. Schmitke, C.R. Wescott and A.M. Klibanov, J. Am. Chem. Soc. 118 (1996)
3360.
[56] E. Torres, R. Tinoco and R. Vazquez-Duhalt, J. Biotechnol. 49 (1996) 59.
[57] M.P.J. van Deurzen, I.J. Remkes, F. van Rantwijk and R.A. Sheldon, J. Mol. Cat. A
Chemical 117(1997)329.
[58] B. Valderrama, M. Ayala and R. Vazquez-Duhalt, Chem. Biol. 9 (2002) 555.
[59] Government of Alberta, Canada, 2002. Department of Energy. (http://www.energy.
gov.ab.ca).
[60] J.W. Bunger and N.C. Li (eds.), Chemistry of Asphaltenes, American Society for
Advanced Chemistry Series 195, 1981.
[61] J.G. Speight, The Chemistry and Technology of Petroleum, Marcel Dekker Inc., New
York, 1998, pp. 412-467.
[62] O.P. Strausz, T.W. Mojelsky and E.M. Lown, Fuel 71 (1992) 1355.
[63] J.C. Bertrand, E. Rambeloarisoa, J.F. Rontani, G. Giusti and G. Mattei, Biotechnol.
Lett. 5(1983)567.
[64] J.F. Rontani, F. Bosser-Joulak, E. Rambeloarisoa, J.C. Bertrand, G. Giusti and R.
Faure, Chemosphere 14 (1985) 1413.
[65] R.C. Wyndham and J.W. Costerton, Appl. Environ. Microbiol. 41 (1981) 791.
[66] E. Premuzic, M.S. Lin, M. Bohenek and W.M. Zhou, Energy Fuels 13 (1999) 297.
[67] G. Thouand, P. Bauda, J. Oudot, G. Kirsh, C. Sutton and J.F. Vidalie, Can. J.
Microbiol. 45 (1999) 106.
[68] D.J. Lacotte, G. Mille, M. Acquaviva and J.C. Bertrand, Chemosphere 32 (1996)
1755.
[69] P.M. Fedorak, K.M. Semple, R. Vazquez-Duhalt and D.W.S. Westlake, Enzyme
Microb. Technol. 15 (1993) 429.
[70] L. Mogollon, R. Rodriguez, W. Larrota, C. Ortiz and R. Torres, Appl. Biochem.
Biotechnol. 70-72 (1998) 765.
[71] H. Garcia-Arellano, E. Buenrostro-Gonzalez and R. Vazquez-Duhalt, Biotechnol.
Bioeng. (2004) (In press)
[72] H. Groenzin and O.C. Mullins, J. Phys. Chem. A 103 (1999) 11237.
[73] H. Groenzin and O.C. Mullins, Energy Fuels 14 (2000) 677.
[74] E. Buenrostro-Gonzalez, S.I. Andersen, J.A. Garcia-Martinez and C. Lira-Galeana,
Energy Fuels 16 (2002) 732.
[75] H. Habe and T. Omori, Biosci. Biotechnol. Biochem. 67 (2003) 225.
[76] R.M. Atlas, Microbiol. Rev. 45 (1981) 180.
[77] R.M. Atlas, Petroleum microbiology, MacMillan Publishing Co., New York,1984.
[78] R.R. Colwell and J.D. Walker, Crit. Rev. Microbiol. 5 (1977) 423.
[79] T.D. Gibson, Science 161 (1968) 1093.
108
[80] R.L. Farrell, K.E. Murtagh, M. Tien, M.D. Mozuch and T.K. Kirk, Enzyme Microb.
Technol. 11(1989)322.
[81] D. Sanglard, M.S.A. Leisola and A. Fiechter, Enzyme Microb. Technol. 8 (1986) 209.
[82] K.E. Hammel, B. Kalyanaraman and T.K. Kirk, J. Biol. Chem. 261 (1986) 16952.
[83] J.A. Field, R.H. Vledder, J.G. van Zelst and W.H.Rulkens, Enzyme Microb. Technol.
18(1996)300.
[84] S.D. Haemmerli, M.S.A. Leisola, D. Sanglard and A. Fiechter, J. Biol. Chem. 261
(1986)6900.
[85] R. Vazquez-Duhalt, D.W.S. Westlake and P.M. Fedorak, Appl. Environ. Microbiol. 60
(1994) 459.
[86] M. Tatarko and J.A. Bumpus, Lett. Appl. Microbiol. 17 (1993) 20.
[87] K.E. Hammel and PJ. Tradone, Biochemistry 27 (1988) 6563.
[88] R. Vazquez-Duhalt, D.W.S. Westlake, and P.M. Fedorak, Appl. Microbiol.
Biotechnol.42(1995)675.
[89] M.H. Gold, J.K. Glenn and M. Alic, Methods Enzymol. 161 (1988) 74.
[90] C. Cripps, J.A. Bumpus and S.D. Aust, Appl. Environ. Microbiol. 56 (1990) 1114.
[91] E.A. Pease and M. Tien, J. Bacteriol. 174 (1992) 3532 .
[92] M.J.J. Kotterman, R. A. Wasseveld and J.A. Field, Appl. Environ. Microbiol. 62
(1996) 880.
[93] J.A. Field, E. de Jong, G. Feijoo-Costa and J.A.M. de Bont, Trends Biotechnol. 11
(1993)44.
[94] J.L. Popp and T.K. Kirk, Arch. Biochem. Biophys. 288 (1991) 145.
[95] M.A. Moen and K.E. Hammel, Appl. Environ. Microbiol. 60 (1994) 1956.
[96] B.W. Bogan, R.T. Lamar and K.E. Hammel, Appl. Environ. Microbiol. 62 (1996)
1788.
[97] B. W. Bogan, B. Schoenike, R.T. Lamar and D. Cullen, Appl. Environ. Microbiol. 62
(1996)2381.
[98] B.W. Bogan and R.T. Lamar, Appl. Environ. Microbiol. 61 (1995) 2631.
[99] M.J.J. Kotterman, R.A. Wasseveld and J.A. Field, Appl. Environ. Microbiol. 62 (1996)
880.
[100] B.W. Bogan and R.T. Lamar, Appl. Environ. Microbiol. 62 (1996) 1597.
[101] M.J. Martinez, F.J. Ruiz-Duenas, F. Guillen and A.T. Martinez. Eur. J. Biochem. 237
(1996) 15412.
[102] T. Mester and J.A. Field, J. Biol. Chem. 273 (1998) 15412.
[103] A. Heinfling, M.J. Martinez, A.T. Martinez, M. Bergbauer and U. Szewzyk, FEMS
Microbiol. Lett. 165 (1998) 43.
[104] A. Heinfling, J. Ruiz-Duenas, M.J. Martinez, M. Bergbauer, U. Szewzyk, and A.T.
Martinez, FEBS Lett. 428 (1998) 141.
[105] F.J. Ruiz- Duefias, M.J. Martinez and A.T. Martinez, Mol. Microbiol. 31 (1999) 223.
[106] F.J. Ruiz- Duefias, M.J. Martinez and A.T. Martinez, Appl. Environ. Microbiol. 65
(1999)4705.
[107] S. Camarero, B. Bockle, M.J. Martinez and A.T. Martinez, Appl. Environ.
Microbiol. 62(1996)1070.
[108] S. Sarkar, A.T. Martinez and M.J. Martinez, Biochim. Biophys. Acta. 1339 (1997)
23.
[109] Y. Wang, R. Vazquez-Duhalt and M.A. Pickard, Can. J. Microbiol. 47 (2001) 277.
109
[110] Y. Wang, R. Vazquez-Duhalt and M.A. Pickard, Curr. Microbiol. 43 (2002) 77.
[111] S. Camarero, S. Sarkar, F.J. Ruiz-Duefias, M.J. Martinez and A.T. Martinez, J.
Biol. Chem. 274 (1999) 10324.
[112] M. Ayala, M.C. Baratto, R. Basosi, R. Vazquez-Duhalt and R. Pogni, J. Mol.
Catalysis B: Enzymatic 16 (2001) 159..
[113] Y. Wang, R. Vazquez-Duhalt and M.A. Pickard, Can. J. Microbiol. 49 (2003) 675.
[114] P.R. Ortiz de Montellano, Cytochrome P450, Structure, Mechanism and Biochemistry,
Plenum Press, New York, 1986.
[115] DJ. King, M.R. Azari and A. Wiseman, Xenobiotica 14 (1984) 187.
[116] M.R. Azari and A. Wiseman, Enzyme Microb. Technol. 4 (1982) 401.
[117] J. Capdevilla, R.W. Estabrook and R.A. Prough, Arch. Biochem. Biophys. 200 (1980)
186.
[118] W. Dehnen, R. Tomingas and J. Roos, Anal. Biochem. 53 (1973) 373.
[119] S.L. Kelly, D.C. Lamb, B.C. Baldwin and D.E. Kelly, Biochem Biophys. Res. Comm.
197(1993)428.
[120] S. Masaphy, D. Levanon, Y. Henis, K. Venkateswarlu and S.L. Kelly, Biotechnol.
Lett. 17(1995)969.
[121] A.P. Koley, J.T.M. Buters, R.C. Robinson, A. Markowitz and F.K. Friedman, Arch.
Biochem. Biophys. 336 (1996) 261.
[122] P. Anzenbacher, T. Niwa, L.M. Tolbert, S.R. Sirimanne and F.P. Guengerich,
Biochemistry 35 (1996) 2512.
[123] A.D. Rahimtula, P.J. O'Brien, H.E. Seifreid and D.M. Jerina, Eur. J. Biochem. 89
(1978) 133.
[124] M. Shou, K.W. Krausz, F.J. Gonzalez and H.V. Gelboin, Carcinogenesis 17 (1995)
2429.
[125] W.L. Backes, M. Hogaboom and WJ. Canady, J. Biol. Chem. 257 (1982) 4063.
[126] W.L. Backes, G. Cawley, C.S. Eyer, M. Means, K.M. Causey and WJ. Canady, Arch.
Biochem. Biophys. 304 (1993) 27.
[127] O-S. Li, J. Ogawa, R.D. Schmid and S. Shimizu, Appl. Environ. Microbiol. 67
(2001)5735.
[128] S.M. Fowler, P.A. England, A.C.G. Westlake, D.R. Rouch, D.P. Nickerson, C. Blunt,
D. Braybrook, S. West, L.L. Wong and S.L. Flitsch, J. Chem. Soc. Chem. Commun.
(1994)2761.
[129] S. Modi, W.U. Primrose, L.Y. Lian and G.C.K. Roberts, Biochem J. 310 (1995) 939.
[130] H. Joo, Z. Lin and F. H. Arnold, Nature 399 (1999) 670.
[131] C.F. Harford-Cross, A.B. Carmichael, F.K. Allan, P.A. England, D.A. Rouch and L-
L. Wong, Protein Eng. 13 (2000) 121.
[132] R. Vazquez-Duhalt, J. Mol. Cat. B: Enzymatic 7 (1999) 241.
[133] R. Vazquez-Duhalt, D.W.S. Westlake and P.M. Fedorak, Enzyme Microb. Technol. 15
(1993)494.
[134] N.L. Klyachko and A.M. Klibanov, Appl. Biochem. Biotechnol. 37 (1992) 53.
[135] E. Torres, J.V. Sandoval, F.I. Rosell, A.G. Mauk and R. Vazquez-Duhalt, Enzyme
[136] R. Akasaka, T. Mushino and M. Hirobe, J. Chem. Soc. Perkin Trans. 1 (1994) 1817.
[137] R. Tinoco and R. Vazquez-Duhalt, Enzyme Microb. Technol. 22 (1997) 8.
[138] JJ. Mieyal, R.S. Ackerman, J.L. Blumer and L.S. Freeman, J. Biol. Chem. 251 (1976)
3436.
[139] Y. Yoshida, K. Kashiba andE. Niki, Biochim. Biophys. Acta 1201 (1994) 165.
110
[140] J.C. Alvarez and P.R. Ortiz de Montellano, Biochemistry 31 (1992) 8373.
[141] G.L. Kedderis, D.E. Rickert, R.N. Pandey and P.F. Hollenberg, J. Biol. Chem. 261
(1986) 15910.
[142] C. Giulivi and K.J.A. Davies, J. Biol. Chem. 265 (1990) 19453.
[143] M. Ortiz-Leon, L. Velasco and R. Vazquez-Duhalt, Biochem. Biophys. Res. Comm.
215(1995)968.
[144] PJ. Thornalley, RJ. Trotta and A. Stern, Biochim. Biophys. Acta 759 (1983) 16.
[145] MJ. Davies, Biochem. Biophys. Acta 28 (1988) 28.
[146] M. Sundaramoorthy, J. Terner and T.L. Poulos, Structure 3 (1995) 1367.
[147] X. Yi, M. Mroczko, K.M. Manjol, X. Wang and L.P. Hager, Proc. Nat. Acad. Sci. 96
(1999) 12412.
[148] L. Gianfreda, F. Xu and J-M. Bollag. Bioremediation J. 3 (1999) 1.
[149] T.F. Thurston, Microbiology 140 (1994) 19.
[150] R. Bourbonnais and M. G. Paice, FEBS Lett. 267 (1990) 99.
[151] R. Bourbonnais, M. G. Paice, I. D. Reid, P. Lanthier and M. Yamaguchi, Appl.
[152] R. Bourbonnais, M. G. Paice, B. Freiermuth, E. Bodie and S. Borneman, Appl.
[153] R. Bourbonnais, D. Leech and M.G. Piace, Biochim. Biophys. Acta 1379 (1998)
381.
[154] K. Li, F. Xu and K.E.L. Eriksson, Appl. Environ. Microbiol. 65 (1999) 2654.
[155] F. Xu, J.J. Kulys, K. Duke, K. Li, K. Krikstopaitis, H.J.W. Deussen, E. Abbate, V.
Galinyte and P. Schneider, Appl. Environ. Microbiol. 66 (2000) 2052.
[156] S. Bohmer, K. Messner and E. Srebotnik, Biochem. Biophys. Res. Commun. 244
(1998)233.
[157] P.J. Collins, M. J. J. Kotterman, J. A. Fiel, and A. D. W. Dobson, Appl. Environ.
Microbiol. 62(1996)4563.
[158] C. Johannes, A. Majcherczyk and A. Huttermann. Appl. Microbiol Biotechnol. 46
(1996)313.
[159] A. Majcherczyk, C. Johannes and A. Huttermann, Enzyme Microb. Technol. 22
(1998)335.
[160] M.A. Pickard, R. Roman, R. Tinoco and R. Vazquez-Duhalt, Appl. Environ.
Microbiol. 65 (1999) 3805.
[161] C. Johannes and A. Majcherczyk, Appl. Environ. Microb. 66 (2000) 524.
[162] Chem Systems. Alkane Activation: Petrochemical Feedstocks of the Future, 1999.
[163] J.A. Labinger and J.E. Bercaw, Nature 417 (2002) 507.
[164] J.C. Murrell, B. Gilbert and I.R. McDonald, Arch. Microbiol. 173 (2000) 325.
[165] H.T. Nguyen. S.J. Elliot, J.H. Yip, S.L. Chan, J. Biol. Chem. 273 (1998) 7957.
[166] B.G. Fox, W.A. Froland, J.E. Dege and J.D. Lipscombs, J. Biol. Chem. 264 (1989)
10023.
[167] D.W, Choi, R.C. Kunz, E.S. Boyd, J.D. Semrau, W.E. Antholine, J.I. Han. J.A.
Zahn, J.M. Boyd, A.M. de la Mora and A.A. DiSpirito, J. Bacteriol. 185 (2003)
5755.
[168] P. Basu, B. Katterle, K.K. Andersson and H. Dalton, Biochem. J. 369 (2003) 417.
[169] A. Schmid, B. Sonnleitner and B. Witholt, Biotechnol. Bioeng. 60 (1998) 10.
[170] T.H.M. Smits, S.B. Balada, B. Witholt and J.B. van Beilen, J. Bacteriol. 184 (2002)
1733.
[171] N.N. Marin, L. Yuste, F. Rojo, J. Bacteriol. 185 (2003) 3232.
Ill
[172] M.W. Peters, P. Meinhold, A. Glieder and F.H. Arnold, J. Am. Chem. Soc. 125
(2003) 13442.
[173] J. Shanklin, C. Achim, H. Schmid, B.G. Fox and E. Miinck, Proc. Natl. Acad. Sci.
USA 94 (1997) 2981.
[174] A. Ratajczak, W. GeiBdorfer and W. Hillen, J. Bacteriol. 180 (1998) 5822.
[175] L.G. Whyte, T.H.M. Smits, D. Labbe, B. Witholt, C.W. Greer and J.B. van Beilen,
[176] A. Tani, T. Ishige, Y. Sakai and N. Kato, J. Bacteriol. 183 (2001) 1819.
[177] A. Bosetti, J.B. van Beilen, H. Preusting, R.G. Lageveen and B. Witholt, Enzyme
[178] D.L. Craft, K.M. Madduri, M. Eshoo and C.R. Wilson, Appl. Environ. Microbiol. 69
(2003) 5983.
[179] A. Glieder, E.T. Farinas and F.H. Arnold, Nat. Biotechnol. 20 (2002) 1135.
© 2004 Elsevier B .V. All rights reserved. 113
Chapter 4
Prospects for biological upgrading of heavy oils and

asphaltenes
K.M. Kirkwood", J.M. Foght", and M.R. Gray"
department of Chemical and Materials Engineering, University of Alberta,

b
Department of Biological Sciences, University of Alberta, Edmonton, Alberta,
Canada, T6G 2E9
1. INTRODUCTION
Increasing supply of heavy crude oils and bitumens, mainly from Canada,
Mexico and Venezuela, has increased the interest in transportation and
conversion of the high-molecular weight fractions of these materials into refined
fuels and petrochemicals. The high viscosity of these crudes requires addition of
a solvent in order to allow pipelining over a significant distance. The cost of
suitable solvents, such as naphtha or natural gas condensate, has led to study of
new methods to reduce the viscosity of heavy crudes. Once they enter a refinery,
processing of heavy crudes and bitumens requires conversion of the vacuum
residue components, including the asphaltenes, into distillable oils. This
upgrading has typically been accomplished with either thermal conversion
(cracking or coking) or by catalytic hydroconversion. Thermal processing can
range from mild cracking, to reduce viscosity, to severe cracking with attendant
formation of coke. These high-temperature processes require expensive
investment in process equipment and supporting infrastructure for supply of
hydrogen and treatment of hydrogen sulfide in cracked off-gases.
In contrast to the available processes, biological processing may offer less
severe process conditions and higher selectivity to specific reactions. This
chapter reviews the characteristics of the molecules in the vacuum residue
fraction of crude oils, and examines the prospects for using biological processes
to improve the value of these materials.
114
2. MOLECULES OF INTEREST
Heavy crude oils pose new upgrading challenges, in addition to the upgrading
needs common to lighter crudes. These problems are related to two types of high
molecular weight molecules present in these oils: waxes and asphaltenes. Waxes
are long-chain paraffinic molecules, or alkanes, which typically cause
operational problems if longer than 40 carbon atoms [3]. Asphaltenes, on the
other hand, are not classified by structure, but are defined as a solubility class,
including material that is soluble in toluene but not in «-pentane (or alternatively
«-heptane). There are two different views on the molecular structure of
asphaltenic material. The first represents asphaltenes as having a single large
condensed polycyclic aromatic core, with aliphatic chains attached on the
periphery (Fig. la) [1, 4-6]. This type of structure, however, does not account
for all of the physical and chemical properties of asphaltenes. The second
Fig. 1. Representative models of asphaltene molecules showing either (a) a single large
condensed polycyclic aromatic core [1] or (b) multiple smaller polycyclic aromatic cores with
aliphatic bridges [2].
115
representation describes asphaltenes as having multiple smaller polycyclic

aromatic cores (2-4 rings) linked by aliphatic bridges of varying lengths (Fig.
lb) [2, 7-10]. Sulfides, ethers, and esters have been identified as common
linking structures in the aliphatic bridges found in asphaltenes [11]. This type of
structure accounts for the observed reactivity of asphaltenes.
These large molecules are problematic for biological transformation.
Transformation rates are limited by the mass transfer of target molecules to the
biocatalyst and, in the case of whole cells, across the cell membrane (reviewed
in Ref. [12]). Interfacial mass transfer can be improved through emulsification,
increasing the interfacial contact area, however emulsification is of limited value
in overcoming the barrier of transport into biological cells unless appropriate
uptake mechanisms are available.
Despite these difficulties, there is evidence in the literature for bacterial
transformation of complex, high molecular weight substrates. Rhodococcus
erythropolis strain IGTS8, for example, was originally isolated from an
enrichment culture with the ability to use coal as its sole source of sulfur. This
mixed culture was able to remove over 90% of the organic sulfur from coal in a
continuous flow reactor [13]. This sulfur would have been covalently bound
within the coal matrix, primarily in thiophenic structures.
The matrix of vulcanized rubber consists of carbon chains crosslinked by
sulfide, disulfide, and polysulfide linkages. Bacterial attack appears to be limited
to sulfur exposed at the surface of solid rubber particles, which is oxidized to
sulfoxides and sulfones, and eventually released as sulfate [14].
The alkane-degrading bacterium R. erythropolis ATCC 13260 (originally
reported as Nocardioides simplex) is able to degrade a high molecular weight
fraction of crude oil [15]. This fraction contains 14.7% sulfur, and was known
from prior work to have hydrocarbon subunits linked by sulfide bridges. R.
erythropolis ATCC 13260 degraded sulfur-bound linear alkanes and steranes in
this oil fraction, leaving oxidized sulfur-bound species such as carboxylic acids.
Sulfur-specific oxidation to sulfones was also observed, but no carbon-sulfur
bond cleavage or desulfurization was reported [15].
In some reports, treatment of heavy crude oils with thermophilic bacteria
led to an apparent enrichment in the lighter fractions of the oil [16-18]. This
shift in composition was attributed to depolymerization of the asphaltene
fraction of the oils, which was defined as the dissociation of small molecules
either physically associated with or weakly chemically bound to asphaltenes. No
significant quantitative change in the asphaltene content was measured. In
addition, uniform removal of the range of sulfur compounds present in the oil
was reported, which is not consistent with known chemical or biological
conversion processes. Problems with sample recovery could account for some of
the observed changes in oil composition, however in the absence of appropriate
116
controls allowing complete material or sulfur balances, definite conclusions

cannot be drawn from this work.
Although these reports are mostly encouraging, and suggest that conversion
of high-molecular weight components of crude oil may be achieved by
biological means, practical experience with asphaltic materials suggests that
conversion rates may be very low. Asphalt (or bitumen in Europe) is widely
used in paving materials, building materials, and waterproofing of foundations
and roofs because of its resistance to degradation by natural organisms in soil
and water, and by photooxidation. In this chapter, we first consider the chemical
structure of the high-boiling components of crude oil, then examine the reactions
that would enhance the value of these materials and consider the evidence for
achieving such transformations by adapting natural biological processes.
3. UPGRADING NEEDS AND OPPORTUNITIES
The chemical goals of heavy oil upgrading encompass molecular weight

reduction of residue fractions to distillate materials, hydrogenation to increase
the hydrogen to carbon (H/C) ratio, and removal of heteroatoms, in particular
sulfur and nitrogen [19]. We will define the potential scope of biological oil
upgrading more broadly, to include all activities which make the material easier
to produce and transport, as well as the chemical changes which increase the
value of the oil. These activities could therefore be applied to in situ treatment,
production, transportation, and processing of crude oils. Five key areas of heavy
oil upgrading where biological treatment could have an impact are viscosity
reduction, composition improvement, deposition control, de-emulsification, and
naphthenic acids removal.
3.1. Viscosity reduction

Heavy oils are currently diluted with light hydrocarbons to reduce viscosity
and allow transportation by pipeline to processing facilities. Natural gas
condensate is typically used as the diluent, and is currently available as a steady
supply. Precipitation of asphaltenes in the pipeline can occur due to the aliphatic
nature of the diluent, but this approach readily achieves the viscosity reduction
needed and is generally accepted. The production of heavy oil is expected to
increase over the next several years and will exceed the availability of the
diluent, so an alternative or supplemental treatment will be required.
The viscosity of heavy oil is a result of interactions among the heaviest
molecules in the oil, the asphaltenes. These interactions include entanglement of
the alkane chains [19] as well as more ordered interactions between the aromatic
clusters leading to structure formation throughout the oil [20]. One potential
biotechnological approach to viscosity reduction is emulsification of the oil
117
using bioemulsifiers such as emulsan. This is discussed in further detail in

Chapter 9.
Breaking the asphaltenes into smaller molecules should also reduce
molecular interactions leading to a reduction in viscosity. Thermal processing
(mild thermal cracking, or visbreaking) can achieve some reduction in viscosity
by breaking up some of the aliphatic structures in the asphaltenes, but the
products can be unstable in downstream processing operations [19]. Microbial
cleavage of aliphatic sulfides to reduce molecular weight and viscosity is the
subject of research in our laboratory and this approach will be discussed further
later in this chapter.
3.2. Composition improvement

Many of the upgrading needs of traditional crude oils are also applicable to
heavy crude oils. These include removal of sulfur, nitrogen, and metals,
aromatic ring cleavage, and hydrogenation. Molecular weight reduction is also
required to improve the fractional composition and value of heavy crude oils.
3.2.1. Heteroatom removal

Sulfur, nitrogen, and metals present in crude oils are problematic for
refining operations since they are poisonous to the catalysts used. Sulfur and
nitrogen removal is also required to meet governmental emissions regulations
when the refined fuels are burned. The application of biotechnology to the
removal of these elements is discussed in Chapters 3 and 4.
3.2.2. Aromatic ring cleavage

The presence of aromatic hydrocarbons has adverse effects on production
and processing of petroleum, and combustion of fuels rich in aromatic
hydrocarbons contributes to soot formation and poor combustion characteristics
(for example, in diesel engines). Aromatics are commonly cracked during
conventional upgrading by high temperature, high pressure catalytic
hydrogenation to saturate and break the aromatic rings, but this is a costly
process in terms of operation and capital. A proposed biological alternative
would employ whole cell biocatalysts and two-phase (oil-water) reactions to
specifically oxidize one or more rings of the aromatic substrates present in crude
oil or middle distillate fractions. Enzymatic ring cleavage without carbon loss
would produce polar compounds soluble in the water phase [21-24]. These
would be recovered for chemical hydrogenation under mild conditions to yield
alkylaromatics with improved combustion characteristics compared to the parent
compounds. Alternatively, enzymatic hydroxylation of the aromatics with
subsequent chemical hydrogenation and hydrogenolysis in the aqueous phase
[25] would yield cycloalkylaromatics sensitive to further thermochemical bond
cleavage. It is proposed that the cost savings of conducting such processes under
118
near-ambient temperature and pressure would make biologically assisted

aromatic ring opening an economically feasible adjunct to conventional
upgrading technology. This potential treatment is reviewed in Chapter 5.
3.2.3. Hydrogenation
Typical H/C ratios for bitumens and residues range from 1.4-1.6 mol/mol.
Hydrogenation is required to increase the H/C ratio of these feeds to a level
suitable for transportation fuels (diesel and jet fuels, around 1.8 mol/mol) [19].
The primary target is the aromatics, including the heterocyclic sulfur and
nitrogen species. The use of microorganisms specifically for aromatic ring
hydrogenation has not been explored, although ring hydrogenation has been
observed in the biodegradation pathways of some aromatic compounds.
The explosive 2,4,6-trinitrotoluene (TNT) is subject to biotransformation in
a variety of anaerobic and aerobic bacterial systems, as well as fungal systems
(reviewed in Ref. [30]). In some aerobic bacteria, the initial reaction is
hydrogenation of the ring, forming hydride- and dihydride-Meisenheimer
complexes (Fig. 2a) [26, 27, 31]. Hydride-Meisenheimer complexes are
similarly formed in the biodegradation of picric acid (2,4,6-trinitrophenol) [32,
33]. For the better-characterized picric acid system, these reactions are catalyzed
by a hydride transferase enzyme, with NADPH serving as the hydride source via
reduced coenzyme F-420 [34, 35].
Fig. 2. Examples of hydrogenation reactions in the biodegradation of (a) TNT [26, 27] and (b)
naphthalene [28, 29].
119
Ring saturation is also observed in the biodegradation of aromatic

hydrocarbons under anaerobic conditions. In the metabolism of benzoate
through the benzoyl-CoA pathway (reviewed in Ref. [36]), stepwise saturation
of the ring precedes ring cleavage and mineralization. Under sulfate-reducing
conditions, naphthalene is activated by carboxylation to form 2-naphthoic acid
[29]. Before ring cleavage and mineralization, 2-naphthoic acid is hydrogenated
starting with the unsubstituted ring, eventually forming decahydro-2-naphthoic
acid (Fig. 2b) [28]. Water is used as the source of protons for these reactions.
Phenanthrene is similarly activated through carboxylation [29], but further
reaction steps have not been identified.
Neither of these systems has been studied for the specific goal of ring
hydrogenation. They are presented here to illustrate that this type of reaction
does occur. Unanswered questions in the existing literature are whether hydride
transferase enzymes exist that are active towards hydrocarbons, as opposed to
nitroaromatics, and whether the anaerobic hydrogenation of activated
naphthalene also occurs in larger or alkylated ring systems. The ability of
bacteria to transfer protons from water to aromatic hydrocarbon derivatives is a
sharp contrast to the hydrogen-using catalytic and thermal processes used in
traditional upgrading, and deserves further research attention.
3.2.4. Molecular weight reduction

Molecular weight reduction is required to convert the residue fraction of
heavy crude oils (materials boiling at temperatures over 525°C) to distillates
(boiling at temperatures under 525°C). Cracking of aliphatic C-S bonds
contributes to molecular weight reduction, but cracking of the C-C bonds found
in alkyl bridges is necessary to achieve the full reduction required. For primary
upgrading of heavy crude oils, thermal treatment is used for cracking operations,
using temperatures over 420°C. The usefulness of chemical catalysts in primary
upgrading is limited due to excessive catalyst fouling and poisoning [19].
Chemically, alkanes are the least reactive of the hydrocarbons.
Nevertheless, aerobic bacterial biodegradation of n-alkanes is well known [37],
and has been documented for chain lengths from Ci (methane) to at least C36
(hexatriacontane) [38]. Branched isoprenoid alkanes such as pristane (2,6,10,14-
tetramethylpentadecane) are also biodegradable [37]. The most common
mechanism involves activation of the alkane through addition of molecular
oxygen to the terminal methyl group by a monooxygenase enzyme to form a
primary alcohol (Fig. 3a). Subsequent oxidation to a carboxylic acid allows
further degradation through central fatty acid metabolic pathways. This method
of degradation of alkyl chains therefore requires a free methyl terminus, and C-
C bond cleavage only occurs two carbons away from the end of the molecule
through P-oxidation [37].
120
Fig. 3. Representative biodegradation pathways of alkanes showing initial activation and

transformation to a carboxylic acid, (a) Aerobic monooxygenation of H-dodecane [37]. (b)
Anaerobic addition ofn-dodecane to fumarate [39]. (c) Aerobic oxidation and cleavage of
cyclododecane [40].
Anaerobic bacteria do not have molecular oxygen available for activation

of non-functionalized hydrocarbons. Alternate activation and biodegradation
mechanisms have been the subject of intense research over the past 15 years (as
reviewed in Refs. [41-44]). Sulfate, nitrate, or iron-reducing bacteria may
activate hydrocarbons through carboxylation or by addition of a C-H from the
hydrocarbon across the double bond of fumarate to form a substituted succinate.
The latter reaction is well-established for toluene, and may also occur for
w-xylene, /?-isopropyltoluene, and ethylbenzene [43]. This reaction has also
been observed as the activation route for «-alkanes including C4-C8 n-alkanes
121
under nitrate-reducing conditions [45, 46] and w-dodecane [39] under sulfate-
reducing conditions. The addition of fumarate to alkanes does not occur at a
terminal methyl C-H, but rather at either a C2 or C3 subterminal methylene C-H
[39, 46], producing a branched dicarboxylic acid (Fig. 3b), which is degraded
through fatty acid metabolism. As with the aerobic pathway described above,
anaerobic alkane degradation therefore proceeds from the terminus of the
molecule.
A subterminal attack on long-chain «-alkanes may occur in some aerobic
bacterial cultures. A mutant Rhodococcus strain, designated KSM-B-3M,
accumulated c«-unsaturated metabolites of «-hexadecane, 1-chlorohexadecane,
and heptadecanonitrile, which were not growth substrates [47]. (The first two
compounds did support growth of the wild-type strain.) In all three compounds
the unsaturation was at position 9. Unsaturated products were also detected for
1-hexadecanol, 1,2-epoxyhexadecane, hexadecyl benzene, and hexadecyl
chloroformate. The mutant had likely lost the ability to cleave the alkane chain
at the unsaturated bond, resulting in the inability to grow on these substrates
[47].
The degradation of phytanyl octadecyl ether by a mixed soil culture and by
Rhodococcus ruber (DSMZ 7512) also showed evidence of an initial
subterminal dehydrogenation [48]. Degradation occurred initially on the linear
side chain, and initial degradation products were the phytanyl ethers of C2 to C8
primary alcohols. The corresponding carboxylic acids appeared next as the
alcohols disappeared from the cultures. The final products were the phytanyl
ethers of acetic acid and propanoic acid. One other metabolite was observed,
phytanyl octadec-9-enyl ether. The formation of unsaturated products was
proposed to be analogous to the dehydrogenation of Cig fatty acids in the cell
membrane, which also occurs at position 9. The alternate degradation pathway
proposed starts with the observed internal dehydrogenation, followed by a
hypothesized olefinic oxidation to a secondary alcohol, oxidation to a ketone,
Baeyer-Villiger oxidation to an ester, ester cleavage, and P-oxidation [48].
A subterminal attack of this type has not been shown for a diterminally
substituted alkyl chain, but a similar pathway has been shown as the mechanism
of degradation for both small and large cyclic alkanes. Both cyclohexane
(reviewed in Ref. [49]) and cyclododecane [40] are oxidized via an alcohol to a
cyclic ketone. The ketones are oxidized to lactones by Baeyer-Villiger
monooxygenases, followed by ester cleavage to an co-hydroxycarboxylic acid
(Fig. 3c) and oxidation to a dicarboxylic acid [40] that can be degraded through
central metabolic pathways. The Baeyer-Villiger monooxygenases appear to
have fairly narrow substrate specificities. Cyclododecanone monooxygenase
from R. ruber strain CD4 could also oxidize cyclopentadecanone, but not
cyclohexanone or cyclooctanone [40]. In growth assays, R. ruber strain SCI,
isolated on cyclododecanone, could also grow on Ci5, C13, C n , and C10 cyclic
122
ketones, but not on C8, C7, or C6 cyclic ketones [50]. Conversely, cyclohexanone
monooxygenases are known to favour shorter chain cyclic ketones. For example,
two enzymes from Brevibacterium sp. strain HCU could oxidize C4-C7 cyclic
ketones, but not C 8 -C| 2 compounds [51].
Molecular weight reduction in the residue fraction of heavy oils requires
cleavage of alkyl bridges, where both ends of the carbon chain are blocked by
attachment to aromatic groups. The more common aerobic and anaerobic
bacterial alkane-degradation pathways are not appropriate for molecular weight
reduction in crude oil, because they only activate the free end of the molecule to
create a fatty acid for central metabolic pathways. More relevant research has
been done with long-chain «-alkanes and cycloalkanes. This work shows that an
alkyl chain can be cleaved through bacterial attack in the absence of a terminal
methyl group. This reaction is more directly analogous to the alkyl bridges
found in high molecular weight crude oil components, and is a promising
avenue for further work.
3.3. Deposition control

Both asphaltenes and waxes may cause deposition problems in the
reservoir, pipelines, and storage and processing equipment. Asphaltenes deposit
due to an increase in the aliphatic content of the oil, while waxes crystallize and
precipitate due to a drop in temperature (e.g. from the reservoir to the surface,
[52]) or an increase in aromaticity of the bulk oil. Changes in solvency occur
due to dilution or to blending of different oils [19]. Both types of compounds
may co-precipitate, through entrapment of one type in a deposit of the other.
Generally, a wax content greater than 2% by weight is found to lead to wax
deposition problems [3]. Deposition prevention is accomplished through
chemical treatment to maintain the molecules in solution, as well as through
temperature and flow control. Waxes are a valuable feedstock for refinery
operations, so prevention of wax deposition is important to preserve the value of
the oil as well as to avoid operational problems. Existing deposits are removed
through circulation of hot water, hot oil, solvents, and surfactants, or through
"pigging" of transfer lines [53]. In the case of asphaltenes, treatments include
addition of aromatic streams to dissolve the deposits, or the addition of
dispersants to prevent flocculation of the asphaltenes into particles that
subsequently deposit on surfaces.
Although biological treatments for deposition control are commercially
available, little scientific literature is available in this area [54, 55]. Three modes
of biological activity are conceivably relevant to deposition control: production
of metabolites (from carbon sources other than the oil) which improve the
solubility of either waxes or asphaltenes, biotransformation of waxes and
asphaltenes to more soluble products (through molecular weight reduction or
functionalization), and biodegradation to remove the problematic compounds
123
either from the oil or from existing deposits. The ability of bacteria to degrade
solid alkanes is limited by mass transfer rates. For instance, liposome
encapsulation was required to achieve biodegradation of hexatriacontane (n-C^)
by a Pseudomonas isolate which did not grow on the crystalline compound [38].
The usefulness of biological treatments for removal of deposits may therefore be
limited to the production of solubilizing agents rather than direct transformation
or degradation of the crystallized molecules. Isolated bacteria and consortia
from paraffin deposits, hydrocarbon contaminated soils and waters, and brine
have been shown to produce biosurfactants, as well as to degrade hydrocarbons
from samples of paraffin deposits and paraffinic oils [56]. In a flow system, a
consortium of these bacteria decreased the paraffin content of a heavy oil. The
treated oil also had a lower freezing point and a decreased low temperature
viscosity, but the effect of these changes on deposition in the flow system was
not reported [56]. To the extent that microorganisms adsorb wax or asphaltic
material, then bacteria could serve to disperse the deposits and prevent
deposition on surfaces, however, no systematic research has been conducted in
this area.
3.4. Emulsion behaviour and de-emulsiflcation

Water-in-oil (W/O) and oil-in-water (O/W) emulsions occur throughout oil
production, transportation, and processing. The water may be from the
formation or may be added through water or steam injection to improve oil
recovery, or addition of wash water in desalting operations. Emulsions may be
produced incidentally through handling or deliberately to improve flow
properties for enhanced oil recovery and transportation [57]. Desirable
emulsions produced for pipeline transportation are O/W emulsions, usually
containing around 30% aqueous phase [58]. Undesirable O/W emulsions are
typically found in waste waters from the oil industry. Although de-
emulsification does recover some oil, treatment is generally driven by
environmental concerns rather than economic incentive. On the other hand,
resolution of W/O emulsions improves the quality of the oil, and is therefore
economically driven [53]. Problems associated with water in oil include
corrosion, scale formation, sludge accumulation in storage tanks, altered
viscosity and flow properties, and reduced distillation efficiency [53].
Regardless of the source, emulsions must be resolved at some point before
refining. This is accomplished through heating, settling, centrifugation,
filtration, electrical dehydration, and chemical treatment. The pipeline
specification includes both solids and water, and is typically a maximum of
0.5% bottom solids and water (BS&W) [58].
Emulsions are formed from two immiscible phases through mixing to
produce a fine dispersion of droplets of one phase in the other, where the
interface between the two phases is stabilized by emulsifying components. The
124
energy added through mixing is essential, since the emulsified state is not
usually thermodynamically stable. Emulsifying agents associate at the interface
of the two phases and impart kinetic stability to the emulsion, either through
reduction of interfacial tension (chemical stabilization), or by providing a barrier
to coalescence (physical stabilization). Resolution of emulsions, or de-
emulsification, proceeds via two steps: flocculation or aggregation of droplets,
and coalescence of droplets to form a continuous second phase. De-emulsifiers
may promote one or both of these phenomena [58].
Crude oil emulsions are complex, and vary from location to location. The
emulsifying agents may be amphiphilic molecules from the oil, especially the
resin fraction, including naphthenic acids. Many crude oil emulsions are
stabilized by fine solids, including clays, scale, or wax crystals [59], or bacteria
themselves [60], which present a barrier to droplet coalescence. Asphaltenes are
especially important in heavy crude oil emulsions. After association with the
interface, asphaltenes agglomerate to form a skin, which prevents coalescence of
droplets. Resins are also believed to play a part in stabilizing this skin [58].
Complex emulsion structures, such as water-in-oil-in-water emulsions, have also
been observed [61]. De-emulsification in the oil industry is challenging due to
the variety of possible emulsion properties, and treatments are currently tailored
to each site and adapted over time [59].
Biological de-emulsification has been studied using a variety of
microorganisms. Whole bacterial cells have received the most research [62-69],
but Streptomyces spores [70], bacterial metabolites [71], and yeast cells [64]
have also been studied. The organisms and emulsion systems used are
summarized in Table 1. The majority of studies have examined model,
chemically stabilized emulsions consisting of water, hydrocarbon, and a
commercial surfactant. This research has allowed some assessment of the mode
of action. De-emulsification ability appears to be associated with the surface of
the bacterial cells. Depending on their hydrophobicity, cells may aggregate at
the oil-water interface, promoting flocculation and coalescence of droplets [72].
Differences in hydrophobicity may account for changes in effectiveness of
microbes in different growth phases, as well as for differing abilities to resolve
O/W or W/O emulsions. In general, it appears that more hydrophilic cells are
required to treat W/O emulsions, while relatively more hydrophobic cells are
able to resolve O/W emulsions [62, 65, 66, 69, 70].
The ability of bacterial cells to de-emulsify both model and oilfield O/W
and W/O emulsions has been demonstrated, but the potential for treating the true
spectrum of real crude oil emulsions has not been rigorously tested. As with
chemical treatments, no single biological treatment will likely be effective for all
chemically stabilized crude oil emulsions. Biological products may still be a
valuable complement to existing chemical technologies.
125
Table 1
Biological systems shown to de-emulsify oil-water emulsions
Organism Emulsion system Comments Refs.
Nocardia amarae O/W emulsions: • Older, more hydrophobic [62, 65, 66,
strain LL-Se6 Alkanes / water cultures more effective 69]
(ATCC 27808) Kerosene / water
Oilfield emulsions
W/O emulsions: • Younger cultures (exponential
Water / kerosene growth phase) more effective
Oilfield emulsions
Corvnebacterium W/O emulsions: • Younger cultures (exponential [64, 691
petrophilum Oilfield emulsions growth phase) more effective
(ATCC 21404)
Micrococcus sp. O/W emulsions: • More effective for O/W [631
Kerosene / water emulsions
W/O emulsions: • Solvent washing increased O/W,
Water / kerosene decreased W/O de-emulsification
Mixed aerobic W/O emulsions: • More effective when grown on [67, 68, 731
bacterial culture Water / kerosene crude oil or motor oil than on
Oilfield emulsions carbohydrates
Streptomvces sp. O/W emulsions: • Only aerial spores were effective [701
strain AA8321 Kerosene / water • Effectiveness increased with
Alkanes / water culture age and hydrophobicity
Diesel / water
Gasoline / water
Paraffin oil / water
Soybean oil / water
Bacillus subtilis W/O emulsions: • Free acetoin in medium [711
Water / crude oil identified as active component
Torulopsis W/O emulsions: • Rate increased with cell [641
bombicola Oilfield emulsions concentration
(ATCC 22214)
The applicability of biotechnology to asphaltene- or solids-stabilized

emulsions has not been studied. Biocatalysis or biologically produced chemicals
may be effective in removing or dispersing asphaltenes or wax crystals,
particularly in combination with suitable cell-surface properties to aid in
dispersion of the solids or in aiding flocculation as appropriate.
126
3.5. Naphthenic acids

Naphthenic acids are found in varying concentrations in crude oils
worldwide. They are a family of carboxylic acids defined by the formula
CBH2n.z02, where n is the carbon number and Z is related to the number of rings.
Characteristic features of naphthenic acids, illustrated by the examples in Fig. 4,
are saturated rings containing five or six carbon atoms, a carboxyl group
separated from the rings by at least one methylene group, and an alkyl
substitution [75]. Naphthenic acids contribute to the total acidity of a crude oil,
typically expressed as the total acid number, or TAN (mg KOH required to
neutralize 1 g of oil). TAN values of 0.5 mg or greater are generally correlated
to high corrosivity, although the total corrosivity of an oil is also affected by
factors such as sulfur content, flow conditions (velocity and turbulence), and
temperature. Naphthenic acid corrosion typically occurs at processing
temperatures between 220°C and 400°C, which corresponds to the boiling range
of these compounds. The effects of corrosive oils are usually addressed through
careful selection of materials of construction [76].
In oil sands processing, hot caustic solutions are used to separate the
bitumen from the sand. Due to the high pH, naphthenic acids preferentially
partition to the aqueous phase, and are discharged with the water into the tailings
ponds. Naphthenic acids are believed to account for the high acute toxicity of
the tailings waters [77]. Research into biodegradation of naphthenic acids has
been pursued out of interest in remediation and reclamation of the tailings
ponds. Early research looked at aerobic degradation of simple model
compounds, including cyclohexane and cyclopentane carboxylic acids, 1- and 2-
methylcyclohexane carboxylic acids, cyclohexane pentanoic acid, 4-
pentylcyclohexane carboxylic acid, and decahydro-2-naphthoic acid [75, 78,
79]. These studies showed that mineralization of these simple compounds was
possible, starting with the carboxylated side chain but also including the ring
structures [78]. Alkylated compounds, representative of true naphthenic acids,
were more resistant to degradation [75, 78]. Direct analysis of the naphthenic
acids found in the tailings waters is challenging, due to the complexity of the
mixture. Biodegradation of naphthenic acids extracted from tailings water was
shown under aerobic conditions through both CO2 production and a reduction in
acute toxicity [78]. Under methanogenic conditions, model compounds
(cyclohexyl propanoic, butanoic, and pentanoic acids, and 6-phenyl hexanoic
acid) were substrates for methanogenesis. Extracted and commercial naphthenic
acids mixtures, however, delayed the onset of methanogenesis, and were not
apparently methanogenic substrates [80].
More recently, gas chromatography with electron impact mass
spectrometry has been used to resolve distinct ion fragments from naphthenic
acid mixtures. The abundances of these ions can be allocated to specific carbon
and Z-numbers, giving a 3-dimensional representation of the distribution of
127
naphthenic acids in a sample. This allowed direct observation of the differences

among samples, and showed changes occurring through biodegradation [81].
Other recent advances in naphthenic acids research include a statistical method
to show significant differences among samples analyzed using the mass
spectrometric technique [74] and a high performance liquid chromatography
method developed for quantitation of total naphthenic acids [82]. These methods
were used to definitively show biodegradation of commercial naphthenic acids
mixtures by aerobic enrichment cultures, accompanied by growth, CO2
production, and elimination of acute toxicity [83]. These commercial mixtures
differ from the naphthenic acids found in oils and tailings primarily in a lack of
compounds with carbon numbers of 22 or greater. It remains to be shown the
extent of biodegradation possible in authentic tailings waters, and whether the
larger naphthenic acids will be affected.
The aim of bioremediation research differs from biological upgrading for
naphthenic acids. Bioremediation requires the complete removal of these
compounds. Biological upgrading would ideally involve biotransformation of
naphthenic acids to compounds that are easier to handle and more valuable. The
reactions observed in the biodegradation of naphthenic acids (side chain and
ring oxidation and mineralization to CO2) are therefore not directly applicable to
oil upgrading. The work done clearly shows that despite their toxicity,
naphthenic acids are ready targets for microbial attack. Future upgrading
research needs to look for more suitable reactions and for systems capable of
catalyzing those reactions.
Fig. 4. Representative naphthenic acids structures (based on Ref. [74]). (R - alkyl group; m
number of carbons in the side chain excluding the carboxyl group)
128
3.6. Opportunities and research gaps

The ability of cells to physically interact with crude oil components at the
interface raises an important issue in studies of bioprocessing. How can we
ensure that proper measurements of oil composition are taken to ensure accurate
assessment of conversion, if oil components can interact strongly with the cell
surfaces by physical processes? The most rigorous measurements will rely on a
material balance on the compound of interest, where the fate of the constituents
in the oil is accurately accounted for in the reacting mixture. This approach
works well with 14C labelled compounds in mineralization studies, but it is less
applicable to complex fractions of heavy crudes. Balances on sulfur provide one
approach, where the disappearance of sulfur compounds from the oil is
confirmed by the appearance of other sulfur compounds in solution, such as
sulfate. Similar balances can be attempted for the other reactions discussed
above, to confirm not only that the substrate is disappearing, but also that its
transformed product is detected stoichiometrically. The lack of such rigorous
controls is likely at the heart of some claims for bacterial conversion of crude oil
components under a range of conditions [16-18].
Microbes have the potential to aid in processing or transportation of heavy
crude oils if interesting reactions can be harnessed, or if the surface properties of
the microbes can be used to aid in flocculating emulsions or dispersing
components. The natural environment of bacteria is either at the oil/water
interface or in the bulk water phase. The most attractive opportunities for
biological upgrading may therefore be in dealing with surface active
components such as naphthenic acids or in flocculating oil-in-water emulsions.
Interesting reactions such as hydrogenation and ring-opening deserve further
study, but are likely to be extremely slow if the substrate consists of large,
complex molecules such as vacuum residues or asphaltenes.
4. VISCOSITY AND ALIPHATIC SULFIDE CLEAVAGE
4.1. Viscosity correlations

Experimental data show that the viscosity of oil is correlated to the average
molecular weight of the material. Fig. 5a shows data compiled from different
sources, including whole oils, bitumen, distillates, and residues. The observable
trends are towards higher viscosity in heavier samples and towards lower
viscosity at higher temperatures. The scatter in Fig. 5a indicates that there are
factors involved other than molecular weight and temperature. Some published
models include properties such as specific gravity to account for this variability
[84, 85]. Although viscosity models fit the data used to generate them, they are
often difficult to extend to other samples due to these other contributing effects.
The general correlation to molecular weight appears to be sound and can be used
as a basis for further analysis of viscosity.
129
Fig. 5. Viscosity data for whole oils, residuals, and distillates [88], oil sand bitumens and
topped crudes [84], oil fractions [85], synthetic crude oils [86], and crude oils and natural
bitumens [20] showing correlation to (a) average molecular weight and (b) asphaltene content.
(Analysis temperatures are indicated in the legends)
130
The correlation of viscosity and molecular weight indicates that there

should also be a correlation between the viscosity and the fractional composition
of the oil. Fig. 5b shows that viscosity increases with the asphaltene content
(weight %), the asphaltenes being among the heaviest molecules in the oil. This
type of correlation has been used to formulate viscosity models based on
logarithmic mixing rules, assigning "pseudo-viscosities" to the different
fractions of the oil and applying a weighting factor to each one [86, 87].
Sulfides, ethers, and esters have been identified as common linking
structures in the aliphatic bridges found in asphaltenes [11]. Nickel boride
desulfurization was used to specifically cleave aliphatic sulfide bonds in two
asphaltene fractions, giving a 4-fold reduction in the molecular weight of the
higher molecular weight fraction. This indicated that aliphatic sulfides were
involved in the linking structures of the molecules, including linkages between
aromatic cores and to smaller structures like alkanes and steranes.
The total sulfur content of asphaltenes includes the sulfide bridges
(aliphatic), cyclic sulfides (aliphatic heterocycles, found as substituted thiolanes
and thianes [89]), and thiophenic sulfur (aromatic sulfur heterocycles). Only
cleavage of the sulfide bridges leads to a reduction in molecular weight, since
removal of the cyclic sulfides and thiophenes leaves the carbon backbone intact.
Table 2
Selected organic sulfur compounds successfully used for enrichment of
microorganisms able to use the compounds as sole sulfur source under sulfur-
limited conditions
Compound Procedures used Refs.
Ametryne and • Culture maintenance alternated between [90]
prometryne selective liquid and non-selective solid media
(herbicides)
Naphthalenesulfonic • Substrate purification by high pressure liquid [91]
acids chromatography
Benzenesulfonic acids • "Scrupulously clean glassware" (procedure
(detergents) not given)
Organic sulfur in coal • Effluent from reactor mutagenized and [92]
reinoculated to accelerate strain evolution
Endosulfan • Use of an Escherichia coli culture to [93]
(an insecticide) scavenge sulfate from medium, followed by
filter sterilization to produce a sulfur-free
medium
131
4.2. Biological sulfur requirements

Sulfur is an essential element for bacterial growth, although total sulfur
requirements are low. Sulfur limitation has been successfully applied to the
enrichment of microorganisms active towards a variety of organosulfur
compounds. Several examples are given in Table 2. All but one of the studies
listed mentioned special procedures used to minimize the effect of contaminant
sulfur.
Aliphatic sulfides have not been studied in oil desulfurization research. The
relatively low carbon-sulfur bond strength results in easy cleavage of these
bonds under thermal treatment compared to thiophenic sulfur, which is only
removed during catalytic hydro treating [19]. Correspondingly, research on
biological desulfurization of oil has focused on the recalcitrant thiophenic
compounds rather than the aliphatic sulfides (see chaper 2). A selection of the
bacterial strains known to desulfurize dibenzothiophene (DBT) is given in Table
3. The substrate ranges of these organisms for organic sulfur sources frequently
include some alkylated species, and sometimes extend to compounds with
aliphatic sulfide bonds as well.
The biodegradation of some aliphatic sulfides not related to oil has been
studied, including dimethyl sulfide (DMS) and analogues of sulfur mustard
(2,2'-dichlorodiethyl sulfide). Biodegradation of some larger sulfides has been
reported as well. These will be presented here to illustrate the types of biological
activity possible with aliphatic sulfides, and to show how they may be relevant
to oil upgrading.
4.3. High molecular weight sulfides

Phytanyl octadecyl sulfide was used as a model compound for sulfur-
bound hydrocarbons found in heavy oil macromolecules [15]. Biodegradation as
a carbon source by R. erythropolis ATCC 13260 occurred only on the linear
octadecyl chain, and not the branched phytanyl chain. Six chain degradation
metabolites were identified (Fig. 6), which suggested two different mechanisms
[15]. Metabolites with an even number of carbon atoms in the linear side chain
were proposed to result from terminal oxidation followed by |3-oxidations
removing two carbon atoms at a time. Metabolites with an odd number of
carbon atoms cannot arise solely from P-oxidations, and an initial mid-chain
oxidation was proposed to occur as well. Oxidation of the sulfur atom to a
sulfone was observed both in the parent compound and in the degraded
metabolites, indicating that sulfur oxidation was independent of the chain
degradation pathways. No evidence of carbon-sulfur bond cleavage was reported
[15].
132
Table 3
Selected dibenzothiophene desulfurizine bacteria and alternate sulfur sources
Bacterial use of a high molecular weight aliphatic sulfide as a sulfur source

has only recently been reported. Rhodococcus sp. strain JVH1 is capable of
using the novel compound fe-(3-pentafluorophenylpropyl) sulfide (PFPS) as its
sole sulfur source for growth [104]. PFPS was specifically designed using ring
fluorination to block any terminal attack on the molecule, necessitating a
subterminal attack to support growth. The desulfurization pathway proposed is
shown in Fig. 7. PFPS is first oxidized to the corresponding sulfoxide and
sulfone (PFPSO and PFPSO2). Carbon-sulfur bond cleavage then yields the
primary alcohol 3-pentafluorophenylpropan-l-ol, which is further oxidized to 3-
pentafluorophenylpropanoic acid. The second product of PFPSO2 cleavage was
proposed to be a sulfmate, analogous to the 4S pathway for DBT
desulfurization, but was not directly observed. Release of the sulfur as sulfite
133
was also hypothesized, but not directly observed. JVH1 was shown to use a
variety of compounds with aliphatic carbon-sulfur bonds as sulfur sources
(including dialkyl sulfides, thiacycloalkanes, and aryl-terminated sulfides), but
not thiophenic compounds. This selective ability to cleave compounds with
aliphatic carbon-sulfur bonds is extremely interesting for research into
biological viscosity reduction in heavy crude oils.
Fig. 6. (a) Structure of phytanyl octadecyl sulfide. (b) Metabolites produced by Rhodococcus
erythropolis ATCC 13260 and proposed reactions in the degradation of phytanyl octadecyl
sulfide [15].
134
Fig. 7. Proposed pathway of PFPS metabolism in Rhodococcus sp. strain JVH1 [104].
Compounds in brackets were not directly observed. (PFPP-OH - 3-pentafluorophenylpropan-
l-ol; PFPP-acid - 3-pentafluorophenylpropanoic acid).
For high molecular weight aliphatic sulfides, the mechanism of attack

appears to depend on the substituent groups. Sulfur-bound alkyl chains are
subject to aerobic degradation, apparently through the same pathway as
«-alkanes. Sulfur oxidation occurred independently, but did not prevent chain
degradation. A 4S-like pathway has been reported for the fluorinated compound
PFPS. Interestingly, DBT-desulfurizing strains only produced PFPSO2, being
apparently unable to cleave the aliphatic carbon-sulfur bonds in PFPS [104].
This illustrated that sulfur-specific desulfurization of aliphatic sulfides and
thiophenes may occur through analogous mechanisms, but that the
desulfurization systems are not necessarily interchangeable.
135
4.4. Dimethyl sulfide

DMS is part of the global sulfur cycle. It is formed in marine sediments
from degradation of dimethylsulfoniopropionate, produced by marine algae and
plants. Some DMS is released from the oceans to the atmosphere where it is
involved in cloud formation, while most is degraded by a variety of marine
microorganisms [105]. The pathway most studied, shown in Fig. 8a, occurs
under aerobic conditions in a variety of species of Hyphomicrobium and
Thiobacillus [106]. Initial cleavage to methanethiol and formaldehyde is
catalyzed by a NADH-dependent monooxygenase. Methanethiol is then cleaved
to a second molecule of formaldehyde and hydrogen sulfide by methanethiol
oxidase. Sulfide is further oxidized to sulfate. Formaldehyde is oxidized to
formate, then to carbon dioxide by formaldehyde dehydrogenase and formate
dehydrogenase.
Thiobacillus sp. strain ASN-1 can degrade DMS under both aerobic and
anaerobic (nitrate-reducing) conditions [107]. Methanethiol, but not
formaldehyde, was produced under aerobic conditions, suggesting a novel
pathway for DMS degradation in this organism. The same degradative pathway
was proposed for this organism under both aerobic and anaerobic conditions
(Fig. 8b), with the terminal electron acceptors being oxygen and nitrate,
respectively. Each methyl group is removed by a methyltransferase and oxidized
to formate. The sulfur is first released as sulfide, which is oxidized to sulfate.
This work was extended to larger sulfides, and it was shown that Thiobacillus
sp. strain ASN-1 could grow on diethyl sulfide, dipropyl sulfide, dibutyl sulfide,
dimethyl disulfide, and dibutyl disulfide [108]. Growth on butanethiol, as well
as on acetate, propionate, and butyrate, suggested that similar reaction
mechanisms were used for the larger compounds as for DMS. Lag periods were
observed when transferring cultures from one sulfide compound to another, but
not from a sulfide to the corresponding thiol, indicating that different enzymes
are used for the degradation of the different sulfides [108].
Anaerobic degradation of DMS has been observed in marine sediments.
Methanogenic consortia from these environments release methane from DMS,
as well as from methanethiol and dimethyl disulfide. Ethane release from the
analogous compounds diethyl sulfide and ethanethiol has also been observed in
marine sediment samples. This release was not observed in killed controls, or in
the presence of 2-bromoethanesulfonic acid, which inhibits methanogenic
bacteria [109].
136
Fig. 6. Dimethyl sulfide biodegradation pathways, (a) aerobic marine thiobacilli and
hyphomicrobia [106]. (b) Thiobacillus sp. strain ASN-1 [107]. (c) Rhodococcus sp. strain
SY1 [95]. (X -cobalamin carrier of methyltransferase)
The terrestrial bacterium Rhodococcus sp. strain SY1, originally isolated

for its ability to use DBT as a sole sulfur source [95], can also degrade DMS as a
sulfur source [96]. A pathway analogous to the sulfur-specific 4S pathway for
DBT was proposed based on the observed products from growth on successive
intermediates (Fig. 8c). The sulfur atom in DMS is first oxidized giving
dimethylsulfoxide and dimethyl sulfone. Growth on dimethylsulfoxide also
showed release of dimethyl sulfone, as well as methanol and methane. Release
of the sulfur as sulfite, which is spontaneously oxidized to sulfate under aerobic
conditions, was proposed but never directly observed.
For the smallest aliphatic sulfide, DMS, degradation primarily occurs
without oxidation of the sulfur atom. This has been observed under aerobic,
nitrate-reducing, and methanogenic conditions. Some larger sulfides may be
subject to similar reactions, with butyl sulfide being the largest compound
tested. A 4S-like oxidative pathway has also been reported, allowing use of
DMS as a sulfur source by a DBT-desulfurizing organism. This suggests that the
DBT-desulfurizing enzymes of some bacteria may also catalyze aliphatic C-S
bond cleavage.
4.5. Sulfur mustard analogues

Sulfur mustard (2,2'-dichlorodiethyl sulfide) was first used as a chemical
warfare agent during World War I. The United States military is currently
137
interested in degradation strategies for this compound for disposal of stockpiles

at American military bases [110]. Biodegradation of sulfur mustard is not
studied directly because of its high toxicity. Proposed disposal processes start
with chemical neutralization followed by treatments such as incineration,
chemical oxidation, or biodegradation. In the neutralization step, sulfur mustard
is hydrolysed to less toxic compounds, primarily the dechlorinated compound
thiodiglycol (TDG) [110], which has been the focus of most biodegradation
studies.
Alcaligenes xylosoxydans ssp. xylosoxydans SH91, a soil bacterium, can
use TDG as its sole source of carbon and energy [111]. TDG is oxidized via [(2-
hydroxyethyl)thio] acetic acid to thiodiglycolic acid in two oxygen-dependent
steps catalyzed by butanol dehydrogenase (Fig. 9a). Thiodiglycol sulfoxide also
accumulates as a dead-end metabolite. Further metabolites have not been shown
analytically, but growth on TDG implies incorporation into cell mass and
mineralization to CO2 and sulfate.
A second strain of A. xylosoxydans, designated PGH10, is able to transform
TDG in minimal medium with citrate or fructose [112]. Degradation is proposed
to follow the same initial pathway as for A. xylosoxydans ssp. xylosoxydans
SH91, based on observation of the same two initial metabolites, [(2-
hydroxyethyl)thio] acetic acid and thiodiglycolic acid. Although this strain can
transform TDG, there is no evidence for the release of carbon or sulfur from the
metabolites. It appears that this strain may not be able to cleave the carbon-
sulfur bond, resulting in its inability to use TDG as a carbon source.
Fungal degradation of TDG has also been studied. The white-rot fungus
Coriolus versicolor (IFO 30340) degraded TDG most effectively in a high-
carbon, low-nitrogen medium, and the brown-rot fungus Tyromyces palustris
(IFO 0507) degraded TDG in sulfur-free medium [113]. Degradation was
measured only as loss of TDG, not as the formation of any metabolites. Both
organisms were also able to degrade the brominated sulfur mustard analog 2,2'-
dibromodiethyl sulfide, again measured as loss of the starting compound.
One further study has investigated a sulfur-specific detoxification strategy
for sulfur mustard. The DBT-desulfurizing bacterium R. erythropolis strain
IGTS8 can use the sulfur mustard analog 2-chloroethyl ethyl sulfide as a sole
sulfur source for growth [114]. Two metabolites were found using gas
chromatography and mass spectrometry. Growing cultures of R. erythropolis
strain IGTS8 accumulated 2-chloroethanesulfinic acid, while 2-chloroethanol
was found in resting cell cultures (Fig. 9b). Neither compound was observed in
killed cell controls.
138
Fig. 7. Metabolites formed in the biodegradation of the sulfur mustard analogues (a)
thiodiglycol [111] and (b) 2-chloroethyl ethyl sulfide [114]. Products in brackets were not
directly observed.
For the sulfur-mustard analogue TDG, both sulfur oxidation and terminal
carbon oxidation were observed in a bacterium using the compound as a carbon
source. These reactions were apparently independent and mutually exclusive
with the sulfoxide produced accumulating as a dead-end metabolite. Carbon-
sulfur bond cleavage was assumed to occur subsequent to terminal carbon
oxidation, but only in the absence of sulfur oxidation. Some fungal strains were
also able to degrade sulfur mustard analogues, although metabolites were not
identified. As with DMS, 2-chloroethyl ethyl sulfide was subject to sulfur-
specific degradation by a DBT-desulfurizing strain, demonstrating again that the
desulfurization enzymes may have a sufficiently broad substrate specificity to
allow attack on both thiophenes and aliphatic sulfides.
5. CONCLUSIONS
Known interactions between microbes and the high molecular weight

components of crude oils include oxidation of aliphatic and aromatic carbon
groups, oxidation of naphthenic acids, and oxidation and desulfurization of
aromatic and aliphatic sulfur groups. Hydrogenation and dehydrogenation
reactions have been demonstrated only on lower-molecular weight components.
All of these reactions are of potential interest for upgrading heavy crude oils and
bitumens, but a major barrier is the transport of reactants to the active site of
reaction, particularly for intracellular enzymes in bacteria. Although membranes
may give significant barriers for bioprocessing of heavy hydrocarbons, the
139
interactions of cell membranes with oil/water interfaces may be of interest in

de-emulsifying oil and in dispersing asphaltenic material to prevent deposition.
Acknowledgments.
The authors would like to thank Dr. P.M. Fedorak and Dr. J.D. Van
Hamme for access to and information on works in press. Funding was provided
by Alberta Energy Research Institute under the COURSE program and by the
Natural Sciences and Engineering Research Council of Canada.
REFERENCES
[I] H. Groenzin and O.C. Mullins, Energy Fuels, 14 (2000) 677.

[2] L. Artok, Y. Su, Y. Hirose, M. Hosokawa, S. Murata, and M. Nomura, Energy Fuels, 13
(1999)287.
[3] N.X. Thanh, M. Hsieh, and R.P. Philp, Org. Geochem., 30 (1999) 119.
[4] J.G. Speight, Fuel, 49 (1970) 134.
[5] T.F. Yen, Prepr. - Am. Chem. Soc, Div. Pet. Chem., 17 (1972) F102.
[6] T.F. Yen, J.G. Erdman, and S.S. Pollack, Anal. Chem., 3 (1961) 1587.
[7] O.P. Strausz, T.W. Mojelsky, F. Faraji, E.M. Lown, and P. Peng, Energy Fuels, 13
(1999)207.
[8] O.P. Strausz, T.W. Mojelsky, E.M. Lown, I. Kowalewski, and F. Behar, Energy Fuels,
13(1999)228.
[9] P. Peng, A. Morales-Izquierdo, E.M. Lown, and O.P. Strausz, Energy Fuels, 13 (1999)
248.
[10] O.P. Strausz, T.W. Mojelsky, and E.M. Lown, Fuel, 71 (1992) 1355.
[II] P. Peng, A. Morales-Izquierdo, A. Hogg, and O.P. Strausz, Energy Fuels, 11 (1997)
1171.
[12] D.C. Bressler and M.R. Gray, Int. J. Chem. React. Eng., 1 (2003) R3.
[13] J.J. Kilbane, Resour. Conserv. Recycl., 3 (1990) 69.
[14] O. Hoist, B. Stenberg, and M. Christiansson, Biodegradation, 9 (1998) 301.
[15] A. Jenisch-Anton, P. Adam, W. Michaelis, J. Connan, D. Herrmann, M. Rohmer, and P.
Albrecht, Geochim. Cosmochim. Acta, 64 (2000) 3525.
[16] E.T. Premuzic, M.S. Lin, and B. Manowitz, Fuel Process. Technol., 40 (1994) 227.
[17] E.T. Premuzic and M.S. Lin, J. Pet. Sci. Eng., 22 (1999) 171.
[18] E.T. Premuzic, M.S. Lin, M. Bohenek, and W.M. Zhou, Energy Fuels, 13 (1999) 297.
[19] M.R. Gray, Upgrading Petroleum Residues and Heavy Oils, New York, NY, 1994.
[20] A.N. Ratov, Pet. Chem., 36 (1996) 191.
[21] J.M. Foght, Q. Wu, P.M. Fedorak, M.A. Pickard, and M.R. Gray, in the Proc. of the
Joint meeting of The Petroleum Society (48th Annual Technical Meeting) and
BIOMINET, Calgary, AB, 1997, Paper 97-13.
[22] Q. Wu, P.M. Fedorak, M.A. Pickard, M.R. Gray, and J.M. Foght, Prepr. Pap. - Am.
Chem. Soc, Div. Fuel Chem., 43 (1998) 515.
[23] Q. Wu, P.M. Fedorak, M.A. Pickard, M.R. Gray, and J.M. Foght, Prepr. Pap. - Am.
[24] Q. Wu, M.R. Gray, M.A. Pickard, P.M. Fedorak, and J.M. Foght, Prepr. - Am. Chem.
Soc, Div. Pet. Chem, 48 (2003) 47.
140
[25] C.L. Coyle, M. Siskin, D.T. Ferrughelli, M.S.P. Logan, and G. Zylstra, Biological
activation of aromatics for chemical processing and/or upgrading of aromatic
compounds, petroleum coal, resid, bitumen and other petrochemical streams, US Patent
No. 6 156 946(2000).
[26] C. Vorbeck, H. Lenke, P. Fischer, and H.-J. Knackmuss, J. Bacteriol., 176 (1994) 932.
[27] C. Vorbeck, H. Lenke, P. Fischer, J.C. Spain, and H.-J. Knackmuss, Appl. Environ.
Microbiol., 64(1998)246.
[28] X. Zhang, E.R. Sullivan, and L.Y. Young, Biodegradation, 11 (2000) 117.
[29] X. Zhang and L.Y. Young, Appl. Environ. Microbiol., 63 (1997) 4759.
[30] A. Esteve-Nunez, A. Caballero, and J.L. Ramos, Microbiol. Mol. Biol. Rev., 65 (2001)
335.
[31] C.E. French, S. Nicklin, and N.C. Bruce, Appl. Environ. Microbiol., 64 (1998) 2864.
[32] P.-G. Rieger, V. Sinnwell, A. Preuss, W. Francke, and H.-J. Knackmuss, J. Bacteriol.,
181 (1999)1189.
[33] C. Behrend and K. Heesche-Wagner, Appl. Environ. Microbiol., 65 (1999) 1372.
[34] S. Ebert, P.-G. Rieger, and H.-J. Knackmuss, J. Bacteriol., 181 (1999) 2669.
[35] G. Heiss, K.W. Hofmann, N. Trachtmann, D.M. Walters, P. Rouviere, and H.-J.
Knackmuss, Microbiology, 148 (2002) 799.
[36] C.S. Harwood, G. Burchhardt, H. Herrmann, and G. Fuchs, FEMS Microbiol. Rev., 22
(1999)439.
[37] R.M. Atlas and R. Bartha, Microbial Ecology: Fundamentals and Applications, 4th ed.,
Menlo Park, CA, 1998.
[38] R.M. Miller and R. Bartha, Appl. Environ. Microbiol., 55 (1989) 269.
[39] K.G. Kropp, LA. Davidova, and J.M. Suflita, Appl. Environ. Microbiol., 66 (2000)
5393.
[40] J.D. Schumacher and R.M. Fakoussa, Appl. Microbiol. Biotechnol., 52 (1999) 85.
[41] J. Heider, A.M. Spormann, H.R. Beller, and F. Widdel, FEMS Microbiol. Rev., 22
(1999) 459.
[42] A.M. Spormann and F. Widdel, Biodegradation, 11 (2000) 85.
[43] F. Widdel and R. Rabus, Curr. Opin. Biotechnol., 12 (2001) 259.
[44] J.D. Van Hamme, A. Singh, and O.P. Ward, Microbiol. Mol. Biol. Rev., 67 (2003) 503.
[45] H. Wilkes, S. Kiihner, C. Bolm, T. Fischer, A. Classen, F. Widdel, and R. Rabus, Org.
Geochem., 34 (2003) 1313.
[46] R. Rabus, H. Wilkes, A. Behrends, A. Armstroff, T. Fischer, A.J. Pierik, and F. Widdel,
J. Bacteriol., 183 (2001) 1707.
[47] K. Koike, K. Ara, S. Adachi, H. Takigawa, H. Mori, S. Inoue, Y. Kimura, and S. Ito,
Appl. Environ. Microbiol., 65 (1999) 5636.
[48] A. Jenisch-Anton, A. Mikolajczak, A. Rabenstein, J. Klindworth, U. Fischer, and W.
Michaelis, Biodegradation, 10 (1999) 383.
[49] Q. Cheng, S.M. Thomas, and P. Rouviere, Appl. Microbiol. Biotechnol., 58 (2002) 704.
[50] K. Kostichka, S.M. Thomas, K.J. Gibson, V. Nagarajan, and Q. Cheng, J. Bacteriol.,
183(2001)6478.
[51] P.C. Brzostowicz, K.L. Gibson, S.M. Thomas, M.S. Blasko, and P. Rouviere, J.
Bacteriol., 182(2000)4241.
[52] P. Singh, A. Youyen, and H.S. Fogler, AIChE J, 47 (2001) 2111.
[53] J.R. Becker, Crude Oil Waxes, Emulsions, and Asphaltenes, Tulsa, OK, 1997.
[54] M.M. Santamaria and R.E. George, in the Proc. of the SPE Annual Technical
Conference & Exhibition, Dallas, TX, 1991, pp. 351-353.
141
[55] F.G. Brown, in the Proc. of the SPE Permian Basin Oil and Gas Recovery Conference,
Midland, TX, 1992, pp. 251-259.
[56] I. Lazar, A. Voicu, C. Nicolescu, D. Mucenica, S. Dobrota, I.G. Petrisor, M. Stefanescu,
and L. Sandulescu, J. Pet. Sci. Eng., 22 (1999) 161.
[57] D.P. Rimmer, A.A. Gregoli, J.A. Hamshar, and E. Yildirim, in L.L. Schramm (ed),
Emulsions: Fundamentals and Applications in the Petroleum Industry, Washington, DC,
1992, pp. 295-312.
[58] L.L. Schramm, in L.L. Schramm (ed), Emulsions: Fundamentals and Applications in the
Petroleum Industry, Washington, DC, 1992, pp. 1-49.
[59] R. Grace, in L.L. Schramm (ed), Emulsions: Fundamentals and Applications in the
[60] L. Dorobantu, Stabilization of oil/water emulsions by hydrophobic bacteria, M.Sc.
thesis, University of Alberta, Edmonton, AB, 2004.
[61] R.J. Mikula, in L.L. Schramm (ed), Emulsions: Fundamentals and Applications in the
[62] W.L. Cairns, D.G. Cooper, J.E. Zajic, J.M. Wood, and N. Kosaric, Appl. Environ.
Microbiol, 43(1982)362.
[63] M. Das, Bioresour. Technol., 79 (2001)
[64] Z. Duvnjak and N. Kosaric, Biotechnol. Lett., 9 (1987) 39.
[65] N.C.C. Gray, A.L. Stewart, W.L. Cairns, and N. Kosaric, Biotechnol. Lett., 6 (1984)
419.
[66] N.C.C. Gray, A.L. Stewart, W.L. Cairns, and N. Kosaric, in C. Ratledge, P. Dawson,
and J. Rattray (eds), Biotechnology for the Oils and Fats Industry, USA, 1984, pp. 255-
268.
[67] N. Nadarajah, A. Singh, and O.P. Ward, Process Biochem., 37 (2002) 1135.
[68] N. Nadarajah, A. Singh, and O.P. Ward, World J. Microbiol. Biotechnol., 18 (2002)
435.
[69] A.L. Stewart, N.C.C. Gray, W.L. Cairns, and N. Kosaric, Biotechnol. Lett., 5 (1983)
725.
[70] S.H. Park, J.-H. Lee, S.-H. Ko, D.-S. Lee, and H.K. Lee, Biotechnol. Lett., 22 (2000)
1389.
[71] K.L. Janiyani, HJ. Purohit, R. Shanker, and P. Khanna, World J. Microbiol.
Biotechnol., 10 (1994) 452.
[72] W.L. Cairns, R. Rumble, and N. Kosaric, in C. Ratledge, P. Dawson, and J. Rattray
(eds), Biotechnology for the Oils and Fats Industry, USA, 1984, pp. 223-239.
[73] O. Ward, A. Singh, and J. Van Hamme, J. Ind. Microbiol. Biotechnol., 30 (2003) 260.
[74] J.S. Clemente, N.G.N. Prasad, M.D. MacKinnon, and P.M. Fedorak, Chemosphere, 50
(2003) 1265.
[75] D.C. Herman, P.M. Fedorak, and J.W. Costerton, Can. J. Microbiol., 39 (1993) 576.
[76] E. Slavcheva, B. Shone, and A. Turnbull, Br. Corros. J., 34 (1999) 125.
[77] M.D. MacKinnon and H. Boerger, Water Poll. Res. J. Canada, 21 (1986) 496.
[78] D.C. Herman, P.M. Fedorak, M.D. MacKinnon, and J.W. Costerton, Can. J. Microbiol.,
40(1994)467.
[79] J.W.S. Lai, LJ. Pinto, E. Kiehlmann, L.I. Bendell-Young, and M.M. Moore, Environ.
Toxicol. Chem., 15 (1996) 1482.
[80] F.M. Holowenko, M.D. MacKinnon, and P.M. Fedorak, Water Res., 35 (2001) 2595.
[81] F.M. Holowenko, M.D. MacKinnon, and P.M. Fedorak, Water Res., 36 (2002) 2843.
[82] J.S. Clemente, T.-W. Yen, and P.M. Fedorak, J. Environ. Eng. Sci, 2 (2003) 177.
142
[83] J.S. Clemente, M.D. MacKinnon, and P.M. Fedorak, Environmental Science &
Technology, (in press)
[84] T. Wakabayashi, Fuel, 76 (1997) 1049.
[85] M.R. Riazi and T.E. Daubert, Oil Gas J., 85(52) (1987) 110.
[86] A. Werner, F. Behar, J.C. de Hemptinne, and E. Behar, Fluid Phase Equilib., 147 (1998)
343.
[87] A. Werner, J.C. de Hemptinne, F. Behar, E. Behar, and C. Boned, Fluid Phase Equilib.,
147(1998)319.
[88] P.J. Closmann and R.D. Seba, J. Can. Pet. TechnoL, 29(4) (1990) 115.
[89] J.D. Payzant, D.D. Mclntyre, T.W. Mojelsky, M. Torres, D.S. Montgomery, and O.P.
Strausz, Org. Geochem., 14 (1989) 461.
[90] A.M. Cook and R. Hiitter, Appl. Environ. Microbiol, 43 (1982) 781.
[91] D. Ziirrer, A.M. Cook, and T. Leisinger, Appl. Environ. Microbiol., 53 (1987) 1459.
[92] JJ. Kilbane, in D.L. Wise (ed), Bioprocessing and Biotreatment of Coal, New York,
NY, 1990, pp. 487-506.
[93] T.D. Sutherland, I. Home, M.J. Lacey, R.L. Harcourt, R.J. Russell, and J.G. Oakeshott,
[94] KJ. Kayser, B.A. Bielaga-Jones, K. Jackowski, O. Odusan, and JJ. Kilbane, J. Gen.
Microbiol., 139(1993)3123.
[95] T. Omori, L. Monna, Y. Saiki, and T. Kodama, Appl. Environ. Microbiol., 58 (1992)
911.
[96] T. Omori, Y. Saiki, K. Kasuga, and T. Kodama, Biosci. Biotechnol. Biochem., 59
(1995)1195.
[97] Y. Izumi, T. Ohshiro, H. Ogino, Y. Hine, and M. Shimao, Appl. Environ. Microbiol., 60
(1994)223.
[98] S.-K. Rhee, J.H. Chang, Y.K. Chang, and H.N. Chang, Appl. Environ. Microbiol., 64
(1998)2327.
[99] J.H. Chang, S.-K. Rhee, Y.K. Chang, and H.N. Chang, Biotechnol. Progr., 14 (1998)
851.
[100]M. Kobayashi, T. Onaka, Y. Ishii, J. Konishi, M. Takaki, H. Okada, Y. Ohta, K.
Koizumi, and M. Suzuki, FEMS Microbiol. Lett., 187 (2000) 123.
[101] J. Konishi, Y. Ishii, T. Onaka, K. Okumura, and M. Suzuki, Appl. Environ. Microbiol.,
63(1997)3164.
[102]T. Furuya, K. Kirimura, K. Kino, and S. Usami, FEMS Microbiol. Lett., 204 (2001)
129.
[103]H.Y. Kim, T.S. Kim, and B.H. Kim, Biotechnol. Lett., 12 (1990) 761.
[104] J.D. Van Hamme, P.M. Fedorak, J.M. Foght, M.R. Gray, and H.D. Dettman, Appl.
Environ. Microbiol., (2004, in press)
[105]H. Fuse, O. Takimura, K. Murakami, Y. Yamaoka, and T. Omori, Appl. Environ.
Microbiol., 66 (2000) 5527.
[106]D.P. Kelly and N.A. Smith, Adv. Microb. Ecol., 11 (1990) 345.
[107]P.T. Visscher and B.F. Taylor, Appl. Environ. Microbiol., 59 (1993) 3784.
[108]P.T. Visscher and B.F. Taylor, Appl. Environ. Microbiol., 59 (1993) 4083.
[109]R.S. Oremland, M.J. Whiticar, F.E. Strohmaier, and R.P. Kiene, Geochim. Cosmochim.
Acta, 52(1988)1895.
[110]D.A. Irvine, J.P. Earley, D.P. Cassidy, and S.P. Harvey, Water Sci. TechnoL, 35 (1997)
67.
143
[111]T.-S. Lee, Biodegradation and Biotransformation of Thiodiglycol, the Main Hydrolysis

Product of Sulfur Mustard, Ph.D. thesis, University of Maryland, College Park, MD,
1998.
[112] V. Garcia-Ruiz, L.E. Martin-Otero, and A. Puyet, Biotechnol. Progr., 18 (2002) 252.
[113]N. Itoh, M. Yoshida, T. Miyamoto, H. Ichinose, H. Wariishi, and H. Tanaka, FEBS
Lett., 412(1997)281.
[114] J.J. Kilbane and K. Jackowski, J. Chem. Technol. Biotechnol., 65 (1996) 370.
Chapter 5
Whole-cell bio-processing of aromatic compounds in crude

oil and fuels
J. M. Foght

Edmonton, Alberta Canada T6G 2E9
1. INTRODUCTION
This review discusses the potential for biological upgrading to improve the
quality of certain crude oils and liquid fuels, using whole cell biocatalysts to
decrease aromaticity and sensitize aromatic heterocycles to subsequent
heteroatom removal. Some specific examples of research directed towards future
applications are presented; however, much of this concept is hypothetical and
requires further research to prove its potential. This chapter presents information
about the aerobic bacteria currently being tested as aromatic ring opening
biocatalysts and speculates on the application of their activities to biological
petroleum upgrading ("bio-processing"). The review does not deal with
reactions achieved with biological desulfurization (BDS), nor with purified
enzymes. For treatments of those topics, see Chapters 2 and 3, respectively.
1.1. Problems posed by aromatic compounds in crude oil and fuels

Aromatic hydrocarbons and heterocycles adversely affect several stages of
petroleum production, handling, and processing. Homologous series of aromatic
hydrocarbons and heterocycles occur in varying proportions in crude oils and
their refined products, depending on the source of the oil and the refining
process applied. Aromatic compounds influence the persistence and toxicity of
oils spilled in the environment [1] and have poor combustion characteristics in
diesel engines, including low cetane number and high particulate matter (soot)
formation. During refining, nitrogen heterocycles (e.g., carbazoles) inactivate
chemical catalysts, interfere with catalytic hydrodesulfurization and consume
large amounts of H2 [2, 3]. As well, combustion of fuels containing S and N
heteroatoms produces SOX and NOX in emissions implicated in acid rain.
Effective and cost-efficient conversion of aromatic hydrocarbons and
146
heterocycles in crude oils and fuels therefore is desirable from an environmental

viewpoint and is of interest to the refining industry.
During conventional refining, expensive high-pressure hydrogenation and
chemical catalysis are used to saturate and "crack" aromatic fused-ring
structures. This thermochemical technology has several problems [4, 5]
including unfavorable reaction kinetics, high consumption of thermal energy and
hydrogen (which contributes to greenhouse gases and other emissions), and
production of less desirable side-products such as gaseous hydrocarbons through
non-specific reactions. The cycloalkane products have better combustion
properties than aromatics, but the quality of their characteristics is still below
those of linear alkane compounds.
1.2. Biological oxidation of aromatic compounds

In contrast to chemical catalysis, biological (enzyme-mediated) reactions
are usually highly substrate-specific and occur at near-ambient temperatures
without the need for expensive high-pressure vessels and hydrogen. They do,
however, require water for activity, necessitating biphasic reaction mixtures of
oil and an aqueous suspension of microbial cells.
In one proposed application, a two-stage process of "Biological Aromatic
Ring Cleavage" (BioARC) is envisioned: the first stage is ring opening of the
aromatic compounds by oxidative enzymatic reactions achieved by pre-grown
whole cell biocatalysts. The two product streams would be a primary stream of
BioARC-treated oil having decreased aromaticity, and a secondary stream of
polar, water-soluble aromatic ring cleavage products (forming, for example,
Compounds IV and V, Fig. 1) that would be recovered by solvent extraction or
reversible sorption. In stage two, chemical hydrogenation of the polar ring
cleavage products under milder conditions than the conventional process would
yield alkyl-aromatics having more favorable combustion characteristics than
either the parent compound or the conventional cycloalkane catalytic products.
Potential side-benefits to the process would be sensitization of S- and N-
heterocycles to subsequent desulfurization and denitrogenation, although there is
no experimental work to support this prediction at present.
A second process described in US Patent # 6,156,946 [6] as "biological
activation of aromatics" uses whole cell biocatalysis to hydroxylate aromatic
and heterocyclic substrates in petrochemicals (e.g., enzymatic steps A and B,
forming Compounds II and III, Fig. 1), followed by aqueous chemical
hydrogenation and hydrogenolysis to produce cyclic alkylaromatics. The
hydrogenated products are proposed to be more susceptible to subsequent
thermochemical cleavage than the parent compounds. This is essentially an
"aromatic activation" process in which the substrates are sensitized to
subsequent hydrogenation or ring cleavage.
147
Fig. 1. Classical "upper pathway" for phenanthrene ring cleavage (adapted from Refs. [7, 8])
showing stepwise oxidation and metabolites from the "aromatic activation" [6] and BioARC
[9] processes. Theoretical products after hydrogenation are indicated. Compounds: I,
phenanthrene; II, cw-3,4-dihydroxy-3,4-dihydrophenanthrene; III, 3,4-dihydroxyphen-
anthrene; IV, 2-hydroxy-2#-benzo[h]chromene-2-carboxylic acid; V, fra«.s-4-(l-hydroxy-
naph-2-yl)-2-oxobut-3-enoic acid; VI, l-hydroxy-2-naphthoaldehyde. Enzymatic steps: A;
aromatic dioxygenase; B, dihydrodiol dehydrogenase; C, extradiol dioxygenase; D,
isomerase; E, hydratase-aldolase.
148
Several species of organic nitrogen found in crude oil must be removed

during refining because they form nitrogen oxides when combusted in fuels. The
carbazoles in particular are resistant to removal by conventional catalytic
hydrodenitrogenation, interfere with hydrodesulfurization, contribute to
corrosion of refinery metals, and poison refinery catalysts [3]. Research has
been initiated into biological denitrogenation, focusing on carbazole as the
model compound because it is the most abundant species and the most
recalcitrant to hydrotreatment. A theoretical approach to bio-denitrogenation is
angular dioxygenation of carbazole, cleaving the heterocycle to yield an
aromatic amine with side groups (e.g., compounds III and IV, Fig. 2).
Depending on the extent of subsequent metabolism, the product(s) would either
be polar enough to be water-washed from the oil, thus reducing its nitrogen
content and generating a metabolite stream for separate processing, or would
remain associated with the oil phase presumably to be subjected to
hydrogenation as a less problematic feedstock.
Key to all of these hypothetical approaches is complete recovery of the
carbon skeleton without loss to either CO2 or biomass. This can be achieved by
using a pre-grown biocatalyst with a truncated enzymatic pathway so that the
biocatalyst cells do not use petroleum hydrocarbon for biomass or oxidize it to
CO2. These three processes are described more completely in Section 5 and
factors influencing their application to biological upgrading are discussed in
general terms in Sections 1.3 to 4.
Bio-processing of aromatic constituents of petroleum may be a feasible
adjunct process, not a replacement, for conventional upgrading processes. It is
possible that some bio-processing applications would be suitable for use in the
field as crude oil is recovered and when it is already in intimate contact with
production water, or applied at the refinery in holding tanks as long as rapid
throughput was not a requirement. It would produce value-added fuels from
crude oils or decreased aromaticity middle distillates at a reduced environmental
cost by lowering the energy expenditure per cubic metre of feedstock. It may
also yield products with a lower nitrogen and sulfur content by sensitizing
heterocycles to subsequent hydrogenation. These improvements would reduce
the contribution of fuel processing and combustion to greenhouse gas emissions
and acid rain, and could be economically feasible if applied appropriately.
1.3. Substrates for aromatic bio-processing

1.3.1. Aromatic hydrocarbons and heterocycles
Crude oils and most refined products contain complex mixtures of aromatic
hydrocarbons and heterocycles. Unsubstituted aromatics include benzene,
naphthalene and phenanthrene, representing the mono-aromatics and di- and
149
tricyclic polyaromatic hydrocarbons (PAH) respectively, as well as higher

molecular weight PAH. Homologous series of alkyl-substituted aromatics are
also present, such as toluene and the isomers of xylene, methyl- and
ethylnaphthalenes [10]. Aromatic heterocycles with N-, S- and O-substitutions
are represented by carbazole, dibenzothiophene and dibenzofuran, among
others, and these also occur as families of alkyl homologues.
Fig. 2. Classical angular attack on carbazole showing main pathway (adapted from Ref. [11])
and some minor products [12]. Compounds: I, carbazole; II, postulated intermediate; III, 2'-
aminobiphenyl-2,3-diol; IV, 2-hydroxy-6-oxo-6-(2'aminophenyl)-hexa-2,4-dienoic acid; V,
2-hydroxypenta-2,4-dienoic acid; VI, anthranilic acid. Other metabolites selected from Ref.
150
[12].
Currently, aromatic bio-processing has focused on the di- and tricyclic
homologues in crude oil and middle distillate fuels (like diesel) for biological
reasons. The monoaromatics can present severe toxicity problems for biocatalyst
cells at quite low concentrations [13]. At the other extreme, aromatics larger
than three fused rings tend to be quite recalcitrant to biological treatment [1] and
therefore are unfavorable substrates for bio-processing. The chemical reason for
targeting di- and tricyclic aromatics is that conventional hydrotreatment yields
cycloalkanes which still have low fuel value compared with straight chain
alkanes; opening one or more aromatic rings to produce side-chain alkyl groups
should improve the fuel value of the products.
The alkyl-substituted homologues exhibit varying susceptibilities to
biological oxidation, with the general rule that increasing molecular weight and
substitution decrease susceptibility to biological attack (e.g., see Refs. [14, 15]
and discussion in Ref. [16] on alkyl substitution of dibenzothiophenes). A
practical biocatalyst must achieve ring opening of many or all of these
substituted aromatics in addition to the unsubstituted parent compounds (i.e.
have a broad aromatic substrate range), but not oxidize non-target hydrocarbons
in the feedstocks such as alkanes.
1.3.2. Suitable feedstocks

An ideal feedstock for aromatic bio-processing has low viscosity and a high
content of di- and tri-cyclic aromatic hydrocarbons with low alkyl substitution.
Experimentally, this has been achieved by preparing "model oils" comprising
pure substrates (such as phenanthrene and dibenzothiophene) dissolved in an
aliphatic carrier such as «-hexadecane, heptamethylnonane [17] or light mineral
oil [18], or by dissolving model compounds in authentic feedstocks such as
middle distillates [19] or crude oil [20]. However, most authentic feedstocks are
less than ideal for bio-processing due to factors including viscosity, average
molecular weight and degree of alkyl and heteroatom substitution. Haus et al.
[21] determined the kinematic viscosity of aliphatic base oils containing a wide
range of aromatic and polar compounds, and found that biodegradability was
closely related to PAH content and decreased with increasing kinematic
viscosity (the absolute viscosity of a liquid divided by its density). Bio-
processing would certainly be affected in the same manner. Conversely, low
viscosity feedstocks with a high content of low molecular weight hydrocarbons,
such as «-alkanes <n-C6 and monoaromatics such as benzene and toluene, may
be toxic to the biocatalyst [13]. In laboratory trials, Kotlar et al. [22] tested two
biocatalysts with a 50% feed of a light gas oil and found no toxicity. In contrast,
Wu et al. [19] noted toxic or inhibitory effects from some authentic distillates
and a light coker gas oil. The inhibition associated with the aromatic fraction of
the distillate could be alleviated somewhat by dilution with the aliphatic carrier
151
heptamethylnonane. A decrease in ring cleavage of model compounds was also

noted with two crude oils compared with non-toxic aliphatic carriers [20]. Thus,
predictably, viscosity and toxicity effects are feedstock- and biocatalyst-specific.
Increasing molecular weight and substitution generally make aromatic
compounds more recalcitrant to microbial attack [14, 15]. Heterocyclic
aromatics can be more stable to microbial biodegradation than their hydrocarbon
analogues, and aromatics with a high degree of alkyl substitution will be
inherently more resistant to biocatalysis than unsubstituted aromatics.
Unfortunately, we are not yet at the stage where the combined effects of a
feedstock's physical and chemical properties can be used to predict its
susceptibility to bio-processing.
2. BIO-PROCESSING OF CRUDE OILS AND DISTILLATES IN OIL-

WATER SYSTEMS
Successful transformation of water-immiscible feedstocks by an aqueous

suspension of biocatalyst cells requires efficient mixing of the water and oil
phases. In actuality, these "two-phase" systems comprise four phases: the two
immiscible liquid phases, a gas phase providing molecular oxygen for the
biocatalysis, and a solid phase representing the biocatalyst cells, either in
suspension or immobilized on a solid support. This reaction mixture has
complex mass transfer kinetics for substrate delivery between liquid-liquid and
solid phases, and for gas-liquid transfer for provision of oxygen.
Maximizing the available interfacial surface area is crucial regardless of
whether the biocatalyst cells require direct contact with the oil droplets or
whether they transform dissolved substrate. In the first case, intimate contact of
hydrophobic biocatalyst cell surfaces (Section 3.3.1) and the oil droplets must be
established at the oil-water interface, and transformation is therefore dependent
upon the available surface area. In the second case, hydrophilic cells encounter
their substrates via partitioning of the hydrocarbons from the oil droplet into the
aqueous phase with subsequent uptake by the cells. This mass transport is also
dependent on oil droplet size (i.e., surface area) and, implicitly, on the water
solubility of the substrates [23]. Unfortunately, the more water soluble aromatic
hydrocarbons (e.g., toluene) also tend to be the most toxic [24], counteracting
the more favorable partitioning coefficients.
2.1. Bioreactor systems

Whether aromatic bio-processing is intended for application at the
refinery or in the field, the reactor system must address two major issues:
providing maximum surface area to minimize mass transport problems, and
providing adequate oxygen to the reaction [25]. However, another potential
problem is associated with the former requirement, namely formation of
152
emulsions and ultimately the separation of the two liquid phases and the
biocatalyst. In fact, many of the same technical problems previously reported
with bench-scale and pilot-scale biodesulfurization processes will likely be
encountered with aromatic ring oxidation. Several types of reactors may be
suitable for petroleum biocatalysis (reviewed in Ref. [26]), with the choice
depending on process economics and product recovery considerations. Two
reactor types are described here: conventional stirred tank bioreactors and
electro-spray reactors.
2.1.1. Two-liquid-phase stirred tank bioreactors

The efficient mixing essential for bio-processing can be achieved in
conventional "two-liquid-phase" stirred tank bioreactors, such as those
described by Villemur and Daugulis and their co-workers [27, 28, 29]. Whereas
other applications of two-liquid-phase bioreactors may require the addition of a
non-degradable, water-immiscible, biocompatible carrier phase to dissolve the
substrate [27], with petroleum the feedstock itself is the solvent for the aromatic
substrates and there should be no need for addition of carrier phases. However,
for toxic feedstocks, it may be necessary to consider diluting the feed with a
non-toxic (e.g. aliphatic) carrier. In this case, it is even more important to ensure
that the biocatalyst cannot oxidize or grow on non-target components of the
feedstock or diluent. Both batch-fed and continuous loop bioreactors may be
suitable for bio-processing, but the former are easier to set up and manipulate.
Oxygen required for biocatalysis is provided by aeration of the two liquid
phases, and therefore may be mass transfer-limited.
Although chemical and microbial surfactants and emulsifiers can increase
the efficiency of droplet formation and stabilize those droplets during
biocatalysis, Marcoux et al. [28] reported that addition of surfactants to a two-
liquid-phase bioreactor did not enhance conversion of high molecular weight
PAH, and rhamnolipids had an inhibitory effect. Another consideration
associated with use of surfactants is whether they will interfere with subsequent
separation of the liquid phases or removal of the biocatalyst cells.
To reduce operational costs associated with water handling, separation,
and disposal, ideally the volume ratios of oihwater should be maximized.
However, this can lead to additional problems of emulsification (e.g., water-in-
oil emulsions), hindering post-treatment separation of the phases. Therefore,
optimum volume ratios likely would have to be determined experimentally for
specific feedstocks and biocatalysts, and emulsions may have to be broken by
centrifugation or by hydrocyclone, such as the patented process for separating
three-phase systems (oil/water/biocatalyst) [30].
153
2.1.2. Electro-spray reactors (emulsion-phase contactors)

An alternative to mechanically mixed bioreactors is the electro-spray
reactor, or emulsion-phase contactor, which creates a shear force between two
immiscible liquids by exploiting the difference in their electrical conductivities
[31]. The shear force disperses the conductive phase (the aqueous suspension of
biocatalyst) into the non-conductive phase (the oil), creating an emulsion with a
droplet size of less than 5 um, using significantly less energy than conventional
stirrers and enhancing mass transfer of oxygen to the liquid phase [32]. This
efficient mixing has the advantage of maximizing surface area for effective mass
transport while minimizing the volume of water required.
Despite these theoretical advantages, an experiment comparing the
performance of electro-spray and batch-stirred bioreactors showed that
biodesulfurization by Rhodococcus erythropolis IGTS8 was equivalent in both
systems, presumably because the biocatalyst produced a surfactant that
augmented the effects of mechanical mixing [33]. Electro-spray reactors may
prove more suitable for use with biocatalysts that do not produce biosurfactants.
However, an additional consideration is the technical difficulty in scaling up
electro-spray reactors to process-scale volumes, compared with conventional
mixers that are more amenable to scale-up.
2.2. Biocatalyst immobilization

The type of bio-processing application chosen will dictate whether
immobilization of the biocatalyst is required. Immobilizing the cells increases
ease of handling for delivery and recovery of biocatalyst, whether simply for
separation from the treated product or for biocatalyst re-use. The immobilizing
matrix must be resistant to the solvent effects of the feedstock and must not
irreversibly adsorb either the substrates or the products, while being sufficiently
mechanically robust to withstand the vigorous mixing required to maintain small
droplet size.
Several biocatalyst support systems are well-known, such as alginate, K-
carageenan, and polyacrylamide, used in numerous biotechnology applications.
Porous nylon [34], resin and urethane pre-polymers [35], and other polymers
[36] have also been described for biodegradation or bio-processing applications.
2.3. Pure culture biocatalysts versus mixed cultures

This review considers only the use of pure culture biocatalysts for bio-
processing, although in theory mixed cultures could be used simultaneously or
sequentially to achieve a series of desirable reactions. For example, a less
conventional approach to bio-processing might include a system similar to one
described by Munoz et al. [37] in which a consortium of an alga and a bacterium
degraded phenanthrene under photosynthetic conditions in a two-phase
partitioning bioreactor without an exogenous supply of oxygen. Another
154
consortium was described in which two PAH degraders incubated together

oxidized higher molecular weight PAH than either alone [38]; this would
expand the substrate range for aromatic bio-processing. Certainly,
biodegradation of complex hydrocarbon mixtures typically relies on microbial
consortia to achieve substrate removal, and there are many instances of guilds of
microbes acting synergistically to degrade aromatic compounds. Currently it
seems impractical to attempt to coordinate biocatalysis by more than one
organism, especially if genetic manipulation is required to prevent
mineralization (Section 4); however, this may be a research direction for the
future.
2.4. Anaerobic processes

Although anaerobic transformations of aromatic hydrocarbons have been
reported for monoaromatics like toluene and w-xylene (e.g., Ref. [39]) and PAH
like naphthalene [40] and phenanthrene [41], at present they are not suitable for
aromatic bio-processing for several reasons: the known anaerobic
transformations are very slow compared with aerobic reactions, relatively few
pure cultures have been reported for anaerobic transformation of di- and tri-
cyclic aromatics or heterocycles (often mixed populations in long-term
anaerobic enrichments are used), correspondingly few catabolic pathways have
been elucidated, and the genetic tools for manipulating such anaerobes are often
cryptic. For these reasons, the potential for anaerobic bio-processing is not
considered in this review, although in the future it may become an attractive
process for long-term in situ ("down-hole") treatment where long residence
times are possible.
2.5. Other process considerations

Several other aspects of aromatic bio-processing will need to be addressed
before a commercial process is developed. Some will be process-specific while
others are common to any large-scale industrial microbiological application. On
the biological side, problems of efficient large-scale production of active
biomass are common to each potential bio-process, but selection of the
biocatalyst formulation (e.g. cell paste, lyophilized powder, immobilized cells,
aqueous suspension) for ease and safety of handling will differ depending on
whether the process is applied in the field or at a refinery. Similarly, there are
common issues of disposing of waste biomass having high chemical and
biological oxygen demand (COD and BOD), but the specifics will differ
depending on the location of the processing unit. Longevity of the biocatalyst
and its suitability for regeneration and re-use also will be application-specific.
On the engineering side, safety issues have been raised for operation of
two-liquid-phase reactors, especially at elevated temperatures with flammable
organic phases or at elevated pressure to provide oxygen to the reaction [42].
155
For the BioARC process, and possibly also for bio-denitrogenation, problems of
separating and recovering the polar products from the aqueous phase must be
addressed, as well as their treatment after recovery (e.g. chemical hydrogenation
or further bio-processing to produce specialty chemicals).
3. ATTRIBUTES OF AROMATIC BIO-PROCESSING BIOCATALYSTS
Aromatic ring cleavage, as exploited in the BioARC process [17, 19], and
hydroxylation, as used in the patented "aromatic activation" process [6], require
multiple, sequential enzymatic steps and various co-factors such as NADH and
FMNH2. Whole cells are the most readily applicable biocatalysts to encompass
these requirements, rather than treatment with a series of pure enzymes.
Fortunately, there is a relatively long history of research into biodegradation of
aromatic hydrocarbons and heterocycles for purposes of bioremediation, which
can be tapped for relevant information. Bio-processing of oil can be seen as a
logical extension of natural biodegradation capabilities. The key difference is
that biodegradation aims to achieve complete oxidation of the substrate
(mineralization to CO2), whereas bio-processing should oxidize or cleave the
aromatics without carbon loss. This generally requires truncation or
modification of existing catabolic pathways through genetic manipulation
(Section 4).
Generally desirable attributes of aromatic bio-processing biocatalysts
include:
1. Rapid growth to high density in an inexpensive medium without

hydrocarbon, generating biomass that either has constitutive or
rapidly inducible biocatalytic activity;
2. Rapid, robust, and sustained aromatic oxidation by the pre-grown
cells as a "resting cell" suspension in an aqueous phase;
3. A broad range of activity against alkyl-aromatic and -heterocyclic
substrates with few undesirable side-reactions;
4. Lack of activity against non-target substrates such as aliphatic
hydrocarbons;
5. A genetic system that can be manipulated to allow isolation of
specific mutants blocked at the desired enzymatic step, thus
preventing loss of carbon and producing the desired oxidized
product;
6. Stability and predictability of the phenotype and the mutation(s)
regardless of whether the catabolic genes are plasmid-borne or
chromosomal;
7. Activity in two-liquid-phase reaction systems where the substrate is
in the water-immiscible phase at a high volume ratio to the cell
156
suspension;
8. Tolerance towards any toxic effects of the feedstock or biocatalytic
products;
9. Non-pathogenicity of the biocatalyst for safe handling and disposal;
10. The ability to scale up active biomass production to commercial scale
operations.
3.1. Examples of candidate bacterial biocatalysts

There is no shortage of bacterial genera that are potential candidates for
achieving aromatic ring cleavage or hydroxylation. A brief list of selected
examples is given in Table 1; further information is reviewed in Ref. [43]. These
strains possess demonstrated activity against PAH and heterocycles, although
their genetic systems may not at present be sufficiently well-known to generate
the desired pathway mutations. Among the candidates are Gram negative genera
such as Pseudomonas and Gram positive genera such as Rhodococcus.
Pseudomonads have long attracted interest as versatile biocatalysts [44],
commonly producing oxygenases that attack hydrocarbons. The rhodococci,
especially those using the "4S" pathway [45], have been of more interest in
biodesulfurization (see also Chapter 2), but in general their genetics are less well
characterized.
Biocatalysts having a hydrophilic cell surface (e.g. Pseudomonas spp.)
depend upon mass transport of PAH from the oil phase to the aqueous phase.
With highly water-insoluble substrates (e.g., larger PAH), this can be a rate-
limiting factor. In contrast, bacteria with hydrophobic surfaces (e.g.,
Rhodococcus erythropolis IGTS8 [45]) may achieve intimate contact with the
substrate by attaching to the oil:water interface and may take up the PAH
substrates directly from the oil droplets. This process has the advantage of
reducing limitations due to substrate mass transport but may stabilize water-in-
oil or oil-in-water emulsions [46], potentially making it difficult to separate the
oil and biocatalyst after treatment. Similarly, bacteria that produce surfactants
may not be suitable as biocatalysts. Although their extracellular products may
enhance droplet formation and "pseudosolubilization" during biocatalysis, they
may also stabilize emulsions, hindering final separation of the treated oil and
aqueous phase.
There is potential for extremophiles, particularly thermophiles, to be used
for bio-processing [16]. The advantages of working at higher bioreactor
temperatures include reduced feedstock viscosity (to improve mixing) and being
able to process the oil as it comes out of the field (often at high temperature)
without having to cool it to a suitable temperature. However, the same
limitations of mass transport and phase separation apply. Salt tolerance is
another consideration in some field applications.
Numerous fungal species produce extracellular PAH-oxidizing laccases
157
and peroxidases (e.g., [47, 48]), and may be suitable for the "aromatic
activation" process. However, fungi that accumulate quinones or trans-
dihydrodiol-PAHs without ring cleavage would not be suitable for the BioARC
process. This chapter does not consider the fungal enzymes, which are reviewed
in Chapter 3.
Table 1
Bacteria with broad aromatic substrate ranges for oxidizing di- and tri-cyclic aromatic
hydrocarbons and heterocycles. This table is not a comprehensive list, but rather an
indicator of the diversity of genera with potential for aromatic bio-processing.
Genus, species and strain names Substrates* References
Gram negative
Alcaligenes denitrificans WW1 N, MeN, P, A [49]
Burkholderia sp. RP007 N,P,A [50]
Comamonas testosteroni N,P [51]
Cycloclasticus pugetii N,P, A [52]
Cycloclasticus sp. A5 N, P, F, D [53]
Neptunomonas naphthovorans NAG-2N-113 N, MeN, P, Ac [54]
Pseudomonads
Pseudomonas putida G7 (NAH7) N,P,A reviewed in [7]
Pseudomonas fluorescens LP6a N, MeN, P, A, F, D [9, 55]
Pseudomonas resinovorans CA10 C, Df [56]
Ralstoniasp. RJGII.123 C [57]
Sphingomonads
Sphingomonas paucimobilis EPA505 N, P, A [58, 59]
Sphingobium yanoikuyae B8/36 N, P, A, Ac, D, C [6]
Sphingomonas sp. ANT 17 N, MeN, P, F, D [60]
Novosphingobium aromaticivorans F199 MeN, P, Ac, F, D, [61,62]
Xanthomonas ampellina C [63]
Gram positive
Bacillus sp. N,P, A [64]
Microbacterium esteraromaticus B21 P [65]
Mycobacterium sp. PYR-1 P,A [66]
Mycobacterium gilvum Bl N, P [65]
Nocardioides sp. KP7 P [67]
Porphyrobacter sp. B51 P [65]
Rhodococcus sp. NCMB12038 N, Ie, Io [68]
Terrabacter (Staphylococcus auriculans) sp. F, Df [69]
DBF63
* N, naphthalene; Me-N, methylnaphthalenes; P, phenanthrene; A, anthracene; Ac,
acenaphthene; F, fluorene; D, dibenzothiophene; C, carbazole; Df, dibenzofuran; Ie,
indene; Io, indole.
158
Suitable new candidate biocatalysts surely exist and could be enriched and
isolated from suitable environments where natural selection processes favor
desirable catabolic phenotypes. Using conventional microbial methods,
enrichment can be facilitated by judicious application of additional selective
pressures in the laboratory, with subsequent genetic manipulation to remove
undesirable properties or add novel phenotypes. To identify a new potential
biocatalyst it may be adequate to screen the mixed community for the desired
genetic complement by DNA:DNA hybridization or PCR, followed by chemical
confirmation of catabolism. However, isolation of truly novel biocatalysts (i.e.,
those with little homology to known aromatic catabolism genes; Section 3.2)
requires classical microbiological selection and biochemical screening
techniques. Notably, Kilbane et al. [70] found that simple chemostat and shake-
flask enrichments of novel bacteria capable of selectively removing the N
heteroatom from quinoline were unsuccessful, and that repeated rounds of
enrichment and mutagenesis were required to select a strain with the desired
catabolic activity.
3.2. Examples of genetic systems for aromatic oxidation

Because of the wide variety of potential bacterial biocatalysts, there is
also a diverse array of genetic systems that encode and regulate aromatic
hydroxylation and ring cleavage activity. In general, the aromatic catabolic
genes are regulated as one or more operons that may be encoded on very large
plasmids (e.g. Novosphingobium (formerly Sphingomonas) aromaticivorans
F199 plasmid pNLl, 184 kb [62]), medium-sized plasmids (e.g., Pseudomonas
putida NCIB 9816-4 plasmid pDTGl, 84 kb [71]) or on the chromosome (e.g.,
Mycobacterium sp. PYR-1 [72] and Sphingomonas sp. ANT 17 [60]). Many of
these bacteria use degradative pathways analogous to that shown in Fig. 1, but
there are increasing numbers of reports of new pathways, such as anthracene
degradation by Mycobacterium sp. LB501T [73]. Unfortunately, detailed genetic
characterization of most of these novel pathways remains to be addressed and
the bulk of the literature deals with descriptions of "re-discovered" classical
PAH catabolic genes.
3.2.1. Gene organization and homology

The genes for PAH oxidation are commonly organized into "upper" and
"lower" pathway operons (Fig. 3), which may be separated by many kilobases
[74]. Using the archetypal example of the naphthalene degradation genes
encoded on plasmid NAH7, the upper pathway {nah operon) comprises
oxidation of naphthalene to salicylate, and the lower pathway {sal operon)
encodes oxidation of salicylate to central metabolites (reviewed in Ref. [71];
Fig. 3). It is the upper pathway genes that are important for the aromatic bio-
processing described here.
159
Functional analogues of the nah genes have been described, including the
ndo [75], dox [76] and pah [77] genes in Pseudomonas spp., phb genes in
Sphingomonaspaucimobilis EPA505 [5S],phn genes in Burkholderia sp. RP007
[50], phd genes in Comamonas testosteroni strains [51] and Nocardioides sp.
KP7 [67], nag genes in Ralstonia {Pseudomonas) sp. U2 [78], and nid genes in
Mycobacterium sp. PYR-1 [72]. Some of these operons are genetically
homologous to the nah genes (nah-\ike), whereas others are unconventional
(non-nah-like) in that their sequences and (or) gene arrangements differ from the
classical nah genes (see comprehensive review in Ref. [7]). For example, some
dox and ndo genes are virtually identical [76], whereas the phdA gene is only
60% homologous to other nah dioxygenase genes [67] and the phn genes not
only show little homology with archetypal ndo genes but are also organized
differently within the operon [50]. In the case of the sphingomonads (reviewed
in Ref. [59]), Sphingobium (Sphingomonas) yanoikuyae Bl and N.
aromaticivorans F199 harbour aromatic catabolism genes that are only distantly
related to the well-studied pseudomonad PAH catabolic genes but appear to be
conserved among other sphingomonads. Lack of cross-hybridization and
sequence similarity among non-homologous PAH-degrading organisms [79, 80]
may hinder molecular screens to identify new, unconventional candidate
biocatalysts in the environment.
In some bacteria the catabolic gene organization is considerably more
complicated than the «a/z-like operons. For example, the chromosomal genes for
aromatic catabolism in several sphingomonads are disjointed and redundant
[59], possibly resulting from repeated lateral gene transfers [7]. Aromatic
catabolism in S. yanoikuyae Bl involves six operons with convoluted pathways
for mono- and polycyclic aromatic catabolism [83]. Some catabolic operons are
flanked by insertion elements or associated with transposons [56, 84], forming
potentially mobile "catabolic cassettes". This could account for their apparent
lateral transfer and occurrence in diverse bacteria, but also has implications for
the genetic stability of some candidate biocatalysts. Repeated insertion into large
plasmids or the chromosome may also account for the presence of multiple
isofunctional enzymes; such redundancy may complicate efforts to engineer
mutants blocked at specific enzymatic steps (Section 4).
Much less has been published about Gram positive PAH-degrading
bacteria, such as Mycobacterium sp. PYR-1 which has both dioxygenase and
monooxygenase activity against PAH and an unusual gene order in the catabolic
operon [72]. The aromatic dioxygenase genes in this strain cluster
phylogenetically with other Gram positive dioxygenase genes, including
Rhodococcus and Nocardioides spp. [72], rather than the well-studied Gram
negative nah genes.
160
Fig. 3. A. Generalized operon organization for classical nah-Mke genes, adapted from Refs. [7,
71].The arrows indicate the direction of transcription, and shading indicates genes for related
enzymes. Aabcd encode the initial dioxygenase; B the dehydrogenase, C the extradiol
dioxygenase, D the isomerase, E the hydratase-aldolase, F the salicylaldehyde dehydrogenase.
Q is uncharacterized. "R" is the divergently transcribed regulatory gene described for P.
putida NAH7 [81]. G, H, I, N, L, J, and K comprise the lower pathway operon for salicylate
degradation, which may be separated from the upper operon by several kilobases. B. Gene
organization of the Pseudomonas sp. LD-2 carbazole catabolic operon, adapted from Ref.
[82]. Genes for related enzyme subunits are indicated by shading, car A genes encode the
multi-subunit terminal oxygenase, carB the weto-cleavage enzyme, and carC the hydrolase.
Although we are progressing towards being able to predict chemical

pathways of microbial degradation [85], anticipating and discovering
unconventional genes encoding this activity will be difficult until we have more
paradigms to draw upon.
3.2.2. Catabolic gene regulation

The nah and sal operons (encoding the upper and lower naphthalene
catabolic pathways) are coordinately controlled in a manner analogous to the
well-studied toluene catabolic genes, as might be expected from their presumed
evolutionary relationship [86]. The NahR regulatory protein transcribed
divergently and constitutively from the lower operon (Fig. 3) is an activator for
both operons [87, 88]. Salicylate, the substrate of the lower pathway operon, is
161
the natural inducer of both operons [81], with 2-aminobenzoate (anthranilic

acid) being a gratuitous inducer [89]. The cloned nah operon can be expressed
and appropriately regulated in Escherichia coli, but its uninduced expression is
lower by an order of magnitude [81]. This has implications for the creation and
use of recombinant biocatalyst strains (Section 4).
Whereas the regulation of genes for catabolism of aromatics like toluene
and naphthalene has been studied extensively, regulation of the unconventional
PAH genes has been less well described and in some cases is unknown except
by inferred analogy to the nah system. This may not be appropriate, and in the
case of catabolic genes dissimilar to the well-known nah-like genes, regulation
is often still cryptic. The dissimilar gene organization of many non-nah-\ike
catabolic operons implies that regulation will differ from the naphthalene and
toluene paradigms. Induction similarly varies among species and operons. For
example, Romine et al. [62] reported that more than one aromatic compound
was required to induce mineralization of naphthalene by N. aromaticivorans
F199. They hypothesized that the gene clustering and complex regulation in this
organism resulted in the ability to simultaneously degrade multiple aromatic
substrates. S. paucimobilis EPA505 has at least two hydrocarbon catabolic gene
promoters that are induced by a partially overlapping suite of aromatic
compounds and some metabolites [58]. Although the catabolic abilities of such
organisms are impressive, this advantage currently may be outweighed by the
sophistication of genetic manipulation required to ensure control over those
activities. Similarly, Marcoux et al. [28] found that addition of low molecular
weight PAH to a mixture of high molecular weight PAH had a greater effect on
oxidation than did addition of known inducers such as salicylate and benzoate.
However, this report contrasts with later observations by the same group [65]
who noted inhibitory effects of naphthalene on degradation of higher PAH by a
consortium. Such cross-talk of inducers and inhibitors is important for
chemically complex feedstocks and products; therefore, the most effective
inducer(s) should be experimentally determined for each biocatalyst in the
context of the feedstock to be upgraded.
In the case of the monoaromatics at least, catabolite repression and
regulation by growth-dependent sigma factors are superimposed on operon-
specific regulation (e.g., [90]). Global regulation of PAH degradation has only
recently been addressed for some organisms (e.g., [62]), although it could be
important for fine-tuning biocatalytic activity and for achieving repeated re-use
of whole cell biocatalysts.
3.3. Cellular factors influencing suitability

3.3.1. Substrate uptake
Several factors inherent to the biocatalyst will affect its rate of activity in
a two-liquid-phase system, including substrate uptake and efflux. Uptake of
162
aromatic hydrocarbons has been proposed to occur through different

mechanisms for different genera. As noted in Section 3.1, biocatalysts with
hydrophobic cell surfaces may attach directly to liquid hydrocarbons [45]. In
contrast, bacteria that utilize dissolved phase hydrocarbons are limited by mass
transfer of substrate as it partitions from the organic phase to the aqueous (e.g.,
[23, 91]) and then to the bacterial membrane. Biosurfactants are produced by
some hydrophilic and hydrophobic genera to aid in uptake by
pseudosolubilization and to enhance partitioning kinetics by increasing the
effective surface area of liquid hydrocarbons in an aqueous phase.
In general, uptake of aromatic hydrocarbons into biological membranes
has been deemed a passive process [92] influenced by the substrate's
octanol:water partition coefficient (log Kow). For example, studies with wild type
biocatalyst P. fluorescens LP6a indicated that uptake of radiolabelled
phenanthrene dissolved in the aqueous phase was accomplished by passive
diffusion [93]. However, recently Miyata et al. [94] provided evidence that
Mycobacterium sp. RJGII-135 demonstrated saturable, active (i.e., energy-
dependent) uptake of phenanthrene. Thus, there are different modes of aromatic
uptake and the energy requirements of active hydrocarbon transport may be a
factor in the efficiency and half-life of a resting cell biocatalyst.
3.3.2. Solvent resistance

The di- and tricyclic aromatic and heterocyclic substrates intended for
bio-processing have high octanol:water (log K<,w) partition coefficients
(summarized in Ref. [95]) and therefore generally are not considered to be toxic
to bacteria (reviewed in Ref. [13]). However, the feedstocks may contain
solvents like toluene and short-chain alkanes. Such hydrocarbons are usually
toxic to microbes because they damage biological membranes, thus disrupting
essential functions [13]. Several mechanisms of intrinsic resistance to such
membrane solvents are known and have been reviewed (see chapter 12)[96].
These include: rapid adjustment of membrane fatty acid composition upon
exposure to solvents, through existing enzymatic activity; longer-term
production of altered membrane lipids more resistant to solvent effects;
formation and sloughing of vesicles containing solvent; increased metabolism of
the solvent when possible; and export of aromatic hydrocarbons achieved by
transmembrane proteins called efflux pumps.
In the latter case, the hydrocarbon flux out of the cell must balance or
exceed partitioning into the cell to confer solvent tolerance. Most hydrocarbon
efflux pumps described to date confer resistance to small aromatic compounds,
particularly toluene, and have varying degrees of homology with antibiotic
efflux pumps. An efflux pump able to transport tri-cyclic aromatics (e.g.,
phenanthrene, anthracene, fluoranthene) has also been described in the wild type
biocatalyst Pseudomonas fluorescens LP6a [93, 97]. Such efflux activity was
163
unexpected, as these non-toxic compounds are carbon sources for the organism.
Why the organism would develop such an efflux system is unknown; it is
possible that the hydrocarbons are gratuitous substrates rather than primary
substrates for the pump. The effect of efflux on aromatic biocatalysis has not yet
been investigated in this organism or any other to date.
3.4. Enzyme specificity

Biocatalysts for aromatic bio-processing must be capable of attacking the
diverse aromatic substrates present in feedstocks, including alkyl- and
heterocyclic homologues. Two general modes of attack occur: oxidation by a
limited number of enzymes with broad substrate range (e.g. P. fluorescens LP6a
[55]), or by suites of enzymes, each with a narrow substrate range but with
overlapping, complementary functions (e.g. S. yanoikuyae Bl [83]).
The details of naphthalene degradation have been known for some time,
whereas the genetic basis for catabolism of phenanthrene was initially unknown.
Menn et al. [98] were the first to show that phenanthrene and anthracene were
also substrates of the naphthalene oxidation pathway, confirming earlier
speculation about broad specificity degradation pathways for aromatic
hydrocarbons and heterocycles (e.g., Ref. [99]). Evidence of broad substrate
specificity has been reported with increasing frequency, with the most
comprehensive inventory having been described for the naphthalene
dioxygenase (NDO) of Pseudomonas sp. NCIB9816 [100]. This multi-subunit
enzyme produces m-dihydrodiols from a wide array of aromatic hydrocarbons
and heterocycles as well as monohydroxylating, desaturating, dealkylating or
sulfoxidizing various aromatic substrates. Depending upon the biocatalytic
approach being used, the substrate ranges of the other enzymes in the pathway
may also be important but they are poorly characterized [8]. Unfortunately, the
substrate ranges of the unconventional dioxygenases have not yet been fully
explored.
Some strains demonstrate a highly restrictive substrate range. For
example, Rhodococcus sp. SI oxidizes only anthracene and 2-methylanthracene
[101] and Nocardioides sp. KP7 grows on phenanthrene but not naphthalene
[67]. In contrast, other bacteria exhibit a broad substrate range (Table 1)
including alkyl and heteroatom substitutions. Interestingly, the substrate range
can vary even between similar catabolic genes: the homologous nah upper
pathway genes encoded on plasmids found in P. putida PpG7 (NAH7) and
NCIB9816 have differing abilities to oxidize 2- and 3-ring PAHs [102].
The ability to attack alkyl homologues of aromatic substrates varies with
the degree and position of substitution. Some bacteria attack only the
unsubstituted aromatic ring whereas others may also indiscriminately oxidize
the alkyl group or the heteroatom [100], generating alcohols, ketones,
sulfoxides, sulfones, etc. These undesirable enzymatic activities may be suitable
164
for site-directed mutagenesis or gene knockout strategies (Section 4) to improve

biocatalyst specificity and eliminate side-reactions.
One of the consequences of having a broad substrate range is the potential
for competitive inhibition of the enzyme by multiple substrates. This will effect
slower kinetics of transformation of individual substrates compared with pure
substrates, as observed by Kobayashi et al. [103] for biodesulfurization of
mixtures of alkyldibenzothiophenes. Another factor is substrate inhibition of the
degradative enzymes. In at least one case, a dihydroxytetralin extradiol
dioxygenase (analogous to step C, Fig. 1) is known to suffer from inhibition by
its own substrate, through an unknown mechanism(s) [104].
Therefore, the potential biocatalytic advantage conferred by broad
substrate specificity may be outweighed by the task of understanding and
manipulating complex genetic systems (gene redundancy and cryptic
organization and regulation) and predicting or preventing potential enzyme
inhibition in the face of mixed aromatic substrates and an even greater
multiplicity of possible products. Until there is more literature available to
elucidate these effects, selection and testing of potential biocatalysts requires
specific experimentation for each application.
4. GENETIC ENGINEERING TO IMPROVE BIOCATALYSTS
For genera or species that are genetically tractable, molecular engineering is an

option for biocatalyst improvement. For some species the molecular tools are
not yet available (e.g., shuttle vectors, transformation and conjugation protocols
for cloning or transposon mutagenesis). Feasible manipulations include pathway
interruption to yield a desired product, expression of catabolic genes in hosts
having solvent tolerance or thermotolerance, escape from feedback inhibition or
catabolite repression, overexpression as a result of increased gene copy number
or improved induction or expression, and the possibility of expanded substrate
range by gene shuffling and site-directed mutagenesis.
4.1. Mutation, cloning and recombinant strains

To prevent the biocatalyst from completely oxidizing aromatic substrates,
it may be necessary to interrupt the substrates' degradative pathway(s). In
simple terms, this can be done by mutating genes for enzymes that act
downstream of the desired product by deletion or insertional inactivation, or by
cloning only the desired genes into a host lacking the complementary functions
of the pathway, or by a combination of the two approaches. It may be desirable
in some cases to create recombinant strains by cloning specific catabolic genes
into host strains with desirable properties such as solvent resistance (see chapter
12). Two examples of these approaches are given below.
To generate the biocatalyst used in the BioARC process, random
165
transposon mutagenesis was performed on wild type P. fluorescens LP6a [55]

using standard methods. The transconjugant with the desired upper pathway
mutation was selected by screening for accumulation of the highly colored ring
cleavage product of dibenzothiophene, analogous to compound V, Fig. 1. The
presumed polar mutation is at step E, Fig. 1. The lower pathway with the NahR
regulatory gene analogue remains intact in the biocatalyst strain, so that
salicylate is a metabolizable inducer of the upper pathway operon. In similar
studies, Kotlar et al. [22] specifically inactivated the nahE gene encoding
naphthalene hydratase-aldolase (analogous to step E, Fig. 1) in S. yanoikuyae
N2, causing it to accumulate aromatic ring cleavage products.
Riddle et al. [82] cloned and subcloned the catabolic genes from the
carbazole-degrading organism Pseudomonas sp. LD2 [12]. The genes encoding
the meta-cleavage and hydrolase enzymes (carBa, carBb, carC; Fig. 3)
catalyzing steps C and D (Fig. 2) were deleted from the fragment, leaving the
multi-subunit carbazole dioxygenase genes {carAal, car Aa2, carAc, carAd;
Fig. 3) intact. The subclone was then transferred into the solvent-resistant,
carbazole non-degrading host P. putida Idaho. The recombinant strain produced
the expected carbazole metabolite 2-aminobiphenyldiol (compound III, Fig. 2)
in two-liquid-phase reactions with carbazole solubilized in 1-methylnaphthalene.
The recombinant strain was active both in the presence and absence of 10%
xylenes (a concentration sufficient to inhibit activity in E. coli). However, the
carbazole oxidation was significantly less than that achieved by the original host
strain with or without xylenes present. Interestingly, the carbazole genes were
not expressed in the related host P. putida KT2440, indicating strain-specific
expression of these catabolic genes. Unpredictable catabolic gene expression has
been noted in classical hosts such as E. coli and even in related host strains (e.g.,
[67, 99]). This is a factor to consider in attempting to generate recombinant
biocatalysts.
4.2. Expansion of substrate range

As more information is published about aromatic dioxygenases and
associated enzymes of the "upper" catabolic pathways for aromatics and
heterocycles, it will become more feasible to contemplate site-directed
mutagenesis to expand biocatalysts' substrate ranges. For example, amino acid
substitutions at the active site of the classical naphthalene dioxygenase of
Pseudomonas sp. NCIB9816-4 revealed some aspects of the enzyme's substrate
specificity [105]. Another recent report [106] identified amino acid residues
associated with substrate specificity of a dioxygenase involved in tetralin
biodegradation.
Some dioxygenases have been shown to have a broad substrate range
[100]. Consequently, it will be valuable to explore the substrate ranges of the
downstream reactions as they may represent "bottlenecks" of substrate
166
specificity. Unfortunately, there is little published information on such enzymes

and their genes, as most research to date has been devoted to the initial oxidation
step.
Some catabolic genes are naturally flanked by insertion elements or are
components of functional or non-functional transposons [56, 84]. This
arrangement may have facilitated evolutionary interchange of catabolic enzyme
subunits, leading to the diverse substrate ranges observed for aromatic
dioxygenases. If an otherwise suitable biocatalyst suffers from narrow substrate
specificity, gene shuffling may be an option to improve the biocatalyst, either by
natural means (selection in chemostats or transformation of suitable hosts) or in
vitro by cloning. For example, Suenaga et al. [107] combined genes for biphenyl
dioxygenase from Pseudomonas pseudoalcaligenes KF707 and Burkholderia
cepacia LD400 to generate a recombinant dioxygenase with an expanded
substrate range.
In summary, for many potential biocatalysts the molecular tools are
available for genetically manipulating their aromatic catabolism genes. The
current and significant stumbling block to tailored aromatic bio-processing is
our incomplete understanding of the diversity and regulation of these genes.
5. EXAMPLES OF AROMATIC BIO-PROCESSING
5.1. BioARC (Biological Aromatic Ring Cleavage)

The BioARC ring opening process utilizes partial oxidation of aromatic
substrates to yield aromatic ring cleavage products analogous to Compounds IV
and V, Fig. 1. At neutral pH these polar products would partition into the
aqueous phase of a two-liquid-phase reaction mixture, to be recovered and
further processed (e.g., by thermochemical hydrogenation to yield alkyl-
aromatics). The bio-processed feedstock and the metabolite stream should then
have lower aromaticity. This concept to date has been tested using Biocatalyst
21-41, a mutant strain of P. fluorescens LP6a [55]. Recently, a recombinant
strain of S. yanoikuyae N2 [22] also has been tested for this activity.
Biocatalyst 21-41 possesses most of the characteristics desirable in an
aromatic bio-processing biocatalyst (Section 3). P. fluorescens is considered
non-pathogenic by Health Canada [108], thus avoiding many potential biosafety
issues. The biocatalyst strain was derived by transposon mutagenesis of wild
type P. fluorescens LP6a [55], which exhibits a broad aromatic substrate range
with the ability to oxidize many di- and tricyclic PAHs and heterocycles [79], as
well as alkyl homologues [9]. This broad substrate phenotype is maintained
stably in the mutant even though the nah-hke catabolic genes are on a 63 kb
plasmid in this strain [55, 79].
To prepare the biocatalyst for use, it is pre-grown overnight without
hydrocarbon to high cell density and induced with either salicylate or 2-
167
aminobenzoate to high activity. The induced culture is harvested for

resuspension in water or phosphate buffer for use as a resting (i.e. non-growing)
cell suspension. The biocatalyst is active in two-liquid-phase systems with
model feedstocks (aromatic substrates dissolved in aliphatic carriers or authentic
distillates [19]) and crude oils [20]. Its activity has been demonstrated in shake-
flasks and bench-scale stirred tank bioreactors [18]. Importantly, it does not
oxidize aliphatic hydrocarbons [9]. The biocatalyst possesses at least one
aromatic efflux pump [97] (Section 3.3.2) although toluene is not a substrate of
this pump and therefore pump activity is not related to classical solvent
resistance. The influence of this pump on PAH biocatalysis activity is currently
unknown. Existing gene knockout mutants will be used to determine whether
suppressing efflux will increase the rate of PAH biocatalysis.
Biocatalyst 21-41 produces the expected ring cleavage products from
dibenzothiophene [109] and phenanthrene as model compounds, as determined
by spectrophotometry and gas chromatography-mass spectrometry. The ratio of
the metabolite(s) produced (e.g. Compounds IV and V, Fig. 1) has not been
determined, as authentic standards are not commercially available. Regardless,
the isomeric composition is not likely to be important during metabolite
recovery or subsequent chemical treatment. In addition to the predicted
products, additional ring cleavage products have been observed, including one
with a molecular mass two units greater than expected (Fig. 4), possibly
representing a partially reduced ring cleavage product [110]. Other side
reactions have been inferred, with oxidation of alkyl substituents to alcohols [9]
and sulfur heteroatoms to sulfoxides [19]. This may be a consequence of the
non-selective oxidations carried out by its Nah-like enzymes [100]. Little
research has been carried out yet on chemical hydrogenation of the observed
oxidation products, and a full study analogous to that of the carbazole oxidation
products ([111], Section 5.3) is needed.
The BioARC process has been considered both for refinery and field
applications, treating distillates (e.g., diesel) or crude oil, respectively. Process
details remain to be optimized regarding biocatalyst immobilization for handling
and separation, re-use and half-life of the biocatalyst, separation of cells from
the liquid phases, and product stream handling as well as determining maximum
feedstock:cell suspension ratios. However, results to date support the potential
of BioARC for petroleum bio-processing.
5.2. "Biological activation of aromatics"

US Patent #6,156,946 [6] describes a process combining biological and
chemical treatment of aromatic substrates including mono-, di- and tricylic
aromatic hydrocarbons as well as their substituted and heterocyclic analogues in
authentic feedstocks. The first step is hydroxylation of the aromatics using
recombinant or mutant bacteria to generate either the dihydrodiol or dihydroxy
168
derivatives (steps A and B, Fig. 1). The second step is chemical dehydration at
high temperature (200 - 600°C) in the presence of CO and absence of O2.
Notably, the chemical hydrogenation or hydrogenolysis treatment described in
the patent does not require prior separation of the aqueous and organic phases
(although the effect of the presence of biocatalyst cells is not addressed). This
treatment theoretically achieves ring cleavage and (or) removal of aromatic N
and S heteroatoms, yielding product streams with lower molecular weight and
decreased heteroatom content. Depending on the reaction conditions, chemical
hydrogenolysis of dihydroxybiphenyls yields phenols, alkyl-phenols,
monohydroxybiphenyls, methylated dihydroxybiphenyls, cyclohexyl- and
cyclopentylbenzenes, alkylbenzenes and dihydronaphthalenes, among other
products. However, no specific examples of heteroatom removal were described
in the patent, nor the efficiency of conversion given for the mixed substrates
present in authentic feedstocks.
Several bacterial strains were suggested to be suitable biocatalysts for the
process, including mutant strains of P. putida F blocked either at the
dehydrogenation step (Step B, Fig. 1) so that cw-dihydrodiols accumulate, or at
the catechol dioxygenase step (Step C, Fig. 1) so that 1,2-dihydroxy compounds
accumulate. Alternatively, dioxygenase genes from strains of Pseudomonas
spp., C. testosteroni or S. yanoikuyae are proposed as sources of genes for
creating recombinant biocatalysts by cloning.
5.3. Bio-denitrogenation
Several species of nitrogen heterocycles are found in petroleum, including
the predominant "non-basic" carbazoles, pyrroles, and indoles, and the "basic"
quinolines and pyridines. Two general modes of attack on these nitrogen
heterocycles are recognized (reviewed in Ref. [2]). Quinolines are sequentially
mono-oxygenated to yield hydroxyquinolines, with subsequent N removal and
further oxidation. Carbazole, however, undergoes an unusual "angular
dioxygenation" at the 1,9 position [112, 113], generating 2'-aminobiphenyl-2,3-
diol, followed by a ring-opened weto-cleavage metabolite and finally anthranilic
acid (Fig. 2), with side-reactions generating additional minor products [12, 57].
The extensive metabolism required to remove the nitrogen heteroatom from
these metabolites would also remove carbon, thus reducing fuel value unless the
metabolites could be recovered for separate processing.
One approach to circumvent this carbon loss would be to block the
angular attack pathway after ring cleavage but before carbon loss, to generate an
aromatic metabolite with a free amine group (Compounds III or IV, Fig. 2), then
hydrogenating this "sensitized" molecule under mild conditions to specifically
remove the nitrogen. Genes encoding the angular attack enzymes have been
cloned, sequenced, and characterized [56, 114, 115], facilitating future genetic
manipulation by analogy with other aromatic-oxidizing systems, and as
169
described in Section 4.1, Riddle et al. [82] deleted genes from the carbazole
operon to achieve metabolite accumulation.
Fig. 4. A. Gas chromatography-mass spectrometry (GC-MS) total ion scan of polar

metabolites accumulated by Biocatalyst 21-41 after incubation with naphthalene. Metabolites
were extracted from acidified medium and derivatized to yield trimethylsilyl derivatives for
GC-MS analysis; o-terphenyl is the surrogate standard. B. Mass spectra of scans 591 and 596
from Panel A, with most favored inferred structures shown. Adapted from Ref. [9].
170
Although this approach appears to have potential for bio-denitrogenation,

Bressler and Gray [111] recently analyzed the products of hydrogenation
reactions using model metabolites possessing oxygen-containing functional
groups as well as the free amine (similar to those expected from angular attack
on carbazole [12]). Although they found that some model metabolites were more
susceptible to catalytic hydrodenitrogenation than carbazole itself, compounds
having both a free amine and an alcohol on an alkyl side chain (e.g., compound
IV, Fig. 2) were susceptible to a variety of rapid, undesirable side reactions
including condensation, polymerization and ring re-closure. To produce a
substrate suitable for efficient hydrodenitrogenation, the biocatalyst would either
have to remove the free amine group after angular attack or generate an aromatic
amine without a side chain leaving group (e.g., an alcohol). This may be a
productive direction for future genetic manipulation of potential biocatalysts by
blocking the carbazole pathway and introducing deaminase activity, or shifting
the catabolic pathway towards more desirable metabolites. The potential for
such an approach has been demonstrated by Riddle et al. [82]. Although their
recombinant strain was not used for bio-denitrogenation studies (the
accumulating metabolite retained the nitrogen heteroatom), the study showed the
potential for genetic manipulation of catabolic genes and expression in
heterologous hosts with desirable attributes. As well, the carbazole metabolite
was sufficiently non-polar that it partitioned into the oil phase, eliminating the
need for recovery of polar metabolites from the aqueous phase.
Bio-denitrogenation biocatalysts must be capable of attacking alkyl-
substituted carbazoles. Fedorak and Westlake [15] used a mixed bacterial
population to study the susceptibility of 26 alkyl carbazoles to degradation; in
general the homologues demonstrated decreasing susceptibility with increasing
substitution, with the C5-carbazoles resistant to attack. This potential problem of
substrate range has not yet been addressed in bio-denitrogenation studies.
Bio-denitrogenation may be easier to accomplish with nitrogen
heterocycles such as quinoline and isoquinoline (reviewed in Ref. [2]). Selective
biodegradation of quinoline, methylquinolines and isoquinoline in shale oil
without oxidation of non-target hydrocarbons has been demonstrated [116].
More recently, Kilbane et al. [70] used rounds of enrichment and mutagenesis to
isolate a strain of Pseudomonas ayucida that selectively used quinoline as a
nitrogen source for growth but not as a carbon source. Resting cells of strain
IGTN9m removed only 5% of total nitrogen from a shale oil, but 68% of the
quinoline in that oil during a 16-h incubation. Sugaya et al. [117] proposed that
bio-denitrogenation of quinoline in crude oil, producing ammonia and water-
soluble metabolites, might be achieved during storage before refining, thus
addressing the need for relatively long retention times.
171
5.4. Feasibility of aromatic bio-processing

This review has not addressed questions of economic feasibility of
aromatic bio-processing, as there is currently insufficient research upon which to
base rational cost estimates. It is hoped that economic advantages can be
realized in terms of reduced capital costs and operating costs, and potential
savings associated with reduced CO2 sequestration costs. Obviously, capital and
operating costs will vary significantly depending on whether the processes are
applied in the field or at the refinery.
The logistics of process scale-up, which are not trivial, are also related to
the type of application. Predictable and reproducible production and disposal of
large masses of biocatalyst are accomplished routinely in the pharmaceutical
industry, and these practices should be transferrable to petroleum biocatalysis in
a refinery situation. However, if the biocatalyst is to be used in the oilfield, it
must be formulated for ease and safety of handling, as well as stability and
robustness under operating conditions which may be neither constant nor ideal.
The emphasis of aromatic bio-processing to date has been on reduction of
feedstock aromaticity and heteroatom content, with the idea that the polar
products would be added back to (or retained in) the organic stream at some
point, so as to maintain full fuel value in the bio-processed feedstock. There is
also unexplored potential for diverting the metabolite stream to alternative
chemical processes, as starting points for synthesis of niche market chemicals.
However, regardless of whether the products are retained, re-introduced or
diverted, it is still necessary that all the bio-processing products have value in
order for the process to be economically viable. As the economics of
conventional petroleum refining shift to accommodate the pressures of meeting
increasingly stringent environmental regulations with feedstocks of declining
quality, aromatic bio-processing may become more attractive as an adjunct to
conventional processing.
6. CONCLUSIONS
Application of aromatic ring cleavage biocatalysis to petroleum bio-processing

is still in the development stage and presents a number of unanswered questions
about process feasibility and economic benefits. We cannot yet confidently
predict the biochemical and genetic behaviour of new aromatic ring cleavage
biocatalysts because of the demonstrated complexity and diversity of known and
potential biocatalysts. This means that each combination of biocatalyst and
feedstock will require application-specific optimization, which slows progress
towards a generalized application.
There is potential for bio-processing to reduce the aromaticity and
heteroatom content of crude oils and refined products. Bioprocessing would
complement conventional refining technologies and result in improved fuel
172
quality at lower capital and operating costs and with reduced environmental
impact. However, the reality awaits further research.
Acknowledgments.
The author acknowledges the financial support of ChevronTexaco (USA),
the National Centre for Upgrading Technology (Canada) and NSERC (Canada)
in developing the BioARC process, and R. Shong, H. Dettman, M. Gray, and K.
Kirkwood for helpful suggestions on the manuscript.
REFERENCES
[I] C.E. Cerniglia, Biodegradation, 3 (1992) 351.

[2] MJ. Benedik, P.R. Gibbs, R.R. Riddle and R.C. Wilson, Trends Biotechnol., 16 (1998)
390.
[3] C. Bianchini, A. Meli and F. Vizza, Eur. J. Inorg. Chem., 2001 (2001) 53.
[4] M. Gray, Upgrading Petroleum Residues and Heavy Oils, Marcel Dekker, New York,
(1994).
[5] C. Song and X.L. Ma, Appl. Catalysis B- Environ., 41 (2003) 207.
[6] C.L. Coyle, M. Siskin, D.T. Ferrughelli, M.S.P. Logan and G. Zylstra, Biological
Activation of Aromatics for Chemical Processing and/or Upgrading of Aromatic
Compounds, Petroleum Coal, Resid, Bitumen and Other Petrochemical Streams. US
Patent No. 6 156 946. (2000).
[7] H. Habe and T. Omori, Biosci. Biotechnol. Biochem., 67 (2003) 225.
[8] R.W. Eaton and P.J. Chapman, J. Bacteriol., 174 (1992) 7542.
[9] J.M. Foght, Q. Wu, P.M. Fedorak, M.A. Pickard and M.R. Gray, Proc. Joint Meet.
Petrol. Soc. and BIOMINET, (1997) Paper 97-13.
[10] J.G. Speight (ed.) The Chemistry and Technology of Petroleum, Marcel Dekker, New
York (1980).
[II] D.C. Bressler and P.M. Fedorak, Can. J. MicrobioL, 46 (2000) 397.
[12] L.M. Gieg, A. Otter and P.M. Fedorak, Environ. Sci. Technol., 30 (1996) 575.
[13] J. Sikkema, J.A.M. de Bont and B. Poolman, Microbiol. Rev., 59 (1995) 201.
[14] P.M. Fedorak and D.W.S. Westlake, Can. J. Microbiol., 27 (1981) 432.
[15] P.M. Fedorak and D.W.S. Westlake, Appl. Environ. Microbiol., 47 (1984) 858.
[16] S. Le Borgne and R. Quintero, Fuel Processing Technol., 81 (2003) 155.
[17] R. A. Efroymson and M. Alexander, Appl. Environ. Microbiol., 57 (1991) 1441.
[18] Q. Wu, P.M. Fedorak, M.A. Pickard, M.R. Gray and J.M. Foght, Prepr. Pap. - Am.
[19] Q. Wu, P.M. Fedorak, M.A. Pickard, M.R. Gray and J.M. Foght, Prepr. Pap. - Am.
[20] Q. Wu, M.R. Gray, M.A. Pickard, P.M. Fedorak and J.M. Foght, Prepr. Pap. - Am.
Chem. Soc, Div. Pet. Chem., 48 (2003) 47.
[21] F. Haus, J. German and G.-A. Junter, Biodegradation, 11 (2000) 365.
[22] H.K. Kotlar, K. Rasmussen, S. Markussen, A. Winnberg, M. Gimmestad and K.
Grande, Proc. 2nd Intl. Conf. Petrol. Biotechnol., (2003) Paper BRC-008.
[23] L.Y. Wick, T. Colangelo and H. Harms, Environ. Sci. Technol., 35 (2001) 354.
[24] A. Inoue and K. Horikoshi, J. Ferment. Bioeng., 71 (1991) 194.
[25] W.A. Duetz, J.B. van Beilen and B. Witholt, Curr. Opin. Biotechnol., 12 (2001) 419.
173
[26] B.L. McFarland, D.J. Boron, W. Deever, J.A. Meyer, A.R. Johnson and R.M. Atlas,
Crit. Rev. Microbiol., 24 (1998) 99.
[27] E. Deziel, Y. Comeau and R. Villemur, Biodegradation, 10 (1999) 219.
[28] J. Marcoux, E. Deziel, R. Villemur, F. Lepine, J.-G. Bisaillon and R. Beaudet, J. Appl.
Microbiol., 88 (2000) 655.
[29] A.J. Daugulis, TIBTech., 19 (2001) 457.
[30] L.-Q. Yu, T.A. Meyer and B.R. Folsom, Oil/Water/Biocatalyst Three Phase Separation
Process. US Patent No. 5 772 901 (1998).
[31] E.N. Kaufman, J.B. Harkins, M. Rodriguez, C. Tsouris, P.T. Selvaraz and S.E. Murphy,
Fuel Processing Technol., 52 (1997) 127.
[32] C. Tsouris, A.P. Borole, E.N. Kaufman and D.W. DePaoli, Ind. Eng. Chem. Res., 38
(1999) 1877.
[33] E.N. Kaufman, J.B. Harkins and A.P. Borole, Appl. Biochem. Biotechnol., 73 (1998)
127.
[34] M.A. Heitkamp and W.P. Stewart, Appl. Environ. Microbiol., 62 (1996) 4659.
[35] M. Naito, T. Kawamoto, K. Fujino, M. Kobayashi, K. Maruhashi and A. Tanaka, Appl.
Microbiol. Biotechnol., 55 (2001) 374.
[36] D.Z. Contreras, J.R. Garcia, A.L. Perez and L.F. Linares, Proc. 2nd Intl. Conf. Petrol.
Biotechnol., (2003) Paper BRC-013.
[37] R. Munoz, B. Guieysse and B. Mattiasson, Appl. Micro Biotech., 61 (2003) 261.
[38] A.J. Daugulis and CM. McCracken, Biotechnol. Lett., 25 (2003) 1441.
[39] M.R. Fries, J. Zhou, J. Chee-Sanford and J.M. Tiedje, Appl. Environ. Microbiol., 60
(1994)2802.
[40] K.J. Rockne, J.C. Chee-Sanford, R.A. Sanford, B.P. Hedlund, J.T. Staley and S.E.
Strand, Appl. Environ. Microbiol., 66 (2000) 1595.
[41] X. Zhang and L.Y. Young, Appl. Environ. Microbiol., 63 (1997) 4759.
[42] A. Schmid, A. Kollmer, B. Sonnleitner and B. Witholt, Bioprocess Engineer., 20 (1999)
91.
[43] J.D. Van Hamme, A. Singh and O.P. Ward, Microbiol. Mol. Biol. Rev., 67 (2003) 503.
[44] L.P. Wackett, Nature Biotechnol., 21 (2003) 136.
[45] D.J. Monticello, Curr. Opinion Biotechnol., 11 (2000) 540.
[46] L. Dorobantu, M.Sc. thesis, University of Alberta, Edmonton, AB, 2004.
[47] M.A. Pickard, R. Roman, R. Tinoco and R. Vazquez-Duhalt, Appl. Environ. Microbiol.,
65 (1999) 3805.
[48] Y. Wang, R. Vazquez-Duhalt and M.A. Pickard, Can. J. Microbiol., 49 (2003) 675.
[49] W.D. Weissenfels, M. Beyer, J. Klein and H.J. Rehm, Appl. Microbiol. Biotechnol., 34
(1991)528.
[50] A.D. Laurie and G. Lloyd-Jones, J. Bacteriol, 181 (1999) 531.
[51] A.K. Goyal and G.J. Zylstra, Appl. Environ. Microbiol., 62 (1996) 230.
[52] S.E. Dyksterhouse, J.P. Gray, R.P. Herwig, J. C. Lara and J.T. Staley, Int. J. Syst.
Bacteriol., 45 (1995) 116.
[53] Y. Kasai, K. Shindo, S. Harayama and N. Misawa, Appl. Environ. Microbiol., 69
(2003) 6688.
[54] B.P. Hedlund, A.D. Geiselbrecht, T.J. Bair and J.T. Staley, Appl. Environ. Microbiol.,
65(1999)251.
[55] J.M. Foght and D.W.S. Westlake, Biodegradation, 7 (1996) 353.
[56] K. Maeda, H. Nojiri, M. Shintani, T. Yoshida, H. Habe and T. Omori, J. Mol. Biol., 326
(2003)21.
174
[57] J. Schneider, R.J. Grosser, K. Jayasimhulu, W. Xue, B. Kinkle and D. Warshawsky,

Can. J. Microbiol., 46 (2000) 269.
[58] S.P. Story, S.H. Parker, J.D. Kline, T.-RJ. Tzeng, J.G. Mueller and E.L. Kline, Gene,
260(2000)155.
[59] O. Pinyakong, H. Habe and T. Omori, J. Gen. Appl. Microbiol., 49 (2003) 1.
[60] C. Baraniecki, J. Aislabie and J. Foght, Microbial Ecol., 43 (2002) 44.
[61] T. Shi, J.K. Fredrickson and D. L. Balkwill, J. Ind. Microbiol. Biotech., 26 (2001) 283.
[62] M.F. Romine, L.C. Stillwell, K.-K. Wong, S.I. Thurston, E.C. Sisk, C. Sensen, T.
Gaasterland, J.K. Fredrickson and J.D. Saffer,. J. Bacteriol., 181 (1999) 1585.
[63] R.J. Grosser, D. Warshawsky and J. R. Vestal, Appl. Environ. Microbiol., 57 (1991)
3462.
[64] H.P. Doddamani and J.Z. Ninnekar, Curr. Microbiol., 41 (2000) 11.
[65] E. Gauthier, E. Deziel, R. Villemur, P. Juteau, F. Lepine and R. Beaudet, J. Appl.
Microbiol., 94(2003)301.
[66] J.D. Moody, J.P. Freeman, D.R. Doerge and C.E. Cerniglia, Appl. Environ. Microbiol.,
67 (2001) 1476.
[67] A. Saito, T. Iwabuchi and S. Harayama, J. Bacteriol., 182 (2000) 2134.
[68] C.C.R. Allen, D.R. Boyd, M.J. Larkin, K.A. Reid, N.D. Sharma and K. Wilson, Appl.
Environ. Microbiol., 63 (1997) 151.
[69] L. Monna, T. Omori and T. Kodama, Appl. Environ. Microbiol., 59 (1993) 285.
[70] J.J. Kilbane, R. Ranganathan, L. Cleveland, K.J. Kayser, C. Ribiero and M.M. Linhares,
[71] K.-M. Yen and CM. Serdar, CRC Crit. Rev. Microbiol., 15 (1988) 247.
[72] A.A.. Khan, R.-F. Wang, W.-W. Cao, D.R. Doerge, D. Wennerstrom and C.E.
Cerniglia, Appl. Environ. Microbiol., 67 (2001) 3577.
[73] R. van Herwijnen, D. Springael, P. Slot, H.A.J. Govers and J.R. Parsons. Appl. Environ.
Microbiol., 69 (2003) 186.
[74] K.-M. Yen and I.C. Gunsalus, Proc. Natl. Acad. Sci. USA, 79 (1982) 874.
[75] S. Kurkela, H. Lehvaslaiho, E.T. Palva and T.H. Teeri, Gene, 73 (1988) 355.
[76] S.A. Denome, D. C. Stanley, E.S. Olson and K.D. Young, J. Bacteriol., 175 (1993)
6890.
[77] H. Kiyohara, S. Torigoe, N, Kaida, T. Asaki, T. Iida, H. Hayashi and N. Takizawa, J.
Bacteriol., 176(1994)2439.
[78] S.L. Fuenmayor, M. Wild, A.L. Boyes and P.A. Williams, J. Bacteriol., 180 (1998)
2522.
[79] J.M. Foght and D.W.S. Westlake, Can. J. Microbiol., 37 (1991) 924.
[80] J. Widada, H. Nojiri, K. Kasuga, T. Yoshida, H. Habe and T. Omori, Appl. Microbiol.
Biotechnol, 58 (2002) 202.
[81] M.Schell, Gene, 36(1985)301.
[82] R.R. Riddle, P.R. Gibbs, R.C. Willson and M.J. Benedik, J. Ind. Microbiol. Biotech., 30
(2003) 6.
[83] G.J. Zystra and E. Kim, J. Ind. Microbiol. Biotech., 19 (1997) 408.
[84] M. Tsuda and T. lino, Mol. Gen. Genet., 223 (1990) 33.
[85] L.P. Wackett, Environ. Microbiol., 6 (2004) 313.
[86] S. Harayama, M. Rekik, A. Wasserfallen and A. Bairoch, Mol. Gen. Genet., 210 (1987)
241.
[87] M. Schell, Proc. Natl. Acad. Sci. USA, 83 (1986) 369.
[88] M. A. Schell and P. E. Wender, J. Bacteriol., 166 (1986) 9.
[89] E.A. Barnsley, J. Bacteriol., 125 (1976) 404.
175
[90] U. Gerischer, J. Mol. Microbiol. Biotechnol., 4 (2002) 111.

[91] R.S. Wodzinski and J.E. Coyle, Appl. Microbiol, 27 (1974) 1081.
[92] J.N. Bateman, B. Speer, L. Feduik and R.A. Harline, J. Bacteriol, 166 (1986) 155.
[93] T. Bugg, T., J.M. Foght, M.A. Pickard and M.R. Gray, Appl. Environ. Microbiol,. 66
(2000) 5387-5392.
[94] H. Miyata, K. Iwahori, J.M. Foght and M.R. Gray, Appl. Environ. Microbiol, 70 (2004)
363.
[95] D.C. Bressler and M.R. Gray, Int. J. Chem. React. Eng, Vol. 1 (2003) R3.
[96] J.L. Ramos, E. Duque, M.-T. Gallegos, P. Godoy, M.I. Ramos-Gonzalez, A. Rojas, W.
Teran and A. Segura, Annu. Rev. Microbiol, 56 (2002) 743.
[97] E.M. Hearn, J.J. Dennis, M.R. Gray and J.M. Foght, J. Bacteriol, 185 (2003) 6233.
[98] F.-M. Menn, B.M. Applegate and G.S. Sayler, Appl. Environ. Microbiol, 59 (1993)
1938.
[99] J.M. Foght and D.W.S. Westlake, Can. J. Microbiol, 36 (1990) 718.
[100] S.M. Resnick and D.T. Gibson, J. Ind. Microbiol, 17 (1996) 438.
[101] S. Tongpim and M.A. Pickard, Can. J. Microbiol, 42 (1996) 289.
[102] Y. Yang, R.F. Chen and M. P. Shiaris, J. Bacteriol, 176 (1994) 2158.
[103] M. Kobayashi, K. Horiuchi, O. Yoshikawa, K. Hirasawa, Y. Ishii, K. Fujino, H.
Sugiyama and K. Maruhashi, Biosci. Biotechnol. Biochem, 65 (2001) 298.
[104] E. Andujar, M.J. Hernaez, S.R. Kaschabek, W. Reinecke and E. Santero, J. Bacteriol,
182 (2000) 789.
[105] R.E. Parales, K. Les, S.M. Resnick, H. Jiang, D.J. Lessner and D.T. Gibson, J.
Bacteriol, 182(2000)1641.
[106] E. Andujar and E. Santer, Microbiology (UK), 149 (2003) 1559.
[107] H. Suenaga, M. Mitsuoka, Y. Ura, T. Watanabe and K. Furukawa, J. Bacteriol, 183
(2001)5441.
[108] Health Canada website "Non-Pathogenic Organisms" <http://www.hc-sc.gc.ca/pphb-
dgspsp/ols-bsl/pathogen/pdf/pathogensfmal-092001.pdf> (accessed Feb 2004)
[109] A. Laborde and D.T. Gibson, Appl. Environ. Microbiol, 34 (1997) 783.
[110] D.T. Gibson and V. Subramanian, In: D.T. Gibson (ed.), Microbial Degradation of
Organic Compounds, Marcel Dekker, New York (1984) pp. 191-252.
[ I l l ] D.C. Bressler and M.R. Gray, Energy Fuels, 16 (2002) 1076.
[112] N. Ouchiyama, Y. Zhang, T. Omori and T. Kodama, Biosci. Biotechnol. Biochem, 57
(1993)455.
[113] H. Nojiri, H. Habe and T. Omori, J. Gen. Appl. Microbiol, 47 (2001) 279.
[114] S.I. Sato, N. Ouchiyama, T. Kimura, H. Nojiri, H. Yamane and T. Omori, J. Bacteriol,
179(1997)4841.
[115] S.I. Sato, J.W. Nam, K. Kasuga, H. Nojiri, H. Yamane and T. Omori, J. Bacteriol, 179
(1997)4850.
[116] J. Aislabie and R.M. Atlas, Fuel, 69 (1990) 1155.
[117] K. Sugaya, O. Nakayama, N. Hinata, K. Kamekura, A. Ito, K. Yamagiwa and A.
Ohkawa, J. Chem. Technol. Biotechnol, 76 (2001) 603.
Chapter 6
Biocatalysis by methane monooxygenase and its

implications for the petroleum industry
T.J. Smith" and H. Daltona
department of Biological Sciences, University of Warwick, Coventry CV4

7AL, United Kingdom*
b
Biomedical Research Centre, Sheffield Hallam University, Howard Street,
Sheffield SI 1WB, United Kingdom
1. INTRODUCTION
Methane monooxygenases (MMOs) are unique among known catalytic systems

in their ability to convert methane to methanol under ambient conditions using
dioxygen as the oxidant.
CH4 + O2 + 2H+ + 2e" -> CH3OH + H2O
Thus, MMOs activate the kinetically inert O2 molecule to react at 30-45 °C, 1
atmosphere pressure with the unreactive hydrocarbon methane (C-H bond
energy = 1 1 9 kcal mol"1) to produce methanol with stoichiometric yield and
high turnover (up to 6.0 s"1 [1]). In addition to this important transformation,
they also catalyse the oxygenation of a diversity of adventitious substrates,
including numerous petroleum constituents and by-products, which have led to
intense interest in their exploitation for biocatalyis and bioremediation. In this
chapter, we review the biological distribution of MMOs, their structural and
catalytic properties and then discuss their applications and potential for
transformation of petroleum-related products. Lastly, we discuss how MMOs
may inspire the synthesis of biomimetic catalysts to facilitate methane to
methanol conversion and other commercially important transformations.
178
2. BIOLOGICAL DISTRIBUTION AND CLASSIFICATION OF

METHANE MONOOXYGENASES
MMOs have been found only in methanotrophic bacteria, which are Gram-
negative organisms that group within the a - and y-proteobacteria and can utilise
methane as their sole source of carbon and energy. Methanotrophic bacteria, and
therefore the MMOs that they produce, are widespread in the environment; they
are found in mesophilic and extreme (e.g. as low as 4 °C [2] and up to 72 °C [3])
conditions and are estimated to be responsible for oxidation of up to 60 % of the
1188 Tg of methane produced globally each year by natural and anthropogenic
sources [4].
Two forms of MMO are known, particulate (pMMO) and soluble
(sMMO). Many methanotrophs, such as Methylomicrobium album BG8 and
Methylomonas methanica, produce only the membrane-associated pMMO.
Others, such as Methylococcus capsulatus (Bath) and Methylosinus
trichosporium OB3b can elaborate either form [5], dependent on the copper-to-
biomass ratio of the culture. Such organisms produce the copper-containing
pMMO at high copper-to-biomass ratios and the nonheme iron-containing
sMMO at low copper-to-biomass ratios [6]. It is relatively straightforward,
during cultivation in laboratory- or industrial-scale bioreactors, to control the
medium copper concentration and culture density to obtain either form of MMO.
An easy colorimetric test, based on the oxidation of naphthalene [7], which is a
substrate of sMMO alone, allows rapid confirmation of the form of MMO that
the culture is expressing.
Although sMMO and pMMO both oxygenate methane to methanol, they
show no similarity in the amino-acid sequences of their protein components,
their requirements for metal cofactors or their location within the cell. They also
differ markedly in their substrate profiles and requirements for electron donors.
It is remarkable that the particulate and soluble MMOs represent evolutionarily
unrelated molecular solutions to a single chemical problem that are frequently
found in the same bacterial cells.
3. SOLUBLE METHANE MONOOXYGENASE
Thanks to more than 20 years of effort with diverse techniques in several

research groups, sMMO is now probably the best characterised non-heme iron
monooxygenase system. Two sMMO systems, from Me. capsulatus (Bath) [8]
and Ms. trichosporium OB3b [1], have been studied in detail. These enzyme
systems show different optimal temperatures (45 and 30 °C, respectively), which
correspond to the optimal growth temperatures of the two species of bacteria. In
their structure, substrate range and mechanistic properties, the two sMMOs are
very similar. This section refers primarily to the Me. capsulatus enzyme,
179
although in many cases almost indistinguishable results have been obtained with
the Ms. trichosporium system.
3.1. sMMO structure

sMMO comprises three components (Fig. 1): (1) a hydroxylase with an
(a|3y)2 structure in which the a, |3 and y subunits have masses of 61, 45 and 20
kDa, respectively; (2) a 39-kDa NAD(P)H-dependent reductase; (3) a third
component known as protein B or the coupling/gating protein, which comprises
a single polypeptide of 16 kDa [8]. The a subunits of the hydroxylase each
contain a u-(hydr)oxo-bridged binuclear iron site [10-12] that is coordinated by
four glutamate and three histidine sidechains and is the site of O2 activation. X-
ray crystallography of the hydroxylases from Me. capsulatus (PDB accession
codes 1MM0, 1MTY) and Ms. trichosporium (1MHY, 1MHZ) has shown that
the hydroxylase is a predominantly a-helical structure in which the binuclear
iron centres reside within the a subunits, in a hydrophobic pocket that is almost
certainly critical in binding substrates [13-15]. Indeed, energy minimisation
calculations have suggested that the most favourable binding site for methane
and other small substrates lies inside this pocket, within 3 A of the binuclear
iron centre (Fig. 2) [16].
The reductase component, which is required for the supply electrons from
NADH or NADPH, contains flavin adenine dinucleotide (FAD) and Fe2S2
prosthetic groups. Protein B has no prosthetic groups. NMR structural analysis
(PDB accession codes 1CKV [17] and 2MOB [18]) has shown that it has a core
ot/p structure with highly mobile regions at the N- and C-termini.
Fig. 1. Components of sMMO. The individual polypeptides of the enzyme are named
according to the genes that encode them [9].
180
Fig. 2. Active site structure of sMMO based on crystallographic data in Ref. [14]. The iron
atoms are shown as large grey spheres and the residues that ligate them as a grey ball-and-
stick representation. The residues forming the hydrophobic active site pocket [ 16] are shown
in white and two solvent molecules that bridge the dinuclear iron site in black.
3.2. sMMO mechanism

In the resting state of the sMMO enzyme, the binuclear iron centre is in
the diferric (Fe m 2 ) oxidation state [10], and must be reduced to the diferrous
(Fe"2) form to allow O2 to bind. The two electrons required for this reduction are
provided from NAD(P)H, passing first via the FAD and then the Fe2S2
prosthetic group of the reductase, which acts as a two-electron/one-electron
transformase, to allow the two-electron oxidation of NAD(P)H to be used to
feed electrons singly into the binuclear iron centre of the hydroxylase [19]. After
binding of O2, the intermediates O, P*, P and Q (Fig. 3) can be detected even in
the absence of substrate. Compound O, whose existence has been inferred from
the independence of formation of compound P on O2 concentration, appears to
be a species in which O2 is bound to the hydroxylase but not covalently attached
to the binuclear iron centre [20]. Formation of compound P* is accompanied by
loss of the g = 16 electron paramagnetic resonance (epr) signal due to the high-
spin Fe"2 state of the binuclear iron centre. In compound P* the binuclear iron
site may be in the FeII[2 or mixed valent (Fe11 Fe m ) state [21]. At this stage
dioxygen is probably covalently bound to the binuclear iron centre in the form
of an unprotonated bridging peroxo species. The compound P* to P transition
requires a proton, probably to protonate the peroxo species before the 0 - 0 bond
scission that occurs upon the decay of compound P. The binuclear iron centre of
compound P has been shown by means of Mossbauer spectroscopy to be in the
epr-silent Fe in 2 form. Compound P decays spontaneously to compound Q,
which is also epr-silent but is bright yellow in colour (absorption maxima 350
and 420 nm) [20]. The binuclear iron centre of compound Q appears based on
Mossbauer spectroscopy to be in the diferryl (FeIV2) oxidation state. Interatomic
181
distance constraints obtained from extended X-ray absorption fine structure

(EXAFS) data are consistent with the bis-|i-oxo bridged 'diamond core'
structure that Lipscomb and co-workers have proposed for the binuclear iron
centre of compound Q (Fig. 3), although other possible structures also fit the
data [22]. Compound Q is the quintessence of sMMO catalysis. It is kinetically
competent to oxygenate methane and other substrates and yet, in the absence of
substrate, is astoundingly stable (ti/2 14 s at 4°C) [23] for so powerful an
oxidant in aqueous solution.
When methane is the substrate, decay of compound Q exhibits a nonlinear
Eyring plot with a point of inflection at 17°C. Other evidence suggests that this
discontinuity is due to a change in rate-determining step and so the reaction of
compound Q with the substrate must comprise at least two distinct steps [23]. In
a study of the Ms. trichosporium enzyme, use of the chromogenic substrate
nitrobenzene allowed detection of an enzyme-bound product species, termed
compound T [24]. It is, however, the preceding events of carbon-hydrogen bond
scission and substrate oxygenation that are fundamental to understanding how
sMMO can activate unreactive hydrocarbons such as methane, and it is the
mechanism of these steps that remain the greatest contention in the sMMO
reaction cycle. In another study of the sMMO from Ms. trichosporium, reaction
of compound Q with methane showed a very large primary deuterium isotope
effect (ca. 50), which was too high to be accounted for solely by the difference
in vibrational zero point energies between CH4 and CD4. One possibility is that
quantum mechanical tunnelling of the hydrogen nucleus is required for in
breakage of the inert C-H bond of methane; indeed, consistent with this
reasoning, the primary deuterium isotope effects observed with less inert
substrates are more modest (e.g. 1 in the case of ethane) [25].
There are three possible mechanisms for C-H bond cleavage: (1) cleavage
may be homolytic, leading to a radical mechanism; (2) cleavage may be
heterolytic, leading to a mechanism in which a carbanion intermediate may be
stabilised by coordination to one of the active-site iron atoms [26]; (3) methane
and an iron-oxygen species may react via a concerted mechanism of bond
breakage and formation [27]. The current evidence and theories are reviewed in
Refs. 28-30. The available data, derived principally from radical clock substrates
and theoretical studies, are complex and not afford a single clear conclusion at
the present time; indeed it is possible that multiple reaction pathways exist and
that different mechanisms operate with different substrates. What is clear,
however, is that sMMO generates within its active site one of the most reactive
oxidising agents in nature that is, as described below, able to oxygenate a range
of petroleum-related substrates.
182
Fig. 3. Principal intermediates during the sMMO catalytic cycle. References are given in the text.
3.3. Substrate profile of sMMO

In addition to its remarkable ability to convert the unreactive methane
molecule to methanol under ambient conditions, sMMO is remarkable in its
diversity of adventitious substrates, often termed cosubstrates, that the enzyme
can oxidise even though the resultant oxidation products cannot be used as
nutrient sources by methanotrophic bacteria. The astonishing range of substrates
known to be oxygenated by sMMO is summarised in Table 1, together with the
principal products obtained. Broadly speaking, singly oxygenated products
predominate with all substrates. Alkanes are hydroxylated, in the case of
aliphatic compounds almost exclusively at the terminal and subterminal
positions. Ring hydroxylation of aromatics occurs primarily at the meta position,
along with a comparable amount of substituent hydroxylation when an alkyl
substituent is present. sMMO oxygenates alkenes to epoxides with retention of
stereochemistry around the C=C double bond. Ethers are cleaved oxidatively to
yield mixtures of alcohols and aldehydes, and pyridine undergoes N-
oxygenation. It appears that the initial oxygenated products formed from some
halogenated substrates decompose rapidly via nonenzymic pathways that result
in the loss of halogen substituents [31]. It is certain that there are many
substrates of sMMO that have simply never been tested with the enzyme.
Indeed, the list of small organic compounds that have been investigated and
found not to be effective sMMO substrates is very short, but includes
tetrachloromethane, iodomethane, trimethylamine [32] and tetrachloroethene
[31]. In addition, terminal alkynes are potent turnover-dependent inhibitors of
183
sMMO; they are probably oxidised to highly reactive ketenes which then attack
catalytically essential groups on the enzyme [33].
4. PARTICULATE METHANE MONOOXYGENASE
4.1. Structure and mechanism of pMMO

Although pMMO appears to be responsible for the bulk of methane
oxidation catalysed by methanotrophic bacteria in the environment, its
instability during purification procedures has meant that a consensus about its
composition and catalytic properties has been slow to emerge. pMMO is well
established as a copper-containing enzyme. The intracytoplasmic membranes of
methanotrophic bacteria that contain it are produced only during high-copper
growth [5] and, in methanotrophs that can express either pMMO or sMMO,
pMMO is induced in response to high copper-to-biomass ratio in the culture [6].
In addition, Cu2+ ions stimulate pMMO activity in vitro and the copper content
of methanotroph membranes has been observed to be directly proportional to the
pMMO activity [36].
One of the chief problems with purification of pMMO has been
solubilising the enzyme without permanent loss of activity. In 1989, it was
observed in our laboratory that solubilisation of pMMO by dodecyl-P-D-
maltoside was reversible, because activity could be restored by adding
phospholipids back to the solubilised enzyme [37]. Dodecyl-(3-D-maltoside
appears to be the most effective detergent tested to date for solubilisation of
pMMO and to the authors' knowledge has formed the basis for all recent studies
of purified pMMO. Shiemke and coworkers [38] observed that pMMO activity
could be detected in solubilised pMMO if duroquinol or decyl plastoquinol were
used as electron donor in lieu of NADH, suggesting that the direct electron
donor into pMMO is a quinol.
Protocols for purification of pMMO typically comprise preparation of the
membrane fraction, washing with buffer containing up to 1 M monovalent salt
and then one or more chromatographic separations, usually including at least
one anion exchange column [39-43]. Some workers have found anaerobic
conditions favourable to retention of activity [43], whilst others have found the
presence of oxygen tolerable [40] or beneficial [41]. Active preparations of
pMMO have generally contained three polypeptides, of about 49, 27 and 22 kDa
[37], among which the 27-kDa subunit is most likely to contain the active centre
because it becomes labelled by the inhibitor acetylene [33, 39]. The 49, 27 and
22-kDa components are encoded by the genes pmoA, B and C, respectively,
which are multicopy genes [44, 45] induced in response to high copper-to-
biomass ratio [46, 47].
184
Table 1
Principal oxidation reactions catalysed by sMMO
Substrate Major detected product(s); relative Specific activity Reference
molar proportions of multiple (nmol of product (type of
products are shown in parentheses. min~ mg" ) assay) 2
Alkanes
Methane Methanol 84 32 (SE)
Ethane Ethanol 68 32 (SE)
Propane Propan-1-ol (39); propan-2-ol (61) 69 32 (SE)
Butane Butan-1-ol (54); butan-2-ol (46) 77 32 (SE)
Pentane Pentan-1-ol (28); pentan-2-ol (72) 73 32 (SE)
Hexane Hexan-1-ol (63); hexan-2-ol (37) 40 32 (SE)
Heptane Heptan-1-ol (22); heptan-2-ol (78) 27 32 (SE)
Octane Octan-1-ol (9); octan-2-ol (91) 9 32 (SE)
2-Methylpropane 2-Methylpropan-2-ol (70); 2- 33 26 (PP)
methylpropan-1-ol (30)
2-Methylpropane 2-Methylpropan-2-ol (70); 2- 33 26 (PP)
methylpropan-1-ol (30)
2,3-Dimethylpentane 3,4-Dimethylpentan-2-ol 20 26 (PP)
Alkenes
Ethene Epoxyethane 148 32 (SE)
Propene Epoxypropane 83 32 (SE)
But-1-ene 1,2-Epoxybutane 49 32 (SE)
czs-But-2-ene cw-2,3-Epoxybutane (47); cis-2- 33 26 (PP)
buten-1-ol (53)
Zrarc.s-But-2-ene ?ra«5-2,3-Epoxybutane (27); trans-2- 39 26 (PP)
buten-1-ol (73)
Al icy die hydrocarbons
Cyclohexane 3 Cyclohexanol 25 34 (SE)
Methylene 1-Cyclohexane-1-methanol (13.7); 26 (PP)
cyclohexane methylene cyclohexane oxide (75.8);
4-hydroxymethylene cyclohexane
(10.5)
p-Pinene 6,6-Dimethylbicyclo[3.1.1 ]hept-2- 26 (PP)
ene-2-methanol (72.3); P-pinene
oxide.
Adamantane 1-Adamantol (50); 2-adamantol (50) 3 26 (PP)
czs-1,4-Dimethyl l-cw-4-Dimethylcyclohexanol (35); 1.3 26 (PP)
cyclohexane 1 -rra«s-4-dimethylcyclohexanol (61);
m-2,5-dimethylcyclohexanol (4)
c/5-l,3-Dimethyl 3,5-Dimethylcyclohexanol (80); 1- 0.5 26 (PP)
cyclohexane cw-3-dimethylcyclohexanol (14); 1-
trans-3 -dimethylcyclohexanol (6);
Halogenated aliphatics
Vinyl chloride 3 748 31 (PP)
Trichloroethene 3 Formate (35); CO (53); glyoxylate 682 31 (PP)
(5); dichloroacetate (5); chloral (6)
185
Substrate Major detected product(s); relative Specific activity Reference

molar proportions of multiple (nmol of product (type of
products are shown in parentheses. mirf'mg" 1 ) 1 assay) 2
l,l-Dichloroethene J Glycolate (80); dichloroacetaldehyde 648 31 (PP)
3
(3)
Trifluoroethene Glyoxylate (53); difluoroacetate (43); 79 31 (PP)
fluoral (5)
Chlorotrifluoro- Oxalate 17 31 (PP)
etheylene 3
Tribromoethylene 3 Formate (80); bromal (5) 9 31 (PP)
Monoaromatics
Benzene Phenol 74 34 (SE)
Toluene Benzyl alcohol; 4-cresol. 53 32 (SE)
Ethylbenzene 3 1-Phenylethanol (30); 4- 18.7 34 (SE)
hydroxyethylbenzene (70)
Styrene 3 Styrene oxide 82 34 (SE)
Pyridine Pyridine N-oxide 29 32 (SE)
Diaromatics
Naphthalene 1-Naphthol, 2-naphthol 7 (W)
Biphenyl 3 2-Hydroxybiphenyl (9); 3- 35 (W)
hydroxybiphenyl (1); 4-
hydroxybiphenyl (90)
2-Hydroxybiphenyl 3 Dihydroxybiphenyls 35 (W)
2-Methylbiphenyl 3 Ring (56) and sidechain (44) 35 (W)
hydroxylated products
2-Chlorobiphenyl 3 Hydroxychlorobiphenyls 35 (W)
2-Bromobiphenyl 3 Hydroxybromobiphenyls (41); 2- 35 (W)
2-Iodobiphenyl 3 Hydroxyiodobiphenyls (18); 2- 35 (W)
Substituted methane derivatives
Chloromethane 84 32 (SE)
Dichloromethane 82 32 (SE)
Chloroform 35 32 (SE)
Bromomethane 66 32 (SE)
Nitromethane 45 32 (SE)
Methanethiol 64 32 (SE)
Methanol 246 32 (SE)
Others
Diethyl ether Ethanol (47); acetaldehyde (53) 32 (SE)
Carbon monoxide Carbon dioxide 61 32 (SE)
Specific activities are as reported in the original publications. Caution is advised when
comparing values from different studies, since various protein preparation and assay
procedures were used by the different authors.
Type of enzyme preparation used for each assay: PP, purified protein; SE, soluble extract;
W, whole cells.
3
sMMO of Ms. trichosDorium OB3b: other entries refer to Me. capsulatus (Bath).
186
Active preparations of purified pMMO have been found to contain

between 2 [41] and 15 [40] equivalents of copper per a(5y protomer. There is
agreement that whilst the polypeptides of the intracytoplasmic membranes
largely comprise pMMO subunits, most of the copper that is associated with
them is relatively loosely bound and can be removed by repeated washing of
membranes before solubilisation, increased detergent concentration or additional
chromatographic separation of solubilised material [41-43]. The loosely
associated copper may be chelated by a low molecular-mass copper-binding
cofactor [39]. This cofactor may be part of a high-affinity copper-uptake system
[48] and its complex with copper may protect pMMO by quenching superoxide
radicals [43]. Chan and coworkers have classified the copper associated with
pMMO into catalytic 'C-copper' and electron transfer, 'E-copper' and have
proposed a model in which the E-copper corresponds to the less tightly
associated copper component [40]. Epr spectroscopy reveals a type 2 copper
centre [36, 39], which accounts for only a fraction of the copper in active
pMMO preparations, at 1 [41] or 2 [36, 42] coppers per apy protomer. It is
possible that much or all of the epr-silent copper is in the +1 oxidation state [36,
49] or could be magnetically coupled in a cluster [41]. Recent EPR results from
our laboratory and Rosenzweig's are best accounted for if the epr-active copper
is ligated by 3 to 4 nitrogen ligands with square planar geometry [41, 42] rather
than the trinuclear copper cluster suggested by Chan's group [36].
There is a correlation between loss of copper from the enzyme and loss of
function; the type 2 copper (II) centre appears to be the most tightly bound
copper associated with the enzyme [41]. Chan's group have reported an NADH-
dependent preparation that contained no significant protein components except
the a, (3 and y subunits of pMMO, and they put loss of NADH-driven activity in
most preparations down to loss of the 'E-copper'. In addition, a 36-kDa FAD-
containing NADH:quinone oxidoreductase has been found in Me. capsulatus
(Bath) [50] and observed to co-purify with pMMO under certain conditions [43].
Thus, whilst the identity of the molecular components required to deliver
electrons from NADH into the pMMO active site remains controversial, there is
consensus that there are components specifically required for this electron
transfer and that, when these are missing, electrons can still be delivered via
duroquinol.
Three out of four groups actively involved in research into the structural
properties of pMMO consider that, in addition to copper, iron forms an
important part of the functional enzyme, at one [41, 42] or two equivalents of
per aPy protomer [43]. Based on EPR spectral properties a tyrosine-ligated Fe3+
ion has recently been proposed [41]. Little is currently known about the catalytic
cycle of pMMO; however, the retention of stereochemistry observed during
oxygenation of cryptically chiral alkanes strongly suggests a primarily concerted
187
mechanism of C-H bond breakage rather than one involving radical or cation
intermediates [51].
4.2 Substrate profile of pMMO

The substrate profile of pMMO (Table 2) is considerably narrower than that of
sMMO. It includes methane and linear short-chain hydrocarbons but excludes
aromatics (benzene, ethyl benzene and styrene), the alicyclic hydrocarbon
cyclohexane [34] and the branched aliphatic 2-methylpropane [52], all of which
are substrates of sMMO. Thus it appears that access to the active site of pMMO
is sterically more restricted than in the soluble enzyme. Consistent with this,
acetylene is a potent suicide substrate of both soluble and particulate enzymes
[33], whereas the bulkier phenylacetylene is much more effective against the
sMMO than pMMO [53].
5. METHANE MONOOXYGENASE IN BIOCATALYIS
As indicated above MMO-containing bacteria are able to catalyse the oxidation

of alkenes to epoxides. Since there is no direct, commercially viable, chemical
route to epoxide formation from alkenes attempts have been made to exploit the
MMO-catalysed reaction on a large scale. In a comprehensive study to produce
epoxypropane from propene we have shown how a two stage system using Me.
capsulatus (Bath) can be used to give high yields of product in a continuous
operation [55]. This methanotroph was chosen because it grows readily at 45°C
and at this temperature the epoxypropane is released into the gas phase. The
product could then be readily condensed from the gas phase at a lower
temperature. A major problem that had to be overcome was the inherent in vivo
toxicity of the epoxides to the cells. In fact epoxypropane was shown to be an
inhibitor of MMO and so caused an inactivation of the cells if grown in a single
stage operation [56]. The problem was solved by using a two stage bioreactor
system in which the first stage was used to catalyse the oxidation of propene to
the epoxide and a second reactor was fed with cells from the catalytic reactor to
reactivate inactivated cells. In the first reactor methanol was used as the electron
donor for the whole cell reaction which had to be fed as a growth-limiting
substrate to minimise its interference in MMO catalysis (methanol is a substrate
for sMMO [see Table 1] but it has a much lower affinity for the enzyme than
propene). In the second stage reactivator, methane was supplied as the growth
substrate along with all nutrients necessary to allow resynthesis of inactivated
cells. Critical to the success of the system were the operating parameters that
included the ratio of the size of the reactors such that a continuous operation was
possible that permitted active cells to be fed back to the first stage to maintain
high catalytic productivity. The continuous recycling of cells from the reactor
vessel to the reactivator (at an aspect ratio of 1:10) gave rates of 20-30 g L"1
188
day"1 over a three week period. Similar experiments with Methylocystis parvum
(OBBP) gave rates of up to 90 g L"' day"1 over a one week period. In a feasibility
study [55] using cells at 30 g L"1 and a production rate of 250 g L"1 day"1 the total
cost of epoxypropane production was estimated at $1.26 kg"1. As such it was not
competitive with the commercial oxirane process which values the product at
around this price; no account in the biological; process was taken for storage,
transport or profit which need to be added to this cost. Consequently, at the
present time this system has not been commercialised although patents for the
process have been filed worldwide.
In principle such an operation could be adapted to produce any of the
sMMO products listed in Table 1 although the separation and purification of the
product will dictate the precise mode of operation. The process has been
evaluated with other substrates including 1-butene and ethane [57] to produce
epoxybutane and ethanal respectively.
pMMO shows moderate stereoselectivity with some reactions (up to 80 %
enantiomeric excess of the R-enantiomer of pentan-2-ol product formed from
pentane) [52] and so may be suitable for eventual development as an
enantioselective catalyst. Interestingly, whilst neither pMMO nor sMMO shows
a high stereoselectivity in epoxide-generating reactions (enantiomeric excesses
for epoxypropane generation by the two enzymes are 18.5 % S [52] and 21 % R
[58], respectively), they show opposite enantioselectivity and so may be suitable
for future genetic development into a pair of enantiocomplementary biocatalysts.
Table 2
Principal oxidation reactions catalysed by pMMO
Substrate Product(s); relative molar proportions of multiple Reference
products are shown in parentheses.
Alkanes
Methane Methanol
Ethane Ethanol 52
Propane Propan-2-ol (ca. 100); propan-1-ol (trace) 52
Butane Butan-2-ol (95); butan-1-ol (5) 52
Pentane Pentan-2-ol (95); pentan-1-ol (5) 52
Alkenes
Propene Epoxypropane 52
But-1-ene 1,2-Epoxybutane (58); but-3-en-2-ol (42) 52
1,3-Butadiene 1,2-Epoxybut-3-ene 52
«s-But-2-ene 2,3-cw-Epoxybutane 52
fr-ans-But-2-ene 2,3-/ra«^-Epoxybutane 52
Chlorinated aliphatics
Trichloroethene' Carbon dioxide 54
sMMO of Ms. trichosporium OB3b; other entries refer to Me. capsulatus (Bath).
189
Both sMMO and pMMO oxidise the major pollutant trichloroethene.

With sMMO, rates comparable to the oxidation of methane have been observed
[31], whilst with pMMO the rate of trichloroethene breakdown is low (1.24
nmol min"1 mg of protein"1 in the pMMO-only methanotroph Mlb. album BG8
[54]). However, the detected products of trichloroethene breakdown by pMMO-
expressing methanotrophs do not include chloral, a pollutant that may be more
harmful than trichloroethene and has been detected in sMMO-mediated
trichloroethene breakdown [31]. Hence bioremediation of trichloroethene by
pMMO may have some advantages. During trials of methane-enhanced
bioremediation of groundwater [59], both sMMO and pMMO genes were
detected and it is likely that both were expressed to some extent in the
population of methanotrophs present.
6. METHANE MONOOXYGENASE BIOMIMETICS
The unique reactivity of MMOs, particularly their ability to convert methane to

methanol under mild conditions, has led to intense interest in the development of
small-molecule mimics of the enzyme active site that may be able to perform
similar reactions. Whilst there is currently no commercially viable biocatalyst
for industrial-scale conversion of methane to methanol, a low molecular-weight
biomimetic that did not further oxidise the methanol produced could be used to
upgrade the gaseous fuel methane to the more easily stored and transported
methanol. In principle, such a mimic could be based on pMMO or sMMO.
However, since there is currently high-resolution structural information for
sMMO alone, this has been the overwhelming focus of attention. The literature
abounds with reports of binuclear iron and other transition metal complexes and
reports of sMMO-like properties. Whilst oxidation of methane to methanol by a
biomimetic that uses dioxygen as oxidant has not been achieved, a number of
important advances have been made. For instance, Jacobsen's group have
prepared a carboxylate bridged binuclear iron complex that efficiently catalyses
the hydrogen peroxide-driven epoxygenation of a range of alkenes [60]. A
number of di-iron complexes have been described that catalyse O2-dependent
oxidation of their own ligands, such as a di-iron complex that can oxidatively
de-N-alkylate additional nitrogen-containing ligands [61]. A di-iron complex in
which the ligands included an N3 ring catalysed O2-dependent oxidation of
triphenylphosphine to the corresponding oxide [62]. Conversion of methane to
methanol catalysed by di-iron complexes under has been reported, albeit at high
pressure with low conversion (< 2 %) and significant side reactions [63]. Future
studies in this area may not only deliver useful novel catalysts for oxygen-driven
reactions but will also cast light on the reaction mechanism of sMMO.
190
7. FUTURE PROSPECTS
The unusual reactivity and broad substrate profiles of MMOs suggest many
possible applications in synthetic chemistry and bioremediation for the enzymes
and biomimetics based on them. Our recent development of a system for site-
directed mutagenesis of the soluble enzyme [64] opens the way for fine-tuning
of the catalytic versatility of sMMO for more precise and profitable
biotransformations than are possible with the wild-type enzyme. Coupled with
the sequencing of the Me. capsulatus genome, which is currently being
undertaken by the University of Bergen and The Institute for Genomic Research,
genetic technology may shortly enable metabolic engineering of novel pathways
incorporating engineered methane monooxygenases for the synthesis of valuable
Pharmaceuticals and other products using methane and other inexpensive
starting materials.
Acknowledgements
We gratefully acknowledge research funding from the Biotechnology and Biological
Sciences Research Council (UK), British Gas (UK), British Petroleum (UK), Idemitsu (Japan)
and the Gas Research Institute (GRI) (Chicago, IL).
REFERENCES
[I] B.G. Fox, W.A. Froland, J.E. Dege and J.D. Lipscomb, J. Biol. Chem., 264 (1989)
10023.
[2] J.P. Bowman, S.A. McCammon and J.H. Skerratt, Microbiology, 143 (1997) 1451.
[3] L. Bodrossy, K.L. Kovacs, I.R. McDonald and J.C. Murrell, FEMS Microbiol. Lett., 170
(1999)335.
[4] W.S. Reeburgh, S. C. Whalen and M. J. Alperin, in J.C. Murrell and D.P. Kelly (eds.),
Microbial growth on Cl compounds, Intercept, Andover UK, 1993, pp. 1-14.
[5] R.S. Hanson and T.E. Hanson, Microbiol. Rev., 60 (1996) 439.
[6] S.H. Stanley, S.D. Prior, D.J. Leak and H. Dalton, Biotechnol. Lett., 5 (1983) 487.
[7] G.A. Brusseau, H.-C. Tsien, R.S. Hanson and L.P. Wackett, Biodegradation, 1 (1990)
19.
[8] J. Colby and H. Dalton, Biochem. J., 171 (1978) 461.
[9] A.C. Stainthorpe, V. Lees, G.P. Salmond, H. Dalton and J.C. Murrell, Gene, 91 (1990)
27.
[10] M.P. Woodland, D.S. Patil, R. Cammack and H. Dalton, Biochim. Biophys. Acta, 873
(1986)237.
[II] A. Ericson, B. Hedman, K.O. Hodgson, J. Green and H. Dalton, J. Am. Chem.
Soc. 110(1988)2330.
[12] J.G. DeWitt, J.G. Bentsen, A.C. Rosenzweig, B. Hedman, J. Green, S. Pilkington, G.C.
Papaefthymiou, H. Dalton, K.O. Hodgson and S.J. Lippard, J. Am. Chem. Soc, 113
(1991)9219.
[13] A.C. Rosenzweig, C.A. Frederick, S.J. Lippard and P. Nordlund, Nature, 366 (1993)
537.
191
[14] A.C. Rosenzweig, H. Brandstetter, D.A. Whittington, P. Nordlund, SJ. Lippard and
C.A. Frederick, Proteins, 29 (1997) 141.
[15] N. Elango, R. Radhakrishnan, W.A. Froland, BJ. Wallar, C.A. Earhart, J.D. Lipscomb
and D.H. Ohlendorf, Protein Sci., 6 (1997) 556.
[16] A.R. George, P. C. Wilkins and H. Dalton, J. Molec. Catal. B, 2 (1996) 103.
[17] K.J. Walters, G.T. Gassner, S.J. Lippard and G. Wagner, Proc.Natl.Acad.Sci.USA, 96
(1999)7877.
[18] S.L. Chang, P.J. Wallar, J.D. Lipscomb and K.H. Mayo, Biochemistry, 38 (1999) 5799.
[19] J. Lund and H. Dalton, Eur. J. Biochem., 147 (1985) 291.
[20] K.E. Liu, A.M. Valentine, D. Wang, B.A. Salifoglou and S.J. Lippard, J. Am. Chem.
Soc, 117(1995)10174.
[21] B. Brazeau and J.D. Lipscomb, Biochemistry, 39 (2000) 13503.
[22] L. Shu, J.C. Nesheim, K. Kauffmann, E. Miinck, J.D. Lipscomb and L. Que Jr., Science,
275(1997)515.
[23] A.M. Valentine, S.S. Stahl and S.J. Lippard, J. Am. Chem. Soc, 121 (1999) 3876.
[24] S.-K. Lee, J.C. Nesheim and J.D. Lipscomb, J. Biol. Chem., 268 (1993) 21569.
[25] B.J. Brazeau, B.J. Wallar and J.D. Lipscomb, J. Am. Chem. Soc, 123 (2001) 10421.
[26] J. Green and H. Dalton, J. Biol. Chem., 264 (1989) 17698.
[27] K. Yoshizawa, T. Yamabe and R. Hoffmann, New J. Chem., 21 (1997) 151.
[28] Y. Jin and J.D. Lipscomb, Biochim. Biophys. Acta, 1543 (2000) 47.
[29] M.H. Baik, M. Newcomb, R.A. Friesner and S.J. Lippard, Chem. Rev., 103 (2003)
2385.
[30] R.J. Deeth and H. Dalton, J. Biol. Inorg. Chem., 3 (1998) 302.
[31] B.G. Fox, J.L. Bourneman, L.P. Wackett and J.D. Lipscomb, Biochemistry, 29 (1990)
6419.
[32] J. Colby, D.I. Stirling and H. Dalton, Biochem. J, 165 (1977) 395.
[33] S.D. Prior and H. Dalton, J. Gen. MicrobioL, 131 (1985) 155.
[34] K.J. Burrows, A. Cornish, D. Scott and I.J. Higgins, J. Gen. MicrobioL, 130 (1984)
3327.
[35] A.S. Lindner, P. Adriaens and J.D. Semrau, Arch. MicrobioL, 174 (2000) 35.
[36] H.-H.T. Nguyen, K.H. Nakagawa, B. Hedman, S.J. Elliott, J.H. Yip, S.J. Jacobs, B.J.
Hales, M.E. Lidstrom and S.I. Chan, J. Biol. Chem., 269 (1994) 14995.
[37] D.D.S. Smith and H. Dalton, Eur. J. Biochem., 182 (1989) 667.
[38] A.K. Shiemke, S.A. Cook, T. Miley and P. Singleton, Arch. Biochem. Biophys., 321
(1995)421.
[39] J.A. Zahn and A.A. DiSpirito, J. Bacteriol., 178 (1996) 1018.
[40] H.-H.T. Nguyen, S.J. Elliott, J.H.-K. Yip and S.I. Chan, J. Biol. Chem., 273 (1998)
7957.
[41] P. Basu, B. Katterle, K.K. Andersson and H. Dalton, Biochem. J., 369 (2003) 417.
[42] R.L. Lieberman, D.B. Shrestha, P.E. Doan, B.M. Hoffman, T.L. Stemmler and A.C.
Rosenzweig, Proc Natl. Acad. Sci. USA, 100 (2003) 3820.
[43] D.-W. Choi, R.C. Kunz, E.S. Boyd, J.D. Semrau, W.E. Antholine, J.-I. Han, J.A. Zahn,
J.M. Boyd, A.M. de la Mora and A.A. DiSpirito, J. Bacteriol., 185 (2003) 5755.
[44] S. Stolyar, A.M. Costello, T.L. Peeples and M.E. Lidstrom, Microbiology, 145 (1999)
1235.
[45] B. Gilbert, I.R. McDonald, R. Finch, G.P. Stafford, A.K. Nielsen and J.C. Murrell,
Appl. Environ. MicrobioL, 66 (2000) 966.
[46] A.K. Nielsen, K. Gerdes and J.C. Murrell, Molec MicrobioL, 25 (1997) 399.
[47] S. Stolyar, M. Franke and M.E. Lidstrom, J. Bacteriol., 183 (2001) 1810.
192
[48] A.A. DiSpirito, J.A. Zahn, D.W. Graham, H.J. Kim, C.K. Lerive, T.S. Derrick, CD.
Cox and A. Taylor, J. Bacteriol, 180 (1998) 3606.
[49] H.-H.T. Nguyen, A.K. Shiemke, S.J. Jacobs, BJ. Hales, M.E. Lidstrom, K.O. Hodgson
and S.I. Chan, J. Am. Chem. Soc, 118 (1996) 12766.
[50] S.A. Cook and A.K. Shiemke, Arch. Biochem. Biophys., 398 (2002) 32.
[51] S.S.-F. Yu, L.-Y. Wu, K.H.-C. Chen, I.I. Luo, D.-S. Huang and S.I. Chan, J. Biol.
Chem., 278 (2003) 40658.
[52] S.J. Elliott, M. Zhu, L. Tso, H.-H.T. Nguyen, J.H.-K. Yip and S.I. Chan, J. Am. Chem.
Soc, 119(1997)9949.
[53] S. Lontoh, A.A. DiSpirito, C.L. Krema, M.R. Whittaker, A.B. Hooper and J.D. Semrau,
[54] S. Lontoh, J.A. Zahn, A.A. DiSpirito and J.D. Semrau, FEMS Microbiol. Lett., 186
(2000) 109-113.
[55] A.O. Richards, S.H. Stanley, M. Suzuki and H. Dalton, Biocatalysis, 8 (1994) 253.
[56] S.H. Stanley, A.O. Richards, M. Suzuki and H. Dalton, Biocatalysis, 6 (1992) 177.
[57] M. Suzuki, H. Dalton, A.O. Richards and S.H. Stanley, Canadian patent # 1,322,734.
(1993).
[58] S. Slade, T.J. Smith and H. Dalton, unpublished observations.
[59] P.W. Baker, H. Futamata, S. Harayama and K. Watanabe, 38 (2001) 87.
[60] M.C. White, A.G. Doyle and E.N. Jacobsen, J. Am. Chem. Soc, 123 (2001) 7194.
[61] D. Lee and S.J. Lippard, J. Am. Chem. Soc, 123 (2001) 4611.
[62] E.Y. Tshuva, D. Lee, W. Bu and S.J. Lippard, J. Am. Chem. Soc, 124 (2002) 2416.
[63] P.-P.J.H.M. Knops-Gerits and W.P. Goddard, Catal. Today, 81 (2003) 263.
[64] T.J. Smith, S.E. Slade, N.P. Burton, J.C. Murrell and H. Dalton, Appl. Environ.
Microbiol, 68 (2002) 5265.
Chapter 7
Biocorrosion
H.A. Videlaa and L.K. Herrerab
department of Chemistry. College of Pure Sciences, INIFTA, University of

La Plata, Argentina
b
Faculty of Engineering, University of Antioquia, Medellin, Colombia
1. INTRODUCTION
Microorganisms influence corrosion by changing the electrochemical

conditions at the metal/solution interface. These changes may have different
effects, ranging from the induction of localized corrosion, through a change in
the rate of general corrosion, to corrosion inhibition [1]. Any biological effect
that either facilitates or impedes one of the anodic or cathodic reactions of the
corrosion process, or permanently separates anodic and cathodic sites, will
increase corrosion. For instance, stimulation of the anodic reaction by acidic
metabolites or the cathodic reaction by microbial production of a cathodic
reactant like hydrogen sulfide, the breakdown of protective films or the increase
in conductivity of the liquid environment will enhance corrosion.
Although the electrochemical nature of corrosion remains valid for
microbial corrosion, the participation of the microorganisms in the process
induces several unique features, mainly the modification of the metal/solution
interface by biofilm formation. Biofilms affect the interaction between metal
surfaces and the environment, not only in biodeterioration processes like
corrosion, but also in several biotechnological processes applied to materials
recovery and handling [2]. Thus, the key to the alteration of conditions at a
metal surface, and hence the enhancement or inhibition of corrosion is the
formation of a biofilm [3]. This can be considered as a gel containing 95% or
more water, made of a matrix of exopolysaccharidic substances (EPS) in which
microbial cells, and inorganic detritus are suspended [4].
Biofilm formation is the result of an accumulation process -not
necessarily uniform in time or space [5]- that starts immediately after metal
immersion in the aqueous environment. A thin film (approximately 20-80 nm
194
thick), due to the deposition of inorganic ions and high relative molecular mass
organic compounds, is formed in a first stage. This initial film is able to alter the
electrostatic charges and wettability of the metal surface facilitating its further
colonization by bacteria. In a short time, (minutes or hours according to the
aqueous environment where the metal is immersed), microbial growth and EPS
production results in the development of a biofilm. This biofilm is a dynamic
system and the different transport processes and chemical reactions occurring at
the biofouled interface will take place from now on, through the biofilm
thickness [6] (Fig. 1).
Microbial colonization of metal surfaces drastically changes the classical
concept of the electrical interface commonly used in inorganic corrosion.
Important changes in the type and concentration of ions, pH values and
oxidation-reduction potential are induced by the biofilm, altering the passive or
active behavior of the metallic substratum and its corrosion products, as well as
the electrochemical parameters used for assessing corrosion rates [7].
Simultaneously with the biological changes that lead to biofilm
accumulation, a sequence of inorganic changes takes place at the metal surface
immediately after its immersion in an aggressive aqueous medium. This
sequence involves the process of metal dissolution and corrosion product
formation.
Fig. 1. Diagrammatic representation for biofilm development (from Ref. [8]).

195
Both biological and inorganic processes occur within the same time
period, but following opposite directions at the metal/solution interface.
Whereas corrosion and corrosion product accumulation occur from the metal
surface towards the solution, biofilm formation is the result of accumulation
processes directed from the bulk towards the metal surface (Fig. 2) [8]).
Fig 2. Sequence of biological and inorganic processes at a biologically and electrochemically

conditioned metal/solution interface (from Ref. [8]).
Thus, a very active interaction between the corrosion product layers and
the biofilms can be expected. The consequent corrosion behavior of the metal
will vary according to the degree of this reciprocal interaction and a concept of a
new biologically conditioned interface must be kept in mind [9]. The approach
for a sound interpretation of any microbial corrosion case must then be
interdisciplinary, and include a thorough process analysis combined with well
defined microbiological and electrochemical methodologies.
Biocorrosion has been focusing increasing attention from different
research areas in the last two decades as an answer to the demand of a wide
variety of industries. Fortunately an increasing intellectual and technical cross-
fertilization of ideas between researchers from different disciplines like
microbiology, electrochemistry and materials science has allowed a considerable
improvement in the understanding of biocorrosion to be reached.
2. SRB INDUCED CORROSION OF STEEL
Biocorrosion of carbon steel in anaerobic environments involving the

presence of SRB have been the focus of most biocorrosion research. Starting
196
with the cathodic depolarization theory (CDT) of von Wolzogen Kuhr & Van
der Vlugt [10] (Fig. 3), a copious list of papers and reviews on the anaerobic
corrosion of iron has been published [11-13]. Bacterial biofilms may develop
anaerobic regions, even in aerobic bulk water environments [2], thus allowing
SRB a very favorable environment for growth. The final result of these
processes within biofilms is to produce a wide variety of sites on the metal
surface that are markedly different from neighboring sites from a
physicochemical standpoint, thus facilitating the initiation of localized corrosion
processes.
Fig. 3. Sequence of reactions of the Cathodic Depolarization Theory. The three elements of
biocorrosion (metal/solution/microorganisms) are involved in different reactions of the whole
process.
As the localized corrosion and breakdown process is strongly dependent

of several experimental factors such as the type and concentration of aggressive
anions present in the medium and the passivating film characteristics, the effect
of sulfur anions has been studied in a series of laboratory experiments using
alkaline [14] and neutral buffered [15] solutions as well as SRB cultures in
saline media [16] under well defined experimental conditions. Taking into
account the results of these studies, a bioelectrochemical interpretation of the
biocorrosion process of carbon steel in anaerobic environments may be
summarized as follows [17]:
i) biogenic sulfides effects on carbon steel passivity breakdown is similar
to that of abiotic sulfides. The characteristics and intensity of sulfide effects on
the corrosion behavior of carbon steel is narrowly related with the nature of the
passive film already present on the metal surface (Fig. 4); ii) in neutral media
sulfide ions lead to the formation of a poorly protective film of mackinawite; iii)
the anodic breakdown of passivity would be the first stage of the corrosion
process.
197
Fig. 4. Scheme of the bioelectrochemical interpretation of the biocorrosion process of carbon

steel in anaerobic environments (from Ref. [17]).
Thus, the role of SRB may be indirect through the production of

aggressive species either as final (sulfides, bisulfides or hydrogen sulfide) or
intermediate metabolic compounds (thiosulfates, polythionates). Physico-
chemical characteristiscs of the liquid environment (pH, ionic composition,
oxygen levels) can modify the SRB effects which could eventually change from
corrosive to passivating; iv) cathodic depolarization effects attributed to SRB
hydrogenase activity or to iron sulfide films would be developed later than
passivity breakdown while corrosion process is in progress; v) the action of
biogenic sulfides can be enhanced by other aggressive anions already present in
the environment (e.g. chlorides) [18] or through microbial consortia within
adhered biofilms on the metal surface [19].
Further studies developed in our laboratory [20-23] allowed us to
conclude that SRB influenced corrosion of steel is markedly affected by the
nature and the structure of the sulfide films produced during the metal
dissolution. The environmental characteristics of the metal/biofilm/medium
interface and its surroundings (pH, ionic composition, oxygen levels, EPS
198
distribution) control the chemical and physical nature of corrosion product

layers and may change their effects on the metal behavior from corrosive to
protective. In marine environments, the impact of sulfur compounds on
corrosion is enhanced by other aggressive anions, such as the widely distributed
chlorides, already present in the medium.
The entrance of oxygen into the anaerobic environment accelerates
corrosion rate, mainly through a change in the chemical nature of iron sulfides
and elemental sulfur production. Both chemical species can provide additional
cathodic reactants to the corrosion process, acting as electron carriers between
the metal and the oxic interface within the biofilm.
2.1. Differences between biotic and abiotic media

Recently, several surface analysis techniques, electrochemical
experiments and microscopy observations were employed to clarify the role of
biotic and abiotic sulfide films in the corrosion behaviour of steel in saline
media [21, 23]. Microbiological experiments were performed under controlled
laboratory conditions using a strain of Desulfovibrio alaskensis (D. alaskensis),
known for its ability to produce EPS, isolated from a soured oil reservoir in
Alaska [24]. Atomic force microscopy (AFM) was applied to image the SRB
biofilms (Fig. 5), current transients were measured to determine the
electrochemical behaviour of steel, and SEM observations coupled with ED AX,
as well as XPS, XRD and electron microprobe analysis (EPMA) were carried
out to examine the structure and composition of biotic and abiotic sulfide films
on the surfaces of steel specimens.
The main conclusions obtained from these studies have been: i) both
abiotic and biotic iron sulfide films are related to the formation of tubercles on
the steel surface. However, for biotic solutions FeS (mackinawite) predominates,
whereas in abiotic media FeS2 (pyrite) is the major iron sulfide present (Fig. 6);
ii) the structure of the outer crust of iron sulfide acts as a barrier for the diffusion
of ions towards and from the pit cavity; iii) biotic films are more adherent to the
metal surface, while abiotic films are flaky and loosely attached; iv) the previous
history of the sulfide film may play a relevant role in the corrosion behaviour of
steel. Depending on the sulfide concentration in the medium and on the presence
or absence of biofilms and EPS, the protective characteristics of the corrosion
product may change. Biogenic layers of corrosion products can offer enhanced
protection by improving the adherence of the film to the metal, but can also
increase corrosion, inducing the presence of heterogeneities at the metal surface.
199
Fig. 5. AFM image of a SRB biofilm on AISI 316 SS (from Ref. [21]).
The differences between biotic and abiotic media containing identical

levels of corrosive compounds (i.e. iron sulfides) may be mainly attributed to the
presence of extracellular polymers and to the heterogeneities created at the metal
surface by the formation of a biofilm. In addition, the physicochemical
parameters of the corrosion product layers may change in the absence or a
presence of a biofilm, rendering these layers either protective or corrosive. This
feature can explain the observed differences in corrosion behavior of steel
between abiotic and biotic environments with respect not only to localized
pitting, but also to the hydrogen attack developed as embrittlement and crack
growth.
2.2. Hydrogen effects and SRB

Although it is widely accepted that SRB can oxidize molecular hydrogen
using their hydrogenase system, the involvement of this enzyme in the corrosion
process remains unclear. The present state of knowledge does not support the
hypothesis that the bacterial uptake of hydrogen is the rate-controlling step for
SRB-influenced corrosion of mild steel [25].
200
Fig. 6. XPS spectra of a biogenic sulfide film (top) and of an abiotic sulfide film (bottom)
(from Ref. [23]).
Hydrogen can have a considerable effect on susceptible alloys when it enters

into the atomic structure of the metal and causes embrittlement. This
phenomenon, including the acceleration of crack formation and growth, is
frequently found in industrial systems (i.e. off-shore oil platforms). In addition,
hydrogen, generated from cathodic protection can be a considerable problem in
areas of stress corrosion or corrosion-fatigue [26]. The amount of hydrogen
available to enter the steel and the ease with which it moves into the metal, is
influenced by the nature of the environment in which the steel is placed (Fig. 7).
201
Fig. 7. Hydrogen permeation through 50D steel, cathodically protected or unprotected and
coated or uncoated, exposed to open seawater and embedded in marine mud. 1 = Grit blasted
plus cathodic protection. 2 = CTE coating plus cathodic protection. 3 = anti-fouling paint. 4 =
grit basted only. 5 = As received (with mill scale, uncoated) (from Ref. [29]).
In seawater, where structures are cathodically protected from corrosion,

hydrogen is available at the metal surface, particularly in anaerobic
environments. Biological activity at the surface, especially bacterial activity, can
enhance the entry of hydrogen into the metal. This action can be direct, by
producing metabolites like sulfides which encourage the entry of hydrogen into
the steel, and indirect, by both disrupting any surface films and due to the need
of increasing cathodic protection when bacteria are present [27]. SRB are the
main bacteria responsible for the enhancement of hydrogen entry as they are
active in the areas where hydrogen will be generated under cathodic protection,
i.e. anaerobic environments such as marine muds (Fig. 7).
It is indisputable that a biologically active environment is able to induce
conditions conductive to the enhancement of corrosion-fatigue crack growth and
hydrogen embrittlement by the activity of SRB [28-30]. A "biologically active
environment" involves not just the presence of microorganisms such as SRB or
a single metabolite i.e. sulfide, but the effect of all the activities of the bacteria
and their interactions with other components of the environment {e.g.
degradation by large fouling organisms, production of EPS, interactions of
202
bacteria and the metal). Indeed, the local environment surrounding the metal
surface is very different from that without bacteria, even if the same levels of
sulfide are detectable.
These areas are particularly important, as pipelines and the bottoms of the
legs of offshore platforms can be buried in marine muds where anaerobic
conditions are predominant and SRB activity is intense (Fig. 8). Moreover, sour
environments, such as those frequently found in oil production activities, are
particularly aggressive due to high levels of hydrogen available at the metal
surface or in a crack, as a consequence of sulfide poisoning of the recombination
reaction at the cathode [31]. In such habitats hydrogen effects can be altered by
the presence of organic molecules on the metal surface and the existence of a
biofilm with its EPS matrix. This feature can explain the differences between
general embrittlement effects (as measured by hydrogen flux) and crak tip
effects (as measured by crack growth). Embrittlement results in the general
lowering of strength of the material causing it to fail in a catastrophic way at a
lower load as it has been reported in high sulfide biological environments more
than in low sulfide abiotic environments, even though the rest of the crack
growth curve is very similar (Fig. 9) [28].
EPS and other organic molecules related to biofilms hinder dissolution
and dissociation reactions and adsorption processes in the crack. Even under low
frequency cyclic loading the crack tip opens and closes rapidly. Thus, any effect
of the environment must occur fast and organic material dragged into the crack
could have a blocking impact.
The results referred here serve to illustrate the complex nature of the
interactions between SRB biofilm and the steel. In many cases, bacterial
metabolism within the biofilm generates sulfides, and consequently, this is the
main cause of the corrosiveness of the environment. However, microbial
metabolic activity is also responsible for the release of EPS which may have a
blocking effect on hydrogen entry into the metal.
3. BIOCORROSION OF ALUMINUM ALLOYS IN FUEL/WATER

SYSTEMS
Microbial contamination of hydrocarbon fuels is the main cause of serious

problems concerning the quality of maintenance of the product, as well as the
corrosion of metals used during the processes of extraction, production,
distribution, and storage of the fuel.
203
Fig. 8. Composite diagram of an offshore structure showing the main sites of biodeterioration
problems: 1. marine fouling; 2. drill cuttings around legs; 3. oil storage and transport; 4.
water-filled legs; 5. production system; 6. seawater injection system; 7. downhole pipework;
8. reservoir problems (from Ref. [31]).
Microoganisms usually present in droplets or thin films are sufficient to

allow microbial growth and the development of biocorrosion. Because of the
electrochemical nature of biocorrosion an aqueous environment is also required.
Fungal and bacterial growth can also occur on side walls not necessarily
adjacent to large waterbottoms. Hyphal growth of fungi usually ramify over
fuel/water interface becoming a site for further water entrapment. As a result of
this microbial propagation action, penetration and breakdown of fuel tank
coatings and passive films occurs on the metal, leading to the onset of localized
corrosion processes generally of the type of pitting attack.
Due to its economic and technological importance, the biodeterioration of
jet fuel and subsequent biocorrosion of aluminum alloys used in the aircraft
industry has been extensively studied [32-35]. Since the end of the 70's several
of our publications have been devoted to elucidate the mechanisms affecting the
whole biocorrosion process [36-39].
204
Fig. 9. Crack growth rates of a RQT 501 steel in biologically active H2S seawater
environments; solid line: crack growth rate in seawater; dotted line: crack growth rate in 520
ppm H2S in abiotic natural seawater (from Ref. [28]).
Different types of microorganisms usually present in soils and natural

waters utilize paraffmic hydrocarbons in the range C,0-CIg (kerosene fraction)
more easily than in the range C5-C9 (gasoline fraction). In a water-free fuel the
microbial detrimental activity is nil. Conversely, as soon as water becomes
available, the growth of microorganisms is possible by utilizing the
hydrocarbons as a source of carbon for their metabolic activity. Chemical
contaminants present in the fuel as well as in bilge water provide some kind of
nitrogen source and the necessary trace elements for growth. Therefore, the
major requirement for microbial activity is the presence of water in the storage
tank. On this respect, during the processing of hydrocarbon products is
practically impossible to avoid the existence of water. Although good
housekeeping procedures could minimize the amount of water accompanying
the fuel, the microorganisms are able to generate their own water phase for
further proliferation [40]. Microbial metabolism generate in this way, small
ecosystems which retain water in certain areas of the storage tank such as
tubercles, slime deposits or even biofilms on the metal surface. The microbial
ecology of aircraft fuel tanks and storage systems correspond to a wide diversity
205
of microorganisms, although in almost all cases the microbial contaminants

reported were fungi and bacteria [41]. The fungus Hormoconis resinae
(formerly Cladosporium resinae), and some species of the genera Aspergillus,
Penicillium and Fusarium have been isolated in jet fuels and fuel storage
systems [42]. H. resinae has also been reported as a contaminant of fuel oils for
marine and terrestrial turbine engines [43].
Due to the high rate of consumption of kerosene fuel by jet engines, even
small amounts of microbial sludges in the fuel becomes hazardous. The
combination of microorganisms and their associated water creates major
problems of fuel contamination, filter plugging, fuel gauge malfunction and fuel
tank corrosion. The corrosion of the tank wall and subsequent leakage of
hydrocarbon fuel can cause important economic losses, as well as different
troubles related with soil and underground water contamination [44]. The
corrosion attack is generally located at the tank bottom where there is an active
microbial population associated with free water.
To allow microbial growth, the environment must provide the basic
elements (carbon, oxygen, hydrogen, nitrogen, phosphorous) and some other
elements needed in much smaller amounts, but nevertheless essential for a
normal metabolism. Although carbon and hydrogen are abundantly supplied by
the hydrocarbon chains of the fuel (usually 95% of the product), an essential
element for their biodegradation is oxygen. At low concentrations of oxygen, the
rate of hydrocarbon oxidation is lower, although in aircraft fuel tanks there is a
periodically replenishment of oxygen during tank refuelling. Usually, the
nitrogen and phosphorous availability in kerosene fuel is the limiting factor for
microbial growth. These elements are generally present as nitrates and
phosphates, either in the water phase or in the additives present in the fuel.
The action of fungal contaminants of jet fuels in corrosion can be accomplished
through:
a) Local increase in the proton concentration, derived from organic acidic
metabolites,
b) Greater oxidizing characteristics of the medium favoring pitting attack,
c) Metabolite production decreasing the surface energy of the interface
passive film/electrolyte
d) Microbial adhesion processes enhancing metal dissolution, and
e) Microbial uptake of corrosion inhibitors (mainly nitrates and
phosphates).
A simplified scheme to illustrate the mechanisms influencing biocorrosion
of aluminum alloys in fuel / water systems is shown in Figure 10 [45].
A protective action of some contaminating bacteria (e.g. S. marcescens),
can be developed when an active degradation of hydrocarbon chains is not
accompanied by important reduction in pH levels. In those circumstances, the
organic anions enhance the passive behavior of aluminum [1].
206
Fig. 10. Simplified scheme of the initiation of pitting attack on aluminum alloys in fuel/water
systems (from Ref. [45]).
4. AEROBIC CORROSION OF IRON
Metal-oxidizing microorganisms create environments for the accumulation of

chloride ions, forming acidic ferric chloride and manganic chloride, which are
corrosive to steel. However, the main mechanism to explain the action of these
microorganisms in biocorrosion is the formation of differential aeration cells
associated with tubercles. A typical aeration cell of this kind is formed when the
tubercle structure at its outer area acts as a cathode due to the continuous oxygen
supply from the water. Oppositely, an anode (anaerobic area ) is formed at the
bottom of the formation, where access of oxygen is limited.
In all cases corrosion acceleration is mainly due to the presence of SRB in
the inner area of the tubercles where anaerobic conditions are created. Thus, it is
highly feasible that the fastidious iron and manganese depositing bacteria
require other organisms to create suitable conditions for their growth. One
important feature of this case of biocorrosion is that once the tubercle (and the
electrochemical cell) is formed, even the dead of the microorganisms does not
207
extinguish the differential aeration mechanism, since a substantial barrier to the

intake of oxygen has been established.
Thus, biocorrosion of cast iron in low flow or stagnant regions of water
distribution pipelines is mainly due to tubercles formed by iron-oxidizing and
manganese-oxidizing bacteria. These bacteria are able to oxidize soluble ferrous
compounds to less soluble ferric compounds (e.g. ferric hydroxide) which
precipitate at the inner wall of the pipelines under the form of tubercles (discrete
hemispherical mounds).
Tubercles create environments suitable for the growth of other hazardous
microorganisms like the SRB which are generally found at the inner (anaerobic)
regions of the deposits. Near the tubercles other aerobic microorganisms like
slime-forming bacteria (e.g. Pseudomonas) are frequently found. In addition,
tubercles are able to impede the penetration of biocides or corrosion inhibitors
diminishing their effectiveness.
Finally, iron-oxidizing bacteria are able to concentrate chlorides and manganese
ions forming acidic ferric chloride and manganic chloride which are highly
corrosive.
5. MICROBIAL INHIBITION OF CORROSION
Corrosion inhibition is the slow down of the corrosion reaction usually

performed by substances which, when added in small amounts to an
environment, decrease the rate of attack by this environment on a metal
(corrosion inhibitors). Microorganisms are able to drastically change the
electrochemical conditions at the metal/solution interface. These changes can
range from the induction or acceleration of corrosion to corrosion inhibition.
There are several examples of microbial effects that could enhance corrosion: i)
stimulation of the anodic reaction by acidic metabolites or the cathodic reaction
by microbial production of a new alternative cathodic reactant (e.g. H 2 S); ii) the
microbial breakdown of protective films or iii) the increase in conductivity of
the liquid environment. However, microbial effects causing corrosion inhibition
has been seldomly mentioned in the literature [46].
Microorganisms can aid to achieve corrosion inhibition according to some
of the following mechanisms: i) neutralizing the action of corrosive substances
present in the environment; ii) forming protective films or stabilizing a pre-
existing protective film on a metal; iii) inducing a decrease in the medium
corrosiveness.
As general key features of microbial inhibition of corrosion the following
points can be summarized [47]:
- Biocorrosion and its counter-process, microbial inhibition of corrosion, are
rarely linked to a single mechanism of to a single species of microorganisms.
208
- Either the corrosive or the inhibitory action of bacteria are developed at

biofilmed metal surfaces where complex biofilm/protective films interactions
occur. Biological activity leads to important changes in the type and
concentrations of ions, pH and oxygen levels inducing significant variations in
the physical and chemical characteristics of the environment as well as in the
electrochemical parameters used to measure the corrosion rate.
- The main mechanisms of corrosion inhibition by bacteria are always linked to
a marked modification of the environmental conditions at the metal/solution
interface by biological activity.
- Microbial corrosion inhibition is frequently accomplished through: i) a
decrease in the cathodic rate by microbial consumption of a cathodic reactant
(e.g. oxygen consumption by respiratory activity); ii) decreasing the medium
aggressiveness in restricted areas of the metal solution interface (e.g. by
neutralizing acidity); iii) providing or stabilizing protective films on the metal
(e.g. biofilm exopolymers with metal binding capacity).
- It must be stressed that in practical situations, the inhibitory action of bacteria
can be reversed to a corrosive action within bacterial consortia structured in
biofilm thickness.
A proper understanding of the identity and role of microbial contaminants
in the specific environment of a metal surface may be used to induce corrosion
inhibition by bacteria as a useful tool to prevent frequent biodeterioration effects
encountered in practice.
6. ELECTROCHEMICAL INTEPRETATION OF BIOCORROSION
The basic concepts of electrochemical corrosion are valid in biocorrosion and

could be used to interpret the acceleration of the corrosion process by
microorganisms in different aqueous media both under anaerobic or aerobic
conditions.
Biocorrosion is rarely interpreted by a single mechanism or rarely caused
by single species of microoganisms [7]. Hence, it is important to be cautious in
the interpretation of data supplied by electrochemical methods. Frequently, these
techniques have been used in complex media where the characteristics and
properties of passive films are not well understood. For instance, the presence of
complex deposits of corrosion products, metabolites and EPS may dramatically
reduce the usefulness of some electrochemical results. Moreover, it has to be
kept in mind that microbial colonization of passive metals can drastically change
their resistance to breakdown by changing locally the type and concentration of
ions, pH values, oxygen gradients and inhibitor levels. These changes should
result in important alterations in the electrochemical behavior of the metal and in
the electrochemical parameters measured in laboratory experiments. In
209
conclusion, whenever electrochemical techniques will be used for the evaluation

of biocorrosion, the actual condition of the metal surface must be considered [9].
In classical electrochemical studies, the interface between a metal and the
surrounding electrolyte has been characterized by a certain distribution of
electrical charges giving rise to the so-called electrical double layer [8]. The
knowledge about the structure of the double layer at the metal/solution interface
is mainly based on experimental data obtained with the dropping mercury
electrode. Thus, the behavior of the interface between mercury and several
aqueous electrolyte solutions could be considered approximately equal to that of
an ideally polarizable interface. However, at the light of the present knowledge
on biocorrosion, it can be easily inferred that this behavior is markedly different
from that corresponding to the complex metal/solution interface associated with
biocorrosion [9]. Consequently, electrochemical concepts used for inorganic
corrosion analysis will have to be adapted to the characteristics of the
biologically conditioned interface.
A detailed description of the electrochemical methods for evaluating
biocorrosion is out of the scope of this chapter. A wide variety of electrochemical
techniques such as corrosion and redox potentials measurements, Tafel and
potentiodynamic polarization, linear polarization and electrical resistance
probes, and several modern electrochemical techniques like alternating current
methods or electrochemical noise have been reviewed by several authors in
relation to their use in biocorrosion evaluation [48-50].
A careful use of several electrochemical concepts and methods should be
coupled when possible, with visual inspection, microscopy, innovative surface
analysis techniques and a sound microbiological analysis to reach an adequate
characterization of the causative microorganisms, their interactions and the
effects of their metabolic activities and products on corrosion.
7. PREVENTION AND CONTROL
One of the classic concepts for maintaining an industrial system free of

biodeteriorating effects is "to keep the system clean". Although this is a very
difficult task to accomplish in practice, there are several general methods that
can be used. These methods can be broadly classified in: i) physical and, ii)
chemical. Among the former, flushing is perhaps the most simple, although of
limited efficacy. A special case is the use of flushing supported by cleaners or
jointly with chemical agents that induce biofilm detachment. Abrasive or non-
abrasive sponge balls are frequently employed in the industry. The former could
present problems related with protective passive films that can be damaged, and
the second is not very effective with thick biofilms.
With reference to chemical methods, the most common approach for
controlling biofouling problems in industrial water systems, is the use of
210
biocides. These substances can be either oxidizing or non-oxidizing toxicants.

Chlorine, ozone and bromine are three typical oxidizing agents of industrial use.
Non-oxidizing biocides are claimed to be more effective than oxidizing biocides
for an overall control of algae, fungi, and bacteria. They have a greater
persistence and many of them are pH independent. Often a combination of
oxidizing and non-oxidizing biocides or two non-oxidizing biocides is used to
optimize microbiological control of industrial water systems. Typical biocides of
the second type are formaldehyde, glutaraldehyde, isothiazolones and quaternary
ammonium compounds.
Increasing legislative requirements and the necessity for greater
environmental acceptability, have contributed to restrict the use of some
traditional biocides and to develop either new compounds or carefully selected
blends of existing biocides.
Table 1
Biocides used in industrial water systems. Properties and usual concentrations.
Chlorine: effective against bacteria and algae; oxidizing; pH dependent; concentration range:
0.1-0.2 mg/1 (continuous treatment)
Chlorine dioxide: effective against bacteria, in a lesser extent against fungi and algae;
oxidizing; not dependent on the pH; concentration range: 0.1-1.0 mg/1
Bromine: effective against bacteria and algae; oxidizing; wide pH range; concentration
range: 0.05-0.1 mg/1
Ozone: effective against bacteria and biofilms; oxidizing; pH dependent; concentration
range: 0.2-0.5 mg/1
Methylene-bis-thiocyanate: effective against bacteria; non-oxidizing; hydrolyses at pH
higher than 8.0; concentration range: 1.5-8.0 mg/1
Isothiazolones: effective against bacteria, algae and biofilms; non-oxidizing; not dependent
on the pH; concentration range: 0.9-10 mg/1
QUATS: effective against bacteria and algae; non-oxidizing; surface activity; concentration
range: 8-35 mg/1
Glutaraldehyde: effective against bacteria, algae, fungi and biofilms; non-oxidizing; wide
pH range; concentration range: 10-70 mg/1
THPS (tetra kis-hydroximethil phosphonium): effective against bacteria, algae and fungi;
low environmental toxicity; specific action againts BRS.
211
Taking into account environmental concerns, the use of ozone for

different types of industrial water systems, presents several advantages with
respect to other biocides [51]. The unique combination of high toxicity of ozone
during treatment, with no toxicant discharge, could make ozone the biocide of
choice for the present decade, if an appropriate balance between positive effects
and costs is reached. Several publications from our laboratory on ozone biocidal
action on sessile and planktonic bacteria, its mechanisms of disinfection and the
optimization of its use can be found in the literature [52].
Among the most promisory non-oxidizing biocides, the THPS (tetra kis-
hydroximethil phosphonium) is a new compound of wide spectrum: effective on
bacteria, fungi and algae. It is being widely used in the oil industry due to its
capacity of dissolving the ferrous sulfide. Its main advantange is its low
environmental toxicity.
8. MONITORING BIOCORROSION
Monitoring programs for biofouling and biocorrosion have been mainly focused
in the assessment of planktonic populations in water samples, and generalized
corrosion by using corrosion coupons or some kind of resistance or polarization
resistance probes.
The main objections to these monitoring programs are: i) the planktonic
population does not properly reflect the type and number of organisms living in
the biofilm and causing biodeterioration problems; ii) susceptibility of
planktonic microorganisms to antimicrobial agents markedly differs from that of
sessile microorganisms within the biofilm, mainly because of the protective
action of their EPS. Thus, the monitoring methods adopted must provide
information of well-established biofilms like those developed in system water.
From the corrosion side, the electrical resistance method is appropriate for
indicating a change in the general corrosion rate, but the results are difficult to
interpret in the presence of localized corrosion like pitting, the most frequent
form of attack found in biocorrosion cases [48]. If biofilms or localized
corrosion are present, the polarization resistance will reveal that something is
happening, but may not give an accurate measure of the corrosion rate. Only the
use of any of these techniques jointly with other electrochemical methods or
parameters assessing localized corrosion hazard can provide valuable data for
monitoring the deleterious effects of biocorrosion and biofouling.
Owing to the variables of dissimilar nature involved in biofouling and
biocorrosion, an effective monitoring program, either for the laboratory or the
field, must necessarily supply information on water quality, corrosive attack,
sessile and planktonic bacteria populations, biofilms characteristics, and
chemical composition of inorganic and biological deposits [53].
212
Fig. 11. RENAprobe sampler device.
Sampling devices for monitoring biocorrosion and biofilms can be

simultaneously used to assess corrosion attack after the removal of biological
and inorganic deposits, giving a wider and more useful information. Sampling
devices fall into two main types: a) directly implanted and, b) side-stream
implanted. Metal coupons, generally made with the same structural material of
the system, present a known surface area, to enable an accurate count of sessile
bacteria per square cm. after biofilm detachment. Coupons are mounted in
holding assemblies which are inserted in the pipework of the laboratory or
industrial system.
Two practical cases of monitoring biocorrosion in industrial waters
developed in our laboratory have been described in the literature [54, 55].
8.1. Monitoring biocorrosion in chemical industry cooling water systems

A monitoring program based on laboratory and field measurements for
assessing biodeterioration on mild steel and stainless steel in recirculating
cooling water systems has been reported [54]. This program was based on: i)
water quality control; ii) corrosion monitoring in the field (weight loss and linear
polarization probe); iii) laboratory corrosion tests (polarization techniques and
corrosion potential vs. time measurements), and iv) use of a multipurpose
sampling device that allows monitoring of sessile populations, biofilms,
corrosive attack, morphology and intensity, and biological and inorganic
deposits analysis. The side stream sampling device (RENAprobe™) allows
realiable biofilm and corrosion sampling without introducing important
modifications in the system (Fig. 11).
213
8.2. Oilfield water injection system

Among the several factors related to biocorrosion hazard in secondary oil
recovery operations are: i) velocity, temperature, oxygen level and redox
potential of the injection water; ii) chemical composition, pH amount of organic
matter and depth of pumping of the injection water and; iii) effectivenes of the
biocide to gain access to the biofilms. Taking into account these factors a
multiple monitoring program for biofouling and biocorrosion was implemented
in an oild field in Argentina [55]. The goal of the program was to perform a
rapid evaluation of the bacteriological status of the system, and to implement an
effective biocide treatment for sessile bacteria.
9. NEW DEVELOPMENTS IN BIOCORROSION
Recent improvements in analytical, microbiological, electrochemical and

microscopical techniques and instrumentation have allowed the development of
new methods for laboratory and field assessment of biocorrosion in industrial
systems.
Chemical analysis within the biofilm by means of microsensors is one of
the most exciting advances in instrumentation [56]. Biofilm systems have been
considered to be diffusion limited [57]. As a consequence, chemical conditions
at the surface and within biofilms can vary dramatically over a distance of a few
micrometers. Thus, the information obtained from bulk water analysis has a
limited value and must be closely analysed before any conclusions are drawn
about the behavior of the metal/solution interface. Direct measurements inside
biofilms are restricted by: i) the small thickness of the biofilm; ii) the diffusion
limitation of concentration profiles across the biofilm; iii) the heterogeneous
nature of the biofilm. The latter aspect is specially important not only in relation
to microbial coverage of the surface but also with respect to biocorrosion. An
example of microsensor technology applied to evaluate vertical profiles of
chemical species in biofilm systems has been reported [58]. Equally important
tools for the study of biofilm structure are the fiber optic microprobe or optrode
used for finding the location of the biofilm/bulk water interface, and the
mapping of electric fields [59] by means of the scanning vibrating microprobe
(SVM).
Advanced microbiological techniques such as DNA probes have also been
applied to biocorrosion and biofouling research [60]. Although these techniques
are restricted to the laboratory at present, their joint utilization with
microbiological field measurements can be highly useful for monitoring
biocorrosion.
Recent developments in microscopy such as the environmental scanning
electron microscope (ESEM), the confocal laser microscope (CSL) and the
214
atomic force microscope (AFM) permit biofilm observation in real time and
without intoducing distortion of the samples. There is an increasing number of
references using these innovative technologies in recent biocorrosion literature
[61-63].
A combination of CSL and microelectrode techniques allowed correlation
of oxygen concentration profiles with biofilm structure [64]. CSL facilitates the
visualization of biofilm structures by eliminating the interference arising from
out of focus objects [65]. Observations performed under flow conditions and
using physiologically active biofilms, provided information to construct a new
conceptual model of biofilm structure.
On the corrosion side, new electrochemical test methods for the study of
localized corrosion phenomena in biocorrosion analysis and monitoring have
been reported [66]. As an example, an electrochemical sensor for monitoring
biofilms on metallic surfaces in real time has been recently presented [67]. The
system provides an immediate indication of the condition of biological activity
on probe surfaces and it is a powerful tool to optimize biocide treatment (Fig.
12).
10. CONCLUDING REMARKS
Biocorrosion is rarely linked to a single mechanism or to a single species of

microorganisms.
Biofilms mediate the interaction between metal surfaces and the liquid
environment. This interaction leads to an important modification of the
metal/solution interface drastically changing the types and concentrations of
ions, pH and oxygen levels.
Biofilm/corrosion product layers interactions at the metal/solution
interface condition the electrochemical behavior of metal surfaces in biological
media.
Acknowledgement
H.A. Videla acknowledges the financial support of the Agencia de Promocion
Cientifica y Tecnologica of Argentina through the project PICT/99 6782 on
Biodeterioration of materials.
215
Fig. 12. Scheme of an electrochemical sensor for monitoring biofilms (from Ref. [67])
REFERENCES
[1] H.A. Videla, Corrosion/96, paper No. 223, NACE International, Houston, TX (1996).
[2] H.A. Videla and W.G. Characklis, Int. Biodeterior. Biodegr., 29 (1992) 195.
[3] R.G.J. Edyvean and H.A. Videla, Interdisc. Sci. Rev., 16 (1991), 267.
[4] G.G. Geesey, Am. Soc. Microbiol. News, 48, (1982) 9.
[5] W.G. Characklis and K.C. Marshall, Biofilms: A Basis for an interdisciplinary
approach. In: Biofilms, W.G. Characklis and K.C. Marshall (eds.) John Wiley and Sons
Ltd, New York, 1990, pp. 3-15.
[6] W.G. Characklis, Biotech. Bioeng., 23, (1981) 1923.
[7] H.A. Videla, Metal Dissolution/redox in Biofilms. In: Structure and Function of
Biofilms, W.G. Characklis and P.A. Wilderer (eds.) John Wiley & Sons, Chichester,
U.K, 1989, pp. 301-320.
[8] H.A. Videla, Microbially induced corrosion: An updated overview. In: Biodeterioration
and Biodegradation 8, H.W. Rossmoore (ed.) Elsevier Applied Science, London, UK,
1991, pp. 63-88.
[9] H.A. Videla, Electrochemical aspects of biocorrosion. In: Bioextraction and
biodeterioration of metals, C.C. Gaylarde and H.A. Videla (eds.) Cambridge University
Press, Cambridge, UK, 1995, pp. 85-127.
216
[10] G.A.H. von Wolzogen Kiihr and L.R. Van der Vlugt, Water (den Haag), 18, (1934) 147
(Translation in Corrosion 17, (1961) 293).
[11] J.D.A. Miller, Metals, in: Microbial biodeterioration, A.D. Rose (ed.), Academic Press,
New York, 1981, pp. 149-202.
[12] R.C. Salvarezza and H.A. Videla, Corrosion, 36 (1980) 550.
[13] G. Gragnolino and O.H. Tuovinen, International Biodeterioration, 20 (1984) 9.
[14] R.C. Salvarezza, H.A. Videla and AJ. Arvia, Corros. Sci., 22 (1982) 815.
[15] R.C. Salvarezza, H.A. Videla and AJ. Arvia, Corros. Sci., 23 (1983) 717.
[16] H.A. Videla, Corrosion of mild steel induced by sulfate-reducing bacteria -a study of
passivity breakdown by biogenic sulfides. In: Biologically Induced Corrosion, S.C.
Dexter (ed.), NACE-8, Houston, TX, 1986, pp. 162-171.
[17] H.A. Videla, Electrochemical interpretation of the role of microorganisms in corrosion.
In: Biodeterioration 7, D.R. Houghton, R.N. Smith and H.O.W. Eggins (eds.) Elsevier
Applied Science, London, UK, 1988, pp. 359-371.
[18] C.A. Acosta, R.C. Salvarezza, H.A. Videla and A.J. Arvia, Electrochemical behaviour
of mild steel in sulfide and chloride containing solutions, Passivity of Metals and
Semiconductors, M. Froment (ed.) 1984, pp. 387-392.
[19] J.W. Costerton and G.G. Geesey, The microbial ecology of source colonization and of
consequent corrosion. In: Biologically Induced Corrosion, S.C. Dexter (ed.) NACE-8,
Houston, TX, NACE, 1986, pp. 223-232.
[20] H.A. Videla, C.L. Swords, M.F.L. de Mele, R.G. Edyvean, P,Watkins and I.B. Beech,
Corrosion/98, paper No. 289, NACE International, Houston, TX (1998).
[21] H.A. Videla, R.G. Edyvean, C.L. Swords, M.F.L. de Mele and I.B. Beech,
[22] H.A. Videla, Biofouling, 15 (2000) 37.
[23] H.A. Videla, C.L. Swords and R.G. Edyvean, Corrosion/2002, paper No. 02557, NACE
International, Houston, TX (2002).
[24] I.B. Beech, V. Zinkevich, R. Tapper and R. Avci, Journal of Microbiological Methods,
36(1999)3.
[25] W. Lee, Z. Lewandowski, P.H. Nielsen and W.A. Hamilton, Biofouling, 8 (1995) 165.
[26] Turnbull (ed) Hydrogen transport and cracking in metals, London Institute of Materials,
UK, 1995.
[27] S.G. Gomez de Saravia, M.F.L. de Mele, H.A. Videla and R.G.J. Edyvean, Biofouling,
11 (1997) 1.
[28] CJ. Thomas, R.G.J. Edyvean and R. Brook, Biofouling, 1 (1988) 65.
[29] R.G.J. Edyvean, J. Benson, CJ. Thomas, I.B. Beech, and H.A. Videla, Corrosion/97,
paper No. 206, NACE International, Houston, TX (1997).
[30] R.G.J. Edyvean, J. Benson, I.B. Beech and H.A. Videla, Microbiological influences on
hydrogen effects on steels. In: Hydrogen Energy Progress XII, vol. 3, J.C. Bolcich and
T.N. Veziroglu (eds.) 1998, pp. 1755-1762.
[31] R.G.J. Edyvean, International Biodeterioration, 23 (1987) 199.
[32] E.C. Hill, Biodegradation of petroleum products. In: Petroleum Microbiology, R.M.
Atlas (ed.), Chap. 15, MacMillan Publishing Co., New York, 1984, p. 579-617.
[33] P. Me Kenzie, A.S. Akbar and J.D. Miller, Fungal Corrosion of Aircraft Fuel Tank
Alloys, Technical paper, The Institute of Petroleum, London, UK (1977) 37.
[34] H.G. Hedrick, Materials Protection, 9 (1970) 27.
[35] D.G. Parbery, Material und Organismen, 6 (1971) 161.
[36] M.F.L. de Mele, R.C. Salvarezza and H.A. Videla, International Biodeterioration
Bulletin, 15 (1979) 39.
217
[37] R.C. Salvarezza, M.F.L. de Mele and H.A. Videla, International Biodeterioration
Bulletin, 15 (1979) 125.
[38] R.C. Salvarezza, M.F.L. de Mele and H.A. Videla, Br. Corros. J., 16 (1981) 162.
[39] R.C. Salvarezza and H.A. Videla, Electrochemical behavior of aluminum in
Cladosporium resinae cultures, in: Biodeterioration 6, S. Barry, D.R. Houghton, G.C.
Llewellyn and C.E. O' Rear (eds.) CAB International Mycological Institute, London,
UK, 1986, p. 212-218.
[40] RJ. Watkinson, Hydrocarbon degradation, in: Microbial Problems and Corrosion in Oil
and Oil Products Storage, E.C. Hill (ed.) The Institute of Petroleum, London, UK, 1984,
pp. 50-56.
[41] R.A. Neihof and M. May, International Biodeterioration Bulletin, 19 (1983) 59.
[42] H.A. Videla, P.S. Guiamet, S.M. do Valle and E.H. Reinoso, Corrosion/88, paper No.
91, NACE, Houston, TX (1988).
[43] D. Allsopp and K.J. Seal, Introduction to Biodeterioration, Edward Arnold, London,
1986.
[44] S. Holmes, Microbiology of hydrocarbon fuels. In: Proceedings 2nd International
Conference on Long-Term Storage Stabilities of Liquid Fuels, Southwest Research
Institute, San Antonio, TX, 1986, 336.
[45] H.A. Videla, Practical Cases. In: Manual of Biocorrosion, Chapter 7, Boca Raton,
Florida: CRC Lewis Publishers, 1996, pp. 179-220.
[46] H.A. Videla, Corrosion inhibition in the presence of microbial corrosion. In: Reviews
on Corrosion Inhibitor Science and Technology, Vol 2, A. Raman and P. Labine (eds.),
Chapter IX, NACE International, Houston, TX, 1996, pp. 1-11.
[47] H.A. Videla, Corrosion Inhibition by Bacteria. In: Manual of Biocorrosion, Chapter 5,
Boca Raton, Florida: CRC Lewis Publishers, 1996, pp. 121-135.
[48] S.C. Dexter, D.J. Duquette, O.W. Siebert and H.A. Videla, Corrosion/89, paper No.
616, NACE International, Houston, TX (1989).
[49] D.J. Duquette, Electrochemical techniques for evaluation of Microbiologically
Influenced Corrosion processes. Advantages and disadvantages In: Argentine-USA
Workshop on Biodeterioration (CONICET-NSF), H.A. Videla (ed.) Aquatec Quimica
S.A., Sao Paulo, Brazil, 1996, pp. 15-32.
[50] F. Mansfeld and B J. Little, Corrosion/90, paper No. 108, NACE International,
Houston, TX (1990).
[51] J.M. Brook and P.R. Puckorius, Corrosion/91, paper No. 212, NACE International.
Houston. TX (1991).
[52] H.A. Videla, M.R. Viera, P.S. Guiamet and M.F.L. de Mele, Corrosion/99, paper No.
186, NACE International, Houston, TX (1999).
[53] H.A. Videla, Detection, Identification and Monitoring. In: Manual of Biocorrosion,
Chapter 6, Boca Raton, Florida: CRC Lewis Publishers, 1996, pp. 137-178.
[54] H.A. Videla, F. Bianchi, M.M.S. Freitas, C.G. Canales and J.F. Wilkes, Monitoring
biocorrosion and biofilms in industrial waters: a practical approach. In: Microbiological
Influenced Corrosion Testing, J.R. Kerns and B.J. Little (eds.) ASTM Publications STP
1232, American Society for Testing and Materials, Philadelphia, PA, 1994, pp. 128-
137.
[55] H.A. Videla, P.S. Guiamet, O.R. Pardini, E. Echarte, D. Trujillo and M.M.S. Freitas,
[56] Z. Lewandowski, Dissolved oxygen gradients near microbially colonized surfaces. In:
Biofouling and biocorrosion in industrial water systems, G.G. Geesey, Z. Lewandowski
and H.C. Flemming (eds.) Lewis Publishers, Boca Raton, FL, 1994, pp. 175-188.
218
[57] W.G. Characklis, Biofilm processes, in: Biofilms, W.G. Characklis and K.C. Marshall
(eds.) John Wiley & Sons, New York, 1990, pp. 195-231.
[58] Z. Lewandowski, T. Funk, F. Roe and B. Little, Spatial distribution of pH at mild steel
surfaces using an iridium oxide microelectrode. In: Microbiologically Influenced
Corrosion Testing, J.R. Kearns and BJ. Little (eds.) ASTM Publications STP 1232,
Philadelphia, PA, 1994, pp. 61-69.
[59] Z. Lewandowski, F. Roe, T Funk and D. Chen, Proc. NSF-CONICET Workshop,

Biocorrosion and Biofouling: Metal/Microbe Interactions, Buckman Laboratories
International, Inc., Memphis, TN, 1993, pp. 52-61.
[60] S. Le Borgne, J. Jan, J.M. Romero and M. Amaya, Corrosion/2002, paper No. 02461,
NACE International, Houston, TX, (2002).
[61] A. Steele, D.T. Goddard and I.B. Beech, Int. Biodet. Biodegrad. 34 (1994) 35.
[62] A. Steele, I.B. Beech and D.T. Goddard, Proc. 1995 International Conference on
Microbialy Influenced Corrosion, American Welding Society-NACE International,
Houston, TX, 1995, 73/1-13.
[63] J.W. Costerton, Structure of biofilms. In: Biofouling and Biocorrosion in Industrial
Water Systems, G.G. Geesey, Z. Lewandowski and H.C. Flemming (eds.) Lewis
Publishers, Boca Raton, FL, 1994, pp. 1-14.
[64] Z. Lewandowski, P. Stoodley and F. Roe, Corrosion/95, paper No. 222, NACE
International, Houston, TX (1995).
[65] D.E. Caldwell, D.R. Korber and J.R. Lawrence, Adv. Microbial Ecol., 12 (1992) 1.
[66] H.A. Videla, Fundamentals of electrochemistry. In: Manual of Biocorrosion, Chapter 4,
Boca Raton, Florida: CRC Lewis Publishers, 1996, pp. 73-120.
[67] G.J. Licina, Corrosion/2001, paper No. 01442, NACE International, Houston, TX
(1995).
Chapter 8
Molecular tools in microbial corrosion

X. Zhu and J.J. Kilbane II
Gas Technology Institute, 1700 S. Mt. Prospect Rd., Des Plaines IL 60018
1. INTRODUCTION
Corrosion is a leading cause for pipe failure, and is a main component of

the operating and maintenance costs of gas and oil industry pipelines [1-5].
Quantifying the cost of corrosion generally in the gas and oil industry, and more
specifically the cost associated with microbial corrosion, is not easily done and
is controversial. Corrosion was estimated in 2001 to cost the gas and oil industry
about $13.4 billion/yr and of this as much as $2 billion/yr may be due to MIC
[6]. Basic research to increase our understanding of the microbial species
involved in microbial corrosion and their interaction with metal surfaces and
with other microorganisms will be the basis for the development of new
approaches for the detection, monitoring, and control of microbial corrosion. A
thorough knowledge of the causes of MIC and an efficient and effective means
of detecting and preventing corrosion is lacking. It is well recognized that
microorganisms are a major cause of corrosion of metal pipes, but despite
decades of study it is still not known with certainty how many species
ofmicroorganisms contribute to corrosion, how to reliably detect their presence
prior to corrosion events, or how to rapidly assess the efficacy of
biocides/mitigation procedures [2-4, 7-12].
Investigations of microbial species present in gas and oil industry
pipelines have traditionally relied upon using samples obtained from pipelines to
grow bacterial cultures in the laboratory [11]. Laboratory growth medium
cannot accurately reflect the true conditions within pipelines and microbiologists
have recognized that the vast majority of microbial species cannot currently be
grown in the laboratory [13-17]. Thus, culture-dependent approaches
underestimate the biocomplexity of microbial communities. Genetic methods
can be used to overcome the difficulties associated with the laboratory
cultivation of bacteria and provide a direct analysis of samples. In the past
decade the use of genetic techniques to detect, identify, and quantify bacteria in
220
the medicine, food, and cosmetic industries has largely replaced microbial
growth tests. These modern biotechnology methods are only beginning to be
employed in the gas and oil industry for problems related to microbiologically
influenced corrosion but it is likely that genetic techniques will be the methods
of choice for monitoring MIC in the future. Initial efforts to introduce the use of
genetic techniques for monitoring MIC or other environmental samples have
involved a type of DNA hybridization test called reverse sample genome
probing (RSGP). There are other hybridization-based genetic techniques
including whole cell in situ fluorescent hybridization, DNA amplification
followed by hybridization (dot-blot hybridization) or gel electrophoresis
(denaturing gradient gel electrophoresis) [17-21] that could also be used to
examine MIC samples, but this has not yet occurred. Another type of genetic
test method that could be used to investigate MIC samples is based on DNA
amplification using the polymerase chain reaction (PCR). PCR-based
approaches include quantitative competitive PCR (cPCR), quantitative real-time
PCR (q-PCR), and reverse transcriptase PCR (RT-PCR) [22-25]. This chapter
summarizes the status of genetic tests to monitor MIC and discusses possible
applications of genetic monitoring techniques for the future.
2. DNA HYBRIDIZATION TECHNIQUES TO MONITOR

MICROBIOLOGICALLY INFLUENCED CORROSION
One of the most commonly used genetic approaches to the monitoring of

bacteria associated with corrosion is reverse sample genome probing (RSGP)
[26-29]. This technique is based upon the hybridization of whole-community
DNA obtained from environmental samples to the DNA of various characterized
bacterial species (reference cultures) that are spotted on the master membrane.
Hybridization experiments take advantage of the fact that DNA is normally
double stranded so that if DNA is denatured (the strands are separated)
complementary strands/DNA sequences will come back together. The additive
nature of the hydrogen bonds formed by complementary nucleic acid bases
provides a strong thermodynamic force favoring the formation of double
stranded DNA by complementary DNA sequences. In RSGP various pure
cultures, most frequently sulfate reducing bacteria (SRB), are isolated from
environments of interest and then grown in the laboratory. About thirty cultures
of SRB have been cultivated from gas and oil production operations so that the
DNA of these cultures can be used in RSGP [30]. RSGP technique was also
used to identify and quantify bacteria communities capable of degrading various
aromatic hydrocarbons in contaminated soil [31-34]. Chromosomal DNA is
purified from each species of bacterial culture that was cultivated in the
laboratory. The DNA is then denatured and added as discrete drops to specific
locations on a membrane, typically a nitrocellulose membrane. The DNA is then
221
chemically /thermally bound to the membrane, which then serves as the master
grid in a subsequent hybridization experiment. To perform RSGP an
environmental sample of interest, such as biomass obtained from a gas and oil
production pipelines, is processed to obtain DNA from the mixture of bacterial
species present. This mixed DNA sample is subjected to a biochemical labeling
process that results in the addition of fluorescent or radioactive compounds to
the DNA mixture [32]. When the labeled DNA mixture is denatured and added
to the master grid complementary DNA strands form double stranded DNA, a
washing step then removes any DNA not bound to the membrane. The amount
of DNA in the mixture that corresponds to each species of bacteria present on
the master grid is determined by quantifying the amount of fluorescence or
radioactivity associated with each spot/location on the master grid. The quantity
of signals in each spot on the master grid reflects the abundance of each
reference organism in the environmental sample. This technique allows the
quantification of many different microorganisms simultaneously.
It can be readily appreciated that there are several drawbacks to RSGP.
The technique requires specialized training and equipment and involves multiple
steps so that several days are typically needed to obtain results. However, a
more important limitation is that the only species of bacteria quantified by
RSGP are those whose genomic DNA is spotted onto the master grid. Since it is
generally accepted that less than 1% of bacterial species in nature can be
cultivated in the lab [35, 36], only a small subset of bacterial species are
available with which to prepare master grids. In microbial corrosion research,
RSGP has been applied almost exclusively to the quantification of SRB. While
SRB are unquestionably capable of causing metal corrosion it is also
unquestionably true that several other types of bacterial groups such as acid
producing bacteria, iron respiring bacteria, denitrifying bacteria, sulfur oxidizing
bacteria, and methanogenic bacteria also can cause metal corrosion (see chapter
7). It would be extremely cumbersome to quantify all of these bacterial types
using RSGP, but it may be possible to develop such tests in the future using
microarray technology. Genetic methods can indeed provide more accurate data
more quickly than microbial growth tests, but methods more convenient than
RSGP are needed. The genetic method of choice for monitoring food and
cosmetic products for microbial contamination, for the detection and
identification of infectious microorganisms in medicine and bio-warfare
monitoring is quantitative real-time PCR. Quantitative PCR has recently been
adapted for use in monitoring microbial populations in gas and oil pipelines, and
may become the most convenient and reliable method to monitor MIC in the
future.
222
3. CHARACTERIZATION OF BACTERIAL AND ACHAEAL

COMMUNITIES IN GAS PIPELINES BY DGGE AND 16S rRNA GENE
SEQUENCING
Before accurate and reliable genetic tests can be developed to quantify microbial
species in a given environment it is helpful to first determine the composition of
the microbial community as completely as possible. Molecular analysis of
bacterial community structure of environmental samples has become a useful
means of examining microbial communities. Molecular analysis involves
extraction and purification of nucleic acids from environmental samples, PCR
amplification of nucleic acids, followed by cloning of PCR products into a
vector and sequencing of cloned PCR products, or followed by community
fingerprinting techniques such as DGGE to separate the amplified PCR products
[14, 17]. Each DGGE band is presumably representative of a specific bacterial
population and the number of distinctive bands is indicative of total community
richness. In addition, separated DDGE bands can also be cloned and sequenced.
The comparison of DNA sequences of 16S rRNA genes with DNA sequence
databases such as GenBank allows the identity of the species of microorganisms
present in environmental samples to be determined. A greater understanding of
the full range of bacterial species present in gas and oil production operation
environments where corrosion is occurring will improve our understanding of
the problem and our ability to detect and control it.
Genetic techniques have been used to characterize the microbial
communities present in natural gas pipelines and a thorough report has recently
been published by the authors [18]. The important observations obtained from
that study were that while the composition of microbial communities from
different pipelines varies significantly, there were some commonly encountered
species or types of microorganisms. Denitrifiers (bacteria that utilize nitrate and
nitrite), such as Comamonas denitrificans, were found to be the most commonly
encountered type of microorganism in gas pipelines. The presence of
denitrifying bacteria in gas pipelines has not previously been reported in the
literature and it is not routinely monitored in microbiological testing of gas
pipeline samples. However, the frequent occurrence of denitrifiers in gas
pipelines, and the proven ability of denitrifiers to contribute to corrosion [37,
38], suggests that denitrifiers should be monitored and that nitrate plays a key
role in metabolism of biofilm microorganisms present in gas pipelines. This is
particularly interesting because gas pipeline liquids do not typically contain
significant levels of nitrate (< 5 mg/L). Even though the absolute concentration
of nitrate may be low, the ability of different members in a microbial community
to oxidize as well as to reduce nitrogen compounds may lead to continuous
cycling of nitrogen within microbial communities, similar to what has been
223
found for sulfur [39]. Moreover, many sulfate reducing bacteria can also utilize
nitrate [38-40].
Other results obtained from characterizing the microbial populations
within gas pipeline samples were that methanogens were frequently present in
pipeline and biofilm samples, and that sulfate reducers were often present at
lower levels than indicated by microbial growth tests. The presence of
methanogens in gas pipelines was unexpected, but it is significant because any
microbial process, such as methanogenesis, that consumes hydrogen is capable
of accelerating corrosion by cathodic depolarization (a process which pulls the
cathodic reduction of protons by removal of the product and thereby accelerates
anodic metal dissolution) [8, 41, 42].
These results highlight the fact that the composition of microbial
communities in gas pipelines has not been thoroughly investigated and prior to
the genetic studies described here the entirety of our knowledge about
microorganisms in gas pipelines was restricted to information gathered by
growing bacteria in various media under laboratory conditions. There are many
types of bacteria, and testing for all types of bacteria using growth experiments
would be very tedious, and in fact it has never been done. We only have data
concerning those types of bacteria that have been tested for, so our view of the
microbial ecology of gas pipelines is biased indeed. A commonly held belief
regarding microbial corrosion is that SRB are the most important contributors to
corrosion. Since traditional microbial growth tests most frequently test for SRB,
and very few other types of microorganisms, it is not surprising that this belief is
widely held. However, our genetic tests of gas pipeline samples were not biased
in looking for any particular type of microorganism, and SRB were not among
the most abundant microorganisms in any of the pipeline samples characterized.
However, when these same pipeline samples were used to inoculate bacterial
growth tests using SRB medium, SRB were invariably found. The differences in
the number of particular types of bacteria detected by microbial growth tests and
by genetic tests were investigated further, and it was found that the species of
SRB that grew in laboratory media inoculated with gas pipeline samples were
not the same species of SRB detected when the gas pipeline samples were
examined directly. In other words, by growing bacteria in a particular medium
an artificial environment is created where nutrient concentrations, and other
factors we have yet to fully appreciate, differ significantly from the conditions
that are present in gas pipelines. Thus, the composition of microbial
communities detected in laboratory growth experiments differs profoundly from
the composition of microbial communities actually present in a gas pipeline.
224
4. LACK OF SPECIFICITY OF MICROBIAL GROWTH TESTS
The traditional means of quantifying bacterial populations present in gas and oil
production samples is to perform microbial growth tests. The growth media used
are not highly selective, and they may allow a range of types of bacteria to grow
and not just the type of bacteria that is intended to be quantified in a given
growth test. However, the accuracy of these microbial growth tests had not
previously been tested to quantify what percentage of bacteria was actually the
type of bacteria targeted. Accordingly tests were performed using traditional
microbial growth media intended for the quantification of specific types of
bacteria, and pure cultures of known bacterial types were then tested to
determine how selective these growth media were. Typical results are shown in
Table 1. Each bacterial culture tested in Table 1 was a pure bacterial culture
isolated from a gas pipeline sample.
It can be seen from the results shown in Table 1 that bacterial growth
media that are intended to support the growth of a particular type of bacteria are
not completely selective and other types of bacteria grow. This is particularly
problematic when the purpose of the microbial growth test is to determine the
quantity of a specific type of bacteria present in a sample. None of the microbial
media tested here, which include all those growth media typically used in
evaluating MIC in the gas industry, allow for the exclusive growth of the
intended type of bacteria. Moreover, only a very small number of pure bacterial
species were tested here, and the results obtained using complex mixed cultures
typical of gas pipelines could show even more widespread growth. The
implication of these results for monitoring MIC is that quantitative results using
microbial growth tests can be misleading as growth observed is not always due
to the type/species of microorganisms that is supposedly being quantified.
Table 1
Growth of pure bacterial cultures in various types of microbial growth media after 6 days
incubation at 30 °C
Growth medium
Microbial type Cultures D N B HAB MET IRB SRB APB
Methanogen Methanoarcina + +
Iron-reducing Shewamlla _ biack ++ +++ . black

Sulfate reducing Desulfovibrio +++ ++ + +++ + ++ ++
Beta-Proteobacteria Acidovorax +++ ++ ++ + +++
Denitrifier Commomonas denitrificans +++ ++-). +++ . ++ +++

225
5. DEVELOPING AN IMPROVED GENETIC METHOD TO

QUANTIFY BACTERIA
This section describes the development of a quantitative PCR (q-PCR) method
to replace traditional microbial growth tests by providing rapid and more
accurate data concerning the quantity of various types of microorganisms
present in gas and oil industry pipeline samples.
Microbial growth tests are rather slow and provide potentially
misleading/inaccurate data. For examples, the previous studies indicated that the
numbers of viable SRB in marine sediments can be underestimated at least
1,000-fold when standard most-probable-number (MPN) techniques are used
with synthetic growth media [43-45], One of the goals of research at GTI has
been the development of a genetic method to quantify bacteria in gas pipeline
samples. We used the DNA sequence data for 16S rDNA genes from bacterial
and archaeal species found in greatest abundance in actual gas pipeline samples
and designed PCR primers and hybridization probes that can be used in
quantitative or real-time PCR to quantify bacteria and archaea. Similarly we also
developed PCR primers that allow us to quantify SRB, denitrifiers, and
methanogens by targeting the dsrAB [40, 46, 47], nirS [48, 49], and mcrA [50,
51] genes respectively. These genes encode enzymes that play crucial roles in
the metabolic pathways of sulfate reduction, denitrification, and methano-
genesis, respectively, and thus constitute highly specific targets to allow a more
precise quantification of these types of bacteria than can be achieved using
microbial growth tests. The PCR primers used in this study were designed on the
basis of DNA sequences of dsrAB, nirS, and mcrA genes most typically found in
gas and oil production operation samples. This was necessary because of the
vast diversity of the microbial kingdom. For example, only a small subset of all
bacterial species that possesses dsrAB genes are typically present in gas and oil
production samples. Therefore, PCR primers designed based on all of the DNA
sequences known for dsrAB genes, such as are found in the GenBank database,
will have to include many degenerate positions in their sequences, and hence
complicate the measurement of effective primer concentrations in the PCR
reactions, and reaction itself. Frequently, primers with many degenerate
positions failed in q-PCR analysis of gas and oil production operation samples,
or yielded results that were not as accurate as when PCR primers specifically
designed to quantify this microbial community were used.
To investigate the use of these genetic tests to quantify bacteria obtained
from gas pipeline samples used to inoculate various types of microbial growth
media, quantitative PCR was performed to quantify the number of bacteria,
archaea, SRB, denitrifying bacteria, and methanogens present in various
samples. The results shown in Table 2 indicate that archaea are generally present
at very low levels even in samples cultivated in methanogen growth medium.
226
The fact that quantification of methanogens by quantifying mcrA gene

sequences uniformly yields higher values than results obtained quantifying
archaea with PCR primers targeting 16S rDNA genes illustrates that currently
available PCR primers that are supposed to be "universal" for archaeal species
are not sufficiently inclusive and many methanogens, particularly
Methanococcus species, are not detected by these archaeal primers [52-54].
Thus more accurate data is obtained by monitoring the mcrA gene that encodes
an enzyme that is essential to methanogenesis [50, 51]. The data in Table 2
make it clear that even when the mcrA gene is used to monitor methanogens,
microorganisms other than methanogens can grow in methanogen growth media.
In several instances there were no detectable methanogens in samples where
significant growth in methanogenic growth media occurred, and in no case did
the percentage of methanogens exceed 30% of the total microorganisms
detected. This confirms the results shown in Table 1 that microbial growth
media is not uniquely selective for the type of bacteria of interest, and that
quantitative results based on growth in microbial growth media can be
misleading.
This point is further illustrated by examining the data in Table 2 regarding
denitrifying bacteria. When quantitative PCR was performed using primers that
target the nirS gene, which encodes an enzyme that is essential to denitrification
in the majority of denitrifying bacteria [48, 49], the percentage of denitrifiers
growing in denitrifying bacteria media ranges from undetectable to 11 %. These
results again indicate that results obtained from microbial growth tests are not
specific and that more accurate results can be obtained using genetic tests.
Similar results were obtained using primers targeting dsrAB genes to quantify
SRB (data not shown).
5.1. Verifying the accuracy of genetic testing methods

To further investigate the accuracy and reliability of genetic tests to quantify
bacteria and types of bacteria in gas pipeline samples, several tests were
performed using pipeline samples that had been spiked with known
concentrations of certain types of bacteria. The results of some of these tests are
shown in Table 3 where seven different gas pipeline samples were analyzed with
and without spiking of a known amount of DNA purified from cultures of
bacteria (Pseudomonas aeruginosa PAO-1), archaea (Archaeoglobus fulgidus),
SRB (Desulfovibno vulgaris), denitrifying bacteria {Pseudomonas aeruginosa
PAO-1), and a methanogen (Methanococcus jannaschii). The first main column
in Table 3 is the gene copy number detected by quantitative PCR after spiking
with known copies of genes in the PCR reaction, and the second column
represents the gene copy number calculated by adding the gene copies spiked to
the reaction to the detected copy number from the un-spiked reaction. It is clear
from Table 3 that the genetic tests were able to accurately quantify target DNA
227
spiked into gas pipeline samples. This is best seen by inspecting the ratio
between the actually detected gene copy number from the spiked PCR reaction
and the calculated copy number per reaction (copy number detected from un-
spiked reaction, plus the copies spiked into the reaction). If all of the DNA
added in spiked samples was accurately quantified the ratio of detected and
calculated would be 1; values above 1 means an overestimate of the actual
concentration, and values below 1 means an underestimate of spiked gene
copies. The average ratios of the detected and calculated were 1.02 ± 0.09, 0.69
± 0.1, 1.2 ± 0.15, 0.87 ± 0.09, and 1.22 ± 0.16 for bacteria, archaea, SRB,
denitrifiers, and methanogens, respectively, which are considered very accurate
for the analysis of complex environmental samples [24].
The data in Table 3 illustrate that genetic tests employing quantitative
PCR techniques provide accurate and reliable data concerning the quantity of
various types of bacteria that may be present in gas pipeline samples.
5.2. Applying the genetic testing methods to quantify various groups of

microorganisms present in pipeline samples
Seven pipeline samples were collected from various natural gas
companies at various geographical locations. The biomass was centrifuged down
and used for DNA extraction using FastDNA SPIN Kit for Soil (Qbiogene,
Carlsbad, CA). The extracted DNA was further purified by phenol/chloroform
extraction to remove inhibitory substances commonly present in this type of
samples. The spiking experiment showed that the extra purification step was
sufficient to eliminate inhibitory effect of PCR amplification (Table 3). The
purified genomic DNA was then amplified and quantified using quantitative
real-time PCR with DNA standards of corresponding target genes. The results
were summarized in Table 4. The results confirmed our previous observation
[18] using DGGE analysis of 16S rRNA gene sequences that denitrifying
bacteria and methanogens are two types of organisms which were commonly
present in relatively high abundance in gas pipeline samples. Denitrifying
bacteria were detected in all seven samples, and the concentration in the sample
was as high as 7.95 x lO^/ml; methanogens were detected in five out of seven
samples, and the concentration was as high as 3.7 x 10^/ml; SRB were detected
in six samples, but at lower concentration in most of samples. The lower archaea
concentration detected using 16S rRNA gene than methanogens based on the
detection of mcrA gene is due to archaea primers, which will exclude many
methanogens, especially Methanococcus species [52-54].
Table 2
Real-time PCR quantification of gas pipeline samples grown in various microbial growth media
Growth Cone, of various microbes in growth medium (/ml) Percentage of total microbes (%)
Sample ID Medium Bacteria Archaea Methanogen Denitrifier Archaea Methanogen Denitrifier
1 MET 4.20E+09 6.75E+04 8.15E+O5 0.00 0.02
3 MET 2.04E+07 1.61E+05 2.81E+06 0.69 12.02
6 MET 1.00E+10 6.75E+04 1.04E+06 0.00 0.01
8 MET 1.52E+11 5.65E+08 3.51E+1O 0.30 18.74
9 MET 1.02E+11 6.65E+09 4.35E+10 4.36 28.52
2 MET 6.92E+07 ND ND ND ND
8 DNB 7.84E+09 2.35E+05 4.26E+07 0.00 0.54

9 DNB 3.80E+10 1.55E+06 4.38E+09 0.00 11.53
4 DNB 1.06E+08 ND 8.50E+05 ND 0.80
2 DNB 7.28E+05 ND ND ND ND
ND = not detected
MET = methanogenic bacteria
DNB = denitrifying bacteria
229
Table 3
Accuracy of real-time PCR quantification of bacteria, archaea, SRB, denitrifiers, and
methanogens in natural gas pipeline samples
Bacteria detected (copy/rnx) Bacteria calculated (copy/rnx) Ratio
Sample ID w/ spiking w/o spiking + spiked copies det:cal
1 7.32E+08 7.97E+08 0.92
2 1.16E+08 1.04E+08 1.12
3 4.56E+06 4.54E+06 1.01
4 1.14E+08 1.05E+08 1.08
5 6.40E+05 6.73E+O5 0.95
8 9.76E+08 1.05E+O9 0.93
9 2.61E+09 2.35E+09 1.11
Spiking: 5.36E+05 copies of 16S rRNA genes of'/'seudomonas aeruginosa PAO-1 per reaction.
Archaea detected (copy/rnx) Archaea calculated (copy/rnx) Ratio

Sample ID w/ spiking copy# w/o spiking + (3.88E+05) det:cal
1 3.07E+05 3.93E+O5 0.78
2 5.3OE+O5 8.36E+O5 0.63
3 2.35E+05 3.88E+O5 0.61
4 3.25E+O5 3.93E+O5 0.83
5 2.19E+05 3.88E+O5 0.56
8 1.6E+07 2.13E+07 0.75
9 1.5E+06 2.29E+O6 0.64
Spiking: 3.88E+05 copies of 16S rRNA genes of Archaeoglobus julgidus per reaction.
SRB detected (copy/rnx) SRB calculated (copy/rnx) Ratio

1 3.O2E+O5 2.47E+05 1.22
2 2.85E+05 2.29E+05 1.24
3 3.09E4O5 2.24E+O5 1.38
4 3.18E+05 2.56E4O5 1.24
5 2.94E+05 2.24E+05 1.31
8 2.37E+06 2.47E+06 0.96
9 3.66E+O5 3.49E+05 1.05
Spiking: 2.24E+05 copies oidsrAB genes oi Desulfovibrio vulgaris per reaction.
Denitrifier detected (copy/mx) Denitrifier calculated (copy/mx) Ratio

1 2.43E+07 3.19E+07 0.76
2 2.78E+05 2.96E+05 0.94
3 1.27E+O5 1.5OE+O5 0.85
4 4.73E+05 5.20E+05 0.91
5 1.38E+05 1.35E+05 1.02
8 6.66E4O5 7.91E4O5 0.84
9 2.02E+07 2.60E+07 0.78
Spiking: 1.34E+05 copies oinirS genes off. aeruginosa PAO-1 per reaction.
Methanogen detected (copy/rnx) Methanogen calculated (copy/rnx) Ratio

1 7.01E+05 5.45E+05 1.29
2 1.48E+06 1.16E+06 1.27
3 5.63E+05 5.1OE+O5 1.10
4 6.37E-H)5 5.22E+05 1.22
5 5.88E+05 5.1OE+O5 1.15
8 1.74E+08 1.12E+08 1.56
9 1.23E+07 1.03E-+O7 1.20
Spiking: 5.1E+O5 copies of mcrA genes oiMethanococcus jannaschii per reaction.
230
Table 4
Quantification of bacteria, archaea, SRB, denitrifiers, and methanogens in natural
gas pipeline samples (/mL)
Sample ID Bacteria Archaea SRB Denitrifier Methanogen
1 1.99E+08 1.59E+03 8.12E+03 7.95E+06 1.18E+04
2 2.58E+07 1.49E+05 1.98E+03 4.05E+04 2.18E+05
3 1.00E+06 5.93E+01 5.29E+01 4.00E+03 ND
4 2.75E+07 1.75E+03 9.51E+O3 1.02E+05 4.14E+03
5 3.42E+04 4.67E+01 ND 1.96E+02 ND
8 2.62E+08 6.97E+06 6.75E+05 1.64E+05 3.70E+07
9 5.88E+08 6.33E+05 4.11E+04 6.48E+06 3.25E+06
ND = not detected
6. CONCLUSIONS
Quantifying various types of bacteria that may be present in gas and oil
production operation samples is a difficult challenge. Traditional tests employ
microbial growth media of various types that are intended to foster the growth of
particular types of microorganisms. Unfortunately, microbial growth media are
not uniquely selective and microorganisms other than the intended type often
grow in the test medium. In some cases, even though significant growth occurs
the concentration of the target population of bacteria is below detection limits.
Thus, if the results of growth in microbial test media are used to determine the
quantity of various types of bacteria present in gas and oil production operation
samples the results obtained can be very misleading. Traditional microbial
growth tests require weeks of incubation, are not accurate, and do not provide
information about what microbial species were actually present in the
environment. An improved means of quantifying various types of bacteria
present in gas pipeline samples is using genetic techniques, such as RSGP and q-
PCR. Other industries such as medicine, food, and cosmetics share with gas and
oil industry the need to detect, identify, and quantify microorganisms. These
other industries have largely abandoned microbial growth tests in favor of
genetic methods. Hybridization test methods such as RSGP are not typically
used in characterizing environmental samples, but future improvements in
microarray technology may change this situation. q-PCR has been adopted by
other industries as the method of choice for rapid quantification of
microorganisms, but this technique has not yet been in the gas and oil industry.
In this chapter data was presented that demonstrated that q-PCR can obtain data
within a few hours that specifically and accurately quantifies bacterial types
present in gas pipeline samples. The gas pipeline samples are analyzed directly
without any cultivation or other manipulation in the laboratory that would alter
the composition of the microbial community. Therefore, q-PCR provides data
231
concerning the actual microbial community present in the gas pipeline. GTI has
developed an accurate and reliable method to quantify bacterial types present in
gas pipeline samples and is now offering this service to the industry.
REFERENCES
[I] E. Buck, G. C. Maddux, and R. L. Sullivan, GRI-96/0056 document number 96-1466

(1996).
[2] S. Farthing, Pipeline and Gas Industry Oct '97 (1997) 43.
[3] J. W. Graves, E. H. Sullivan, Materials Protection 5 (1996) 33.
[4] V. P. Kholodenko, S. K. Jigletsova, V. A. Chugnov, V. B. Rodin, V. S. Kobelev, S. V.
Karpov, Appl. Biochem. Microbiol. 36 (2000) 594.
[5] B. G Pound, (1998) Gas Research Institute.
[6] G. H. Koch, M. P. H. Brongers, N. G. Thompson, Y. P. Virmani, and J. H. Payer,
FHWA-RD-01-156, published online, Federal Highway Administration, Washington
D.C. (2001).
[7] K. M. E. Emde, D. W. Smith, and R. Facey, Water Res. 26 (1992) 169.
[8] P. Angell, Curr. Opin. Biotechnol. 10 (1999) 269.
[9] J. F. Batista, R. F. Pereira, J. M. Lopes, M. F. Carvalho, M. J. Feio, M. A Reis,,
Biodegradation 11 (2000)441.
[10] D. H. Pope, T. P. Zintel, B. A. Cookingham, R. G. Morris, D. Howard, R. A. Day, J. R.
Frank, and G E. Pogemiller, Corrosion '89 (1989), NACE paper 192.
[II] D H Pope, R. M. Pope, (1998) Gas Research Institute.
[12] L. N. Strickland, R T. Fortnum, B. W. DuBose, Corrosion '96 (1996), NACE paper 297.
[13] B. L. Maidak, J. R. Cole, T. G Lilburn, C. T. Parker, Jr., P. R Saxman, J. M. Stredwick,
G. M. Garrity, B. L. Li, G. J. Olsen, S. Pramanik, T. M. Schmidt, and J. M. Tiedje,
Nucleic Acids Res. 28 (2000) 173.
[14] G. Muyzer, E. C. de Waal, and A. G Uitterlinden, Appl. Environ. Microbiol. 59 (1993)
695.
[15] A. L. Reysenbach, G S. Wickham, and N. R. Pace, Appl. Environ. Microbiol. 60 (1994)
2113.
[16] W. A. Williams, J. H. Lobos, and W. E. Cheetham, Int. J. Syst. Bacteriol. 47 (1997)
207.
[17] X. Y. Zhu, T. Zhong, Y. Pandya. R. D. Joerger, Appl. Environ. Microbiol. 68 (2002)
124.
[18] X. Y. Zhu, J. Lubeck, J. J. KilbaneH, Appl. Environ. Microbiol. 69 (2003) 5354.
[19] X. Y. Zhu, and R. D. Joerger, Poult. Sci. 82 (2003) 1242.
[20] E. F. DeLong, L. T. Taylor, T. L. Marsh, C. M. Preston, Appl. Environ. Microbiol. 65
(1999)5554.
[21] S. J. Giovannoni, M. S. Rappe, K. L. Vergin, N. L. Adair, Proc. Natl. Acad. Sci. USA
93(1996)7979.
[22] S. Y. Lee, J. Bollinger, D. Bezdicek, A. Ogram, Appl. Environ. Microbiol. 62 (1996)
3787.
[23] A. Felske, A. D. L. Akkermans, and W. M. de Vos, Appl. Environ. Microbiol. 64 (1998)
4581.
[24] J. R. Stults, O. Snoeyenbos-West, B. Methe, D. R. Lovley, D. P. Chandler, Appl.
232
[25] C. F. Brunk, and N. Eis, Appl. Environ. Microbiol. 64 (1998) 5064.

[26] G. Voordouw, Y. Shen, C. S. Harrington, A. J. Telang, T. R. Jack, D W. S. Westlake,
Appl. Environ. Microbiol. 59 (1993)4101.
[27] G. Voordouw, J. K. Voordouw, T. R. Jack, J. Foght, P. M. Fedorak, D. W. S. Westlake,
[28] G. Voordouw, J. K. Voordouw, R. R. Karkhoff-Schweizer, P. M. Fedorak, D. W. S.
Westlake, Appl. Environ. Microbiol. 57 (1991) 3070.
[29] D. W. S. Westlake, J. M. Foght, P. M. Fedorak, G. Voordouw, T. R. Jack, Corrosion
control for low-cost reliability: 12th international corrosion congress 5B (1993) 3794.
[30] A. J. Telang, S. Ebert, J. M. Foght, D. W. S. Westlake, G. E. Jenneman, D. Gevertz, G.
Voordouw, Appl. Environ. Microbiol. 63 (1997) 1785.
[31] E. A. Greene, J G. Kay, K. Jaber, L. G. Stehmeier, G. Voordouw, Appl. Environ.
Microbiol. 66 (2000) 5282.
[32] Y. Shen, L. G. Stehmeier, and G. Voordouw, Appl. Environ. Microbiol. 64 (1998) 637.
[33] G. Voordouw, S. M. Armstrong, M. F. Reimer, B. Fouts, A J. Telang, Y. Shen, D.
Gevertz, Appl. Environ. Microbiol. 62 (1996) 1623.
[34] C. Hubert, Y. Shen, and G. Voordouw, Appl. Environ. Microbiol. 65 (1999) 3064.
[35] V. Torsvik, J. Goksoyr, and F. L. Daae, Appl. Environ. Microbiol. 56 (1990) 782.
[36] M. S. Osburne, T. H. Gnossman, P. R. August, 1. A. MacNeil, ASM News 66 (2000)
411.
[37] M. Nemati, G. E. Jenneman, and G. Voordouw, Biotechnol. Prog. 17 (2001) 852.
[38] B. A. Till, L J. Weathers, and P. J. J. Alvarez, Environ. Sci. Technol. 32 (1998).
[39] C. M. Santegoeds, T. G. Ferdelman, G. Muyzer, D. de Beer, Appl. Environ. Microbiol.
64(1998)3731.
[40] A. Dhillon, A. Teske, J. Dillon, D. A. Stahl, and M. L. Sogin, Appl. Environ. Microbiol.
69(2003)2765.
[41] D. Thierry, and W. Sand, In, Corrosion Mechanisms in Theory and Practice: 2nd edition
(P. Marcus ed.) Marcel Dekker, New York (2002) 563.
[42] J. Horn, and D. Jones, Manual of environmental microbiology (2002) (Hurst, C. J.,
Crawford, R. L., Knudsen, G. R, Mclnerney, M. J., Stetzenbach, L. D., Ed), pp. 1072-
1083 ASM Press, Washinton, DC.
[43] F Vester, and K. Ingvorsen, Appl. Environ. Microbiol. 64 (1998) 1700.
[44] N. B. Ramsing, H. Fossing, T. G. Ferdelman, F. Andersen, B. Thamdrup, Appl.
[45] G. R. Gibson, R. J. Parkers, and R. A. Herbert, J. Microbiol. Methods 7 (1987) 201.
[46] R. Karkhoff-Schweizer, D. Huber, and G. Voordouw, Appl. Environ. Microbiol. 61
(1995)290.
[47] M. Wagner, A. J. Roger, J. L. Flax, G. A. Brusseau, D. A. Stahl, J. Bactenol. 180 (1998)
2975.
[48] G. Braker, A. Fesefeldt, and K.-P. Witzel, Appl. Environ. Microbiol. 64 (1998) 3769.
[49] G. Braker, H. L. Ayala-del-Rio, A. H. Devol, A. Fesefeldt, J. M. Tiedje, Appl. Environ.
Microbiol. 67(2001)1893.
[50] P. E. Luton, J. M. Wayne, R. J. Sharp, P. W. Riley, Microbiology 148 (2002) 3521.
[51] T. Lueders, M. W. Fnedrich, Appl. Environ. Microbiol. 69 (2003) 320.
[52] A.-L. Reysenbach, K. Longnecker, and J. Kirshtein, Appl. Environ. Microbiol. 66
(2000), 3798-3806.
[53] K. Takai, andK. Honkoshi, Appl. Environ. Microbiol. 66 (2000), 5066-5072.
[54] M. T. Suzuki, L. T. Taylor, and E. F. DeLong, Appl. Environ. Microbiol. 66 (2000)
4605.
Chapter 9
Potential applications of bioemulsifiers in the oil industry

H. Bacha and D.L. Gutnick
Department of Molecular Microbiology and Biotechnology, Tel-Aviv

University, Tel-Aviv, 69978, Israel
a
Present address, Biotechnology Research Laboratories, Taro Pharmaceuticals
U.S.A., 3 Skyline Drive, Hawthorne, New York, 10532, U.S.A.
1. INTRODUCTION
Virtually hundreds of different chemicals, both synthetic and naturally occurring

materials, exhibit properties of specific and preferential interaction at surfaces
and interfaces. This generally arises from the presence, in such molecules of
both hydrophilic and hydrophobic moieties leading to the ability of these
materials to orient themselves at the interface. The effects of such materials in
modifying the particular surface and consequently affecting its properties
represent its surface activity, and such materials are generally referred to as
surfactants. Surfactants vary widely in structure and function and thus are used
in virtually all major industries. Surfactants are very versatile and have found
uses as detergents lowering interfacial tension, emulsifiers, dispersants,
deemulsifiers, wetting agents, foam retardants, stabilizers, gelling agents etc.
They are also used widely in consumer products as well major constituents in
the household cleaning and soap industry [1], Moreover, the market for
chemical surfactants continues grow. For example, the world wide consumption
for surfactants was 18 million tons in 1970, 25 million tons in 1990, and about
40 million tons in the year 2,000 [2]. It has been forecasted that the surfactant
market in the U.S. and Europe will increase at about 2.5% per year between
2000 and 2005 [3]. These surfactants, generally products of the petrochemical
industry are used in almost all modern industrial processes. Chemical surfactants
are generally produced as by-products of the petrochemical industry and consist
primarily of alkylbenzene sulfonates, alkyl phenol ethoxylates, synthetic fatty
alcohols and their derivatives. These products are believed to account for 70-
75% of the surfactant consumption in the industrialized countries [4].
234
While chemical surfactants are both inexpensive and efficient, they may
have very negative effects on the environment. Increasing awareness on the part
of the consumer, coupled with the potential for legislation governing excessive
use of such chemicals offers new opportunities for biotechnological alternatives.
The potential advantages of such products include, 1. biodegradability resulting
in lower levels of pollution, 2. selectivity and specificity towards hydrocarbon
substrates, 3. potential for using recombinant DNA technology to engineer
changes in surfactant structure and function, 4. compatibility with chemical
products leading to novel formulations, 5. natural products may have unique
characteristics which cannot be produced by simple chemical synthesis. One
such product, which is used in the oil industry is xanthan gum, a microbial
polymeric exopolysaccharide with unique, sheer thinning rheological properties
[5]. In this Chapter we will focus primarily on microbial biosurfactants as
bioemulsifiers and their potential applications in the petroleum industry.
2. LOW MOLECULAR WEIGHT BIOSURFACTANTS
As is the case with chemical surfactants of non-biological origin, biosurfactants

exhibit a wide variety of surface activities ranging from reduction of surface
tension, formation of water-in-oil and oil-in-water emulsions, adsorption to and
coating of surfaces, surface wetting, flocculation of solids, foaming and
defoaming at air/liquid interfaces and emulsion breakage. Moreover, many of
these compounds can exhibit more than a single activity. Since there is a very
wide range of structures and producing organisms it is often useful to consider
the lower molecular mass biosurfactants as a group distinct from the polymeric
bioemulsifiers [6-9]. In this section we will refer to the biosurfactants as low
molecular mass detergents whose primary activity is to lower interfacial tension,
while bioemulsifiers are generally higher molecular mass polymers.
Biosurfactants generally are smaller and exhibit molecular masses in the range
of a few thousands of Daltons. As detergents, their primary surface activity is to
lower interfacial tension at the air/water interface. As is the case with chemical
surfactants, biosurfactant formulations may include biosurfactants in
combination with chemical or biological co-surfactants, solvents, biocides and
other ingredients. Their production can be the result of denovo synthesis [10], or
in some cases the result of biotransformations resulting from transacylation or
other post synthetic processing [11]. From the point of view of large-scale
production, microbial systems have been considered as the major sources of
biosurfactants largely owing to the availability of fermentation technology for
their production. Moreover, in certain cases, it is possible to manipulate the
biosynthetic pathways and their regulation using modern recombinant DNA
technology [12]. Since, in recent years a number of original papers and excellent
235
reviews dealing with low molecular weight biosurfactants have been published
[7,9,13-16], the subject will be treated only briefly in this section.
2.1. Structure-function features of low molecular weight biosurfactants
2.1.1. Chemical characterization of biosurfactants

As illustrated in Table 1A and IB, biosurfactants exhibit a wide variety of
structures including fatty acids and alcohols, phospholipids, lipopeptides, which
can be either cyclic or linear, and glycolipids consisting of disaccharides linked
via esters or ethers to fatty acids. While most of these compounds are
extracellular, an interesting exception has been reported, in which the cells
themselves exhibit emulsifying activity [13].
Lipopeptides. In the lipopeptides, produced by several species of Bacilli
and other organisms, the hydrophilic moiety is represented either by a cyclic
oligopeptide, as in surfactin [17] or a 2, 4-diaminobutyric acid intercalated
cyclic oligopolypeptide, as in polymyxin [18]. Another class of biodetergents
was isolated from a variety of microorganisms belonging to the genera
Pseudomonas. They share a common cyclic oligopeptide structure represented
by charged amino acids such as serine, aspartic acid and glutamine, while non-
charged amino acids as leucine, and isoleucine are present as well [19].
Interestingly, amino acids which are normally not found in proteins, such as L-
norvaline, L-norleucine, L-allylglycine, etc. have been found in some
lipopeptide biodetergents strongly suggesting that NRPS's (Non-Ribosomal
Protein Synthases) play a role in the generation of these amphipathic peptides
[20]. The amino acid number that constitutes the oligopeptide varies among the
species, but they share the common lactonized structure formed as a result of an
ester bond. For example, serrawetin W2 contains 5 amino acids, surfactin is a
heptapeptide; WLIP, viscosinamide, massetolides and viscosin contain 9 amino
acids, while tensin, hodersin, pholipeptin, lokisin, arthrofactin and amphisin
contain 11 amino acids (Table IB). In all of these compounds, a hydrophobic 3-
hydroxydecanoyl moiety serves as the lipophilic component [19].
Glyco- and flavolipids. Certain microorganisms produce biodetergents
consisting disaccharides as the hydrophilic moieties and fatty acids linked via
ester or ether linkages. In the case Pseudomonas aeruginosa the sugar residue is
rhamnose to form a rhamnolipid [21], while Arthrobacter parqffineus produces
a trehalose lipid [22]. Similarly, the yeast Torulopsis produces a biodetergent
consisting of two molecules of glucose. Interestingly, here the acyl moiety is
introduced by biotransformation of an exogenously added source of fatty acid
[31]. Another fungal glycolipid contains sophorose as the hydrophilic moiety.
Recently a new product was discovered during a screen for surfactant producers
from soil samples in the Southwestern desert of the U.S. [24]. The new
surfactant, produced by a Flavobacteria contains a hydrophilic group consisting
236
of citric acid and two cadaverine molecules. The hydrophobic moiety contains
two acyl groups of from 6-10 carbons.
237
238
239
2.1.2. Surface activity of low molecular weight bio surfactants

As is the case with chemical surfactants, low molecular weight
biosurfactants generally act to lower the surface tension at the air/water
interface, and in some cases lower the interfacial tension at the interface
between two immiscible liquids (water/oil). The biosurfactant reduction in
surface tension at the air/water interface is generally a reduction from 70 down
to below 30 mN/m. Of course an additional feature is the critical micelle
concentration (CMC), which is the highest effective concentration of the
surfactant. Many of the biosurfactants exhibit CMC's in the order of hundreds of
milligrams per liter, while significantly lower concentrations are required for the
lowering of surface tension. As pointed out previously [7, 9, 13-14], there is not
necessarily a correlation between the capacity for surface tension reduction and
the capacity for forming and stabilizing emulsions.
Part of this may relate to hydrocarbon substrate specificity associated with
emulsification, and the affinity of the particular surfactant for the hydrocarbon,
water interface. For example, the flavolipids lower surface tension and stabilize
hydrocarbon in water emulsions [24], while many of the glycolipids are less
effective at emulsification.
240
241
242
243
244
245
246
2.2. Searching for low molecular weight biosurfactants
2.2.1. Screening methods

Most lower molecular weight biosurfactants are microbial in nature and
are produced primarily by bacteria and fungi [9]. Since the growth of
microorganisms on oil is almost always accompanied by the emulsification of
the oil in the aqueous media, the most common mode of enrichment for such
surfactant-producing microorganisms is to search for organisms capable of
growth on either pure or crude petroleum products. In addition, the ability to
grow on a variety of renewable waste products has been employed to isolate
surfactant producers [21]. The advantage of this approach is that lowers the cost
of the growth substrate and consequently the cost of production. Since the
growth of hydrocarbon-degrading microorganisms is generally accompanied by
the emulsification of the hydrocarbon growth substrate in the aqueous growth
media, it was initially assumed that the natural role of the surfactant is to enable
the efficient assimilation of the hydrophobic growth substrate by the surfactant-
producing microorganism [49]. In some cases this was indeed demonstrated
suggesting a potential application of biosurfactants in stimulating hydrocarbon
bioremediation [16]. However, many of the surfactant producers such as
Acinetobacter radioresistens or Acinetobacter sp. A2 cannot utilize
hydrocarbons as growth substrates. This is also the case for many of the cyclic
lipopeptides produced by the Bacilli. The results clearly suggest that detergency
247
and/or capacity for emulsification need not be directly related to the natural role
of the surface-active molecule. A standard measure for surface activity is the
reduction of surface tension. This method may vary somewhat from instrument
to instrument and is somewhat cumbersome since frequently measurements
must be made at several concentrations. Generally, a good candidate for a
surfactant lowers the surface tension from 70mN/m to below 30mN/m. An
additional screen based on surface activity was to search for organisms, which
produced materials which lyse eukaryotic cells [50-53]. The problem with this
approach was that it also enriched for microbial pathogens. Moreover, it was
shown to have the weakest correlation with other modes of screening and to
yield the highest number of false positives [14]. Another useful and effective
screen involves examining the capacity of a drop of culture broth from a
putative surfactant producer to collapse an aqueous droplet formed on a
hydrophobic surface [54-55]. The surfactant in this case increases the contact
angle and droplet collapse can be estimated as a function of surfactant
concentration and related to standard materials. The droplet collapse method is
rapid and yields relatively low numbers of false positives [14]. The oil spreading
technique involves placing a droplet of a surfactant containing solution on a
surface coated with a liquid hydrocarbon. The surface activity causes oil
spreading leaving a clear zone at the point of application the diameter of which
is a qualitative measure of surface activity [56]. Recently the three screening
methods, lysis of blood agar, droplet collapse and oil spreading were compared
with surface tension measurements for 205 natural isolates. The oil spreading
technique appeared to give the highest correlation with the surface tension
lowering, although there was strong negative correlation between clear zone
diameter and droplet collapse suggesting that the two procedures measured
similar activities and could be correlated well with surface tension
measurements. It was suggested that an effective protocol for screening natural
isolates is to use the droplet collapse method and subsequently employ the oil
spreading technique for more quantitative preliminary evaluations [14].
The wide variability in structures (see Table 1A and IB) and the
production and secretion of bio surfactants by organisms, which do not grow on
hydrocarbons indicates that there is probably no general biological role for all
biosurfactants [57]. Moreover, unless the search is for a closely related analog
whose synthesis may be catalyzed by similar gene products, it is unlikely that
modern methods in molecular ecology will be productive in identifying new
organisms or products. Recently, natural isolates from arid soil samples from the
Southwestern U.S. were screened for their ability to produce extracellular
materials in the culture broth, which lowered surface tension at the air/water
interface [12], According to this screen over two percent of the isolates
produced surfactants when grown on a rich medium. Interestingly, biosurfactant
producers were found in both uncontaminated soils as well as in soils showing
248
elevated levels of hydrocarbons or heavy metal ions. About 45 biosurfactant

producing strains were isolated. In one case, a new organism a Flavobacterium
was isolated which produced a new biosurfactant with a unique structure (Table
1A).
2.2.2. A role for some glycolipids

The idea that bio surfactants increased the surface to volume ratio of the
carbon source by physically breaking the oil down into smaller droplets has been
proposed as a biological role for the surfactants in oil-degrading microorganisms
[7]. However, this might be a strategy for organisms, which grow at the
oil/water interface, but it cannot be true for those organisms which produce
surfactants but which don't grow on hydrocarbons [12].
Rhamnolipids. Although a general role for bio surfactants in general
microbial behavior is unlikely, a biological role for a specific glycolipid,
rhamnolipid, has been implicated in biofilm formation by the cystic fibrosis
pathogen, Pseudomonas aeruginosa [58]. Mutants defective in rhamnolipid
biosurfactant production can adhere to the surface of tissues, but cannot organize
themselves into a structured biofilm [59]. The biosynthesis of rhamnolipid
proceeds via two gene products, RhlA and Rhl B, the genes for which are
present in an operon whose transcription is cell-density dependent. The control
of the operon is via a transcriptional regulator termed RhlR whose activity is
controlled by an autoinducer, butanoyl homoserine lactone to activate
transcription. The cell density dependence is due to a process termed quorum
sensing in which the autoinducer is produced in low levels, and a threshold
concentration is required to interact with RhlR, thereby converting the protein
into a transcriptional activator. In the absence of the autoinducer, the RhlR
protein appears to function as a repressor of the operon [60].
Serrawettin W2. Serratia liquifaciens MG1 exhibits a swimming motility
on low agar (<0.4%) and a swarming motility on harder agar surfaces.
Interestingly, it was shown that the swarming is achieved on the harder agar due
in part to the production of a surface-active lipopeptide termed serrawettin W2
(Table 1). This peptide surfactant lowers the surface tension on the harder agar
surface enabling the rapid swarming across the surface of the agar. As in the
case of the regulation of the rhamnolipid in P. aeruginosa, surfactant production
is sensitive to quorum sensing by N-butyl homoserine lactone at concentrations
of about 150 nM and at higher concentrations (900 nM), by hexanoyl
homoserine lactone, both of which are produced in S. liquifaciens [61]. The
synthesis of these autoinducers is catalyzed by the product of the gene swrl.
When the autoinducer reaches the threshold level it interacts with the regulator
SwrA which is the transcriptional activator of the biosynthetic regulon [61]. The
peptide biosynthesis is catalyzed by a general complex termed Srf, which is
involved in synthesizing the peptide bonds, and adding the C-10 hydroxy fatty
249
acid side chain to provide the hydrophobic tail. Homologous of this complex are
found in Bacilli, which produce cyclic lipopeptides such as surfactin.
3. BIOEMULSIFIERS
Bioemulsifiers are another group of biosurfactant. Many surface-active agents

exhibit multiple surface activities such as reduction of surface tension and
reduction of interfacial tension. However, this is not always the case.
Bioemulsifiers may be poor detergents, while the ability to lower interfacial
tension may not be sufficient to stabilize water in oil or oil in water emulsions
[62-64]. Bioemulsifiers are microbial products, which form and stabilize water
in oil, or oil in water emulsions. Their discovery was based initially on the
observation that growth of microorganisms on various oils is often accompanied
by the emulsification of the hydrophobic carbon source in the aqueous growth
medium [8-9,49, 65]. Although some low molecular weight biosurfactants are
also bioemulsifiers, here we will focus primarily on the polymeric
bioemulsifiers since their higher viscosity enhances their stabilization properties
[66-69]. Moreover, the polymeric bioemulsifiers exhibit a preference for the
oil/water interface and do not form micelles which enables them to be used at
lower concentrations (i.e. higher ratios of oil: emulsifier).
3.1. Protein-Polysaccharide Interactions

As shown in Table 2 most of the isolated and characterized bioemulsifiers
show a composition where a protein fraction is part of a complex, which
contains a polysaccharide backbone, however the role of the nature of the
protein-polysaccharide interaction at the oil/water interface remains to be
clarified. What follows below is a brief consideration of the interactions of
proteins and polysaccharides at interfaces. This very general treatment may help
to understand some of the complex interactions governing bioemulsifier
function.
3.1.1. Proteins at oil-water interfaces

Proteins often act as surface-active agents e.g. p*-lactoglobulin [83] and
elastin [84], In order to exhibit this surface activity they should be able to be
transported from the bulk solution and to concentrate at the interface of the
system. Subsequently, they should be able to penetrate into the surface layer.
Each stage is accompanied by a specific energy barrier, which, once overcome,
results in a successive lowering of interfacial energy [85]. This property of the
protein stems from its flexibility, which in turn is conferred by the peptide bonds
[86]. At the interface the protein may unfold to varying extents, reorient,
rearrange and spread to form a continuous cohesive film [87-89].
250
Table 2
Composition of biosurfactants
MW* Composition Reference
Compound Organism
C+ P § FA*
Exopolysaccharides
Acinetohacter >3xlO5 15 20 [70]
calcoaceticus MM5
Bacillus sp. IAF343 44 2 [71]
Bacillus cereus IAF 346 44 2 [71]
Halomonas euhhalina H28 35 4 [72]
Alasan Acinetobacter 9xl0 5 20 [73]
radioresistens K53
Biodispersan Acinetobacter 51400 70 30 [74]
calcoaceticus A2
Emulsan Acinetobacter venetianus lxlO6 70 15 12 [75]
RAG-1
Liposan Candida lypolytica ATCC 27600 83 17
8662 [76]
Lipids
Rhodococcus sp. Q15 14-18 [77]
Saccharomyces uvarum 18 [78]
Lipopeptides
Bacillus liqueniformis JF-2 1035 15 [79]
Amphisin Pseudomonas sp. DSS73 1395 10 [19]
Arthrofactin Pseudomonas sp. MIS 38 1354 7 [56\
Bacitracin Bacillus liqueniformis 1600 [34]
Hodersin Pseudomonas sp. 1409 10 [80]
Lokisin Pseudomonas sp. DSS41 1355 10 [80]
Lychesin A Bacillus liqueniformis 1030 12-17 [38]
BAS50
Serrawettin Serratia liquefaciens MG1 732 8 [42]
Surfactin Bacillus subtilis ATCC 923 12 [43]
21332
Tensin Pseudomonas fluorescens 1410 10 [45]
96.578
Viscosin Pseudomonas viscosa 1126 10 [46]
Viscosinamide Pseudomonas fluorescens 1126 10 [47]
DR54
Glycolipids
Pentasaccharide Nocardia 750 18-20 [27]
corynebacteroides SMI
Rhamnolipids Pseudomonas strains 800 10 [27]
Sophorose Candida bombicola ATCC 1084 [28]
22214
Candida bogoriensis 22 [81]
Other biosurfactants
Aerobactin Aerobacter aerogenes 621 1371 6 [23]
Mannoprotein Saccharomyces cerevisiae 44 17 [82]
+ §
in Dalton. Carbohydrates in %. Proteins in %. * Fatty Acids in linear carbon length.
251
The activity of proteins surfactants is due in part to their backbone, which

allows the polypeptide to be flexible and disordered in aqueous solution.
Accordingly, particular domains present in a protein can rearrange at the
interface to form an ordered structure while other domains remain unassociated
with the interface. In the latter case, three dimensional loops and tails may be
generated [90-93]. The association of specific domains in some ordered
arrangement at the interface can be detected in a variety of ways including direct
examination using physical and chemical analyses [94]. Many of the proteins,
which associate with the interface, appear to be in their native state. For
example, even an unstructured protein such as P-casein may not conform to the
simple picture of loops and tails occurring in the interfacial area. With globular
proteins such as lysozyme or fS-lactoglobin, so much secondary and even tertiary
structure is retained that the adsorbed monolayer has been considered as a two-
dimensional system of "interacting deformable particles" [95], Although the
situation in the interfacial region is dynamic, the overall effect is generally
considered as an irreversible adsorption. This is because segments of the protein
molecules can attach to the interface at many contact points, but it is unlikely
that the protein will detach from all these points of contact at the same time or to
the same extent [96]. If additional polymer is available this may also move
towards the interface and associate with the free loops and tails of the bound
material. In addition to proteins, there have been several reports of the
emulsification properties of peptides obtained by enzyme hydrolysis [97]. These
peptides were obtained either by synthesis [98] or by enzymatic proteolysis (e.g.
elastin digested by either pepsin [84] or papain [99]; P-casein digested by
chymosin [100]; BSA digested by trypsin [101], and gelatin digested by papain
[102].
3.1.2. Polysaccharides at oil-water interfaces

Polysaccharides are widely used in food products to modify texture [103],
to control water mobility [104], and to improve moisture retention [63]. The
contribution of polysaccharides to the stability of emulsions has traditionally
been related primarily to their rheological effects on the continuous phase [64].
Polysaccharides are known to have less surface activity in comparison to
proteins [63]. This is due to their relative rigidity and low flexibility and to the
repetition of the monomer units in the backbone [105]. Polysaccharides can be
either stabilizers or emulsifiers depending on their behavior at liquid-liquid
interfaces. Their activity as emulsifiers is evident when they adsorb onto the oil-
water interface. Such property is characteristic, in particular, of semi-rigid-chain
polysaccharides having an elongated chain conformation, for example, alginates
and pectins [106]. These anionic polysaccharides possess surface activity
resulting in an increase in the negative charge on the oil droplets giving rise to
electrostatic repulsion [107-110]. According to a different explanation, it is
252
assumed that the surface activity of these polysaccharides is mainly caused by a

combination of non-hydrated moieties such as certain carboxylates and methoxy
carbonyl groups, with hydrated moieties such as a polydioxypyranosyl main
chain [109].
Several stabilizing mechanisms have been proposed to explain the
functionality of polysaccharides as stabilizing agents. One of them is their
hydrophilicity accompanied by their high molecular weight, which leads to
formation of a macromolecular barrier in the aqueous medium between
dispersed droplets. The gums of Portulaca oleracea [111], and Opuntia ficus
[112] are representatives of this group. Another mechanism of stabilization
derives from the high intrinsic viscosity of certain polysaccharides in water [64]
as in the case of xanthan gum, produced by Xanthomonas campestris [113].
These remarkably viscous polymers actually retard phase separation by
enhancing the viscosity of the aqueous phase [114].
3.1.3. Polysaccharide-protein interactions

Interactions between surfactant molecules and synthetic or natural polymers
have been studied extensively over the past several decades. These interactions
have received considerable attention because of their ability to impart significant
changes to the interfacial, rheological, and physicochemical properties of
polymer systems, with important implications in various pharmaceutical,
biomedical, food processing, and photographic applications [115-118]. In recent
years, the study of the interaction between proteins and polysaccharides has
grown in importance. The reason for this is that protein-polysaccharide solutions
may exhibit special properties, which provide abundant topics for scientific
investigation as well as the possibility of practical applications. Different classes
of interactions are found between polysaccharides and proteins. Such
interactions depend directly on the nature, pH, ionic strength, temperature,
concentration and other minor variables, which generate attraction or repulsion
between the polysaccharide and the protein in solution.
In systems containing protein-coated oil droplets, an attractive protein-
polysaccharide interaction may enhance emulsion stability by forming a thicker
and/or strong steric stabilizing layer, or it may act inversely destabilizing the
emulsion by forming polymer bridges between flocculated droplets [119-121]. A
repulsive protein-polysaccharide interaction may stabilize the emulsion by
immobilizing protein-coated droplets in a polysaccharide gel network. This
phenomenon is due to the formation of hydrogen bridges and local electrostatic
interactions (dipoles) of oppositely charged groups [106]. Incompatibility of
proteins with all types of polysaccharides becomes more pronounced with an
increase in salt concentration. Presumably, this is mainly due to intensification
of protein self-association (aggregating with itself) [122].
253
Protein and polysaccharide molecules might be brought together at an

emulsion droplet surface by two distinct pathways: 1) covalent coupling by
chemical reaction to form a protein-polysaccharide hybrid molecule [119, 123-
129] ; or 2) non-covalent association to form a protein-polysaccharide complex
[94,96, 105, 121, 130-132].
1) Covalent coupling: this is the case in which two kinds of biopolymer
molecules are inseparably linked together. The main advantage of a covalent
protein-polysaccharide hybrid over a non-covalent complex is the retention of
molecular integrity and solubility over a wide range of solution conditions [94].
A covalent linkage between protein and polysaccharide is an attractive
interaction, which can be specific, strong, and relatively permanent. This
interaction might be observed using either natural or artificial hybrids. Natural
hybrids are represented by galactomannans [133] and by gum arabic, which
contains 2% covalently bound protein [134]. Examples of synthetic covalent
hybrids are the interactions between gelatin [135] or whey proteins [105] with
propylene glycol alginate under mildly alkaline conditions; ovalbumin-dextran
under dry-heat storage [124] or casein-carrageenan in mixed gelling system
[105, 136].
2) Non-covalent coupling: In this case the association may be relatively
weak and hence readily reversed. A non-covalent complex may dissociate or
precipitate on changing temperature or pH. Weak interactions between proteins
and polysaccharides can be repulsive or attractive. Repulsive interactions are
always non-specific and are of transient duration. They usually arise from
excluded volume effects and/or electrostatic interactions. Net repulsive protein-
polysaccharide interactions are most likely to be found in mixtures of proteins
with non-ionic polysaccharides or with anionic polysaccharides at a pH above
the protein isoelectric point [137]. This interaction is observed in systems
containing mixtures of gelatin-dextran [138] or casein-guar gum [137]. Other
non-covalent hybrids between a specific genetic recombinant protein or its
peptides and various polysaccharides have been found in the case of the
recombinant form of the cell surface esterase from Acinetobacter venetianus
RAG-1 [139-140].
Attractive biopolymer interactions may be strong or weak. Strong
attractive interactions may occur between positively charged proteins (pH<p/)
and anionic polysaccharides, especially at low ionic strength, and weak
attractive interactions may occur between uncharged or negatively charged
proteins (pH>p/) and polysaccharides [141]. Alternatively, weak attractive
interactions arise as a result of averaging over a multitude of individual specific
chemical interactions between groups on the biopolymers. Examples include 1)
ionic interactions in BSA-dextran sulfate [95, 123] or BSA-K-carrageenan [142];
2) van der Waals interactions as in the case of lysozyme with different
polysaccharides [143]; 3) hydrogen bonding as in gelatin complexed with
254
different polysaccharides [131]; and 4) hydrophobic interactions as in poly

methacrylic acid [144] or in mixed gels of gelatin with sodium alginate [145].
The protein-polysaccharide interactions may change from net repulsive to
net attractive, or vice-versa, on changing the temperature [144] or the solvent
conditions (pH, ionic strength) [146]. Therefore, many polysaccharides carrying
negatively charged carboxyl or sulfate groups, can form strong electrostatic
association, so called "complex coacervation" [147] between the two kinds of
polyelectrolytes of opposite net charge at pH values below the isoelectric point
of proteins (typically pH~5). Nevertheless, the same polyelectrolytes fail to form
strong electrostatic complexes at neutral pH. In this case local electrostatic
interactions may still occur between anionic polysaccharide molecules and
positively charged patches on proteins [105, 148]. Such interactions can be
reinforced by non-ionic attraction (e.g. hydrogen bonding) leading to protein-
polysaccharide complex formation.
3.2. Microbial sources of bioemulsifiers

Table 3 A and 3B list a number of bacterial and fungal strains, which have
been shown to produce extracellular bioemulsifiers. The emulsification process
causes droplet formation of one of the phases and a subsequent increase in
emulsion turbidity, which is easily monitored. In addition, the bioemulsifiers
also stabilize the emulsions by retarding droplet coalescence.
3.3. The emulsan paradigm

Arguably, the most extensively studied polymeric microbial bioemulsifier
is emulsan, the amphipathic, polyanionic polysaccharide produced by the oil-
degrading Acinetobacter venetianus RAG-1 [65, 149-151]. In addition, this
product represents an interesting case study encompassing all stages of
development, technology transfer and product development. Although many of
the features of this model system have been presented in other reviews [6-8,
151-153], newer developments regarding the genetics, biosynthesis and
regulation of its production are only now emerging, and will be discussed here.
Similarly, the development of novel approaches to generating new amphipathic
bioemulsifier formulations [140] has now been developed and will also be
presented.
3.3.1. Physical and chemical characteristics

Emulsan was initially isolated from the cell-free broth of a stationary phase
culture of A. venetianus RAG-1 grown on a light crude oil. Subsequently, the
non-dialyzable polymer was produced on a variety of water soluble carbon
sources with optimum yields obtained on ethanol as sole source of carbon and
energy [65, 75, 154-156]. The bioemulsifier (106 Da) consists of complex
255
between a polyanionic heteropolysaccharide (75%) and a non-covalently

associated protein mixture (25%) [75].
As shown in Fig. 1, the linear backbone of the polysaccharide consists of a
trisaccharide subunit consisting of a 1:1 ratio of the amino-sugars, D-
galactosamine, D-galactosamine uronic acid and 2,5-dideoxy, 2,5-diamino
glucose. The amphipathic properties of the biopolymer arise from the presence
of about 20% by weight fatty acids present in the form of amides and esters.
The fatty acid composition can be modified by including fatty acids in the
growth media [161-164], although the commercial product produced on ethanol
minimal media consists of a mixture of 2, and 3 hydroxy dodecanoic acids,
palmitic acid, 2-hydroxybutyric acid and acetic acids [7(55]. The apparent pK of
the polymer is about 3.05 owing to the galactosamine uronic acid.
In addition to the lipoheteropolysaccharide composition of the
bioemulsifier, the extracellular emulsan complex also contains between 10 and
25% by weight protein. A portion of the protein mixture can be removed by any
of a number of procedures including sepharose gel filtration chromatography,
polymer precipitation in the presence of the quaternary ammonium detergent
cetyl, trimethyl ammonium bromide [766], hot phenol extraction [75] and
proteolysis. As discussed below, the deproteinized polymer, termed apoemulsan
[154] retains a portion of the emulsifying activity towards some of the
substrates, but is generally inactive in the emulsification of less polar materials
such as pure n-alkanes such as hexadecane. The role of this protein in mediating
emulsification by apoemulsan will be discussed in a different section below.
Fig. 1. The trisaccharide structure of the emulsan subunit.

256
Table 3A
Production and surface activity of bioemulsifiers
Compound Organism C-source Yield* ST Reference
Exopolysaccharides
Acinetobacter calcoaceticus BD4 glucose 0.6 [157]
Acinetobacter calcoaceticus MM5 tetradecanc 0.06 \70]
Bacillus cereus IAF 346 sucrose 0.5-1.2 53 \7l]
Bacillus sp. IAF343 sucrose 0.5-1.2 28 [71]
Halomonas eurihalina H28 crude oil [72]
Alasan Acinetobacter radioresistens K53 ethanol 2.2 [73]
Biodispersan Acinetobacter calcoaceticus A2 ethanol 4 [74]
Emulsan Acinetobacter venetianus RAG-1 ethanol 15 [158\

Liposan Candida lypolytica ATCC 8662 hexadecane [76]
Lipids
Rhodococcus sp. Q15 glucose, acetate 36 [77]
Nocardia erythropolis ATCC 4277 hexadecane, 29 [159\
kerosene
Saccharomyces uvarum n-dccane 20 [78]
Lipopeptides
Bacillus liqueniformis JF-2 glucose 25-34 [79]
Cory'nebacterium lepus kerosene 0.35 30 [160]
Amphisin Pseudomonas sp. DSS73 glucose 27 [80]
Arthrofactin Pseudomonas sp. MIS 38 L-broth 24 [56]
Bacitracin Bacillus liqueniformis 27 [34]
Hodersin Pseudomonas sp. glucose 27 [80]
Lokisin Pseudomonas sp. DSS41 glucose 27 [S01
Lychesin A Bacillus liqueniformis BAS50 glucose 0.16 28 [38]
Scrrawettin Serratia liquefaciens MG1 glucose 28 [42]

Surfactin Bacillus subtilis ATCC 21332 glucose 3-4 27 [43]
Tcnsin Pseudomonas fluorescens 96.578 glucose 27 ISO]
Viscosin Pseudomonas viscosa glucose 26.5 [19\
Viscosinamide Pseudomonas fluorescens DR54 glucose 27 [80]
ing/I.
Surface Tension in mN/m.
257
The molecular weight of emulsan is about 10 Da [75]. Nevertheless, the

polymer exhibits a rather low reduced viscosity in water of about 570 cc/gram
[150] and appears to be a long rod of axial ratio about 50:1. The low viscosity of
the biopolymer complex is due in part to the presence of the associated-protein.
Emulsan is Newtonian in its flow properties, despite the rather high molecular
weight, a feature which makes it easier to produce as a fermentation product.
3.3.2. Emulsan activity

As an emulsifier emulsan does not dramatically affect either surface
tension or interfacial tension between two immiscible phases. Nor does it form
micelles, which is characteristic of detergents. Rather its activity appears to
depend on its high affinity for the oil/water interface. Being a water-soluble
polymer emulsan both forms and stabilizes oil-in-water emulsions resulting
from the orientation of the biopolymer at the oil droplet surface [169]. This
orientation was first inferred from the binding characteristics of the positively
charged rhodamine cation, which bound emulsan in an emulsion, but did not
bind to the polyanionic polymer when emulsan was in solution. This binding of
cations to emulsan preferentially at the oil/water interface was observed with
several metal ions such as C d \ Zn+ , UO2 , Mn , and a host of others [169-
170]. In many cases the binding was more extensive than anticipated from the
charge density of the polymer suggesting that the polymer had assumed a
different conformation when oriented at the interface. Under conditions of high
energy input, when the formation of small oil droplets is the result of vigorous
agitation, the high affinity of emulsan for the interface results in its coating of
the droplets preventing coalescence due to charge repulsion of the negatively
charged uronic acids. This stabilization does not require the presence of protein,
and can be achieved with apoemulsan at correspondingly low ratios of
emulsifier to oil between 1:100 to 1:1000 parts of emulsan to oil. The resulting
emulsions are very stable, and withstand high-speed centrifugation. In fact,
rather than breaking the emulsion into two phases, the centrifugation causes
formation of a cream layer which itself is an oil-in-water emulsion consisting of
a bulk aqueous phase which constitutes only about 30-50% of the cream,
depending on the ratio of emulsan to oil. This cream layer, termed an
emulsanosol, is itself an oil-in-water emulsion readily dispersible in the aqueous
phase. Emulsanosols or apoemulsanosols can be prepared with a variety of pure
and cruder oils, and are stable for years. Their potential applications will be
discussed later in the Chapter.
In addition to stabilization of oil/water emulsions, emulsan also forms
emulsions at lower rates of agitation. In these systems the presence of proteins
plays a major role since emulsion formation is quite specific for the hydrocarbon
substrates [150], In general, emulsan has been shown to emulsify relatively
polar mixtures of hydrocarbons such as those containing both an aliphatic and an
258
aromatic substrate. Hexadecane alone is a very poor substrate, and in the

absence of protein, emulsan is completely ineffective with non-polar materials
[139]. Interestingly, activity towards hexadecane and other non-polar waxes and
sludges can be reconstituted in the presence of a single recombinant protein, the
cell-surface esterase of A. venetianus RAG-1. The results in Table 4 illustrate
the effect of addition of the cell surface esterase of RAG-1 on the emulsification
of a variety of substrates, none of which is emulsified very well by emulsan
itself. Interestingly, the addition of the recombinant esterase to emulsan itself
had a dramatic effect on the emulsification towards very hydrophobic substrates
[139].
Table 3B
Production and surface activity of bioemulsifiers
Compound Organism C-source Yield' ST Reference
Glycolipids
Rhodococcus erythropolis Mihagol L 32 [25]
DSM 43215 Mihagol S 8
Flavolipid Flavobacterium sp MTN11 glucose 0.1 26 [24]
Pentasaccharide Nocardia corynebacteroides nCi4_ 15 2.8 26 [27]
SMI
Rhamnolipids Pseudomonasputida 21BN hexadecane, 1 29 [167]
glucose
Sophorose Candida bombicola ATCC glucose, 70 31 [168]
22214 safflower oil
Torulopsis petrophilum glucose 43 [31]
Candida bogoriensis glucose 2 [SI]
Others biosurfactants
Aerobactin Aerobacter aerogenes 621 glucose 1 [23]
Mannoprotein Saccharomyces cerevisiae glucose 8" [82]
Synthetic surfactants
Cetyl Triethyl Ammonium Bromide (CTAB) 30
Linear alkylbenzene sulfonate 47
Sodium dodecyl sulphate 37
Tween 20 30
Water 72
ing/1.
Surface Tension in mN/m.
±
in gr/ wet cell gr.
259
Table 4
Enhancement of apoemulsan activity on different hydrophobic substrates by recombinant
esterase
Hydrophobic substrate Emulsifying Activity Ratio
Anthracene 966
Crude oil 4.3
Dicyclohexane 8.9
Diesel oil 5.7
Eicosane 1800
Fluoranthene 593
Heptadecane 28
Immersion oil 10.4
2-Methyl Naphthalene 1984
Mineral oil 6.6
Octadecane 2250
Petroleum refinery sludge 2
Pyrene 420
Soya oil 1260
Squalene 600
Tetracosane 506
Apoemulsan emulsifying activity in presence of recombinant esterase: emulsifying

activity.using emulsan alone.
3.3.3. Emulsan as a microbialproduct

Physiology of production. As a fermentation product emulsan is produced
primarily on a minimal medium containing ethanol as a sole source of carbon
and energy [65]. During early exponential growth the emulsan capsule is present
on the cell surface as a minicapsule, which is subsequently released from the cell
surface as the cells approach stationary growth. Turnover experiments using
radioactive substrate indicated that product release is accompanied by de-novo
synthesis [10, 171-172]. Moreover, polymer release is accompanied by a prior
release of a protein complex including the cell surface esterase mentioned
above.
When the cells were starved for carbon in the presence of
chloramphenicol the esterase was released but the emulsan remained associated
with the cell surface. Addition of fresh nutrients did not give rise to polymer
release under these conditions, presumably because the esterase could not be
made in the presence of chloramphenicol. The results strongly implicated cell
surface esterase in the release process. Interestingly, when emulsan production
was studied in a resting cell system in which the cells were immobilized on a
celite column, emulsan was shown to accumulate on the surface of the cells in
response to increased shear forces [173].
260
Emulsan biosynthesis. In order to study the biosynthesis of emulsan, an

insertional plasmid was used to generate mutants-defective emulsan production.
Such mutants are visualized either by virtue of their translucent colonial
morphology (Fig. 2) or according to their resistance to particular RAG-1 specific
phage, ap3, which uses the cell bound emulsan minicapsule as a phage receptor
[174]. A mutant blocked in emulsan biosynthesis is thus resistant to phage ap3,
but sensitive to a second phage, ap2, which does not require the emulsan
receptor. In this regard, the phage ap3 does not bind to the released, water
soluble emulsan complex, but does bind to the concentrated emulsanosol, again
supporting the idea that the emulsan polymer undergoes a conformational
change at the oil/water interface [174]. These data also support the important
notion that the conformation of the biopolymer on the cell surface is similar to
its conformation at the oil/water interface.
Natural role for emulsan. Cells defective in emulsan do not grow on
crude oil in liquid culture, although they do grow on these hydrocarbons when
they are provided in the vapor phase [174]. Gutnick and Shabtai showed that the
cell bound polymer protected cells of RAG-1 from the toxic effects of catalytic
surfactants such as cetyl-trimethyl ammonium bromide [166]. In fact, among
mutants resistant to the cationic detergent were those, which actually produced
more biopolymer than the corresponding wild-type RAG-1. Based on findings
that cells lacking emulsan on their surface more avidly to hydrophobic materials
[175-177], it is possible that the natural role of the biopolymer is to aid in
actually removing the cells from the oil, when utilizable carbon is no longer
available to the organism. This would expand the versatility of the organism by
allowing it to release itself and search for new more productive surfaces for
metabolism. In support of this hypothesis was the finding that emulsan could be
used to remove other organisms from hydrophobic surfaces [178]. Another
potential role for emulsan relates to its role as a microbial capsule prior to its
release from the cell surface. Ophir and Gutnick [179] showed that capsule
producing organisms exhibit at least a ten fold higher resistance to desiccation
than do isogenic strains lacking the capsule. This resistance depended on the
adherence of the capsule to the cell surface and could not be restored by addition
of fresh biopolymer from the outside. Emulsan producers were shown to exhibit
such resistance to desiccation.
Emulsan biosynthetic pathway. The genes encoding the biosynthetic
pathway for apoemulsan have recently been localized to a 27 kbp cluster termed
the wee regulon [180]. The entire cluster was sequenced and shown to encode 23
putative open reading frames arranged in two divergent operons separated by a
non-translated region (Fig. 3). Mutations in any of the genes resulted in a defect
in emulsan production. These defects could all be complemented by a wild-type
allele. Interestingly, most of the genes for emulsan biosynthesis were
homologous to genes discovered in the biosynthesis of most polysaccharides in
261
spite of the fact that A. venetianus RAG-1 is thus far the only natural isolate
which has been shown to produce emulsan [180]. According to convention, the
specific genes for emulsan biosynthesis were termed wee; the first letter
signifies that the gene is a biosynthetic gene from a polysaccharide biosynthetic
cluster, the second land third letters identifying the specific product (emulsan,
exopolysaccharide). In accordance with convention, some gene products exhibit
similar functions in all organisms, and thus are allowed to retain their original
names. Figure 3 summarizes a hypothetical biosynthetic pathway for
apoemulsan, with, the rightward operon encoding proteins involved in precursor
synthesis and activation, aminoglycosyl transferases for assembling the
trisaccharide subunit on the inner side of the cytoplasmic membrane, a
polymerase, decorating enzymes for the acylation of the aminosugars, a
translocase which moves the polymer from the cytoplasmic face of the
membrane to the outer or periplasmie face, and enzymes involved in subsequent
translocation of the polymer through a specific channel or porin to the outer
surface of the cell [180].
Regulation and the production of viscoemulsan. Located in the
intercistronic region are two putative d promoters. The leftward operon
consists of three repeating frames wza, wzb and wzc, which encode a porin, a
protein tyrosine phosphate phosphatase and a protein tyrosine kinase,
respectively [180-181]. Knockout mutants in any of these genes resulted in
defects in emulsan production. Both Wzc and Wzb proteins of RAG-1 were
cloned and over-expressed in E. coli. The Wzc Ptk was shown to be an
autophosphyorylase in which a tyrosine (s) in the C-terminal portion of the
protein is phosphorylated and subsequently dephosphorylated by the
phosphatase [182]. Similarly, the phosphotyrosine of Wzc from RAG-1 was
shown to be dephosphorylated by Wzb [181]. According to other reports, the
phosphorylated form of Wzc is expected to negatively regulate polymer export
through the porin Wza. Elevated levels of extracellular biopolymer production
would then be initiated with the activity of Wzb, the phosphatase, which
removes the phosphates, permiting the enhanced export. Consistent with the
hypothesis was the finding that knockout mutants in the phosphatase were also
emulsan deficient. However, the results did not explain why knockouts in Wzc
would be emulsan deficient as well. Apparently there is a requirement for the
Wzc protein even in its non-phosphorylated state. The Wzc protein contains a
series of five tyrosine residues in close proximity to each other at the C
terminus. When these tyrosines were deleted, the resulting protein was made but
could be phosphorylated and surprisingly, a high molecular mass
polysaccharide, termed viscoemulsan, was produced [181]. This product appears
to contain the same constituents as emulsan, but is not active as an emulsifier
(Nakar, In preparation). The introduction of a wild-type allele of wzc gave rise
to the production of a wild-type allele of emulsan suggesting that the protein
262
tyrosine kinase may act to control the size of the exported polymer. It is also of
interest that the Wzc protein is required for viscoemulsan production even
though it cannot be phosphorylated [181] suggesting that there is an additional
role for the protein. A model to describe the role of phosphorylation and
dephosphorylation is shown in Fig. 4 [181]. According to this model Wzc, Wzb,
Wza proteins and others interact in a multienzyme complex to control the export
of the exopolysaccharide. The process is initiated by dephosphorylation of the
protein relaxing the control on the porin diameter and enabling larger amounts
of polymer to be translocated to the external surface of the cell. Under
conditions of rapid growth and high ATP, all of the tyrosine residues are
phosphorylated and polymer production is low. In fact, emulsan production does
not take place in rich media, although its biosynthesis has been shown to occur.
The polymer can be detected immunologically. The manipulation of the export
process coupled with the modifications of the biosynthetic genes offers new
approaches to the generation of new and novel products and is currently in
progress.
3.3.4. Engineering novel derivatives of emulsan

The production of new viscous derivatives such as viscoemulsan
represents one approach to engineering new biopolymers. Kaplan and co-
workers have used nutritional modification to modify fatty acid composition and
surface active properties of the resulting derivatives of emulsan [161-164].
Moreover, as described above, the surface activity of apoemulsan-containing
formulations can be enhanced by the addition of a particular cell surface
enzyme, the cell surface esterase of RAG-1 [139].
Fig. 2. Colonial morphology of parental emulsan-producing RAG-1 and a translucent,

emulsan-defective mutant, TR3.
263
Fig. 3. The wee cluster for the biosynthesis of emulsan. The scale of the cluster size is in
kilobases. The black arrows represent putative orf sequences. White arrows represent partially
sequenced orf s. Putative promoter sites are indicated with thin black arrows. The names of
the genes are shown below the corresponding orf s. Orf s labeled solely with capital letters
are putative pathway specific genes encoding Wee A-K respectively.
Novel surface-active breakdown products. In order to stabilize water/oil

emulsions emulsan must be a polymer [176]. The evidence supporting this
comes from the activity of a particular emulsan depolymerase isolated from a
bacterial isolate capable of using galactosamine as a sole source of nitrogen.
When apoemulsan was incubated with this crude enzyme for different periods of
time, it was found that cleavage of less than five percent of the glycosidic bonds
were necessary to inactivate the biopolymer. The results support the idea that the
interaction of biopolymer at the oil/water interface is a weak, and that the
stabilization is brought about by many points of weak interactions at the
oil/water interface. Interestingly, subjecting apoemulsan to exhaustive digestion
by the depolymerase generated a series of small acylated aminooligosaccharides
consisting of between three and six aminosugars [183]. These materials were all
found to act as small molecular weight detergents, although they did not
effectively stabilize emulsions. They were found to be active towards more
hydrophobic substrates such as hexadecane.
Emulsification enhancing proteins and peptides (EEPs). As described
above, emulsification of oils by emulsan and apoemulsan is strongly enhanced
by the addition of a cell surface recombinant esterase from RAG-1 [139].
Potential principles governing this type of interaction have been presented
above. The cloning of the enzyme in E. coli was first accomplished by selecting
recombinants of E. coli, which could grow on simple triglycerides as sole
sources of carbon and energy [184-186]. The clone was required to generate
metabolizable substrates such as glycerol and simple fatty acids such as acetate
in order to grow. Cloning, sequencing, over-expression and mutagenesis
experiments demonstrated that the esterase is a serine protease [187]. However,
using a threading program in which primary sequence was related to predicted
structures it was found that the protein in fact more resembled an a, P hydrolase
enzyme such as acetyl cholinesterase [187]. Interestingly, this was also found
264
for an esterase from another member of the genus Acinetobacter, the strain A.
calcoaceticus BD4 [188] and its miniencapsulated derivative BD413. While this
enzyme shows strong sequence and structural homology to the RAG-1 enzyme,
it did not display any emulsification enhancement when added to apoemulsan
[139].
Specificity towards hydrocarbons. As shown in Table 4 the recombinant
esterase protein enhances emulsification of apoemulsan towards a variety of
pure and crude hydrophobic substrates. EEP activity was observed with mutants
of the esterase defective in catalytic activity, suggesting a role for the protein
other than as an enzyme.
Esterase exhibits EEP activity towards other polysaccharides.
Surprisingly, the interaction of the recombinant RAG-1 esterase with the water
soluble, rhamnose-containing exopolysaccharide from A. calcoaceticus BD4 led
to the formation of a new bioemulsifier complex. In sharp contrast, the esterase
from BD4 did not enhance emulsifying activity of apoemulsan towards
hydrophobic substrates [140]. Remarkably, the recombinant esterase from RAG-
1 exhibited EEP activity with over 25 different natural biopolymers, none of
which exhibited any emulsifying activity in the absence of the protein. In these
cases, the enhancement was not dependent on catalytic activity of the
recombinant protein (Bach and Gutnick, in preparation). The results point to a
new approach to generation of amphipathic emulsifiers, which is no longer
dependent on fermentation to produce the polymer emulsifier. Among the
inexpensive materials, which can be converted into bioemulsifiers using this
unique formulation with the RAG-1 esterase are cellulose, dextran, starch,
xanthan, alginic acid, and a variety of plant and bacterial polysaccharides
including the inactive viscoemulsan described above (Table 5). The mode of
action of the EEP remains to be elucidated although evidence is discussed below
demonstrating that there is a unique motif in the RAG-1 esterase, which is
missing from other homologues.
Mapping the EEP domain. Initial observations showed that limited
proteolysis of the recombinant esterase yielded a fragment of about 10 kDa,
which retained the ability to enhance emulsification of hydrophobic substrates
such as hexadecane. Accordingly, a series of site directed mutants were
generated and over-expressed to produce different fragments of the esterase.
Since the fragments were rapidly degraded even in strains of E. coli lacking Clp
or Lon proteases, fragments were prepared which were fused in frame to the C-
terminus of the maltose binding protein [189]. The various constructs are shown
in Fig. 5. The each over-expressed fusion was tested with apoemulsan using the
model hydrophobic substrate, hexadecane as a substrate for emulsification.
Virtually all the enhancing activity was localized to the C-terminal third of the
esterase. It was of interest that the maltose binding protein itself exhibited no
EEP activity. Moreover the fusion protein containing the active polypeptide was
265
no less active than the intact enzyme. Removal of the terminal 15 amino acids
from the C-terminus completely abolished the EEP activity. Sequence analysis
showed that this 15 amino acid C-terminal peptide is unique to the RAG-1
esterase and probably accounts for the unique characteristics of this protein.
However, as shown in Table 2, many organisms produce emulsifiers consisting
of protein/polysaccharide complexes [7, 9, 190]. In most cases the protein
requirement has yet to be clarified and it is possible that there are other proteins
or peptides, which exhibit unique EEP activity. Regardless, EEP technology
offers a new approach to bioemulsifier production and paves the way for new
families of inexpensive, non-toxic, amphiphiles.
Fig. 4. Hypothetical model for the role of protein tyrosine kinase (Wzc) and protein tyrosine
phosphatase (Wzb) in emulsan export. 1. Dephosphorylated Wzc allows for polymerization
and translocation of emulsan. 2. Phosphorylation of Wzc halts the process, thereby
determining the size of the exported polymer. 3. Emulsan release and beginning of a new
round of polymerization, translocation and release. Wza-translocation channel; Wzb-protein
tyrosine phosphatase; Wzc-Protein tyrsoine kinase; Wzx-polymerase; Wzy-translocase.
266
3.4. Other polymeric bioemulsifiers
3.4.1. Alasan
Acinetobacter radioresistans radioresistens KA53 produces a
bioemulsifier complex (10' kDa) consisting of three proteins and a
polysaccharide [73]. The emulsifying activity was associated primarily with the
AlnA protein. Interestingly, the N-terminal sequence of a recombinant form of
the AlnA protein produced in E. coli showed strong homology to the outer
membrane protein, OmpA [191]. The recombinant form of AlnA was more
active as an emulsifier than the complex. It was also shown to solubilize
polyaromatic hydrocarbons and at higher concentrations of the substrate, to form
hexamers [192]. The crude alasan complex also formed alkane/water emulsions
at an optimum pH of 5. This activity was significantly enhanced after heating at
100°C. Interestingly, the alasan producing strain does not grow on hydrocarbons
or on oil substrates and the biological role of this complex remains to be
elucidated.
Table 5
Enhancement of the emulsifying activity of different polysaccharides by recombinant
esterase in the presence of hexadecane
_ , , , Emulsifying activity
Polysacchande ,.., , , , . .
(U/mg polysacchande/mg esterase)
Agarose 963
Alginic acid 496
Apoemulsan 5430
BD-4 exopolysaccharide 3396
Carrageenan 3345
Cellobiose 626
Cellulose 766
Chitin 540
Colamc acid 2050
Dextran 583
Emulsan 6752
Ficoll 400 263
Gum Arabic 1895
Pectin 1830
Polyvinyl Pyrrolydone 1950
Potato starch 544
Pullulan 3400
Stewartan 1196
Xanthan 2720
Xylan 1854
267
Recently another strain of A. radioresistans SI3, which grows on

aromatic substrates was analyzed for changes in membrane protein composition
in response to changes in the growth substrates [193]. Two-dimensional gel
electrophoresis of protein extracts from S13 revealed elevated levels in an Omp
A-like alasan ortholog in response to growth on phenol.
3.4.2. Liposan
Liposan is a polymeric bioemulsifier produced by the yeast Candida
lipolytica ATCC 8662 [76]. The protein polysaccharide complex consists of
83% polysaccharides and 17% protein. When grown on hexadecane organism
appeared to colonize the hexadecane droplets. Liposan emulsified alkanes with a
chain length between C6 and C18 with the emulsifying activity increasing with
increasing chain length. Liposan has also been shown to emulsify various crude
oils such olive and corn oils, gas oil, kerosene, paraffin, halowax 1000 and a
series of aliphatic and aromatic hydrocarbons.
3.4.3. Biodispersan
This polymer is produced by Acinetobacter sp. A2. The extracellular
product exhibited a molecular mass of 51,400. This polyanionic polysaccharide
is a dispersant, which disperses limestone and aids lowers the energy required
for grinding limestone to form a powder, which is an ingredient in paper [74,
194].
Fig. 4. Generated esterase constructs fused in frame to the C-terminus of the maltose binding
protein.
268
3.4.4. Exopolysaccharide-protein complex from Acinetobacter calcoaceticus

BD4
A.calcoaceticus BD4 produces a thick rhamnose-containing exopoly-
saccharide protein complex [157, 195]. The same biopolymer is produced by a
strain, which produces similar quantities of extracellular polysaccharide even
though it accumulates lower amounts of the capsular polysaccharide on the cell
surface. The protein-polysaccharide complex was shown to emulsify mixtures of
alkanes and aromatic substrates, but was inactive against pure alkanes.
Interestingly, the complex was not particularly active in the presence of crude
oil. Of interest, however, was the finding, that unlike the emulsan complex,
which retained partial emulsifying activity even after removal of the protein
fraction, the BD4 product was completely dependent on both the protein and the
polysaccharide fractions. The polysaccharide could be prepared without the
protein by physically shearing it from the cell surface, and the activity restored
by the addition of the protein fraction [196]. The protein fraction was specific
for the BD4 polysaccharide, and was able to enhance the emulsifying activity of
apoemulsan (H.Bach, in preparation). However, as was the case with a host of
bacterial exopolysacchride [140], the recombinant esterase from RAG-1 was
able to reconstitute emulsifying activity to the polysaccharide from A.
calcoaceticus BD4.
4. POTENTIAL APPLICATIONS
4.1. General comments

The petroleum industry consumes millions of tons of surfactants each
year in a large number of applications (see chapters 4 and 19). Surfactants are
used in oil field applications (drilling muds, enhanced oil recovery),
environmental and equipment clean-up and maintenance, viscosity reduction
and oil transportation, emulsion breakage and dewatering of crude oil prior to
refining, and more recently, water/oil based fuels. Generally, the surfactant
packages include a combination of surface-active agents, and frequently include
compatible solvents and specialized chemicals depending on the quality of the
specific oils and sludges. As described in this Chapter, biological products and
processes can be employed in all of these applications. In this section we will
consider only a few larger scale experiments using biosurfactants. The
successful trials indicate that at least in terms of product efficacy, these
materials have potential. However, their profitability has yet to be unequivocally
demonstrated. In this section we will discuss a few large- scale field trials
pointing to the possibility of employing biosurfactants and emulsifiers in the oil
industry.
Unlike biotechnology products for other sectors such as health care or
Pharmaceuticals, where the cost of development and even the cost of obtaining
269
approval from the regulatory agencies, are offset by the high prices and
profitability of the product, the cost of applications in the oil industry must be
kept relatively competitive with products of the chemical industry. This is a
particularly difficult constraint considering that production of many of the
biotechnological products may involve large-scale fermentation processes,
which exert a considerable impact on the cost of the product, particularly if
extensive downstream processing is required. Several approaches may be used
to enhance the cost effectiveness of biosurfactants.
4.1.1. Searching for the "world beater"

Occasionally, the search for natural products yields a compound with
unique properties unlike others either natural or developed by the chemical
industry. Such products are termed "world beaters" because their properties are
unique and unmatched by products of chemical synthesis. Antibiotics represent
a classical example of natural materials of major chemotherapeutic importance
whose activities are unmatched by chemical synthesis, although they are
generally modified by various chemical transformations [ 197]. Another example
of a microbial product with unique properties is the exopolysaccharide product
of the plant pathogen Xanthomonas campestris, xanthan, is a major polymer
whose sheer thinning properties and high reduced viscosity make it a major
industrial product in foods as a thickening agent, in drilling muds, and as a
gelling agent for use in oil field fracturing programs [5]. Moreover, xanthan
viscosity is exploited in oil field flooding during enhanced oil recovery, since
the extraordinary high viscosity enables it to actually "push" the released oil out
of the well. In most cases, however, the natural biosurfactant exhibits
characteristics, which are promising but not necessarily unique.
4.1.2. Cutting the cost of production

The cost of production of biosurfactants is frequently a function of the cost
of fermentation and subsequent downstream processing. This is particularly true
for production of products in which the carbon source must be a hydrocarbon,
which, is often the case with low molecular weight glycolipids [9, 16, 53, 198-
199]. A key approach in this system involves enhancing the product yield by
upgrading and optimizing the fermentation [21]. In the latter case the
fermentation of rhamnolipids has been upgraded such that lOOg/liter was
produced from 160g of soybean oil as a carbon source, a remarkable conversion
of substrate to product. In addition, production on various industrial waste
products has also lowered the cost of production [200-202]. Assuming that the
product biosurfactant is sufficiently active, this presents a way of upgrading the
waste material by producing a product of higher added value. This approach
may be particularly advantageous in the oil industry, since the crude product
270
need not be purified to any significant extent and may not require expensive
downstream processing [203].
Enhanced productivity can in principle be obtained by transferring
biosynthetic genes into an organism such E. coli K-12 which is easy to grow and
which utilizes a "friendlier" source of carbon and energy [204-205].
At least in pilot scale, most biosurfactants are produced in batch
fermentations. Cooper and co-workers developed a semi-continuous approach to
producing biosurfactants via self-recycling system [206]. Similarly, emulsan
was produced in a similar protocol in which the product was allowed to grow
and accumulate in early stationary phase followed by the removal of 90% of the
cells and emulsan, which was harvested downstream. The fermentor was filled
with fresh media and the culture again entered exponential and early stationary
growth, the major portion of the emulsan recovered and the cycles repeated in
the same fermentor for several semi-continuous production runs. The cost of
production is thus significantly reduced (Cooper, D., Personal communication).
Another way to cut the cost of production is to upgrade the producing
strain in order to enhance overall productivity [8, 199], This approach has been
used in the case of the emulsan producing strain A. venetianus RAG-1 [166].
The positive selection for emulsan overproducers was based on the fact that the
emulsan polyanion binds the toxic cation cetyl-trimethylammonium bromide
(CTAB). Among the mutants of RAG-1 resistant to CTAB, were those such as
strain A. venetianus CTR49, which overproduce the extracellular polyanion and
are thus significantly more resistant to the CTAB than the parent. In the
laboratory, mutants of this variety produced up to twice as much emulsan per
gram of ethanol carbon source than the wild-type.
4.1.3. Upgrading the product

Metabolic engineering of new products. Another approach to generating a
viable technology employing biosurfactants is to employ physiology,
formulation and/or recombinant DNA technology to generate modified products
with improved properties. One such approach involved preparing emulsan from
RAG-1 cells grown in the presence of various fatty acids [161-164] in order to
modify the nature of the acyl groups present in the side chains of the
bioemulsifier. We have carried out similar experiments and have found that the
emulsan produced is significantly more active towards hydrophobic substrates
such as hexadecane alone. Of course, the enhanced efficacy still needs to be
weighed against the increased cost of the fermentation due to the inclusion of
fatty acids in the media.
Formulation packages. Emulsifiers and surfactants are generally
incorporated into surfactant packages, which include a mixture of surfactants
designed to lower interfacial tension between water and oil phases. Also, in
cleaning applications, the formulations may also contain a biopolymer to
271
stabilize the emulsion [207-209] and prevent coalescence of the phases, and they
can also contain a compatible solvent. The solvent need not be a water based
solvent, but it should be able to dissolve both the low molecular weight
surfactants as well as the biopolymer. Materials such as pine oil, liquid
terpenoids, dimethyl sulfoxide, and various light crude oils have all been
included in various surfactant packages [210]. In addition to solubilizing all of
the components into a pumpable mixture, the solvent addition has also been
shown to enhance the cleaning of oil contaminated tanks by removing the last
remnants of sludge and other flammable materials from the walls of the
container rendering the tank not only clean, but also gas-free. The choice of
suitable components for various surfactant packages must also take into account
other potential components, which must be included. For example, if routine
cleaning operations include rinses with anticorrosive materials, the formulation
package must be designed on the basis of compatibility with such components.
Similarly, emulsion based fuels may need to be formulated together with
materials which lower sulfur emissions. Specially designed surfactant
formulations may also require compatibility with a variety of materials including
flame retardants, biocides etc.
4.1.4. EEP technology

In section 3.3.2 the ability of a recombinant cell surface esterase from
RAG-1 to enhance the emulsification of apoemulsan towards a variety of
hydrophobic substrates was described [139]. The remarkable feature of this
system was the finding that the RAG-1 esterase and several of its derivatives
were able to interact with a host of polysaccharides to generate a series of
amphipathic complexes, which exhibited strong emulsifying activity [140]. This
surprising activity, has paved the way for the generation of a whole suite of
bioemulsifiers, in which the polymeric component need not be produced as a
fermentation product. In fact, it may be possible to upgrade waste materials such
as crude celluloses, starch, pectins, etc. to bioemulsifier when combined with a
specific peptide derived from the esterase. This peptide can be produced as an
over-expressed protein fusion following cloning in a suitable vector [Bach and
Gutnick, in preparation], or it can be generated via proteolysis of the esterase
itself. The feasibility of employing EEP technology for emulsification in oil
industry applications will become clearer once larger scale field trials are
conducted and evaluated.
4.2. Bioemulsification, cleaning and sludge recovery
4.2.1. Tank clean ing

Oil storage containers accumulate enormous quantities of sludges and
bottom sediments. Previous work from this and other laboratories have
272
described bioremediation techniques in which the container is basically turned

into a fermentor and microbial biodegradation is used to clean the tanks [211-
212]. Bioremediation of oily wastes is covered in other chapters in this volume.
Another approach is to generate stable oil/water emulsions using a surfactant or
biosurfactant package in order to recover the sludge waste material in a
homogenous, combustible form. With this approach the costs of the tank
cleaning may be partially offset by the added value of the recovered emulsion.
Petroferm U.S.A. developed an emulsan-based biosurfactant package designed
to clean oil tanks and render them gas-free. Tanks in the order of several to tens
of thousands of cubic meters were cleaned using this system after first designing
several pumping and mixing devices. The oil-in-water emulsions generated in
these large-scale field trials were stable and could withstand high-speed
centrifugation to generate emulsanosols [207]. As will be discussed below, these
crude emulsanosols containing up to 30% water can be used as an emulsion-
based fuel.
Another emulsan-based application was developed by Dr. Mary Ann
Jones at the research station of the U.S. Navy in Washington. This application is
designed to rapidly and efficiently clean the filters normally used in the engine
rooms of ships for sludge removal from bilges. The emulsan-based formulation
includes, in addition to crude emulsan, light crude oil as a solvent. The system is
not designed for oil recovery, but rather for rapid filter cleaning. A rough
estimate suggests that the emulsan-based cleaning process could cost as little as
$300 per ship cleaning [Jones, Personal communication].
4.2.1. Emulsion-based fuels

Stable oil-in-water or water-in-oil emulsions, if sufficiently homogeneous
can be burned for energy, provided that the water content does not exceed about
30%. In fact, this is the water content of a stable emulsanosol prepared on an oil
such as a light Texas or Iranian crude oil. A large scale experiment was
conducted in which about fifty barrels of an emulsanosol was prepared from a n
emulsion consisting of about 70% high vacuum residuals emulsified in 30% by
weight water. The combustion of this material was tested at an experimental
laboratory in MIT. The combustion was identical to the combustion of a light
fuel oil in terms of burn temperature, efficiency and light off. Interestingly, the
emulsion of nitrogen gases was slightly lower suggesting that in the presence of
water there may be less NOx emission. The results of this large-scale field trial
[207-210, 213] have paved the way for a potential application in which sludge
from storage tanks can be emulsified, and the emulsion (i.e. the emulsanosol)
can be recycled as a source of energy. The system allows for the upgrading for
otherwise unusable materials such as vacuum residuals to be recovered as a
source of energy offsetting the cost of the tank cleaning. In addition, the fact that
the oil phase consists of small droplets suggests that the burn efficiency may be
273
somewhat higher than with a regular hydrocarbon fuel since the surface area is
larger. What is even more interesting is that the quality of the burn was
indistinguishable from that of a light high quality fuel oil suggesting that
emulsion based fuels can be a viable alternative for some applications [207-
210,213].
Interestingly, these larger scale burn experiments were performed with a
biosurfactant containing formulation. However, at Petroferm U.S.A. specialty
chemical formulations were developed some of which did not contain the
emulsan, but was composed of chemical surfactants, a solvent and other
components. Emulsion based fuels have become more and more popular,
because they permit the efficient combustion of various hydrocarbons which are
normally difficult to burn. Arguably, the best studied system is the water-
bitumen emulsion system, termed Oriemulsion which has been commercialized
and is currently exported from Venezuela throughout the world. The emulsion is
chemical based, and resembles the initial Petroferm formulations.
4.3. Viscosity reduction and oil transportation

As discussed above, the stability of emulsan-based oil-in-water emulsions
results from the coating of the oil droplets with emulsan in an oriented
conformation; hydrophobic moieties coming into contact with the oil surface,
and the hydrophilic components oriented towards the aqueous phase. This
results in a homogeneous suspension in which the viscosity of the oil component
is significantly reduced at room temperature. The extent of the viscosity
reduction is a function of the water composition of the bulk phase, which for an
emulsanosol can be as high as 30%. Under these conditions the viscosity of a
high viscosity oil from the Orinoco basin in Venezuela, (Boscan crude) was
reduced from >50,000 Cp to about 85 Cp in the form of an emulsanosol
generated with an emulsan based surfactant package [207-210, 213] at a ratio of
1 part surfactant to 500 parts oil. About fifty barrels of this emulsion was
transported through an experimental pipeline of 1.25 inches for 96 h during
which the mixture was subjected to over 500 pump transits. There was no effect
on the low viscosity, although the shear forces on the emulsion might have been
expected to produce an inversion from an oil-in-water to a water-in-oil
emulsion. Moreover, even after the system was shut down and the emulsion
allowed to stand undisturbed for 48 h, there was no breakage or inversion of the
emulsion and the low viscosity was maintained through the pipeline for an
additional 48 h. The results support the use of surfactant packages to generate
oil-in-water emulsions for pipelining highly viscous oils. In fact, this is the basis
of the Orimulsion technology.
274
There is little doubt that biosurfactants and bioemulsifiers exhibit characteristics

and activities, which are applicable in the oil industry. As noted, in some cases
product efficacy has been tested in large scale and successful trials have been
recorded. However, the impact of such materials on the oil industry is likely to
be far less than in other industrial sectors, where the price of the final product is
high relative to the costs of production. Ongoing efforts to isolate new
biosurfactants, genetically modify existing ones, enhance productivity of the
producing organism and otherwise lower the cost of production, and formulate
new and more effective biosurfactant packages should yield a host of products
and applications in the future.
Moreover, for some applications such as enhanced oil recovery or
bioremediation, the biotechnology associated with in-situ biosurfactant
production accompanying the growth of microorganisms, discussed elsewhere in
this book, can be a useful and economically competitive strategy. Finally, the
incorporation of biosurfactants in novel formulations suitable for sludge
liquification and viscosity reduction leading to economically feasible techniques
for waste recovery and recycling.
REFERENCES
[1] C.A. Houston, (2002) Colin A. Houston & Associates, Inc, NY.
[2] J.L. Salanger, Surfactants-Types and Uses, Laboratorio FIRP, Merida, Venezuela,
(1999).
[3] R. Modler and Y. Ishikawa, Surfactants, Specialty Chemicals Update Program - SRI
Consulting: CA. (2001).
[4] G. Bognolo, Sustainable Surfactants, Renewable Feedstocks for the 21 st Century,
Competitive Industrial Materials from Non-Food Crops, Sand Hutton, York, UK,
(1998).
[5] I.W. Sutherland, Biotechnol. Adv., 12 (1994) 393.
[6] D.L. Gutnick and W. Minas, Biochem. Soc. Trans., 15(1987) Suppl. 22S.
[7] E. Rosenberg and E.Z. Ron, Appl. Microbiol. Biotechnol., 52 (1999) 154.
[8] Y. Shabtai, O. Pines and D.L. Gutnick, Develop. Indust. Microbiol., 26 (1985) 291.
[9] J.D. Desai and I.M. Banat, Microbiol. Molec. Biol. Rev., 61 (1997) 47.
[10] C. Rubinovitz, D.L. Gutnick and E. Rosenberg, J. Bacteriol., 152 (1982) 126.
[11 ] H.J.A. Asmer, S. Lang, F. Wagner and V. Wray, J. Am. Oil Chem. Soc, 65 (1988)
1460. [12] A.A. Bodour, K.P. Drees and R.M. Maier, Appl. Environ. Microbiol., 69
(2003) 3280.
[13] A. A. Bodour and R.M. Miller-Maier, Biosurfactants: types, screening methods, and
applications., Encyclopedia of Environmental Microbiology, Wiley, NY, 2002, pp.75O-
770.
[14] N.H. Youssef, K.E. Duncan, D.P. Nagle, K.N. Savage, R.M. Knapp and M.J.
Mcmerney, J. Microbiol. Meth., 56 (2004) 339.
275
[15] A. Fiechter, Trends in Biotechnol., 10 (1992) 208.

[16] I.M. Banat, R.S. Makkar and S.S. Cameotra, Appl. Microbiol. Biotechnol., 53 (2000)
495.
[17] K. Arima, A. Kakinuma and G. Tamura, Biochem. Biophys. Res. Commun., 31 (1968)
488.
[18] T. Suzuki, K. Hayashi, K. Fujikawa, and K. Tsukamoto, J. Biol. Chem., 57 (1965) 226-
227.
[19] D. Sorensen, Cyclic Lipopeptides from Pseudomonas spp., The Marine Chemistry
Section, Department of Chemistry, Faculty of Science, University of Copenhagen,
Copenhagen, 2002, pp. 1-61.
[20] MA. Marahiel, T. Stachelhaus and H.D. Mootz, Chem. Rev., 97 (1997) 2651.
[21] S. Lang and D. Wullbrandt, Appl. Microbiol. Biotechnol., 51 (1999) 22.
[22] T. Suzuki, K. Tanaka, J. Matsubara and S. Kmoshita, Agric. Biol. Chem., 33 (1969)
1619.
[23] F. Gibson and D.I. Magrath, Biochim. Biophys. Acta, (1969) 65.
[24] A.A. Bodour, C. Guerrerobarajas, B.V. Jiorle, ME. Malcomson, A.K. Paull, A.
Somogyi, L.N. Trinh, R.B. Bates and R.M. Maier, Appl. Environ. Microbiol., 70 (2004)
114.
[25] A. Kretschmer, H. Bock and F. Wagner, Appl. Environ. Microbiol., 44 (1982) 864.
[26] K. Brandenburg, B. Lindner, A. Schromm, M.H.J. Koch, J. Bauer, A. Merkli, C.
Zbaeren, J. Gwynfor Davies and U. Seydel, Eur. J. Biochem., 267 (2000) 3370.
[27] J. Kim, M. Powalla, S. Lang, F. Wagner, H. Lunsdorf and V. Wray, J. Biotechnol., 13
(1990)257.
[28] S. Itoh and S. Inoue, Appl. Environ. Microbiol., 43 (1982) 1278.
[29] P. Rapp and H.E. Gabriel-Jurgens, Microbiology, 149 (2003) 2879.
[30] http://www.cyberlipid.org.
[31] D.G. Cooper and D A Paddock, Appl. Environ. Microbiol., 46 (1983) 1426.
[32] D. Sorensen, T.H. Nielsen, C. Christophersen, J. Sorensen and M. Gajhede, Acta Crys.,
C57 (2001) 1123.
[33] M. Morikawa, H. Daido, T. Takao, S. Murata, Y. Shimonishi and T. Imanaka, J.
Bacteriol., 175(1993)6459.
[34] J.D. Epperson and L. Ming, Biochemistry, 39 (2000) 4037.
[35] M. Moormann, U. Zahringer, H. Moll, R. Kaufinann, R. Schmid and K. Altendorf, J.
Biol. Chem., 272 (1997) 10729.
[36] M. Doi, S. Fujita, Y. Katsuya, M. Sasaki, T. Taniguchi and H. Hasegawa, Arch.
Biochem. Biophys., 395 (2001) 85.
[37] J.A. Tnschman, PR. Jensen and W. Fenical, Tetrahedron Lett., 35 (1994) 5571
[38] MM. Yakimov, K.N. Timmis, V. Wray and H.L. Frederickson, Appl. Environ.
Microbiol, 61 (1995)1706.
[39] M. Fridkin, I. Ofek, H. Tsubery and S. Cohen, http://www.weizmann.ac.il.
[40] H. Ui, T. Miyake, H. Iinuma, M. Imoto, H. Naganawa, S. Hatton, M. Hamada, T.
Takehuchi, S. Umezawaand K. Umezawa, J. Org. Chem., 62 (1997) 103.
[41] I. Kuiper, L. Lagendijk, R. Pickford, J.P. Derrick, EM. Lamers, J.E. Thomas-Oates, JJ.
Lugtenberg and G.V. Bloemberg, Molec. Microbiol., 51 (2004) 97.
[42] P.W. Lindum, U. Anthoni, C. Christophersen, L. Eberl, S. Molin and M. Givskov, J.
Bacteriol., 180(1998)6384.
[43] M. Kowall, J. Vater, B. Kluge, T. Stein, P. Franke and D. Ziessow, J. Colloid Interface
Sci., 204 (1998)1.
276
[44] A. Segre, R.C. Bachmann, A. Ballio, F. Bossa, I. Grgurina, N.S. Iacobellis, G. Marino,
P. Pucci, M. Simmaco and J.Y. Takemoto, FEBS Lett., 255 (1989) 27.
[45] T.H. Nielsen, C. Thrane, C. Christophersen, U. Anthoni, and J. Sorensen, J. Appl.
Microbiol., 89 (2000) 992.
[46] T.R. Burke, M. Knight and B. Chandrasekhar, Tetrahedron Lett, 30 (1989) 519.
[47] T.H. Nielsen, C. Christophersen, U. Anthoni, and J. Sorensen, J. Appl. Microbiol., 86
(1999)80.
[48] R.J. Mortishire-Smith, J.C. Nutkins, L.C. Packman, C.L. Brodey, P.B. Rainey, K.
Johnstone and D.H. Williams, Tetrahedron, 47 (1991) 3645.
[49] J.E. Zajic and W. Seffens, CRC Critical Rev. Biotechnol., 1 (1984) 87.
[50] P.G. Carnllo, C. Mardaraz, S.J. Pitta-Alvarez and A.M. Giuletti, World J. Microbiol.
Biotechnol., 12(1996)82.
[51] I.M. Banat, Biotechnol. Lett, 15 (1993) 591.
[52] H. Yonebayashi, S. Yoshida, K. Ono and H. Enomoto, Sekiyu Gakkaishi, 43 (2000) 59.
[53] C.N. Mulligan, D.G. Cooper and R.J. Neufeld, J. Ferment. Technol., 62 (1984) 311.
[54] D.K. Jain, D.L. Collins-Thompson, H. Lee and J.T. Trevors, J. Microbiol. Methods., 13
(1991)271.
[55] A.A. Bodour and R.M. Miller-Maier, J. Microbiol. Meth., 32 (1998) 273.
[56] M. Morikawa, Y. Hirata and T. Imanaka, Biochim. Biophys. Acta, 1488 (2000) 211.
[57] E.Z. Ron and E. Rosenberg, Environ. Microbiol., 3 (2001) 229.
[58] C. Fuqua, MR. Parsek and E.P. Greenberg, Ann. Rev. Gen., 35 (2001) 439.
[59] D.G. Davies, MR. Parsek, J.P. Pearson, B.H. Iglewski, J.W. Costerton and E.P.
Greenberg, Science, 280 (1998) 295.
[60] G. Medina, K. Juarez, B. Valderrama and G. Soberon_Chavez, J. Bactenol., 185 (2003)
5976.
[61] L. Eberl, S. Molin and M. Givskov, J. Bacteriol., 181 (1999) 1703.
[62] E. Dickinson (ed.), Food Emulsions and Foams, The Royal Society of Chemistry,
London, 1987, pp. 1-29.
[63] G.O. Philips, D.J. Wedlock and P A Williams (eds), Gums and Stabilizers for the Food
Industry, IRL Press, Oxford, 1988, pp. 249-263.
[64] A. Stephen (ed.), Food Polysaccharides and their Applications, Marcel Dekker, N.Y.
1995, pp. 501-515.
[65] A. Reisfeld, E. Rosenberg and D.L. Gutnick, Appl. Microbiol., 24 (1972) 363.
[66] A. Oberbremer, R. Miller-Hurting and F. Wagner, Appl. Microbiol. Biotechnol., 32
(1990)485.
[67] Y. Zhang and R.M. Miller, Appl. Environ. Microbiol., 58 (1992) 3276.
[68] M.F. Van Dyke, P. Couture, M. Brauer, H. Lee and J.T. Trevors, Can. J. Microbiol., 39
(1993)1071.
[69] M.A. Providenti, C.A. Flemming, H. Lee and J.T. Trevors, FEMS Microb. Ecol.,17
(1995) 15.
[70] M. Marin, A. Pedregosa and F. Laborda, Appl. Microbiol. Biotechnol., 44 (1996) 660.
[71] D.G. Cooper and B.G. Goldenberg, Appl. Environ. Microbiol., 53 (1987) 224.
[72] C. Calvo, F. Martinez-Checa, F.L. Toledo, J. Porcel and E. Quesada, Appl. Microbiol.
Biotechnol., 60 (2002) 347.
[73] S. Navon-Venezia, Z. Zosim, A. Gottlieb, R. Legmann, S. Carmeli, E.Z. Ron and E.
Rosenberg, Appl. Environ. Microbiol., 61 (1995) 3240.
[74] E. Rosenberg, C. Rubinobitz, R. Legmann and E.Z. Ron, Appl. Environ. Microbiol., 54
(1988)323.
277
[75] A. Zuckerberg, A. Diver, Z. Peeri, D.L. Gutnick and E. Rosenberg, Appl. Environ.
Microbiol.,37(1979)414.
[76] M.C. Cirigliano and G.M. Carman, Appl. Environ. Microbiol., 50 (1985) 846.
[77] G. Whyte, S.J. Slagman, F. Pietrantonio, L. Bourbonniere, S.F. Koval, J.R. Lawrence,
W.E. Inniss and C.W. Greer, Appl. Environ. Microbiol., 65 (1999) 2961.
[78] S.J. Hamza, A.S. Abbas and A.A. Halob, J. Islamic Academy Sci., 7 (1994) 1.
[79] S. Lin, M.A. Minton, M.M. Sharma and G. Georgiou, Appl. Environ. Microbiol., 60
(1994)31.
[80] T.H. Nielsen, D. Sorensen, C. Tobiasen, J.B. Andersen, C. Christophersen, M. Givskov
and J. Sorensen, Appl. Environ. Microbiol., 68 (2002) 3416.
[81] A.J. Cutler and R.J. Light, J. Biol. Chem., 254 (1979) 1944.
[82] D.R. Cameron, D.G. Cooper and R.J. Neufeld, Appl. Environ. Microbiol., 54 (1988)
1420.
[83] M. Shimazu, M. Saito and K. Yamauchi, Agric. Biol. Chem., 49 (1985) 189.
[84] M. Hatton and K. Takahashi, Food Hydrocol., 7 (1993) 325.
[85] C.W. Cumperand AE. Alexander, Trans. Faraday Soc, 46 (1950) 235.
[86] E. Dickinson, B.S. Murray, G. Stamsby and D.M.W. Anderson, Food Hydrocol., 2
(1988)477.
[87] M. Shimazu, T. Kamiya and K. Yamauchi, Agric. Biol. Chem., 45 (1981) 2491.
[88] IE. Kinsella, Cnt Rev. Food Sci. Nutr, 21 (1984) 197.
[89] R.D. Wamska and J.E. Kinsella, J. Agric. Food Chem., 33 (1985) 1143.
[90] E. Dickinson and M. Lai, Adv. Molec. Rel. Intern. Processes, 17 (1980) 1.
[91] J.A. Defeyter and J. Benjamins, J. Colloid Interf. Sci., 9 (1982) 289.
[92] E. Dickinson and S.E. Euston, Molec. Phys., 68 (1989) 407.
[93] E. Dickinson and S.E. Euston, J. Chem. Soc. Faraday Trans., 86 (1990) 805.
[94] E. Dickinson (ed), Food Polymers, Gels and Colloids, The Royal Society of Chemistry,
Cambridge, 1991, pp. 113-122.
[95] G.O. Philips, D.J. Wedlock and P. A. Williams (eds.), Gums and Stabilizers for the Food
Industry, Oxford University Press, Oxford, 1992, pp. 351-362.
[96] N.S. Hettiarachchy and G.R. Ziegler (eds.), Protein, Functionality in Food System,
Marcel Dekker, N.Y., 1994, pp. 225-259.
[97] K. Watanabe, H. Toyokawa, A. Shimada and S. Arai, J. Food Sci., 46 (1981) 1467.
[98] M. Saito, M. Ogasawara, K. Chikuni and M. Shimizu, Biosci. Biotechnol. Biochem., 59
(1995)388.
[99] Z. Mozaffar and Z.U. Haque, Food Hydrocol., 5 (1994) 573.
[100] S.W. Lee, M. Shimizu, S. Kaminogawa and K. Yamauchi, Agric. Biol. Chem., 51
(1987) 161.
[101] M. Saito, K. Chikuni, M. Monma and M. Shimizu, Biosci. Biotechnol. Biochem., 57
(1993)952.
[102] M. Watanabe, N. Fujn and S. Arai, Agric. Biol. Chem., 46 (1982) 1587.
[103] E. Dickinson and D. Lorient (eds.), Food Macromolecules and Colloids, The Royal
Society of Chemistry, Cambridge, 1995 pp. 1-19.
[104] G. Williamson, C.B. Faulds, J.A. Matthews, V.J. Morris and GJ. Brownsey, Carbohyd.
Polymers, 13(1990)387.
[105] J.R. Mitchell and D.A. Ledward (eds), Functional properties of food macromolecules,
Elsevier Applied Science, London, 1986, pp. 315-3 54.
[106] Y.A. Antonov, N.P. Lashko, Y.K. Glotova, A. Malovikova and O. Markovich, Food
Hydrocol., 10(1996)1.
[107] V.B. Tolstoguzov and E.E. Braudo, J. Disp. Sci. Technol., 6 (1985) 575.
278
[108] K.D. Schwenke and R. Mothes (eds.), Food Proteins. Structure and Functionality, VCH,
Weinheim, 1993, pp. 203-209.
[109] G.O. Philips, P. A. Williams and D.J. Wedlock (eds.), Gums and Stabilisers for the Food
[110] V. Grinberg and V.B. Tolstoguzov, Food Hydrocol., 11 (1997) 145.
[Ill] N. Garti, Y. Slavin and A. Aserin, Food Hydrocol., 13 (1998) 145.
[112] N. Garti and M.E. Leser, Polym. Adv. Technol, 12 (2001) 123.
[113] W. Liu and T. Sato, Carbohydr. Research, 160(1987)267.
[114] J. Greener, B.A. Contestable and M.D. Bale, Macromolecules, 20 (1987) 2490.
[115] E. Dickinson (ed.), Food Emulsions and Foams, The Royal Society of Chemistry,
London, 1987, pp. 1-29.
[116] G.O. Philips, P.A. Williams and D.J. Wedlock (eds.) Gums and Stabilizers for the Food
[117] G.L. Hasenhuettle and R.W. Hartel (eds.), Food Emulsifiers and their Applications,
Thomson Publishing, 1997.
[118] S.E. Hill, A.P. Ledward and J.R. Mitchell (eds.), Functional Properties of Food
Macromolecules, Aspen Publisher, Maryland, 1998, pp. 252-277.
[119] Y. Cao, E. Dickinson and D.J. Wedlock, Food Hydrocol., 5 (1991) 443.
[120] E. Dickinson, V.B. Galazka and D.M.W. Anderson, Carbohyd. Polymers, 14 (1991)
373.
[121] E. Dickinson and P. Walstra (eds.) Food colloids and polymers: stability and
mechanical properties, The Royal Society of Chemistry, Cambridge, 1993, pp. 3-15.
[122] V. Ya, V. Grinberg and V.B. Tolstoguzov, Food Hydrocol., 11 (1997) 145.
[123] A. Kato, T. Sato and K. Kobayashi, Agric. Biol. Chem., 53 (1989) 2147.
[124] A. Kato, Y. Sasaki, R. Furuta and K. Kobayashi, Agric. Biol. Chem., 54 (1990) 107.
[125] S. Nakamura, A. Kato and K. Kobayashi, J. Agric. Food Chem., 40 (1992) 735.
[126] M. Hattori, S. Imamura, K. Nagasawa and K. Takahashi, Biosci. Biotech. Biochem., 58
(1994) 174.
[127] M. Hattori, S. Imamura, K. Nagasawa and K. Takahashi, J. Agric. Food Chem., 42
(1994)2120.
[128] E. Dickinson and V.B. Galazka, Food Hydrocol., 5 (1995) 281.
[129] K. Nagasawa, K. Takahashi and M. Hattori, Food Hydrocol., 10 (1996) 63.
[130] J.M.V. Blanshard and JR. Mitchell (eds.) Polysaccharides in Food, Butterworths,
London, 1979, pp. 205-217.
[131] J.R. Mitchell and D.A. Ledward (eds.), Functional Properties of Food Macromolecules,
Elsevier Applied Science, London, 1986, pp. 385-415.
[132] M. Ahmed and E. Dickinson, Colloid Surface, 47 (1990) 353.
[133] N. Garti and D. Reichman, Food Hydrocol., 8 (1994) 155.
[134] R.C. Randall, G.O. Philips and P.A. Williams, Food Hydrocol., 2 (1988) 131.
[135] J.E. Mckay, G. Stainsby and E.L. Wilson, Carbohyd. Polymers, 5 (1985) 223.
[136] K. Nishinari and E. Doi (eds.) Food Hydrocolloid-structures, properties and functions,.
Plenum Press, N.Y., 1982, pp. 211-224.
[137] Y.A. Antonov, J. Lefevre and J. Doublier, Appl. Polymer Sci., 71 (1999) 471.
[138] P. Harris (ed.), Food Gels, Elsevier Applied Science, London, 1990, pp. 291-359.
[139] H. Bach, Y. Berdichevsky and D. Gutnick, Appl. Environ. Microbiol., 69 (2003) 2608.
[140] D.L. Gutnick and H. Bach. US Patent No. 6 512 014 (2003).
[141] S. Ohashi, F. Ura, M. Takeuchi, H. Lida, K. Sakaue, T. Ochi, S. Ukai and K. Hiramatsu,
Food Hydrocol., 4 (1990) 323.
[142] E. Dickinson and K. Pawlowsky, Food Hydrocol., 12 (1998) 417.
279
[143] K. Maenaka, G. Kawai, K. Watanabe, F. Sunada and I. Kumagai, J. Biol. Chem., 269
(1994)7070.
[144] J.W. Goodwin (ED), Colloidal Dispersions, The Royal Society of Chemistry, London,
1982, pp. 99-128.
[145] M.A. Muchkin, E.S. Wajnermann and V.B. Tolstoguzov, Nahrung, 20 (1976) 313.
[146] R.G. Chilvers and V. J. Morris, Carbohyd. Polymers, 7 (1987) 111.
[147] H.R Kruyt (ed.), Colloid Science: reversible systems, Elsevier, Amsterdam, 1949,pp.
232-258.
[148] D.A. Imeson, A.P. Ledward and JR. Mitchell, J. Sci. Food Agnc, 28 (1977) 661.
[149] E. Rosenberg, A. Zuckerberg, C. Rubinovitz and D.L. Gutnick, Appl. Environ.
Microbiol.,37(1979)402.
[150] E. Rosenberg, A. Perry, D.T. Gibson and D.L. Gutnick, Appl. Environ. Microbiol., 37
(1979)409.
[151] D. Gutnick, Biopolymers, 26 (1987) S223.
[152] N. Kosaric (ed), Biosurfactants and Biotechnology, Marcel Dekker, N.Y., 1987, pp.
211-246.
[153] K.J. Towner, E. Bergogne-Berezin and C. A. Fewson (eds.) The Biology of
Acinetobacter, Plenum Press, N.Y., 1991, pp. 411-441.
[154] D.L. Gutnick, E. Rosenberg and Y. Shabtai. US Patent No. 4 234 689 (1980).
[155] D.L. Gutnick, E. Rosenberg, I. Belsky and Z Zosim. US Patent No. 4 311 830 (1982).
[156] D.L. Gutnick, E. Rosenberg, I. Belsky and Z. Zosim. US Patent No. 4 395 3 54 (1983).
[157] N. Kaplan, E. Rosenberg, B. Jann and K. Jann, Eur. J. Biochem., 152 (1985) 453.
[158] D.L. Gutnick, Personal Communication, (2004).
[159] C.R. Macdonald, D.G. Cooper and J.E. Zajic, Appl. Environ. Microbiol., 41 (1981) 117.
[160] D.G. Cooper, J.E. Zajic and D.F. Gerson, Appl. Environ. Microbiol., 37 (1979) 4.
[161] A. Gorkovenko, J. Zhang, R.A. Gross, A.L. Allen and D.L. Kaplan, Can. J. Microbiol.,
43(1997)384.
[162] A. Gorkovenko, J. Zhang, R.A. Gross, D.L. Kaplan and A.L. Allen, Carbohydr. Polym.,
39(1999)79.
[163] A.K. Johri, W. Blank and D.L. Kaplan, Appl. Microbiol. Biotechnol., 59 (2002) 217.
[164] J. Zhang, A. Gorkovenko, R.A. Gross, A.L. Allen and D. Kaplan, Int. J. Biol.
Macromol., 20 (1997) 9.
[165] I. Belsky, D.L. Gutnick and E. Rosenberg, FEBS Lett, 101 (1979) 175.
[166] Y. Shabtai and D.L. Gutnick, Appl. Environ. Microbiol., 49 (1985) 192.
[167] B.K. Tuleva, G.R Ivanov and N.E. Chnstova, Naturforsch, 57C (2002) 356.
[168] A.M. Davila, R. Marchal and J.P. Vandecasteele, Appl. Microbiol. Biotechnol., 47
(1997)496.
[169] Z. Zosim, D.L. Gutnick and E. Rosenberg, Biotechnol. Bioeng., 25 (1983) 1725.
[170] D.L. Gutnick and H. Bach, Appl. Microbiol. Biotechnol., 54 (2000) 451.
[171] S. Goldman, Y. Shabtai, C. Rubinobitz, E. Rosenberg and D. Gutnick, Appl. Environ.
Microbiol., 44(1982)165.
[172] Y. Shabtai and D.L. Gutnick, J Bacteriol., 161 (1985)1176.
[173] S.D. Wang and D.I.C. Wang, Biotechnol. Bioeng., 36 (1990) 402.
[174] O. Pines and D. Gutnick, J. Bacteriol., 157 (1984) 179.
[175] O. Pines, Y. Shoham, E. Rosenberg and D.L. Gutnick, Appl. Microbiol. Biotechnol., 28
(1988)93.
[176] Y. Shoham, M. Rosenberg and E. Rosenberg, Appl. Environ. Microbiol, 46 (1983) 573.
[177] M. Rosenberg and E. Rosenberg, J. Bacteriol., 148 (1981) 51.
280
[178] M. Rosenberg, A. Perry, E.A. Bayer, D.L. Gutnick, E. Rosenberg and I. Ofek, Infect
Immun., 33(1981)29.
[179] T. Ophir and D. Gutnick, Appl. Environ. Microbiol., 60 (1994) 740.
[180] D. Nakarand D.L. Gutnick, Microbiology, 147 (2001) 1937.
[181] D. Nakarand D.L. Gutnick, J. Bactenol, 185 (2003) 1001.
[182] Mijakovic, S. Poncet, G. Boel, A. Maze, S. Gillet, E. Jamet, P. Decottigmes, C.
Grangeasse, P. Doublet, P. LeJVlarechal and J. Deutscher, Embo J., 22 (2003) 4709.
[183] Y. Shoham, E. Rosenberg and D.L. Gutnick, US Patent No. 4 704 360 (1983).
[184] P.G. Reddy, R. Allon, M. Mevarech, S. Mendelovitz, Y. Sato and D.L. Gutnick, Gene,
76(1989)145.
[185] R. Alon, Esterase from the oil-degrading Acinetobacter lwoffii RAG-1: expression of
the est gene and protein characterization. Ph.D. Thesis, Tel-Aviv University (1993).
[186] R.N. Alon and D.L. Gutnick, FEMS Microbiol. Lett, 112 (1993) 275.
[187] R.N. Alon, L. Mirny, J.L. Sussman and D.L. Gutnick, FEBS Lett, 371 (1995) 231.
[188] R.G. Kok, J.J. Van Thor, I.M. NugterenJRoodzant, MB. Brouwer, M.R. Egmond, C.B.
Nudel, B. Vosman and K.J. Hellingwerf, Mol. Microbiol., 15 (1995) 803.
[189] H. Bach, Y. Mazor, S. Shaky, A. Shoham Lev, Y. Berdichevsky, D.L. Gutnick and I.
Benhar, J. Mol. Biol., 312 (2001) 79.
[190] N. Sar and E. Rosenberg, Curr. Microbiol., 9 (1983) 309.
[191] A. Toren, E. Orr, Y. Paitan, E.Z. Ron and E. Rosenberg, J. Bactenol., 184 (2002) 165
[192] A. Toren, E.Z. Ron, R. Bekerman and E. Rosenberg, Appl. Microbiol. Biotechnol, 59
(2002) 580.
[193] E. Pessione, M.G. Giuffrida, L. Prunotto, C. Barello, R. Mazzoli, D. Fortunate, A. Conti
and C. Giunta, 3 (2003) 1070.
[194] E. Rosenberg, C. Rubinobitz, A. Gottlieb, S. Rosenhak and E.Z. Ron, Appl. Environ.
Microbiol., 54 (1988) 317.
[195] W.H. Taylor and E. Jum, J. Bacteriol., 81 (1961) 688.
[196] N. Kaplan, Z. Zosim and E. Rosenberg, Appl. Environ. Microbiol., 53 (1987) 440.
[197] O. Abian, C. Mateo, G. FernandezLorente, J.M. Guisan and R. Fernandez_Lafuente,
Biotechnol. Progr., 19 (2003) 1639.
[198] F. Arendt, M. Hinsenveld, and W.J. van den Brink (eds), Contaminated soil '90,
Kluwer Academic Publishers, Dordrecht, The Netherlands, 1990, pp. 491-492.
[199] C.N. Mulligan, T.Y.K. Chowand B.F. Gibbs, Appl. Microbiol. Biotechnol, 31 (1989)
486.
[200] GL. Ghurye, C. Vipulanandan and R.C. Willson, Biotechnol. Bioeng., 44 (1994) 661.
[201] M.E. Mercade, M.A. Manresa, M. Robert, M.J. Espuny, C. De Andres and J. Guinea,
Bioresource Technol., 43 (1993) 1.
[202] M.E. Mercade and M.A. Manresa, J. Am. Oil Chem. Soc, 71 (1994)61.
[203] H.J. Daniel, R.T Otto, M. Binder, M. Reuss, and C. Syldatk, Appl. Microbiol.
Biotechnol., 51 (1999)40.
[204] U.A. Ochsner, A. Fiechter and J. Reiser, J. Biol. Chem., 269 (1994) 19787.
[205] U.A. Ochsner, J. Reiser, A. Fiechter and B. Witholt, Appl. Environ. Microbiol., 61
(1995)3503.
[206] W.C. Mccaffrey and D.G. Cooper, J. Ferment. Bioeng., 79 (1995) 146.
[207] M.E. Hayes, K.R. Hrebenar, P.L. Murphy, L.E. Futch Jr., J.F. Deal III and P.L. Bolden
Jr. US Patent No. 4 684 372 (1987).
[208] M.E. Hayes, K.R. Hrebenar, P.L. Murphy, L.E. Futch Jr., J.F. Deal m and P.L. Bolden
Jr. US Patent No. 4 793 826 (1988).
281
[209] ME. Hayes, K.R Hrebenar, PL Murphy, L.E. Futch Jr., J.F. Deal in and PL. Bolden
Jr.US Patent No. 4 821 757 (1989).
[210] ME. Hayes, K.R. Hrebenar, J.L. Minor and L.M. Woodworth, US Patent No. 4 886 519
(1989)
[211] D.L. Gutnick and E. Rosenberg, Ann. Rev. Microbiol, 31 (1977) 379.
[212] I.M. Banat, N. Samarah, M. Murad, R. Home and S. Benerjee, World J. Microbiol.
Biotechnol, 7(1991)80.
[213]M.E. Hayes, K.R. Hrebenar, J.F. Deal in and PL. Bolden Jr. US Patent No. 4 666 457
(1987).
Chapter 10
Anaerobic hydrocarbon biodegradation and the prospects

for microbial enhanced energy production
J.M. Suflitaab, LA. Davidovaab, L.M. Giegab, M. Nanny ac and R.C. Princed
a
Institute for Energy and the Environment, bDepartment of Botany and
Microbiology, cSchool of Civil Engineering and Environmental Science,
d
ExxonMobil Research and Engineering Co., Annandale, NJ 08801, USA.
1. INTRODUCTION
Hydrocarbon-based energy underpins the economic, social, and political fabric

of the world and demand for oil is expected to grow unabated for the foreseeable
future. It is forecast that global oil consumption will increase annually by an
average of more than 4 x 107 barrels per day to eventually reach 4.3 x 1010
barrels per year by 2020 [1]. This is projected to be about a 58% increase over
current usage by 2025 [2]. The U.S. Geological Survey recently predicted that
about 3 trillion barrels of oil remain to be recovered worldwide (with 1 trillion
barrels already harvested), half from proven reserves and half from undeveloped
or undiscovered sources [3]. However, as proven reserves get exploited and
develop into mature fields, secondary and tertiary recovery technologies will
increasingly be relied upon to obtain the remaining residual oil. This is
particularly true for the U.S. and other oil-importing countries that tend to rely
more heavily on mature, domestic energy sources. With demand far surpassing
energy production in the U.S., there is heightened interest in diversifying energy
sources, tapping unconventional energy supplies and the development of new
technology to more fully exploit domestic reserves.
Although oil is expected to remain the dominant energy fuel in the next
20 years, the use of natural gas as a substantial energy source has risen
significantly in the past 10 years [4]. In fact, natural gas is projected to be the
fastest growing primary energy source and an increasingly important alternative
to oil [2]. Natural gas, consisting mainly of methane (>95%) but also with small
amounts of other short-chain hydrocarbons (C2 to C4), can be harvested from
284
large gas fields, sometimes associated with oil reservoirs, or be obtained from
unconventional sources such as shale, coalbeds, or tight sands. The use of
natural gas is becoming increasingly popular due to its abundance across the
globe. Further, the lower price of natural gas relative to oil makes it an attractive
energy source [5]. Natural gas is also a cleaner energy source than oil or coal,
and thus can help reduce greenhouse gas emissions [6]. Natural gas combustion
produces only about 56% and 71% of the CO2 associated with the equivalent
amount of energy produced from coal or oil, respectively [Energy Information
Administration (1999). Natural Gas 1998: Issues and Trends (http://www.eia.
doe.gov/oil_gas/natural_gas/analysis_publications/natural_gas_1998_issues_and
_trends/it98.html). Moreover, methane use results in less NOX, SO2, and
particulates per equivalent amount of energy generated, relative to other sources.
Despite the increasing use of natural gas and its attendant environmental
advantages, world reliance on oil is unlikely to wane in the near future given the
existing energy infrastructure and the aforementioned dependence of many
societies on this energy form. However, there is a biotechnological link between
oil and natural gas that is the product of the relatively recent recognition that
many hydrocarbons are susceptible to anaerobic biodegradation and can be
converted to methane and carbon dioxide [7-9]. Unlike the well-documented
patterns of aerobic oil biodegradation [10], anaerobic hydrocarbon metabolism
was essentially dismissed as ecologically insignificant for many years. This
view has been completely altered in recent years with the growing appreciation
for the metabolism of hydrocarbons coupled with the consumption of electron
acceptors other than oxygen.
Not surprisingly then, the majority of world oil reserves are believed to be
biodegraded to at least some degree, but it was generally accepted that aerobic
oxidation processes were largely responsible for such alterations [11, 12].
Recent evaluations of many petroliferous formations have convincingly argued
that it is actually anaerobic processes that predominate in oil and gas reservoirs,
sometimes leading to the production of biogenic methane [12-14]. Geological
evidence has suggested that such methanogenic processes occur very slowly
over millennia, and are most important in reservoirs shallower than 4 km and at
temperatures of less than 80°C [12, 15, 16]. Microbial decay of oils in deep
subsurface reservoirs can clearly reduce oil quality, and a better understanding
of the microbial principles behind such decay will be important to help
distinguish between degraded, low-value oils and untouched, high-value oils
[11]. However, if methanogenesis continues to be identified as an important
process in deep reservoirs worldwide, the recovery of methane gas as an
alternate form of energy from otherwise unrecoverable or biodegraded sources
might have far-reaching economic and environmental implications.
The purpose of this chapter is to review evidence for anaerobic
hydrocarbon biodegradation and to provide an overview of some of the more
285
generalizing metabolic features. We will explore whether these reactions can be

predicted and identify some of the implications for the ability of anaerobes to
convert hydrocarbons into methane and and thereby generate useful energy.
2. ANAEROBIC HYDROGEN METABOLISM

Crude oils are enormously complex mixtures containing tens of thousands of
individual components [17]. Even condensates and refined products include a
dizzying array of constituent hydrocarbons. Once released in the environment,
the relative concentrations of these chemicals change over time reflecting
individual susceptibilities to various fate processes such as sorption,
volatilization, dispersion and biodegradation [18]. One way of assessing the
susceptibility of such complex mixtures to anaerobic biodegradation is to
consider individual chemical classes of hydrocarbons and determine how
metabolism varies with structural complexity (see below).
Two decades ago, it was generally considered that aliphatic and aromatic
hydrocarbons could only be mineralized in the presence of oxygen. Oxygen
served as both a respiratory electron acceptor and as a co-substrate for mono- and
dioxygenases catalyzing initial hydrocarbon activation steps [19, 20]. We now
know that microbial metabolism is much more diverse than this and many classes
of hydrocarbons are amenable to microbial attack under a variety of anaerobic
conditions. Substrates include BTEX (benzene, toluene, ethylbenzene and
xylenes) compounds [21], polycyclic aromatic hydrocarbons [22], saturated and
branched alkanes [23, 24] and alicyclic hydrocarbons [25, 26]. Knowledge of the
mechanisms used by anaerobes to catalyze such transformations are summarized
here.
Toluene has historically served as a model substrate to study anaerobic
alkylbenzene biodegradation. Initial work with denitrifying strains of Thauera
and Azoarcus [27, 28] showed that the first step of toluene degradation occurred
by the addition of the aryl methyl carbon to the double bond of fumarate to yield
benzylsuccinic acid. This remarkable reaction is catalyzed by a novel glycyl
radical-containing enzyme, benzylsuccinate synthase [29, 30]. Another
denitrifying strain EbNl, [31], the sulfate-reducing isolates Desulfobacula
toluolica and strain PRTOL1 [31, 32], the iron reducer Geobacter
metallireducens [33], a defined methanogenic consortium [34] and an anaerobic
phototroph [35] have also been shown to carry out this initial toluene
transformation reaction. Subsequent transformations of benzylsuccinate lead to
the formation of benzoyl-CoA and succinyl-CoA [28, 36, 37]. Benzylsuccinate
gets thioesterified to a CoA derivative in a succinyl-CoA-dependent reaction and
the resulting product is then oxidized to is-phenylitaconyl-CoA. The latter
compound presumably undergoes modified [3-oxidation yielding benzoyl-CoA
and regenerates succinyl-CoA. The succinyl-CoA-(i?)-benzylsuccinate CoA-
286
transferase and (7?)-benzylsuccinyl-CoA dehydrogenase responsible for these

bioconversion steps have recently been purified and characterized [37, 38].
Benzoyl-CoA, a central metabolite in the anaerobic oxidation of numerous
aromatic compounds, is presumably further oxidized to acetyl-CoA and CO2 as
described by Harwood et al. [39].
Another addition reaction for anaerobic toluene decay has also been
reported. Toluene activation by the denitrifying strain Azoarcus tolulyticus Tol-4
occurs in two consecutive steps with a 2-carbon fragment (presumably acetyl-
CoA) to ultimately form benzylsuccinate [40]. Evidence supporting this
mechanism includes the detection of cinnamic acid in cultures receiving toluene
and the production of radiolabeled benzylsuccinic acid from incubations
amended with trans-cvcmsmic acid and 14C-acetate. Benzylsuccinate formation is
believed to be preceded by the formation of hydrocinnamoyl-CoA and
cinnamoyl-CoA intermediates. Studies on this toluene pathway are rare so
speculation on how common it might be is difficult.
Biodegradation of m-xylene and o-xylene has been observed in
enrichments under sulfate-reducing [41-43], nitrate-reducing [44] and
methanogenic conditions [45]. The anaerobic biodegradation of p-xylene has
only been demonstrated under sulfate- [41, 43] and nitrate-reducing conditions
[46]. A handful of nitrate- and sulfate-reducing bacteria capable of m-xylene and
o-xylene biodegradation have been isolated [47-51]. Under denitrifying
conditions, m- and o-xylene are activated by fumarate addition reactions by
Azoarcus sp. strain T [28, 52]. Tentative identification of £-(3-methylphenyl)-
itaconyl-CoA and 3-methylbenzoate from /w-xylene suggested a pathway
analogous to that of anaerobic toluene oxidation [52]. Biochemical, molecular,
and genetic studies of benzylsuccinate synthase from Azoarcus sp. strain T
proved that this enzyme catalyzed the initial reaction in anaerobic
biodegradation of both toluene and m-xylene [52, 53]. Further, partially purified
benzylsuccinate synthase transformed all three xylene isomers to their
methylbenzylsuccinate analogs [54]. Under sulfate-reducing conditions, whole-
cell suspensions of the toluene-grown sulfidogenic culture PRTOLl transformed
o-xylene to (2-methylbenzyl)succinate. However, the cell could not grow on o-
xylene suggesting the involvement of benzylsuccinate synthase [55] in the
transformation of the parent molecule. In sulfate-reducing enrichments, 2-, 3-,
and 4-methybenzylsuccinates were positively identified as metabolites of 0-, m-,
and p-xylene, respectively [43] indicating that parent molecules were
anaerobically attacked in a comparable fashion when sulfate served as a terminal
electron acceptor.
There are multiple pathways for ethylbenzene metabolism under
anaerobic conditions. Three pure denitrifying cultures were isolated for their
ability to completely mineralize ethylbenzene [49, 56]. In these organisms,
ethylbenzene is initially activated via dehydrogenation to yield 1-phenylethanol
287
[31,49] by a remarkable molybdenum enzyme that shows clear sequence

similarities to the aerobic dimethyl sulfoxide reductase family of molybdopterin-
containing enzymes [57]. In this reaction the oxygen atom in the hydroxyl group
originates from water [56]. 1-Phenylethanol is then further transformed to
acetophenone, and ultimately to benzoyl-CoA [36]. Under sulfate reducing
conditions, ethylbenzene was converted to the corresponding ethylbenzyl-
succinic acid (3-phenyl-l,2-butane-dicarboxylic acid) by a putative fumarate
addition reaction [43]. Recent studies with a pure sulfate-reducing bacterium
confirmed this mode of ethylbenzene metabolism [58], with the fumarate
addition occurring at the methylene carbon of the side chain rather than at the
terminal methyl group.
Benzene is the least reactive of all aromatic hydrocarbons and was
believed to be entirely recalcitrant under anaerobic conditions. This too has
proven to be incorrect. Benzene can be degraded under nitrate- [59- 61], sulfate-
[62-65], and Fe(III)-reducing conditions [66-68] and with coupling to methane
production [69-71]. To date only two pure benzene-degrading strains, both
affiliated with the Dechloromonas genus, have been described [60]. Both strains
can oxidize benzene with nitrate as an electron acceptor. Physiological studies
and 13C-labeling data suggested benzene activation by an initial methylation
reaction to form toluene [72]. However, alkylation reactions are not entirely
consistent with previous information on anaerobic benzene decay. For example,
13
C-phenol and 13C-benzoate have been detected as intermediates in an
enrichment culture incubated with 13C-benzene under sulfate-reducing
conditions. 13C-Benzoate was also found in comparable methanogenic and
Fe(III)-reducing enrichments [73], suggesting that the hydroxylation of benzene
to phenol is one of the initial steps in anaerobic benzene decay. The conversion
of phenol to benzoate could then occur by the carboxylation of phenol to form/>-
hydroxybenzoate followed by the reductive removal of the hydroxyl group to
form benzoate. 13Carbon and deuterium labeling studies confirmed that the
carboxyl carbon of the benzoate intermediate is derived from one of the carbon
atoms of benzene [73, 74]. Therefore, it seems plausible that multiple
mechanisms for anaerobic benzene decay also exist amongst the anaerobes.
The anaerobic biodegradation of w-alkanes has also been demonstrated
recently. Several sulfate-reducing and denitrifying bacterial strains [42, 75-78]
as well enrichment cultures [44, 79] are capable of the complete conversion of
«-alkanes to carbon dioxide. A broad range of «-alkanes may be susceptible to
anaerobic biodegradation, including long-chain alkanes ranging from Ci5 to C34
[80]. All of the sulfate-reducing alkane-degrading strains isolated to date are
short oval-shaped rods belonging to -subclass of the Proteobacteria. None of
them has been fully characterized and the complete metabolic pathway for
anaerobic alkane decay remains speculative. Recently, a product resulting from
the initial metabolic activation of a model alkane was identified. When
288
deuterated dodecane was added to an alkane-degrading sulfate-reducing culture,

alkylsuccinic acid derivatives with complete deuterium retention were formed,
demonstrating that the primary attack on the parent substrate occurred by
addition to the double bond of fumarate [79]. The same mechanism was
subsequently demonstrated for a denitrifying Azoarcus-like strain [81]. This
bioconversion represents a remarkable reaction that superficially resembles the
anaerobic biodegradation of toluene. However, alkanes are not activated at a
terminal methyl position like toluene or the xylene isomers. Rather the succinyl
moiety is attached subterminally at the C2 (and less frequently the C3) position
of alkanes [79, 81, 82]. The enzymology of this reaction is still under
investigation, though EPR spectroscopy suggests a radical mechanism
comparable to that for toluene decay [81]. Another reported alkane activation
mechanism involves direct carboxylation with inorganic bicarbonate. In studies
with the sulfate-reducer strain Hxd3, So et al. [83] showed that 13C-bicarbonate
was added to alkanes at the C-3 position, followed by the elimination of the two
adjacent terminal carbon atoms yielding a fatty acid one carbon shorter than the
parent alkane. As of this writing, the fumarate addition mechanism seems to be
more widespread for «-alkane activation as it has now been shown for three
disparate anaerobic cultures [79, 81, 84, 85]. The metabolic steps following the
formation of alkylsuccinates are not yet completely clear, but an important
contribution has recently been published [86]. Studies performed with the
denitrifying strain HxNl grown with deuterated «-hexane or deuterated fumarate
revealed the formation of a suite of fatty acids [86]. The identification of 4-
methyloctanoic acid with deuterium in the C-3 position suggested that the
product of fumarate addition, hexylsuccinate (or [l-methylpentyl]succinate),
could possibly undergo rearrangement of the carbon skeleton prior to further
oxidation. Based on the identification of other transient metabolites, such as 4-
methyloct-2-enoic and 3-hydroxy-4-methyloctanoic acids, a hypothetical
pathway was proposed which allowed for the regeneration of fumarate. Recent
studies with 13C-hexane transformation in a sulfate-reducing culture provided
evidence supporting the proposed model [82]. Mass spectral and nuclear
magnetic resonance (NMR) data indicate that multiple 13C nuclei originating
from l-13C-hexane become incorporated into a variety of metabolites. The
labeling patterns argue that 13C-fumarate is produced and recycled during
hexane biodegradation. 4-Methyloctanoic acid, an important metabolite of the
proposed pathway, was positively identified in the organic extracts of 12C- and
13
C-hexane-amended culture supernatants by both mass spectral analysis and
13
C-NMR. Similarly, 3-hydroxy-4-methyloctanoic acid, was also tentatively
identified [82]. Common metabolites formed during anaerobic alkane utilization,
as well as the presence of multiple 13C-carbons in the metabolites, strongly
suggest that this sulfate-reducing culture and a denitrifying strain, HxNl, not
289
only initiate alkane degradation via fumarate addition, but most probably share
the entire degradation pathway.
Polycyclic aromatic hydrocarbons (PAHs) are also susceptible to
anaerobic decay. Naphthalene can be completely mineralized by pure cultures of
sulfate-reducing and denitrifying bacteria [87, 88]. Enrichments from coal-tar
contaminated sediments and garden soil were reported to mineralize [14C]-
naphthalene with soluble Fe(III) and insoluble FeOOH, although not more than
15% of added radioactive substrate was recovered as 14CO2 [89]. Anaerobic
degradation of phenanthrene was also demonstrated in sediments [22, 90] and by
a sulfate-reducing enrichment culture [91]. Studies with marine sediments also
indicated the loss of 2 to 5-ringed PAHs under anaerobic conditions, with the
smaller PAHs degrading more rapidly than the heavier molecular weight
counterparts [90]. It has been shown that unsubstituted PAHs, such as
naphthalene and phenanthrene, are initially attacked by carboxylation to form 2-
naphthoic acid and phenanthrenecarboxylic acid, respectively [91, 92]. The
carbon in both cases arises from inorganic CO2. 2-Methylnaphthalene is
converted to 2-naphthoic acid following the anaerobic oxidation of the methyl
group [93]. A mechanism for the activation of 2-methylnaphthalene is the
addition of fumarate to the methyl group [92, 94]. The product of this reaction,
naphthyl-2-methyl-succinic acid, is subsequently oxidized to 2-naphthoic acid
which further decomposes by ring reduction reactions to form the fully saturated
decalin-2-carboxylic acid prior to ring cleavage and ultimate mineralization [91,
92, 95].
Alicyclic hydrocarbons can comprise a substantial fraction (often up to
~12% wt/wt) of the organic molecules in petroleum mixtures. Despite this
quantitative importance, little is known about the metabolic fate of this class of
materials. Recently, a study of the anaerobic metabolism of a model alicyclic
hydrocarbon, ethylcyclopentane, revealed that it too was initially activated by
fumarate addition to form ethylcyclopentylsuccinic acid [25]. Wilkes et al. [84]
recently observed that when the denitrifying strain HxNl was incubated with
crude oil, a series of C4 to C8 «-alkanes as well as cyclic alkanes, were activated
to their corresponding alkylsuccinates and methyl-branched fatty acids. Further,
cyclopentane, cyclohexane, and their methyl-and ethyl substituted congeners
were rapidly consumed in live incubations of sulfate-amended anoxic sediment
enrichments from a gas condensate-contaminated aquifer [26]. Though alicyclic
biodegradation was more extensive under sulfate-reducing conditions, there was
biodegradation of simpler alicyclic compounds under methanogenic conditions.
In parallel methanogenic incubations, 90% of cyclopentene and methyl-
cyclopentene was lost in 100 days [26]. Thus, this class of materials is also
susceptible to methanogenic biodegradation.
290
3. CHEMICAL THEORY AND THE SUSCEPTIBILITY OF

HYDROCARBONS TO FUMARATE ADDITION REACTIONS
The section above attests to the diversity of anaerobic hydrocarbon

biodegradation reactions. Given the complexity of even simple hydrocarbon
mixtures, it is clearly impractical, if not impossible, to assay all component
molecules for their susceptibility to anaerobic biodegradation. While relatively
little is known of the enzymatic processes and reaction mechanisms responsible
for such bioconversions, insight can be gleaned from a consideration of the
molecular structure of the reported metabolites. A common theme among many
classes of hydrocarbons is the importance of fumarate addition reactions that
lead to metabolites containing a succinic acid functional group. The site of
fumarate addition to the hydrocarbon is likely a key determinant in the
susceptibility of the parent molecules to anaerobic destruction and to accurate
predictions of the metabolites that might reasonably be anticipated. Thus,
accurate prediction of the fumarate addition site based upon chemical principles
is essential.
In a very general sense, fumarate addition to a hydrocarbon substrate can
be thought of as a four-step process. The first step involves complexation of the
hydrocarbon substrate and enzyme, and is controlled by factors such as diffusion
of the hydrocarbon substrate to the enzyme and the thermodynamics of the
enzyme-substrate complex. The second step involves oxidation of the
hydrocarbon substrate, hypothesized by us to occur via a hydrogen atom transfer
(HAT). This is a one-electron oxidation of the substrate through abstraction of a
hydrogen radical (H) transforming the hydrocarbon substrate into a free radical
intermediate [96]. In the case of alkyl aromatic compounds, a second oxidation
mechanism is also possible, one that occurs through electron transfer (ET) [96].
ET involves the removal of an electron from the -orbitals of the aromatic ring,
resulting in an aromatic radical cation. The aromatic radical cation then looses a
proton (H+) to the surrounding matrix and is transformed into the hydrocarbon
radical intermediate. It is hypothesized that the position of the radical in the
hydrocarbon is controlled by the stability of the hydrocarbon radical
intermediate, and therefore the site of fumarate addition can be readily predicted
based upon chemical rules governing free radical stability. The third step
involves addition of fumarate to the hydrocarbon radical, leading to the next step
which is the release of the newly formed metabolite from the enzyme. The
ensuing discussion focuses on the hypothesis that the structure of succinic acid
metabolites suggests that anaerobic fumarate addition reactions proceed via a
radical mechanism; either by a hydrogen abstraction transfer or an electron
transfer mechanism. As a result, metabolite structure can be predicted for
alkane, alicyclic and alkyl aromatic hydrocarbons based upon the chemical rules
that govern free radical stability.
291
As noted, fumarate addition to «-alkanes predominantly occurs at the

subterminal C2 position and to a lesser degree at the C3 position. Fumarate
addition to these positions, rather than the less sterically-hindered terminal
methyl position, strongly suggests a HAT mechanism. In the HAT mechanism,
the site of homolytic cleavage {reaction 1) is a function of the C-H bond
dissociation energy.
RH->R+H (1)
The bond dissociation energy (AH2g8) is a function of the stability of the

free radical intermediate (R*). For alkane compounds, the relative stability of
radical intermediates is of the order: tertiary > secondary > primary [97]. Thus,
the ease of abstracting a hydrogen radical from carbon atoms in an alkane
follows the same relative trend as the bond dissociation energy as seen for
alkanes in Table 1.
The relative stability pattern in alkanes results from hyperconjugation,
that is, delocalization involving a bonds. The greater the number of
hyperconjugative forms that can be generated for a free radical intermediate, the
greater the stability of that intermediate [97]. The bond dissociation energy data
in Table 1 presents a 1 to 3 kcal mol"1 difference between n-alkane methyl and
methylene groups. More importantly, however, is the fact that a 1 kcal mol"1
difference exists between the terminal methyl group and the subterminal C2
carbon for both pentane and hexane, thus illustrating the favorable reactivity of
the subterminal C2 carbon relative the terminal methyl group. Observation of
fumarate addition to the C3 carbon is not surprising since the difference between
the bond dissociation energies of the C3 and C2 methylene carbons is expected to
be minimal, at least much less than 1 kcal mol"1.
The site of fumarate addition to alicyclic compounds should follow the
HAT mechanism similar to w-alkanes, although a decrease in ring strain due to
the loss of a hydrogen atom from the alicyclic ring will slightly lower the bond
dissociation energies relative to the w-alkane analog. This decrease in bond
dissociation energy is observed for cyclopentane which is 3.6 kcal mol"1 less
than that of the C2 carbon in «-pentane. Alkylation of cyclopentane to form
methyl- and ethylpentane produces a tertiary carbon in the ring at the site of
alkyl attachment. As predicted by the HAT mechanism, the tertiary carbon is
more stable as a free radical than the secondary ring carbons, displaying bond
dissociation energies of 93.7 kcal mol"1. Thus, based upon the bond dissociation
energies, and assumption of a HAT mechanism, it is predicted that similar
carbons, i.e., secondary and tertiary carbons, will be more reactive in alicyclic
alkanes as compared to the corresponding «-alkane. Thus, the most favorable
site for fumarate addition to an alkylated alicyclic compound will be at the
tertiary carbon followed by secondary carbons on the alicyclic ring. Based upon
292
the relatively higher bond dissociation energies, fumarate addition to the alkyl
side chain of an alkylated alicyclic hydrocarbon is therefore unexpected.
Table 1
Bond Dissociation Energies (AH298) at 298 K for various hydrocarbons in the reaction
RH -> R* + H* Bolded hydrogen atom represents abstracted hydrogen.
zl//2»«(kcal mol"1) reference

Alkanes
CH3-H (methane) 104.99 +/- 0.03 [98]

CH3CH2-H (ethane) 101.1+/-0.4 [99]
(CH3)2CH-H (propane) 98.6 +/- 0.4 [99]
CH3CH2CH2CH3 (H-butane) 98.2 +/- 0.5 [99]
(CH3)3C-H (wo-butane) 96.5 +/- 0.4 [99]
fl-CjHu-H («-pentane) 100.2 [100]
CH3CH2(CH2)2CH3 («-pentane) 99.2 [100]
«-C6H]3-H («-hexane) 99.0 [100]
CH3CH2(CH2)3CH3 («-hexane) 98.0 [100]
Alicvclic Alkanes
CP-H (cyclopentane) 95.6 +/- 1 [101]

CPH(CH3) (methylcyclopentane) 93.7 [102]
CPH(CH2CH3) (ethylcyclopentane) 93.7 [102]
Alkyl Aromatics
C6H5-H (benzene) 112.9+/-0.5 [103]

C6H5CH2-H (toluene) 89.8 +/- 0.6 [104]
C6H5CH2 CH3 (ethylbenzene) 85.4+/- 1.5 [105]
C6H5CH2 CH2CH3 (n-propylbenzene) 87.5 [106]
Y-C6H5CH(CH3)2
(zso-propylbenzene - substituted)
Y = 2,5 dimethyl 86.7 [107]
Y = 4-;-butyl 83.5 [107]
C6H5C(CH3)2CH2-H(?-butylbenzene) 98.7 [107]
Naphthalene-H
(Ci position) 112.2+/-1.3 [108]
(C2 position) 111.9+/-1.4 [108]
Naphthalene-CH2-H
(CH3 at Ci position) 85.1 +/- 1.5 [105]
(CH3 at C2 position) 85.6 [107]
293
For alkyl aromatic compounds, the benzylic hydrogen atoms, i.e.,

hydrogen atoms bonded to the carbon atom directly attached to the aromatic
ring, require the least energy to abstract due to resonance stabilization created by
delocalization of the free radical within the aromatic 7t-orbital system of the
aromatic ring. Table 1 illustrates how resonance stabilization decreases the bond
dissociation energy of the benzylic hydrogen to a range of 83.5 - 89.8 kcal mol"1
for a variety of alkyl aromatic compounds relative to the bond dissociation
energy for hydrogen atoms bonded directly to the aromatic ring (113 kcal mol"1)
or to other carbons present in the alkyl functional group (e.g., 98.7 kcal mol"1 for
/-butylbenzene). Therefore, in light of a radical mechanism, it is predicted that
fumarate will add to the benzylic carbon atom (as long as the benzylic carbon is
not quaternary and a benzylic hydrogen is available for abstraction) regardless
of the alkyl functional. Moreover, in consideration of the stability afforded
through hyperconjugation in the alkyl group, the C-H bond dissociation energy
will be lower for a tertiary benzylic carbon compared to a secondary benzylic
carbon. Such stabilization is demonstrated in the 0.8 to 4 kcal mol"1 decrease in
the bond dissociation energy of various substituted z-propylbenzene compounds
relative to n-propylbenzene.
The metabolites of the TEX hydrocarbons have been observed to contain
the succinic acid functional group at the benzylic carbon; no addition of
fumarate to the methyl group of ethylbenzene has been detected. In light of a
radical mechanism, the lower reactivity of benzene relative to the TEX
compounds is supported by the relative bond dissociation energies. In fact, the
lack of detection of succinic acid benzene metabolites and the detection of
phenol and benzoate as intermediates (above), suggests that alternative
mechanisms exist for oxidizing benzene for less energy than the 112.9 kcal mol"1
required for hydrogen radical abstraction from the aromatic ring. Similarly for
naphthalene and alkylnaphthalene, the bond dissociation energy for abstracting a
hydrogen radical from an unsubstituted naphthalene is relatively high, 111.9 to
112.2 kcal mol"1, while the comparable reaction from the methyl group of
methylnaphthalene is 85 kcal mol"1. These differences in bond dissociation
energies may account for the fact that naphthyl-2-methyl-succinic acid has been
detected in cultures and in the field (above) but the succinic acid metabolites of
naphthalene have not.
4. GEOCHEMICAL INDICATORS OF METHANOGENIC OIL

BIODEGRADATION
As discussed above, there is ample evidence that anaerobic microbial processes

occur under reservoir conditions. There is even evidence, albeit indirect, that
such processes are occurring in situ. The most widely used indicator for
294
biological methanogenesis comes from the carbon isotopic abundance signature

of the methane in natural gas deposits (e.g. Hunt [109]). Most methane is
thought to arise from thermogenic decomposition of biomass, kerogen, and oil
[110], but biological processes are also important. Bacteria prefer the lighter 12C
isotope over 13C, thus microbially-produced methane is isotopically lighter (C13
of -110 to -60 %o) than thermally-produced gas (C13 of -60 to -15 %o). The
microbial process has classically been thought to occur in the relatively shallow
subsurface, and to be from relatively recently buried biomass rather than from
material that has undergone burial and catagenesis to petroleum. However,
several reservoirs have now been found to have methane with isotopic
signatures suggestive of a biogenic origin [14, 111], and this is certainly
consistent with microbial methanogenesis from petroleum at depth.
Unfortunately, ready interpretation of isotopic enrichment, already complicated
by the likely mixing of thermogenic, biogenic, and abiogenic [112] sources, is
further confounded by the discovery of anaerobic methane oxidation [113], a
microbial activity in which the lighter methane isotope is clearly preferred [114].
It is thus clear that supporting evidence is needed to confirm a microbial origin
for methane in many cases.
This evidence might come from the oil itself. The consideration above
indicates that anaerobes prefer to transform some hydrocarbons relative to
others. For example, an anaerobic microbial consortium was able to degrade
dimethyl-cyclopentanes and cyclohexanes under sulfate-reducing but not under
methanogenic conditions and the activity under the former conditions was
limited to specific isomers [26]. Perhaps the results of such preferences can be
identified in oils from candidate reservoirs? Alternatively it may be possible to
detect by-products of anaerobic biodegradation in waters associated with oil
reservoirs, or in the oil itself. The former is proving very useful in identifying
anaerobic biodegradation in contaminated aquifers, where succinate derivatives
of w-alkanes, cyclic alkanes, and alkylaromatic hydrocarbons as well as
naphthoic acids have been detected [115-117]. Detecting these compounds in
produced waters would be good evidence that anaerobic hydrocarbon
biodegradation was proceeding underground.
Are there compounds in the oil that may act of fingerprints of
biodegradation? Crude oils often contain naphthenic acids, carboxylic acids with
one or more saturated ring structures, and at least some are believed to be the
results of partial biodegradation of oil components [17]. Electrospray ionization
mass spectrometry is proving to be an excellent tool for determining the
molecular identity of naphthenic acids [118-120], and as more potential
biodegradation intermediates are identified it will be important to see whether
such compounds are present in crude oils. Dicarboxylic acids, such as the
succinate derivatives indicative of anaerobic hydrocarbon metabolites, have not
295
yet been identified in oils, but they may be so polar that they primarily partition
to the aqueous phase.
5. PROSPECTS FOR HYDROCABON METHANOGENESIS

With notable exceptions, it is becoming increasingly clear that fumarate addition
reactions represent an important mechanism for the initial activation of
structurally-diverse hydrocarbons by anaerobic microorganisms. Indeed, recent
surveys for such anaerobic metabolites at hydrocarbon-impacted sites identified
a variety of alkylbenzylsuccinates and alkylsuccinates in situ, as well as putative
PAH metabolites such as naphthoic acids and tetrahydronaphthoic acids [43,
115-117, 121, 122].
Based on such observations, one can envision that the same type of
biochemical reactions might occur in oil reservoirs. However, it has long been
accepted that the microbial food web in oil fields is based on aerobic
hydrocarbon-oxidizing bacteria [109, 123-125]. According to this "aerobic"
model, low molecular weight polar compounds such as fatty acids, organic
acids, and alcohols resulting from aerobic hydrocarbon decay serve as substrates
for fermentative, acetogenic, and sulfate-reducing bacteria. Further metabolic
transformations of these compounds produce H2 and acetate, that can then be
used by methanogenic bacteria to produce methane. While this aerobic-
anaerobic successional model of oil decomposition in reservoirs dominated
popular thinking for many years, a reevaluation is needed in light of new
knowledge. Recent geochemical considerations and microbiological data
strongly indicate that oil biodegradation in the deep terrestrial subsurface
proceeds mainly through anaerobic metabolism [11, 12, 16]. Biodegraded oils in
deep anoxic horizons are often accompanied by hydrocarbon gasses of
biological origin [14, 126]. Accordingly, isotopically light methane with 813C
from -45% to -59 % indicative of a biological origin and in situ rates of methane
production in the range from 1.3 to 80 nmol liter "' day"1 were observed in oil
fields under various environmental conditions [127-130]. In the latter studies,
methane precursors were considered to be low molecular weight compounds that
originated from aerobic oil decomposition and migrated to anoxic layers.
Recent studies have now shown that petroleum hydrocarbon
biodegradation can be directly coupled to methane production. For example, the
production of methane from the decay of toluene, o-xylene, benzene, alkanes,
and some alicyclic compounds has been documented [26, 34, 45, 61, 70, 71] . In
incubations of gas condensate-contaminated sediments amended with artificially
weathered oil, the entire «-alkane fraction (Q3-C34 range) was completely
consumed under both sulfate-reducing and methanogenic conditions. In the
sulfate-free incubations, «-alkane degradation was accompanied by methane
accumulation [9]. In other studies, individual alkanes such as hexadecane and
296
pentadecane were converted to CH4 by enrichment cultures and in sediment

incubations [7, 8]. Based on our current understanding of methanogens, the
conversion of hexadecane to CH4 might require as many as three groups of
microorganisms: acetogenic (or syntrophic) bacteria converting hexadecane to
acetate and H2, and acetoclastic and hydrogenotrophic archaea producing CH4
from acetate or H2 and CO2, respectively. Molecular characterization of a
hexadecane-degrading methanogenic community confirmed this possible
composition. It revealed three clones closely related to syntrophic bacteria of the
genus Syntrophus, one clone closely related to the genus Methanosaeta, an
acetoclastic methanogen, and two clones related to Methanospirillum and
Methanoculleus, which comprise hydrogenotrophic methanogens [7]. Similarly,
Watanabe et al. [131] found a substantial diversity of methanogens in the
groundwater under an oil storage cavern in Japan.
Though the most often described alkane-degrading bacteria are the
sulfate-reducing bacteria, they can conceivably participate in methane
production from hydrocarbons even in the absence of sulfate. These bacteria are
known to couple with methanogens to form syntrophic associations wherein
electron transfer occurs between the bacteria. In effect, the methanogen serves
as the electron acceptor for the sulfate reducers. Thus, phylogenetic analysis of
two alkane-degrading sulfate-reducing bacteria revealed that they were closely
related to Syntrophobacter (from 92 to 95% identity), a genus that is known to
degrade fatty acids in syntrophic co-culture with methanogens [132]. In
consistent fashion, a defined co-culture of one of these organisms cultivated
with Methanospirillum hungateii in the absence of sulfate could produce
methane from dodecane (unpublished results). It is therefore not unreasonable to
presume that similar microbial associations can exist in petroliferous subsurface
formations and catalyze hydrocarbon conversions to methane and CO2.
Of course, the rate of bioconversion is an extremely important when
considering the prospects for microbial enhanced energy recovery. As noted,
some researchers believe that such reactions, while clearly possible, take
geologic time due to the limited diffusion of nutrients. While this may be true
along oil migration paths, evidence to the contrary in other locales suggests that
the rates need not be slow. For instance, it has been demonstrated that
subsurface bacteria from oil-bearing sediments could convert hexadecane to
methane quite rapidly and without a lag. Thus 10% of added 14C-hexadecane
was converted to I4CH4 in about 15 d [8]. The in situ rates of methanogenesis
can also be quite high in deep high temperature oil reservoirs. The rates of
methanogenesis measured in formation waters of the Jurassic horizon (2299 m
deep; 84°C) exceeded 80 nmol of CH4 liter "' day"1. Hybridization of 16S rRNA
obtained from formation water with group-specific phylogenetic probes revealed
the presence of thermophilic methanogens and heterotrophs [130]. Laboratory
incubations of formation waters and raw production fluids from two deep high-
297
temperature petroleum reservoirs in California demonstrated active methane

production at in situ temperatures (70-83°C). Total community DNA analysis
revealed archaeal phylotypes closely related to thermophilic methanogens and
sulfidogenic archaea as well as bacterial thermophiles such as Thermatoga sp.,
Thermococcus sp., Thermoanaerobacter sp. and Desulfothiovibrio sp. [133].
These findings contrast with the belief of low metabolic activity in the deep hot
subsurface and the cessation of oil biodegradation due to the paleosterilization
of formations that have at some time experienced temperatures greater than
80°C [12, 134].
The bulk of the accumulated microbiological evidence suggests that oil-
degrading subsurface microbial communities can be quite metabolically
versatile. However, it is unreasonable to presume that the same community
structure exists in all subsurface locales. The environmental conditions during
oil diagenesis may have effectively eliminated critical bacterial components of
obligate consortia responsible for oil methanogenesis. Clearly, the presence of
hydrocarbons in the terrestrial subsurface attests to the fact that such consortia
are far from ubiquitous in distribution.
6. HYDROCARBON METHANOGENESIS AND IMPLICATIONS FOR

ENERGY RECOVERY
Although oil is the dominant source of energy on a global scale,

conventional oil production technologies are only able to recover about one-
third of oil in reservoirs [135]. As a result, large quantities of residual oil remain
trapped in reservoir rock pores, mainly due to capillary or subterranean forces in
the vicinity of a well bore [135, 136]. Thus, enhanced oil recovery (EOR)
methods have been developed to help overcome these forces and make oil move
(see chapter 15). These technologies may be based on thermal, chemical, gas-
miscible, or microbial technologies. It is estimated that EOR strategies can
potentially add up to 60 billion barrels of oil in the near term though the
increased use of existing domestic fields [137]. Understanding the multiphase
flow properties of subsurface reservoir rocks and the forces that entrap oil is key
for successful EOR and will help determine which technique may apply best for
a given reservoir. The processes involved are complex and have been reviewed
[135].
It has long been recognized that gasses dissolved in oil lower its viscosity
and cause swelling. This is a major driving force for oil mobilization. In fact,
gas-based EOR processes have been touted as the current, most profitable
technology for recovering the large amounts of remaining oil in mature fields
[135, 136]. Carbon dioxide has long been used effectively to drive enhanced oil
recovery, and represents about 25% of EOR operations in the U.S. [6, 136, 138].
A secondary outcome in the use of CO2 to recover oil has far-reaching
298
environmental implications too; CO2 can be stored in reservoirs to help in the

reduction of greenhouse gas emissions [6, 136, 138]. Since the combustion of
fossil fuels is the largest contributor of greenhouse gas emissions, the recycling
(capturing and subsequent sequestering) of anthropogenic CO2 into spent or
even active reservoirs offers a promising way to both decrease the potential for
global warming and increase oil recovery and profits. It has been estimated that
fossil fuel reservoirs can store up to 900 billion metric tonnes of CO2 worldwide
[138].
As outlined in the Introduction, natural gas is abundant worldwide, but like oil,
natural gas fields can only be harvested to residual amounts or pressures making
further gas unrecoverable. Carbon dioxide can also serve as an EOR gas for
natural gas recovery by way of re-pressurization of reservoirs [139]. The use of
CO2 as a cushion gas for natural gas storage is also being considered [140]. Of
course, CO2 sequestration into natural gas fields for either recovery or as a
cushion gas also offers the environmentally-friendly advantage of reducing
greenhouse gas emissions [138]
Although gas-based EOR with CO2 is best-understood and most widely
used, the viscosity lowering of a crude by other gasses including nitrogen, flue
gas, and dissolved methane and their relevance for EOR has also been
considered [136, 141-143]. Indeed methane gas associated with oil can
potentially help reduce its viscosity and thus enhance its recovery [141, 142]. In
previous sections, we have discussed the prospect that methane gas found
associated with oil reservoirs can be present as a by-product of anaerobic,
microbial consumption of oil produced over millennia. In fact, there is evidence
suggesting that many "dry gas" fields have arisen due to the microbial
degradation of oil [13, 14]. In fields characterized by light hydrocarbons, C2 to
C5 alkanes are presumably biodegraded to methane, helping to re-establish a
"gas cap" [12, 15]. In theory, such "biogenic gas" could feasibly reduce oil
viscosity to the point where it can be more easily recovered. In practice, gas
pressure accumulations over geological time-scales have no doubt aided in
conventional oil recovery but of course it remains unclear whether these gasses
were thermally- or biologically-produced.
Given the success of gas-based energy recovery, and the recent discovery
that microorganisms can convert hydrocarbons into methane gas at substantial
rates (i.e. faster than geological time scales), one could envision combining the
principles of microbial- and gas-based-EOR to help recover residual oil in
mature fields. Although not yet widely used in the oil industry, advances in
microbial-EOR technologies have proved promising to recover residual oil (see
chapter 15)[135]. Although too numerous to describe here, some MEOR
technologies have explored the use of bacterial inoculation into wells to produce
gaseous by-products which can help mobilize trapped oil [135]. By analogy,
spent reservoirs might be inoculated with the appropriate microbial communities
299
to produce methane gas that could help decrease the viscosity of oil and aid in
further recovery. What if such an inoculation procedure resulted in at least some
fraction of the available energy being recovered as usable methane gas? Such
speculative technology is quite far from being addressed or realized, especially
from an economic point of view, but initial laboratory experimentation on this
topic has been promising (Fig. 1). Samples (10 g) taken from a field in Nowata,
OK that had undergone secondary oil recovery procedures (water flooding) were
used to test the importance of a methane-producing oil-degrading inoculum
enriched from a gas-condensate contaminated aquifer [9]. When residual oil
core samples were ground or broken into small portions, the oil-degrading
inoculum was effective in stimulating methanogenesis relative to a variety of
controls. The latter included a heat-inactivated preparation, an oil-unamended
control, and production water from the same field that received the inoculum (
Fig. 1).
Interestingly, the rate of methanogenesis was much greater with the
residual oil core samples than that observed when a standard oil or even when
the formation (Nowata) crude alone served as a substrate for the inoculum.
While the reasons for this result are under investigation, it is clear that such
inocula may play a potential role for the enhanced recovery of methane from oil
trapped in mature reservoirs.
Fig. 1. Methane production from residual oil in core samples inoculated with a methanogenic
bacterial enrichment capable of anaerobic hydrocarbon metabolism. Symbols: Oil unamended
control (•); Nowata crude oil (•); Production water (X); An artificially weathered Alaska
north slope oil standard (A); Crushed core (o); Pebbled core (•). Heat inactivated and
uninoculated controls are not depicted, but were uniformly negative.
300
7. MICROBIAL ENHANCED ENERGY RECOVERY AND CARBON

DIOXIDE
Figure 1 clearly indicates that a hydrocarbon-degrading methanogenic bacterial
inoculum can attack oil deposited in rocks and covert it to natural gas. In fact,
this metabolism is much faster than comparable incubations amended with an
equivalent amount of oil from the same formation (estimated amount of oil in
core was 0.0 lg oil/g rock based on 30-40% residual saturation). These
observations lead to numerous questions that center on the rate and efficiency of
oil bioconversion, the role of inocula in the process, the nutritional environment
presented by petroliferous formations, the diversity of hydrocarbons susceptible
to microbial attack, the biotechnological control of such bioconverstions and
many other fundamental and practical considerations. Careful exploration of
these issues in the future will help define the utility of enhanced energy recovery
efforts at a time when the need for such considerations is particularly acute.
Tomes have been written on the eventual transitioning of global energy
use patterns and their potential impact on the environment. Yet, it seems clear
that any energy form will have an impact on the environment and that fossil fuel
use will remain the predominant energy form for decades to come. Global
climate change concerns are forcing worldwide reductions in atmospheric CO2
emissions. Since methane consumption produces a fraction of the CO2 per BTU
generated relative other fossil fuels, a greater reliance on methane will help
reduce the rate of increase in global carbon dioxide emissions. The
biotechnological link between the consumption of hydrocarbons for the
production of methane may be a way of enhancing the recovery of energy in an
environmentally responsible fashion, mostly from mature domestic reserves that
are otherwise unprofitable or too technically difficult to exploit. It is our hope
that this article helps spur such considerations.
REFERENCES
[1] National Research Council (NRC). (2003). Committee on Oil in the Sea: Inputs, Fates,
and Effects, Ocean Studies Board and Marine Board, Divisions of Earth and Life
Studies and Transportation Research Board, NRC. In: Oil in the sea III: Inputs, fates,
and effects. The National Academies Press, Washington, D.C.
[2] Energy Information Administration (EIA). (2004).
www.eia.doe.gov/oiaf/ieo/index.html.
[3] C. Hall, P. Tharakan, J. Hallock, C. Cleveland, and M. Jefferson. Nature 426 (2003)
318.
[4] J.B. Curtis and S.L. Montgomery. AAPG Bull. 86 (2002) 1671.
[5] S.M. Al-Fattah and R.A. Startzman. SPE Journal, May (2000) 62-72.
[6] S. Bachu and S. Stewart. J. Can. Petrol. Technol. 41 (2002) 32.
301
[7] Zengler, K., H.H. Richnow, R. Rosello-Mora, W. Michaelis, and F. Widdel. Nature 401
(1999)266.
[8] Anderson, R.T. and D.R. Lovley. . Nature 404 (2000) 722.
[9] Townsend, G.T., R.C. Prince, and J. M. Suflita. . Environ. Sci. Technol. 37 (2003)
5213.
[10] Prince, R. C. (2002). Petroleum and other hydrocarbons, biodegradation of. hi
Encyclopedia of Environmental Microbiology; Bitton, G. Ed.; John Wiley, New York,
pp. 2402-2416.
[11] Roling, W.F.M., I.M. Head, and S.R. Larter. Res. Microbiol. 154 (2003) 321.
[12] Head, I. M., D.M. Jones, and S.R. Larter. Nature 246 (2003) 344.
[13] Sakata, S., Y. Sano, Maekawa, T., and Igari, S.-I. Org. Geochem. 26 (1997) 399.
[14] Pallasser R.J. Org. Geochem. 31 (2000) 1363.
[15] Wenger, L.M., C.L. Davis, G.H. Isaksen. (2001). Multiple controls on petroleum
biodegradation and impact on oil quality. SPE paper 71450.
[16] Larter, S., A. Wilhelms, I. Head, M. Koopmans, A. Aplin, R. DiPrimo, C. Zwach, M.
Erdmann, N. Telnaes. Org. Geochem. 34 (2003) 601.
[17] Tissot, B.P. and D.H. Welte. 1984. Petroleum Formation and Occurrence. Springer-
Verlag, Berlin.
[18] Prince, R. C , R.M. Garrett, R.E. Bare, M.J. Grossman, G.T. Townsend, J.M. Suflita, K.
Lee, E.H. Owens, G.A. Sergy, J.F. Braddock, J.E. Lindstrom, and R.R. Lessard. Spill
Sci. Technol. Bull. 8 (2003) 145.
[19] Britton, L. 1984. Microbial degradation of aliphatic hydrocarbons, hi D. T. Gibson
(Ed.), Microbial degradation of organic compounds, Marcel Dekker, Inc., New York,
pp. 89-129.
[20] Gibson, D.T. and V. Subramanian. 1984. Microbial degradation of aromatic
hydrocarbons, hi T.D. Gibson (ed.), Microbial degradation of organic compounds,
Marcel Dekker, Inc., New York, pp.181-252.
[21] Krumholz, L.R., M.E. Caldwell, and J. M. Suflita. (1996). Biodegradation of "BTEX"
hydrocarbons under anaerobic conditions, hi Bioremediation Principles and
Applications. (R.L. Crawford and D. L. Crawford, Eds.) Cambridge University Press,
Great Britain, pp. 61-99.
[22] Coates, J.D., R.T. Anderson, and D.R. Lovley. Appl. Environ. Microbiol. 62 (1996)
1099.
[23] Widdel, F., and R. Rabus. Curr. Opin. Biotechnol. 12 (2001) 259.
[24] Spormann, A.M. and F. Widdel. Biodegradation 11 (2000) 85.
[25] Rios-Hernandez, L.A., L.M. Gieg, and J.M. Suflita. Appl. Environ. Microbiol. 69
(2003) 434.
[26] Townsend, G.T., R.C. Prince, and J.M. Suflita. FEMS Microbiol. Ecol. (2004). (in
press).
[27] Biegert, T., G. Fuchs, and J. Heider. Eur. J. Biochem. 238 (1996) 661.
[28] Beller, H.R. and A.M. Spormann. J. Bacteriol. 179 (1997) 670-676.
[29] Leuthner, B., C. Leutwein, H. Schulz, P. Horth, W. Haehnel, E. Schiltz, H. Schagger,
and J. Heider. Mol. Microbiol. 28 (1998) 615.
[30] Krieger C.J., W. Roseboom, S.P.J. Albracht, and A.M. Spormann. J. Biol. Chem. 276
(2001) 12924.
[31] Rabus, R. and J. Heider. Arch. Microbiol. 170 (1998) 377.
[32] Beller, H.R. and A.M. Spormann. Appl. Environ. Microbiol. 63 (1997) 3729.
[33] Kane, S.R., H.R. Beller, T.C. Legler, and R.T. Anderson. Biodegradation 13(2002)
149.
302
[34] Beller, H.R. and E.A. Edwards. Appl. Environ. Microbiol. 66 (2000) 5503.
[35] Zengler, K., J. Heider, R. Rosello-Mora, and F. Widdel. Arch. Microbiol. 172 (1999)
204.
[36] Heider, J., A.M. Spormann, H.R. Beller, and F. Widdel. FEMS Microbiol.Rev. 22
(1999) 459.
[37] Leutwein C. and J. Heider. Arch. Microbiol. 178 (2002) 517.
[38] Leutwein, C. and J. Heider. J. Bacteriol. 183 (2001) 4288.
[39] Harwood, C.S., G. Burchhardt, H. Herrmann, and G. Fuchs. FEMS Microbiol.Rev. 22
(1999)439.
[40] Chee-Sanford, J.C., J.W. Frost, M.R. Fries, J. Zhou, and J. Tiedje. Appl. Environ.
Microbiol. 62 (1996) 964.
[41] Edwards, E.A., L.E. Wills, M. Reinhard, and D. Grbic-Galic. Appl. Environ. Microbiol.
58(1992)794.
[42] Rueter, P., R. Rabus, H. Wilkes, F. Aeckersberg, F. A. Rainey, H.W. Jannasch, and F.
Widdel. Nature 372(1994)455.
[43] Elshahed, M.S., L.M. Gieg, MJ. Mclnerney, and J.M. Suflita. Environ. Sci. Technol.
35(2001)682.
[44] Rabus, R., H. Wilkes, A. Schramm, G. Harms, A. Behrends, R. Amann, and F. Widdel.
[45] Edwards, E.A. and D. Grbic-Galic. Appl. Environ. Microbiol. 60 (1994) 313.
[46] Haner, A., P. Hohener, and J. Zeyer. Appl. Environ. Microbiol. 61 (1995) 3185.
[47] Dolfing J., J. Zeyer, P. Binder-Eicher, and R.P. Schwarzenbach. Arch. Microbiol. 154
(1990)336.
[48] Fries, M.R., J. Zhou, J. Chee-Sanford, and J.M. Tiedje. Appl. Environ. Microbiol. 60
(1994)2802.
[49] Rabus, R. and F. Widdel. Arch. Microbiol. 163 (1995) 96.
[50] Hess, A., B. Zarda, D. Hahn, A. Haner, D. Stax,, P. Hohener, and J. Zeyer. Appl.
[51] Harms, G., K. Zengler, R. Rabus, F. Aeckersberg, D. Minz, R. Rossello-Mora, and F.
Widdel. Appl. Environ. Microbiol. 65 (1999) 999.
[52] Krieger, C.J., H.R. Beller, M. Reinhard, and A.M. Spormann. J. Bacteriol. 181 (1999)
6403.
[53] Achong, G.R., A.M. Rodriguez, and A.M. Spormann. J. Bacteriol. 183 (2001) 6763.
[54] Beller, H.R. and A.M. Spormann. (1999). FEMS Microbiol. Lett. 178 147.
[55] Beller, H.R., A.M. Spormann, P.K. Sharma, J.R. Cole, and M.Reinhard. Appl.
[56] Ball, H.A., H.A. Johnson, M. Reinhard, and A.M. Spormann. J. Bacteriol. 178 (1996)
5755.
[57] Johnson, H.A., D.A. Pelletier, and A.M. Spormann. J. Bacteriol. 183 (2001) 4536.
[58] Kniemeyer, O., T. Fischer, H.Wilkes, F.O. Glockner, and F. Widdel. Appl. Environ.
Microbiol. 69 (2003) 760.
[59] Burland, S.M. and E.A. Edwards. (1999). Appl. Environ. Microbiol. 65 529.
[60] Coates, J.D., R. Chakraborty, J.G. Lack, S.M. O'Connor, K.A. Cole, K.S. Bender, and
L.A. Achenbach. Nature 411 (2001) 1039.
[61] Ulrich, A C. and E.A. Edwards. Environ. Microbiol. 5 (2003) 92.
[62] Edwards, E.A. and D. Grbic-Galic. Appl. Environ. Microbiol. 58 (1992) 2663.
[63] Lovley, D.R., J.D. Coates, J.C. Woodward, and E.J.P. Phillips. Appl. Environ.
Microbiol. 61 (1995)953.
303
[64] Coates, J.D., R.T. Anderson, J.C. Woodward, E.J.P. Phillips, and D.R. Lovley.
Environ. Sci. Technol. 30 (1996) 2784.
[65] Weiner, J. and D.R. Lovley. Appl. Environ. Microbiol. 64 (1998) 775.
[66] Lovley, D.R., J.C. Woodward, and F.H. Chapelle. Nature 370 (1994) 128.
[67] Anderson, R.T., J.N. Rooney-Varga, C.V. Gaw, and D.R. Lovley. Environ. Sci.
Technol. 32(1998)1222.
[68] Caldwell, M.E., R.S. Tanner, and J.M. Suflita. Anaerobe 5 (1999) 595.
[69] Grbic-Galic, D. and T. Vogel. Appl. Environ. Microbiol. 53 (1987) 254.
[70] Kazumi, J., M.E. Caldwell, J.M. Suflita, D.R. Lovley, and L.Y. Young. Environ. Sci.
Technol. 31(1997)813.
[71] Weiner, J. and D.R. Lovley. Appl. Environ. Microbiol. 64 (1998) 1937.
[72] Coates, J.D., R. Chakraborty, and M.J. Mclnerney. Res. Microbiol. 153 (2002) 621.
[73] Caldwell, M.E. and J.M. Suflita. Environ. Sci. Technol. 34 (2000) 1216.
[74] Phelps, CD., X. Zhang, and L.Y. Young. Environ. Microbiol. 3 (2001) 600.
[75] Aeckersberg, F., F. Bak, and F. Widdel. Arch. Microbiol. 156 (1991) 5-14.
[76] Aeckersberg, F., F. Rainey, and F. Widdel. Arch. Microbiol. 170 (1998) 361.
[77] So, CM. and L.Y. Young. Appl. Environ. Microbiol. 65 (1999) 2969.
[78] Ehrenreich, P., A. Behrends, J. Harder, and F. Widdel. Arch. Microbiol. 173 (2000) 58.
[79] Kropp, K.G., LA. Davidova, and J.M. Suflita. Appl. Environ Microbiol. 66 (2000)
5393.
[80] Caldwell, M.E., R.M. Garrett, R.C. Prince, and J.M. Suflita. Environ. Sci. Technol. 32
(1998)2191.
[81] Rabus, R., H. Wilkes, A. Behrends, A. Armstroff, T. Fischer, and F. Widdel. J.
Bacteriol. 183(2001)1707.
[82] Davidova, I., L. Gieg, K. Kropp, M. Nanny, J. Suflita. (2004) (submitted for
publication)
[83] So, CM., CD. Phelps, and L.Y. Young. Appl. Environ. Microbiol. 69 (2003) 3892.
[84] Wilkes, H., S. Ktihner, C. Bolm, T. Fischer, A. Classen, F. Widdel, and R. Rabus. Org.
Geochem. 34(2003)1313.
[85] Callaghan, A.V., L.M. Gieg, K.G. Kropp, J.M. Suflita, and L.Y. Young. (2003).
Fumarate addition during hexadecane degradation by the sulfate-reducer AK-01.
American Society for Microbiology 103-rd General Meeting, Washington, D.C,
Abstract Q-038, p. 521.
[86] Wilkes, H., R. Rabus, T. Fischer, A. Armstroff, A. Behrends, and F. Widdel. Arch.
Microbiol. 177(2002)235.
[87] Galushko, A., D. Minz, B. Schink, and F. Widdel. Environ. Microbiol. 1 (1999) 415.
[88] Rockne, K.J., J.C. Chee-Sanford, R.A. Sanford, B.P. Hedlund, J.T. Staley, and S.E.
Strand. Appl. Environ. Microbiol. 66 (2000) 1595.
[89] Ramsay, J.A., H. Li, R.S. Brown, and B. Ramsay. Biodegradation 14 (2003) 321.
[90] Rothermich, M.M., L.A. Hayes, and D.R. Lovley. Environ. Sci. Technol. 36 (2002)
4811.
[91] Zhang, X. and L.Y. Young. Appl. Environ. Microbiol. 63 (1997) 4759.
[92] Annweiler, E., W. Michaelis, and R.U. Meckenstock. Appl. Environ. Microbiol. 68
(2002) 852.
[93] Sullivan, E.R., X. Zhang, C. Phelps, and L.Y. Young. Appl. Environ. Microbiol. 67
(2001)4353.
[94] Annweiler, E., A. Materna, M. Safinowski, A. Kappler, H.H. Richnow, W. Michaelis,
and R.U. Meckenstock. Appl. Environ. Microbiol. 66 (2000) 5329.
[95] Zhang X, Sullivan E. R, and L.Y Young. Biodegradation. 11 (2000) 117.
304
[96] Baciocchi, E., F. D'Acunzo, C. Galli, and O. Lanzalunga. J. Chem. Soc. Perkin Trans.
2.2(1996)133.
[97] March, J. 1968. Advanced Organic Chemistry Reactions, Mechanisms, and Structure.
McGraw-Hill Book Company, New York, New York.
[98] Ruscic, B., M. Litorja, and R. Asher. J. Phys. Chem. A. 103 (1999) 8625.
[99] Saekins, P.W., M.J. Pilling, J.T. Niiranen, D. Gutman, and L.N. Krasnoperov. J. Phys.
Chem. 96(1992)9847.
[100] Pedley, J.B, R.D. Naylor and S.P. Kirby. 1986. Thermodynamic Data of Organic
Compounds. 2nd ed., Chapman and Hall, New York
[101] Castelhano, A.L. and D. Griller. J. Amer. Chem. Soc. 104 (1982) 3655.
[102] Tumanov, V.E. and E.T. Denisov. Neftekhimiya. 41 (2001) 109.
[103] Wenthold, P.G., and R.R. Squires. J. Am. Chem. Soc. 116 (1994) 6401.
[104] Ellison, G.B., G.E. Davico, V.M. Bierbaum, and C.H. DePuy. Int. J. Mass Spectrum.
Ion Processes. 156 (1996) 109.
[105] McMillen, D.F. and D. M. Golden. Ann. Rev. Chem. 33 (1982) 493.
[106] Denisov, E.T. and T.G. Denosova. (2000). Handbook of Antioxidants. CRC Press, New
York.
[107] Kromkin, E.A., V.E. Tumanov, and E.T. Denisov. Neftekhimiya. 42 (2002) 3.
[108] Reed, D.R. and S.R. Kass. J. Mass Spectrom. 35 (2000) 534.
[109] Hunt, J.M. 1979. Petroleum Geochemistry and Geology, 2nd Ed. W.H.
Freeman and Company, New York, pp. 413.
[110] Domine, F., R. Bounaceur, G. Scacchi, P.-M. Marquaire, D. Dessort, B. Pradier, and O.
Brevart. Org. Geochem. 33 (2002) 1487.
[ I l l ] Sassen, R., A.V. Milkov, E. Ozgul, H.H. Roberts, J.L. Hunt, M.A. Beeunas, J.P.
Chanton, D.A. DeFreitas, and S.T. Sweet. Org. Geochem. 34 (2003) 1455.
[112] Sherwood Lollar, B., T.D.Westgate, J.A. Ward, G.F. Slater, and G. Lacrampe-
Couloume. Nature 416 (2002) 522.
[113] Nauhaus, K., A. Boetius, M. Kriiger, and F. Widdel. Environ. Microbiol. 4 (2002)
296.
[114] Orphan, V.J., C. H. House, K.U. Hinrichs, K. D. McKeegan and E. F. DeLong.
Science 293 (2001) 484.
[115] Beller, H.R. Biodegradation. 11 (2000) 125.
[116] Gieg, L.M. and J.M. Suflita. Environ. Sci. Technol. 36 (2002) 3755.
[117] Griebler, C, M. Safinowski, A. Vieth, H.H. Richnow, and R.U. Meckenstock.
Environ. Sci. Technol. 38 (2004) 617.
[118] Barrow, M.P., L.A. McDonnell, X. Feng, J. Walker, and P.J. Derrick. Anal. Chem. 75
(2003) 860.
[119] Gabryelski, W. and K.L. Froese. Anal. Chem. 75 (2003) 4612.
[120] Lo, C.C., B.G. Brownlee, andNJ. Bunce. Anal. Chem. 75 (2003) 6394.
[121] Phelps, C D , J. Battistelli, and L.Y. Young. Environ. Microbiol. 4 (2002) 532.
[122] Martus, P. and W. Puttman. Sci. Total Environ. 307 (2003) 19.
[123] Belyaev, S.S., K. Laurinavichus, A.Y. Obraztsova, S.N. Gorlatov, and M.V. Ivanov.
Microbiologiya. 51 (1982) 997.
[124] Nazina, T.N, E.P. Rozanova, and S.I. Kuznetsov. Geomicrobiol. J. 4 (1985) 103.
[125] Palmer, S.E. (1993). Effect of biodegradation and water washing on crude oil
composition. In Engel M.H, Macko S.A. (Eds.), Organic Geochemistry. Plenum Press,
New York, pp. 511-533.
[126] James, A.T. and B.J. Burns. Bull. Am. Assoc. Petrol. Geol. 68 (1984) 957.
305
[127] Ivanov, M.V., S.S. Belyaev, A.M. Zyakun, V.A. Bondar, and K.K. Laurinavichus.
Geokhimiya. 11 (1983) 1647.
[128] Borzenkov LA., S.S. Belyaev, Y.M. Miller, LA. Davydova, and M.V. Ivanov.
Microbiology 66 (1997) 104.
[129] Nazina, T,N., A.E. Ivanova, LA. Borzenkov, S.S. Belyaev, and M.V. Ivanov.
Geomicrobiol. J. 13 (1995) 181.
[130] Bonch-Osmolovskaya, E.A., M.L. Miroshnichenko, A.V. Lebedinsky, N.A. Chernyh,
T.N. Nazina, V.S. Ivoilov, S.S. Belyaev, E.S. Boulygina, Y.P. Lysov, A.N. Perov, A.D.
Mirzabekov, H. Hippe, E. Stackebrandt, S. L'Haridon, and C. Jeanthon. Appl. Environ.
Microbiol. 69(2003)6143.
[131] Watanabe, K., Y. Kodama, N. Hamamura, and N. Kaku. Appl. Environ. Microbiol. 68
(2002) 3899.
[132] Davidova, I.A., K.G. Kiopp, K.E. Duncan, and J.M. Suflita. (2002). Anaerobic
biodegradation of n-alkanes by sulfate-reducing bacterial cultures. Abstracts of the
International Symposium on Subsurface Microbiology. Copenhagen, Denmark,
September 8-13.
[133] Orphan,, V.J., S.K. Goffredi, E.F. Delong, and J.R. Boles. Geomicrobiol. J. 20 (2003)
295.
[134] Connan, J. 1984. Biodegradation of crude oils in reservoirs. In J. Brooks, and D.H.
Welte (Eds.), Advances in Petroleum Geochemistry. Academic Press, London, pp. 89-
129.
[135] Mclnerney, M.J. and D.W.S. Westlake. 1990. Microbial enhanced oil recovery. In:
Microbial Mineral Recovery, H.H.L. Ehrlich and C.L. Brierley (Eds.), McGraw-Hill,
NY, pp. 409-445.
[136] Rao, D. J. Can. Petrol. Technol. 40 (2001) 11.
[137] Aycaguer, A.-C, M. Lev-On, and A.M. Winer. Energy & Fuels 15 (2001) 303.
[138] Oldenburg, CM., K. Pruess, and S.M. Benson. Energy & Fuels 15 (2001) 293.
[139] Oldenburg, CM. Energy & Fuels 17 (2003) 240.
[140] Killesreiter, H. Erdoel und Kohle, Erdgas, Petrochemie 38 (1985) 405.
[141] Frauenfeld, T.W.J., R.K. Ridley, R.K., and D.M. Nguyen. J. Petrol. Technol. 40
(1988)333.
[142] Mayne, CJ. and R.W. Pendleton. Soc. Petrol. Eng. AIME 1 (1986) 131.
[143] Alvarez, M.R, M.F. Hilton, and H.L. Oil & Gas J. 82 (1984) 95.
Chapter 11
Using nitrate to control microbially-produced hydrogen

sulfide in oil field waters
R.E. Eckford and P.M. Fedorak
Department of Biological Sciences, University of Alberta, Edmonton, Alberta,

Canada T6G 2E9
1. INTRODUCTION
The presence of hydrogen sulfide (H2S) in oil fields can be the result of abiotic
or biotic processes. In the later case, sulfate-reducing bacteria (SRB) are the
culprits that produce this nocuous gas, leading to "souring" that is defined as the
process whereby petroleum reservoirs experience an increase in the production
of H2S during the economic production life of the field [1]. The increase in H2S
content leads to a decrease in the economic value of the gas and oil, as well as
operational problems associated with the H2S.
This microbial process in wastewaters and oil field waters can be
controlled by another group of microbes, known as nitrate-reducing bacteria
(NRB). Their metabolic activities stop sulfate reduction by SRB, and in many
cases the NRB can actually consume sulfide, thus decreasing H2S concentration
in the waters. Jenneman et al. [2] have referred to these sulfide-consuming
bacteria as "sulfide bioscavengers". Hitzman and Sperl [3] used the term
"biocompetitive exclusion" to describe the microbial process in which NRB use
volatile fatty acids and out-complete SRB to prevent or decrease sulfide
production, and enhance oil recovery.
This chapter will review (a) H2S in the petroleum industry, (b) the
metabolism of SRB leading to sulfide production, (c) the occurrence, types and
activities of NRB that might be found in oil field waters, (d) some laboratory
studies that have elucidated the mechanisms by which NRB control sulfide
produced by SRB, (e) some oil field experiences with nitrate injection to control
sulfide in wastewaters, surface waters and oil field waters, and (f) some of the
U.S. patents that apply to this microbial process.
308
Although nitrite, rather than nitrate, addition has been studied, this chapter
focuses solely on the use of nitrate to control sulfide in oil field waters. This is a
proven biotechnology that is under-utilized by the petroleum industry.
2. H2S AND THE PETROLEUM INDUSTRY
2.1. Formation of H2S

Kerogen is the organic source material from which petroleum is formed
and released [4-5]. The formation of petroleum occurs in the deeper subsurfaces
as burial continues and temperature and pressure increase [5]. First oil, then gas
is expelled from kerogen as the maturation process continues. Significant oil
generation occurs between 60° and 120°C, and significant gas generation occurs
between 120° and 2v25°C [5]. During the maturation process, H2S is also
released.
Machel [6] wrote, "The association of dissolved sulfate and hydrocarbons
are thermodynamically unstable in virtually all diagenetic environments. Hence,
redox-reactions occur, whereby sulfate is reduced by hydrocarbons either
bacterially (bacterial sulfate reduction) or inorganically (thermochemical sulfate
reduction)." Temperature is the major factor determining which process occurs.
The microbiological process is common at temperatures for 0 to 60 or 80°C,
whereas, the thermochemical process occurs at temperatures greater that 100° to
140°C [6]. Because temperature increases with burial depth, H2S found at
shallow depths is usually the result of bacterial sulfate reduction whereas, H2S
found at greater depths is the result of thermochemical sulfate reduction [7].
However, there are shallow pools that contain higher than expected
concentrations of thermochemically generated sulfide [8]. These are believed to
be the result of thermochemical sulfate reduction occurring downdip and
migrating upward to a shallow reservoir [8].
At the time of discovery, the H2S concentration in an oil field depends
upon its maturation history and/or the migration of H2S into the oil field.
However, during oil recovery from some oil fields, an increase in H2S
concentration (souring) can occur as a result of pressurizing the formation by
injecting water into the reservoir. This process, know as waterfiooding, is
discussed in section 3. Three well-documented examples of oil field souring are
given in the following paragraphs.
Cochrane et al. [9] describe the souring of the Ninian field in the North
Sea. This field was discovered in 1974, and after several years of operation,
injection of sea water was used to maintain the production rate. This was
followed by an increase in sulfide production attributed to bacterial sulfate
reduction. The reservoir temperature was initially between 100° to 120°C, but in
the areas adjacent to the injection well bores, the temperature was cooled to as
low as 40°C, which was conducive to bacterial sulfate reduction.
309
Frazer and Boiling [10] described the souring of the Kuparuk River field
on the North Slope of Alaska. The field was initially sweet, but after injection of
Beaufort Sea water, detectable levels of H2S began to appear at the producing
wells. The connate water contained essentially no sulfate. However, the sulfate
in the sea water stimulated bacterial sulfate reduction in the reservoir that had a
temperature of about 70°C.
The Skjold oil field in the North Sea soured upon the onset of
waterflooding [11]. Oil and gas production began from this field in 1982 and sea
water injection began in April 1985. In September 1985, the first recorded H2S
production was measured to be 1.8 ppm in the gas phase. In 2002, the
concentrations varied from 10 to 1000 ppm [11]. In late 1999, this field
produced 1150 kg H2S d"1.
These examples clearly demonstrate that waterflooding can stimulate
bacterial sulfate reduction, leading to souring. Although these examples refer to
offshore oil fields, souring also occurs in land-based oil fields using
waterflooding [12-15]. As a result of the bacterial production of toxic H2S, the
value of the oil decreases as the oil field sours.
2.2. H2S toxicity and properties

H2S is a very dangerous gas, even though it occurs in nature. Its
characteristic rotten egg smell is generally obvious at 0.13 ppm by volume and
quite noticeable at 4.6 ppm [16]. Unfortunately the smell sense becomes quickly
fatigued and can fail to warn of higher concentrations. Collapse, coma and death
from respiratory failure may occur within a few seconds after one or two
inspirations of the undiluted H2S [17]. The U.S. Occupational Safety and Health
Administration has established the acceptable ceiling concentration of 20 ppm
(by volume) for H2S with an acceptable maximum peak above the acceptable
ceiling concentration of 50 ppm for an 8-h shift [16].
The specific gravity of H2S is 1.19; therefore it will collect in low places
and accumulate under poorly ventilated conditions [18]. H2S is soluble in water
and oil. It is a weak acid existing in aqueous solutions as H2S, HS~, or S~ (pKa
values of 7.04 and 11.96). Aqueous solutions of H2S absorb O2 leading to the
formation of elemental sulfur [17].
2.3. Detrimental effects of H2S

Besides its toxicity, H2S is a nuisance in the petroleum industry because it
contaminates gas and stored oil, it corrodes iron in the absence of air (anaerobic
corrosion), and it precipitates as amorphous ferrous sulfide (FeS), plugging and
diminishing the injectivity of water injection wells [18]. In addition, fluids with
water and H2S, may cause sulfide stress cracking of susceptible metals. This is
affected by metal composition, pH, H2S concentration, total pressure, total
tensile stress, temperature and time [19].
310
Two types of cracking known to occur in wet H2S environments are

sulfide stress corrosion cracking and hydrogen-induced cracking (see chapters 7
and 8). The former occurs in steels of relatively high strength and in welds of
welded steel structures. A crack propagates under working stress or residual
stress vertically to the stress axis [20]. This type of corrosion is most damaging
to drillpipe and well production facilities [21]. Hydrogen-induced cracking
occurs parallel to the surface when no external stress is applied. It is also known
as hydrogen blistering because of the blisters that appear on the surface of the
metal [20].
General corrosion attack by H2S is influenced by the presence of CO2, O2
and brine, [18, 21]. It is related to the alloy composition and strength of steel
[21]. H2S forms FeS scale, which is cathodic to the metal, promoting localized
attack under the scale, as well as the penetration of H2 into the metal [21-22].
Figure 1 shows the process whereby an anode and cathode pair are generated by
the action of SRB acting on sulfates in the presence of iron. The cathode is
depolarized as the SRB consume H2. At the anode, iron (Fe) is oxidized to Fe2+
which combines with H2S produced by the SRB, giving FeS. This process
results in a loss of structural material. Heterotrophic SRB also play a role in the
deposition of FeS (Fig. 1).
Fig. 1. Iron metal corrosion mediated by SRB in a biofilm. The process is caused by the
consumption of H2 causing cathodic depolarization. Adapted from Ref. [18].
311
Removal of dissolved gases (O2, H2S and CO2) from drilling and
produced fluids is necessary to minimize corrosion damage. H2S in oil base
drilling fluid is removed by gas separators and vacuum degassers, and then
neutralized. Controlling corrosion in H2S-containing environments requires
proper selection of materials, including the use of low-hardness steels,
application of inhibitors and complete exclusion and removal of O2 from water
used in petroleum production [21]. Clearly, the presence of H2S greatly
increases the cost of exploration for oil and natural gas, and the cost of
production and storage of petroleum.
Plugging (or biofouling) of injection wells is also caused by SRB. The
sulfide they produce, precipitates soluble iron in the injection or formation water
forming colloidal FeS [23]. This colloidal material becomes associated with
bacterial cells and oil, forming a gummy mass that can clog reservoirs and plug
injection wells. The activities of SRB can also produce calcite (CaCO3) that can
add to the plugging problem.
3. OIL RECOVERY AND WATERFLOODING
Under primary oil recovery, typically less than 30% of the original oil is
produced, so that improved or enhanced methods are used to recover some of
the remaining oil [24]. These processes, known as secondary and tertiary
recovery methods, include the addition of energy into the reservoir and are
accomplished by injecting some type of fluid through injection wells. This is
referred to as enhanced oil recovery and involves water injection, gas injection,
steam injection, combustion, miscible fluid displacement and polymer injection
[24]. In this paper, only water injection or waterflooding will be discussed.
Waterflooding involves pumping water into the reservoir to stimulate
production. The injected water provides pressure to force the oil out of the rock
and to sweep it toward producing wells as shown in Fig. 2. Waterflooding has
been attempted in almost every type of reservoir, with its greatest success in
relatively homogenous reservoirs having sufficient permeability to allow water
injection at a reasonable rate [24]. Up to 60% of the oil can be recovered with
waterflooding [5]. Water handling can become a major operational procedure.
For example, in some western Canadian oil fields, the proportion of water in the
oil-water emulsion brought to the surface can be 95% by volume [15]. That is,
the volume of water handled is 19 times greater than the volume of oil produced.
Water used as injection water can be of three types: formation water, sea
water or fresh water. Formation water is subsurface brackish or brine water
produced from a petroleum or non-petroleum producing formation. Sea water
may also include water from a salty (non-potable) lake. Fresh water, containing
312
less than 2000 ppm dissolved solids, is primarily water that can be made potable
by flocculation, filtration and chlorination [25].
Because oil field reservoir rocks are porous, they are susceptible to plugging by
solids suspended in or precipitated from an injection fluid [26]. This makes
water quality testing necessary to determine parameters such as: amount and
composition of suspended solids, clay sensitivities, presence of bacteria,
compatibility of two or more waters, and compatibility of the injection solution
with reservoir rock. An example of incompatible waters occurs when sulfate
scales, such as barium sulfate, calcium sulfate or strontium sulfate are formed by
mixing waters containing sulfate with waters containing barium, calcium or
strontium ions [26]. As well, the gases O2, H2S and CO2 found in injection
waters and implicated in corrosion [25-26], must be monitored. Water quality
testing, should be continued after the enhanced oil recovery operation hasstarted,
to ensure that the system is maintained at optimum conditions [25]. Water
treatment methods are outlined by Rose et al. [27].
Fig. 2. A simple waterflooding operation. Oil, gas and water are collected from the production
wells and the produced water is separated from the oil and gas. The produced water is
combined with source water and injected into the oil-bearing rock to pressurize the formation
and sweep the oil to the producing wells.
313
Water should be free of bacteria that can cause corrosion [25-26], or

plugging of equipment and injection wellbores [25]. The presence of bacteria
can be problematic because they reproduce rapidly over wide ranges of pH,
temperature, pressure and anoxia in the reservoir. Bacteria found in oil field
injection waters that cause problems are SRB, iron-reducing bacteria and slime-
formers [25, 27]. Of special concern are the SRB.
Source waters used in waterflooding can increase the activities of SRB
souring for several reasons [1]. The source water, especially sea water, may
contain sulfate to serve as a terminal electron acceptor and may introduce SRB,
nutrients such as short chain fatty acids and ammonium into the reservoir. Large
volumes of source water may reduce the salinity and temperature in the
formation near the injection well, providing an environment that is more
conducive to the growth of SRB and oil field souring.
4. SULFATE-REDUCING BACTERIA
Ask any person who works in the oil field or who is involved with the transport
or storage of crude oil to name some bacteria, and most will immediately
respond "sulfate-reducing bacteria" or "SRB". These bacteria are well-known,
and in the oil field environment, they are a nuisance because their metabolic
activities produce H2S that can sour reservoirs, create plugging through FeS
formation and induce corrosion [28]. SRB have the unique ability to utilize
sulfate as a terminal electron acceptor. This is an anaerobic respiratory process
used to generate energy for the biosynthetic reactions involved in cell growth
and maintenance [29].
The SRB are a diverse group of prokaryotes that are found in many
anaerobic environments. These bacteria have been the subject of several books
[30-33] and countless articles. The phylogeny of SRB has recently been
reviewed [34], and based on rRNA sequences, they fall into four groups: Gram-
negative mesophiles, Gram-positive endospore-formers, thermophilic bacteria,
and thermophilic Archaea.
4.1. Overview of the metabolism of SRB

The dissimilatory H2S-producing SRB have little energy available to
them. The upper limits of energy conservation from sulfate reduction are set by
thermodynamics. For example, if a potent electron donor like H2 is oxidized, the
free energy change of the overall reaction, under standard conditions at neutral
pH, is -38 kJ (mole H2)A (reaction 1), which is 6-fold lower than with O2 as a
terminal electron acceptor (reaction 2) [35].
4H2 + SOzf + 2H+ -> H2S + 4H2O G°' = -38 kJ (mol H2)"1 (1)
314
4H2 + 202 -> 4H2O G°' = -237 kJ mol H2)"1 (2)
As late as the 1970's, only a few genera of SRB were recognized, and
these were known to use only a few growth substrates, most notably lactate,
pyruvate or H2. Now it is apparent that SRB are capable of using various
compounds for electron donors.
Based on their metabolic capabilities, heterotrophic SRB fall into two
groups: those that cannot oxidize acetate, and those that carry out complete
oxidation of acetate to C0 2 [36]. Reaction (3) illustrates the overall reaction of
lactate-utilizing SRB that cannot oxidize acetate. One mol of acetate
accumulates for each mol of lactate that is consumed.
2CH3CHOHCOO" + S04 = + 2H+ -> 2CH3COO" + 2H2O + 2CO2 + H2S

G°' = -77 kJ (mol lactate)"1 (3)
The complete oxidation of acetate is given by reaction (4), showing that less
energy is available per mol of acetate than per mol of lactate (reaction 3).
CH3COO" + SOzf + 3H+ -» 2CO2 + H2S + 2H2O

G°' = -41 kJ (mol acetate)"1 (4)
Increased understanding of the metabolic diversity of SRB now indicates

that nearly 100 organic compounds can be used by various SRB [37]. These
substrates include fatty acids up to C2o; aromatic hydrocarbons such as toluene,
xylenes, ethylbenzene, and naphthalene; «-alkanes from (C6 to C2o); and simple
oxidation products of hydrocarbons such as benzoate, phenol, and cresol [38-
40]. These substrates are present in native crude oils or partially degraded crude
oils. Thus, if there is an ample supply of sulfate in water contacting crude oil in
an anaerobic environment, there is the potential for SRB to actively produce
H2S, using many different organic compounds (or H2) as an energy source.
The ability to reduce sulfate links this diverse group of bacteria. However,
it is now apparent that various SRB can reduce other chemical species including
Fe(III), nitrate, some chlorinated aromatics, sulfur oxyanions and O2 [37].
Molecular oxygen can be reduced by most SRB. In this case, the stoichiometry
(for example, 2H2 consumed per O2 reduced) indicates that O2 can be
completely reduced to water.
SRB are also capable of fermentative growth or utilization of other
electron acceptors, such as sulfite, thiosulfate and elemental sulfur [12, 35] and
tetrathionate [12]. Many SRB are able to ferment organic substrates in the
absence of sulfate. For example, Desulfotomaculum orientis can carry out
315
fermentation using homoacetate. Also many SRB can perform a unique

fermentation of inorganic sulfur compounds which are disproportionated to
sulfate (a more oxidized compound) and sulfide (a more reduced compound).
For example, thiosulfate is transformed to equal amounts of sulfate and sulfide,
and sulfite is disproportionated to 3/4 sulfate and 1/4 sulfide [35].
Some species of SRB are able to utilize nitrate as an electron acceptor.
When nitrate is used as an electron acceptor, SRB produce ammonia, but not N2,
as an end product. Nitrite is formed as an intermediate of nitrate reduction and
can be reduced by many sulfate reducers unable to reduce nitrate. In the
presence of both sulfate and nitrate some SRB will preferentially use one or the
other as an electron acceptor, and some SRB will reduce both concomitantly
[35].
4.2. Activities of SRB in anaerobic environments

When microorganisms get into stagnant or closed water systems,
dissolved O2 is quickly and completely consumed. Despite the absence of O2,
organic matter may undergo biological decomposition by microbial activities,
including fermentation. The degradation reactions by which most fermentative
bacteria gain energy are disproportionations of the organic matter, part
converted to CO2, and part converted to reduced products, such as fatty acids,
H2, and alcohols [18]. If sulfate is abundant in these anaerobic environments, the
fermentation products are used by SRB. Sulfate serves as the terminal electron
acceptor, and the reducing power from the decomposed organic matter results in
the formation of H2S.
SRB grow in anaerobic muds found in fresh water or sea water
environments [41]. They are also indigenous members of the microbial
community in ground waters, marine environments, coastal sediments, marine
hydrothermal vents associated with volcanic or tectonic activity, and hot springs
[42]. SRB can flourish in environments wherever decomposable organic matter
gets into anaerobic, sulfate-containing waters. Here H2S is produced and
evidenced by visible blackening of the sediment when FeS forms from iron
minerals [18]. Marine and estuarine saltmarsh sediments, saline and hypersaline
lakes and ponds, as well as oil field waters with high sulfate content are the most
permanent and significant habitats of SRB [43]. Large amounts of sulfate are
required for this process, so that the consequence resulting from the growth of
SRB is the dissemination of massive quantities of H2S [29].
Many SRB use simple, low molecular weight compounds, and therefore
depend on fermentative bacteria to cleave and ferment complex organic matter.
SRB convert only about 10% of the total substrate carbon to cellular material, so
that the bulk of the substrate has to be decomposed for providing energy. Thus,
316
SRB, make themselves conspicuous by the formation of their metabolic product,

H2S, rather than by formed cell mass [18].
How do SRB become so closely linked to oil recovery processes? Some
think that SRB are imported with surface or ground waters. This hypothesis is
illustrated by a gradual increase of sulfide production after the beginning of
operations in oil fields [18]. Azadpour et al. [42] reported that SRB were absent
in thirteen core samples of petroliferous formations obtained from a wide variety
of geographical locations, depths and types of formations. Produced waters from
six of the wells were also tested and five were positive for SRB. Acetate-
utilizing SRB of the genus Desulfobacter were found in an oil field sea water
injection system [44]. In culture, they produced extensive biofilm and exhibited
high levels of hydrogenase activity, which suggests a sessile habit and a role in
the cathodic depolarization mechanism of microbially influenced corrosion.
Others have suggested that deep terrestrial subsurface reservoirs contain active
and diverse populations of microorganisms including SRB [12]. Thermophilic
SRB isolated from oil field waters in the Norwegian sector of the North Sea
were thought to be indigenous to the reservoir [45]. See chapter 14 and Ref. [46]
for discussion of microorganisms and oil reservoirs.
4.3. Controlling SRB in oil fields using biocides

Virtually all oil field water systems contain some bacteria [27], and
biocides are widely used to kill or inhibit the activities of these microorganisms,
including SRB. There are two general types of biocides: oxidizing and non-
oxidizing. Typically, oxidizing biocides (such as chlorine, sodium hypochlorite,
chlorine dioxide, chloroamines and bromine) are used in fresh water systems,
whereas non-oxidizing biocides (including aldehydes, quaternary amines,
halogenated organics, organosulfur compounds, and quaternary phosphonium
salts) are used in many different types of water systems [47].
Biocide application in large waterflooding systems presents problems
such as high cost, environmental risks [18], and worker safety. The use of
biocides is most successful in controlling unwanted activities in surface
facilities. When used to eliminate bacteria in injection water or kill SRB in the
formation, the degree of difficulty and expense increases significantly [12].
Nonetheless, application of biocides is the most common method of controlling
microbial activities in the oil field. Jack and Westlake [48] reviewed the control
of SRB in the petroleum industry.
5. NITRATE-REDUCING BACTERIA
5.1. Types of NRB

There are two major groups of bacteria that could be stimulated by the
presence of nitrate in anaerobic environments. These are chemoorganotrophs
317
(heterotrophs) that use organic compounds as electron donors and as their

carbon source for growth (Fig. 3), and chemolithotrophs (autotrophs) that
typically use reduced inorganic sulfur species as electron donors and CO2 as
their carbon source for growth (Fig. 4). The latter group is also known as the
"colorless sulfur bacteria". Figures 3 and 4 show some of the characteristics of
these NRB and their end products from nitrate reduction. These figures broadly
represent the types of bacteria that might be stimulated by nitrate, although
some, such as Thiobacillus denitrificans, and Paracoccus pantotrophus, (Fig. 4)
do not appear to have been described as oil field bacteria. Pseudomonas stutzeri
is given as an example of a heterotrophic NRB that might be stimulated by
nitrate (Fig. 3). A nitrate-respiring bacterium, that has a 100% similarity to P.
stutzeri, was isolated from an enrichment from water injectors in a North Sea oil
field [49].
Among the heterotrophs in Fig. 3 are facultative anaerobes (such as some
Pseudomonas and Bacillus species), that prefer to grow using O2 as their
terminal electron acceptors, but will grow using nitrate as their terminal electron
acceptor in the absence of O2. These are known as denitrifying bacteria, yielding
N2 as the major endproduct of nitrate respiration. There have been countless
studies of denitrifying bacteria in soils and wastewater treatment, but these
bacteria have been largely ignored in oil field studies. Denitrifying bacteria have
been shown to degrade a variety of hydrocarbons (for review see Refs. [39-40]),
and with the abundant supply of dissolved hydrocarbons in produced waters,
these heterotrophs may be stimulated by nitrate injection into a reservoir.
Another group of heterotrophic, facultative anaerobes is the ammonium-
producing, NRB, such as Citrobacter spp. (Fig. 3), other members of
Enterobacteriaceae, and a few other genera [50]. We have found no
investigations that have described ammonium production in oil field waters by
this group of facultative anaerobes. However, Telang et al. [51] mentioned an oil
field isolate (designated NH15b) that was tentatively identified as a Citrobacter
sp. or Salmonella sp. These would have the potential to reduce nitrate to
ammonium.
Using a MPN method with medium that is selective for heterotrophic,
ammonium-producing, NRB, we have observed that these NRB were detected,
but not abundant, in western Canadian oil field waters nor were their numbers
greatly increased when nitrate was added to laboratory incubations of produced
waters [Eckford and Fedorak, unpublished data]. Recently, the strictly anaerobic
ammonium-producing, nitrate-reducing bacterium, Denitrovibrio acetiphilus
was isolated from an oil reservoir model column, and it was shown to produce
ammonium in medium that contained acetate and nitrate [52].
Some SRB (Desulfovibrio spp.) have also been included as heterotrophs
that might be stimulated by the addition of nitrate (Fig. 3) because a few of these
318
reduce nitrate to ammonium [53-56]. In the presence of nitrate, some SRB will
preferentially use nitrate, and some will use both concomitantly [54].
Thiobacillus denitrificans is listed as one of the chemolithotrophs in
Fig. 4. In general, this species is not tolerant to high sulfide concentrations, but
Sublette and Woolsey [57] enriched Thiobacillus denitrificans strain F that
initially tolerated up to 1.75 mM sulfide, and later up to 2.5 mM sulfide [58].
This strain has been used in studies to demonstrate its ability to reduce H2S
concentrations in porous rock cores [59-60] and in sour produced waters
[58,61].
Gevertz et al. [62] described two novel bacterial isolates that are obligate
chemolithotrophs, using nitrate as a terminal electron acceptor, and sulfide as an
energy source. Both grow under anaerobic conditions. One isolate is a denitrifier
that closely resembles Thiomicrospria denitrificans, and it has been called
Thiomicrospria strain CVO (Fig. 4). The other isolate was called Arcobacter
strain FWKO B, and it reduces nitrate to nitrite.
Fig. 3. Examples of some heterotrophic bacteria that could be stimulated by the presence of
nitrate in anaerobic environments that contain suitable organic substrates.
319
Fig. 4. Examples of some chemolithotrophic bacteria that could be stimulated by the presence
of nitrate in anaerobic environments. See text for details.
Injection of nitrate into an oil field might also stimulate the activity of
bacteria similar to P. pantotrophus [63] (formerly Paracoccus denitrificans [64]
and Thiosphaera pantotropha strain GB17 [65]). This bacterium was isolated
from a denitrifying effluent treatment system. It is a facultative anaerobe and
facultative autotroph (Fig. 4) that uses nitrate as an electron acceptor. It grows
autotrophically with sulfide as an electron donor, or heterotrophically with a
variety of organic compounds (including acetate which is commonly found in
produced waters [66-67]) as electron donors [65]. We are not aware of any
research that has detected facultative chemolithotrophs in oil field waters.
The bacteria shown in Fig. 4 all have the capability of oxidizing sulfide
while reducing nitrate. These are referred to as nitrate-reducing, sulfide-
oxidizing bacteria (NR-SOB). Greene et al. [68] compared the sulfide tolerance
of four species of NR-SOB. In their liquid medium, sulfide was oxidized by
Thiobacillus denitrificans strain F at concentrations less than 0.5 mM, by
Thiomicrospira denitrificans and Arcobacter sp. strain FWKO B at up to 3 mM,
and by Thiomicrospira strain CVO at up to 15 mM.
320
Although only a few NR-SOB have been identified in oil field waters,
Loka Bharathi et al. [69] isolated over 100 strains of anaerobic colorless NR-
SOB from sea water and a sulfide-rich creek. Their data showed that different
isolates oxidized sulfide at different rates. For example, one isolate oxidized all
of the sulfide in the medium within 9 days, whereas another isolate oxidized
only 2.9% of the sulfide in the same time. Thus, it is likely that different NR-
SOB in the produced water from oil fields would oxidize sulfide at different
rates.
5.2. NRB in oil field waters

The presence of NRB in oil field waters has not be studied extensively.
This group of microorganisms was not even mentioned in a review entitled
"Microbiology of petroleum reservoirs" [46]. Several investigations have
enumerated NRB in oil field waters using most probable number (MPN)
methods with different media formulations. Some of the results are summarized
in Table 1, in chronological order. One of the first enumeration studies [70] used
molasses or sucrose as electron donors in the media to count heterotrophic NRB
in samples taken as near the wellheads as possible. Very low numbers ( 4 L"1)
were found in these samples.
Most of the other media formulations preferentially, but not exclusively,
cultured autotrophs. For example, the medium used by Davidova et al. [14]
(Table 1) contained only inorganic compounds except for yeast extract, with
thiosulfate serving as the electron donor. This would preferentially grow
microorganisms that are similar to Thiobacillus denitrificans. Other
investigations in Table 1 used sulfide as the electron donor with filter-sterilized
produced water from the oil field that was being studied [51, 71]. The filtered
produced water undoubtedly contained some dissolved organic compounds, so it
would support the growth of heterotrophic NRB and autotrophic NRB. The
medium used by Telang et al. [72] in Table 1, contained only inorganic
compounds except for acetate, with sulfide serving as the electron donor.
Telang et al. [72] in Table 1 described the isolation and characterization
of two autotrophic NR-SOB from an oil field in Saskatchewan, Canada. One
was designated Thiomicrospira strain CVO (formerly Campylobacter strain
CVO, [51 ]) and the other was designated Arcobacter strain FWKO B. The DNA
from these two isolates has been used extensively with a method known as
reverse sample genome probing (RSGP), first described by Voordouw et al.
[73]. Using RSGP, Telang et al. [51] (Table 1), demonstrated that the abundance
of strain CVO increased after the waterflooded oil field was treated with nitrate.
This molecular technique corroborated the increase in NR-SOB numbers
determined by the MPN method. The high specificity of the RSGP for NR-SOB
precluded the detection of other NRB in samples from four additional oil fields
321
from western Canada and west Texas [72], although culture methods detected
NRB (Table 1).
Eckford et al. [74], in Table 1, surveyed five oil fields in western Canada
for various types of NRB. Different media formulations were used to selectively
enumerate thiosulfate-oxidizing NRB, heterotrophic NRB, or NR-SOB. None of
the 18 water samples contained detectable numbers of thiosulfate-oxidizing
NRB. As was observed by Adkins et al. [70], the numbers of NRB were very
low or non-detectable near the wellheads [74]. However, NRB were detected in
source and preinjection waters, and in samples from water storage tanks and free
water knock out units. Although much of the work on NRB in oil field waters
has neglected the heterotrophic NRB, the numbers of heterotrophic NRB were
greater than the numbers of autotrophic NRB in 12 of the 15 samples compared.
In one oil field, heterotrophic NRB were found, but no autotrophic NRB were
detected (Ref. 74, Table 1).
NRB were detected in biofilms on coupons in the anaerobic part of the
water injection system of the Veslefrikk field in the North Sea [75], (Table 1).
The medium used to enumerate these attached bacteria contained organic acids
as carbon sources, providing counts of heterotrophic NRB. These numbers
increased dramatically after nitrate injection (Table 1).
The literature surveyed in Table 1 represents 15 different oil fields that
have been examined for NRB. Each of the oil fields contained detectable
numbers of NRB at one or more sampling locations. Thus, each field had a
microbial community containing NRB with the potential to be stimulated by
nitrate amendment.
6. CONTROLLING MICROBIAL PRODUCTION OF SULFIDE WITH

NITRATE ADDITION
6.1. Microbial mechanisms leading to the control of sulfide concentrations

after nitrate addition
There appear to be five mechanisms by which sulfide concentrations can
be controlled in the presence of nitrate and sulfate. The first involves the
competition between heterotrophic NRB and SRB for a common electron donor.
For example, acetate serves as an electron donor for NRB [76] and for several
genera of SRB [34]. Equations (5) and (6) illustrate that if acetate is available,
nitrate reduction yields more energy per mol of electron donor or acceptor than
does sulfate reduction [77].
322
Table 1
Detection and enumeration of NRB in oil field waters.
Refs. Oil fields Methods Comments
70 Oklahoma, MPN with molasses Samples collected near wellheads.
USA and sucrose as electron Medium would detect heterotrophic
donor NRB. MPN values were 4mL"\
71 Saskatchewan, Single-bottle MPN Oil field water contained about 120

Canada using filter-sterilized mg sulfide L"1. Method likely selected
oil field water for NR-SOB. Initial count, 104 ml/ 1 ,
supplemented with Count after nitrate injected into
nitrate reservoir, 108 mL"1.
51 Saskatchewan, Single-bottle MPN Oil field water contained about 100

Canada using filter-sterilized mg sulfide L"1. Method likely selected
oil field water for NR-SOB. Initial counts as low as 0
supplemented with mL"1. Counts after nitrate injected into
nitrate reservoir, as high as 108 mL"1.
51 Saskatchewan, RSGP NR-SOB strain CVO became

Canada dominant community member after
nitrate injection into reservoir.
72 Western Canada Single-bottle MPN Method likely selected for NR-SOB,

and west Texas, using medium with but may have grown heterotrophic
USA sulfide, acetate and NRB. Counts from 102 mL"1 to 106
nitrate mL"1 in five samples examined.
72 Western Canada RSGP NR-SOB strains CVO and FWKO B

& west Texas, detected in only one of five samples
USA examined.
14 Oklahoma, MPN with inorganic Method likely selected for thiosulfate-

USA and salts, yeast extract, and oxidizing NRB, but may have grown
Alberta, Canada thiosulfate as the heterotrophic NRB. Counts were
electron donor. typically <500 mL"1.
74 Alberta and MPN with three differ- No thiosulfate-oxidizing NRB

Saskatchewan, ent media. One selected detected in any of 18 samples. Other
Canada for thiosulfate-oxidi- NRB detected in 16 samples. Number
zing NRB, one selected of heterotrophic NRB greater than
for heterotrophic NRB, number of NR-SOB in 12 of 15
and one for NR-SOB. samples.
75 Veslefrikk in MPN with acetate, Sampled biofilms on coupons in water

the North Sea butyrate, caproate and injection system. Prior to nitrate
lactate as carbon injection, 103 NRB cm"2, after 18
sources. months of nitrate injections, a 60,000-
fold increase in NRB occurred
323
5CH3COO" + 8NO3" + 3H+ -> IOHCO3" + 4N2 + 4H2O

AG0' = -495 kJ (mol N O 3 ) " ' or
AG°' = -792 kJ (mol acetate)"1 (5)
CH3COO" + SO4= - • 2HCO3~ + HS"

AG°' = -47 kJ (mol acetate or SO4= )"' (6)
Thus, heterotrophic NRB out-compete heterotrophic SRB for electron

donors, thereby suppressing sulfide production. Oil field waters contain
dissolved organic compounds including short-chain fatty acid anions like
acetate, propionate and butyrate [12, 46, 67], as well as aromatic compounds
such as toluene and phenols that are substrates for heterotrophs. This mechanism
would stop sulfide production, but it would not remove sulfide that is present in
the reservoir or produced waters.
A second mechanism results from the increased redox potential of an
aqueous environment caused by the activities of denitrifying bacteria [78]. The
production of N2O (and maybe NO), two oxidizing agents, raises the redox
potential to above -100 mV, which is too high for the growth of SRB [31].
Nitrate reduction in laboratory experiments causes the redox indicator,
resazurin, to turn from colorless to pink [78-79]. Resazurin is 50% oxidized at
-51 mV [80]. This alteration of the redox potential in an aqueous environment
inhibits sulfide production.
A third mechanism results from the stimulation of NR-SOB in the
presence of nitrate. Two processes come into play in this case. Some NR-SOB
are denitrifiers and they produce N2O from nitrate, thereby elevating the redox
potential of the medium [78]. In addition, the NR-SOB use sulfide as their
electron donor, and oxidize it to elemental sulfur or sulfate [2]. Thus, these two
processes combine to inhibit sulfate reduction and remove sulfide that is present
in the aqueous environment. The activities of the NR-SOB have the potential to
stop sulfide production, and to remove essentially all of the sulfide in the
aqueous environment.
A fourth mechanism is nitrate reduction by SRB. Some SRB reduce
nitrate to ammonium [53-56]. The importance of this mechanism in controlling
sulfide production is largely unexplored. Jenneman et al. [78] point out that
when SRB reduce nitrate, ammonium is formed rather than N2O or N2. The
formation of N2O would be detrimental to the SRB as discussed above.
The fifth mechanism is the production and accumulation of nitrite during
nitrate reduction. Myhr et al. [49] demonstrated that the activity of the dominant
sulfate-reducing strain found in their laboratory experimental system was
inhibited by 120 uM nitrite. However, some species of SRB contain nitrite
324
reductase which reduces nitrite to ammonium [81], thereby protecting these

species from the nitrite produced by NRB [68].
6.2. Control of sulfide in wastewaters

Long before nitrate addition was considered for controlling sulfide in oil
field waters, it was used to control odors in wastewaters and receiving surface
waters. For example, in 1931, a combination of sodium nitrate and chlorinated
lime was used to control odors from Coney Island Creek in New York [82]. This
creek was described as "one of the vilest bodies of water in the United States"
[82] as a result of receiving 6,000,000 gallons (23,000,000 L) of sewage and
industrial wastewater. After the first day of chemical application, there was a
marked decrease in odor. During the month-long treatment, 10 tons (9 Mg) of
sodium nitrate were applied to the creek, and the sulfide concentrations in the
water decreased sharply.
Table 2 summarizes five studies, in chronological order, in which nitrate
was used to control odors and sulfide production in wastewaters. The first four
entries in Table 2 enhance the activities of native NRB by adding nitrate. Two of
these were large scale projects that involved nitrate applications to a river [83]
and to a sludge storage lagoon [84] for odor control. The other three studies
were laboratory-scale investigations using sewage sludge [78], oily sludges from
naval operations [85], and aqueous solutions of sulfide [58]. The latter report
described work in which Thiobacillus denitrifwans was initially used to oxidize
sulfide, and later Thiobacillus denitrificans strain F was used because of its
tolerance to higher sulfide concentrations.
Each of the attempts to control odor or sulfide production listed in Table 2
was successful. One of the studies [84] observed that nitrate amendment led to
increased redox potential followed by a reduction in odor. The increased redox
potential was observed in another study [78] and this was attributed to the
microbial production of N2O. The increase in redox potential to above -100 mV
would inhibit growth of SRB.
6.3. Laboratory studies using cores or columns

Using nitrate to control sulfide production in a petroleum reservoir
involves adding nitrate to the injection water and pumping it into the oil-bearing
formation. To be effective, the nitrate must migrate into the reservoir and be
consumed by NRB. The NRB may be present in the oil field or water handling
system, or they might be deliberately added to the oil field to stimulate nitrate
reduction. Several laboratory studies have been done to assess the effectiveness
of this process using cores or a column of sand. Five of these studies are
summarized in Table 3, in chronological order.
325
Table 2
Laboratory and field studies using nitrate to control sulfide production in wastewaters
Ref. Summary
83 Three pulp mills discharged sulfite wastes into the Androscoggin River in Maine
U.S.A. This resulted in H2S production in the river and odor problems in nearby
towns. In 1949, a total 641 tons (582 Mg) of NaNC>3 were added to the river. This
controlled H2S production and odors. Most of the nitrate was reduced to ammonium.
84 To control odor, waste sodium nitrate liquor (containing both nitrate and nitrite) was
added to a storage lagoon that held aerobically digested waste activated sludge.
Initially, the redox potential of the water was near -lOOmv, but after several months of
nitrate addition, it rose to near +300 mV. There was low odor potential when the
redox was above +100. Acetate concentrations decrease in the lagoon, and N2
production from denitrification provided mixing within the sludge.
78 Laboratory studies were done with a 10-fold dilution of sewage sludge amended with
20 mM sulfate and one of three electron donors: glucose, acetate, or H2. The addition
of 59 mM nitrate completely inhibited sulfide production. Nitrate, nitrite and N2O
were detected in the inhibited samples, and the oxidation of the redox indicator,
resazurin, was attributed to the presence of N2O. The numbers of SRB decreased with
prolonged incubation of the oxidized medium.
85 Oily sludge from a settling tank at the U.S. Navy Craney Island Fuel Depot in
Virginia was placed in serum bottles and amended with nitrate, stimulating indigenous
NRB. Sulfate reduction was diminished with 50 mM nitrate, and sulfide accumulation
was prevented with as little as 16 mM nitrate. Nitrite and nitrous oxide were products
of nitrate reduction. Sulfide was oxidized to sulfur or sulfate. The results indicated
that nitrate would be useful for preventing sulfide formation in oily wastes produced
onboard marine vessels.
58 This paper reviewed bench-scale processes developed for the sulfide removal from
gases and aqueous solutions by Thiobacillus denitrificans. When H2S was introduced
to batch anoxic or aerobic cultures of T. denitrificans, the H2S was immediately
metabolized. Oxidation of H2S to sulfate was accompanied by growth. T. denitrificans
was immobilized by co-culture with floc-forming heterotrophs and this mixture was
used to treat water that was contaminated with sulfide. The sulfide-active floe was
stable for 5 months of operation with no external organic carbon required to support
the growth of the heterotrophs. T. denitrificans strain F, which tolerates higher sulfide
concentrations, was also used in some studies.
326
Table 3
Laboratory studies using nitrate to control sulfide production columns or cores
Ref. Summary
59 This study investigated the efficacy of nitrate and the sulfide-tolerant Thiobacillus
denitrificans strain F in controlling H2S concentrations in cores of sandstone.
Formation water from a gas storage facility in Redfield, Iowa, U.S.A. was injected
into two core systems, with hydraulic retention times (HRTs) of 3.2 h and 16.7 h.
With the addition of nitrate alone, no thiobacilli were cultured from the core system,
but nitrate was consumed and the concentrations of sulfide in effluent decreased by
about 40% in the core with the shorter HRT, and 98% with the longer HRT. Thus, an
indigenous microbial community capable of oxidizing sulfide while using nitrate as
the electron acceptor was present. Inoculation with strain F reduced the effluent
sulfide by about 80% in the core with the shorter HRT.
60 The test materials for this study included core material from the St. Peter formation at
Redfield, Iowa, U.S.A. and water from the same formation, supplemented with
acetate and enriched with SRB to 107 cells ml/ 1 . The core material did not contain
large numbers of organisms capable of using nitrate, and no strain F-like organisms
were detected. When nitrate and strain F were injected into the core, sulfide
concentrations decreased, demonstrating the ability of strain F to control sulfide in
the core.
86 This work examined controlling microbial souring in anaerobic upflow columns

containing crushed Beria sandstone maintained at 60°C. Produced waters from the
Ninian North Sea and the Kuparuk North Slope oil fields were used as sources of
microorganisms, and these gave similar results. A highly anaerobic medium that
contained short-chain organic acids found in the produced waters was pumped
through the columns. Nitrate injection stimulated indigenous microbes and inhibited
souring at thermophilic temperatures. Initially, 3.6 mM nitrate was needed to inhibit
souring but later 0.36 mM nitrate prevented further souring. Nitrate was reduced to
nitrite, with no N2O, N2 or ammonium detected.
2 Brine from an oil field near Coleville, Saskatchewan, Canada was filtered,
supplemented with phosphate and nitrate and pumped into a porous (1288 mD)
ceramic core 19.1 cm long. When 5 mM nitrate was shut in the column, all of the
sulfide was removed in 3 d and the numbers of NRB increased. Under various flow
regimes, with sulfide-containing brine, sulfide removal was between 87 and 100%.
Elemental sulfur, bacteria and CaCC>3 were produced, but there was no significant
permeability changes across the core following all treatments.
49 Separate enrichments of aerobic oil-degrading bacteria, NRB, SRB and methanogens

were inoculated into a 200-cm column packed with oil-soaked silica sand. The
column was flooded with air-saturated synthetic sea water and operated under
different influent regimes for nearly 1100 d. Injecting 0.5 mM nitrate led to the
complete elimination of H2S. Inhibition of the SRB was attributed to the nitrite
produced from nitrate reduction. Three strains of heterotrophic NRB were isolated
from the column and none used H2S or S° as electron donor.
327
Four of the five studies in Table 3 detected NRB in the cores or produced
waters used in the experimental systems. In the fifth study, [49] the investigators
inoculated the column with a mixture of enrichment cultures, including NRB.
Two of the studies, Refs. 59 and 60, focused on the activities of thiobacilli.
None were detected in the cores or waters, similar to the findings of Eckford and
Fedorak [74]. Inoculating these two cores with Thiobacillus denitrificans strain
F stimulated sulfide reduction when nitrate was injected into the cores (Refs. 59-
60, Table 3).
Two of the studies [2, 86], (Table 3) relied solely on the formation water
as the source of NRB. One study supplemented the medium with short-chain
organic acids [86], whereas the other study did not supplement with organic
compounds [2]. Thus, these studies likely enriched for different nutritional types
of NRB. Nonetheless, souring was inhibited in both studies. Indeed, sulfide
production was controlled in each of the five studies summarized in Table 3.
6.4. Laboratory studies using natural microbial communities in produced

waters
Produced waters from various oil fields have been used as sources of
planktonic microorganisms in studies of the ability of nitrate to control sulfide
formation in these waters. Table 4 summarizes four of these investigations in
chronological order. In each study, sulfide removal was stimulated by nitrate
addition. In three of the reports, no organic supplementation was required to
stimulate sulfide removal. However in one case [87], two of the four oil field
waters did not respond to amendments with inorganic nutrients (nitrate and
phosphate). Sulfide removal was only stimulated after the addition of acetate or
formate plus vitamins or yeast extract, indicating that in some cases
heterotrophic NRB play an important role in the process of sulfide removal.
Eckford and Fedorak [15] demonstrated that heterotrophic NRB can be
stimulated by simply adding nitrate. This is illustrated in Figs. 5 and 6. A
produced water sample was collected from the free water knock out at the
Coleville field that has a severe souring problem. This water was used for a
serum-bottle microcosm study. Initially, the microcosm contained 2.7 mM
sulfide which increased to 3.1 mM by day 1 and then dropped below detection
by day 3 in the nitrate-amended microcosm (Fig. 5a). The sulfate concentration
increased noticeably over the first 14 d of incubation, with a total increase of 3.5
mM by day 38, closely matching the 3.1 mM decrease in sulfide. The nitrite
concentration was at a maximum of 1.8 mM on day 3 and then gradually
decreased to 0.2 mM by day 38. Figure 5b shows the results of chemical
analyses of a microcosm that was not supplemented with nitrate. The sulfide
increased to 4 mM by day 5, and the sulfate remained fairly steady at from 0.68
mM to 0.54 mM throughout the testing period. Neither nitrate nor nitrite was
detected in the microcosms. The huge increase in numbers of heterotrophic NRB
328
(Fig. 6a) during the time that sulfide was removed (Fig. 5a) suggests that these
bacteria play a role in this process. However, their role has not be elucidated.
Table 4
Laboratory studies on controlling sulfide production in produced waters by adding nitrate to
stimulate natural microbial communities.
Ref. Summary
71 Anaerobic enrichments were prepared by supplementing nitrate and phosphate to

& brine samples collected from an oil field near Coleville, Saskatchewan, Canada.
2 Within 24 to 48 h after supplementation, complete oxidation of 3 to 4 mM sulfide
was observed. Elemental sulfur was formed and the stoichiometry of the reaction
was 5HS" + 2NO3~ + 7H+ -> 5S° + N2 + 6H2O.
87 Waters from four west Texas oil fields were used to determine which amendments
were required to stimulate sulfide removal. In two of the samples, addition of 40
mM nitrate and phosphate was not sufficient to promote microbial removal of
sulfide over a 28-d incubation. However, sulfide removal was observed when
acetate or formate plus vitamins or yeast extract were added to these two waters that
had been supplemented with nitrate and phosphate. These results illustrate the
importance of heterotrophic activity in sulfide removal.
14 Two waterflooded, souring oil fields in Oklahoma, U.S.A. and Alberta, Canada were
studied. SRB and NRB were found in produced waters from both oil fields. The
majority of the sulfide production appeared to occur after the oil was pumped
aboveground, rather than in the reservoir. Sulfide production was greatest in the
water storage tanks in the Alberta field. Laboratory experiments showed that adding
5 and 10 mM nitrate to produced waters from the Oklahoma and Alberta oil fields,
respectively, decreased the sulfide content to negligible levels and increased the
numbers of NRB.
15 Produced waters from three sulfide-containing western Canadian oil fields were
amended with nitrate only. In less than 4 d, the sulfide was removed from the waters
from two of the oil fields (designated P and C), whereas nearly 27 d were required
for sulfide removal from the water from the third oil field (designated N). Nitrate
stimulated large increases in the numbers of the heterotrophic NRB and NR-SOB in
the waters from oil fields P and C, but only the NR-SOB were stimulated in the
water from oil field N. These data suggest that the stimulation of the heterotrophic
NRB is required for rapid removal of sulfide from some oil field produced waters.
329
Fig. 5. Chemical analyses of microcosms that contained produced water from the Coleville oil
field in Canada. Nitrate amended (a), unamended (b). From Ref. 15.
Bacterial enumerations were done on samples from the nitrate-amended

and the unamended microcosm. The MPN results are shown in Fig. 6. Initially,
the number of NR-SOB (2.1xlO5 ml/ 1 ) was much greater than the number of
heterotrophic NRB (4.3x102 ml/ 1 ). There was no increase in the numbers of
heterotrophic NRB (Fig. 6a) or NR-SOB (Fig. 6b) in the unamended microcosm.
In contrast, there was a rapid increase in the numbers of heterotrophic NRB and
NR-SOB by day 7 (Figs. 6a and 6b) in the nitrate-amended microcosm. The
numbers of heterotrophic NRB and NR-SOB increased 22,000-fold and 440-
fold, respectively. These proliferations occurred during the time when nitrate
consumption was the most rapid, and sulfide was depleted from the microcosm
(Fig. 5a). At day 7, the numbers of heterotrophic NRB and NR-SOB were
9.3xl0 6 ml/ 1 and 9.3xl0 7 ml/ 1 , respectively. Over the remainder of the
330
incubation, the heterotrophic NRB numbers remained high, whereas the NR-
SOB numbers dropped to near their original count (Figs. 6a and 6b). The SRB
numbers did not change in the nitrate-amended microcosm and showed a slight
increase in the unamended microcosm with a maximum at day 7 (Fig. 6c).
Fig. 6. Heterotrophic NRB (a), NR-SOB (b) and SRB (c) counts is samples from microcosms
that contained produced water from the Coleville oil field in Canada (Fig. 5). Error bars show
95% confidence intervals. From Ref. [15].
331
Laboratory studies have led to field application of nitrate or changes to

field operations. For example, the work described in references [2, 71] (Table 4)
preceded the experimental injection of nitrate into the Coleville field in
Saskatchewan, Canada [13, 71, 88], and results from laboratory studies
encouraged the implementation of nitrate injection in a North Sea oil field [75].
Based on laboratory investigations, Davidova et al. [14] observed that the rate of
sulfide production was higher in aboveground samples than in samples collected
from wellheads. At an Alberta oil field, they observed high sulfate reduction
activity in water storage tanks that had retention times of 2 to 3 d, and they
calculated that 80 kg of microbially-produced sulfide was injected into the
reservoir daily from these storage tanks. Operators of this oil field have now
eliminated the long retention time in the storage tanks, which has helped to
reduce souring.
6.5. Laboratory studies using co-cultures of bacteria

To assess the microbial dynamics and processes that occur when nitrate is
added to communities containing NRB and SRB, Voordouw and co-workers
have done several studies in which pure cultures of bacteria were mixed and
monitored (Table 5). Their work focused on the activities of the autotrophic
NR-SOB Thiomicrospira strain CVO and Arcobacter strain FWKO B. In all
cases, the NR-SOB proliferated with the addition of nitrate, and in most cases,
they removed sulfide from the medium and caused the cessation of sulfate
reduction. However, sulfate reduction was not stopped in co-cultures in which
the SRB produced nitrite reductase [68] (Table 5). Nitrite formed during nitrate
reduction is inhibitory to SRB. However, the inhibition is only transient when
SRB, that produce nitrite reductase, reduce nitrite to ammonium [81]. This work
[68] (Table 5) clearly demonstrated that the activities of these NR-SOB cannot
control sulfide production by all SRB, although the NR-SOB can oxidize the
sulfide that is formed by the SRB.
Rates of corrosion have also been studied in co-culture experiments
(Table 5) [89]. The addition of strain CVO and nitrate to a culture of
Desulfovibrio sp. strain Lac6 accelerated the corrosion rate to 0.07 mm y"1.
Lacatena et al. [90] also measured corrosion rates, but they worked with an
undefined, mixed enrichment culture in produced water. In the absence of nitrate
in produced water, the corrosion rate was 0.46 mm y"1, but with nitrate in the
produced water, the corrosion rate dropped sharply to 0.03 mm y"1. Data from
nitrate injection into a North Sea oil field showed that prior to nitrate injection
the corrosion rate was 0.7 mm y"1, but after 4 months of nitrate injection, the rate
dropped to 0.2 mm y"1 [75]. Thus, the co-culture experiments (Table 5, Ref. 89)
gave results that differed from those obtained with undefined mixed cultures
[90] and full scale operations [75].
332
Table 5
Laboratory studies using co-cultures and nitrate to control sulfide production.
Ref. Summary
72 Mixtures of strains CVO and FWKO B were incubated in medium with different
concentration of sulfide. Using RSGP, it was demonstrated that CVO dominated in
co-cultures with low (1 mM) sulfide, but FWKO B dominated with high (15 mM)
sulfide. CVO or FWKO B were co-cultured with Desulfovibrio strain Lac6. Sulfide
drop from 1 mM to 0 mM in 24 h in the presence of CVO. Over a 277-h incubation,
sulfide remained between 1 and 2 mM in the presence of FWKO B.
91 Strain CVO was added to cultures of Desulfovibrio strain Lac6 that were growing in
various concentrations of nitrate or lactate. In pure culture, sulfate reduction by the
Desulfovibrio sp. was unaffected by the nitrate concentrations up to 10 mM. Sulfide
concentrations decreased rapidly after the addition of CVO. This effect was due to the
increase in the redox potential of the medium, as indicated by the oxidation of
resazurin.
89 The influence of nitrate-mediated control of sulfide production on metal corrosion was

studied with strain CVO and a Desulfovibrio strain Lac6. The corrosion rate in
cultures of the Desulfovibrio sp. without or with nitrate was 0.01 mm y"'. The addition
of CVO to the nitrate-containing culture increased the corrosion rate to 0.07 mm y"1.
The same trend was observed when CVO and nitrate were added to a consortium of
SRB from a produced water. The increased rate of corrosion was attributed to the
formation of thiosulfate and polysulfide during the oxidation of sulfide.
68 Strain CVO was grown in co-cultures with four different Desulfovibrio strains. Two
of these did not have nitrite reductase, and their growth was stopped in the presence of
CVO as it produced nitrite and elevated the redox potential of the medium. However,
two of the strains had nitrite reductase, and they reduced the nitrite formed by strain
CVO. The SRB decreased the redox potential and continued to produce sulfide. This
illustrated that the action of strain CVO cannot inhibit SRB that possess nitrite
reductase.
6.6. Oil field observations

There have been few reports of field tests or full-scale application of
nitrate injection to control sulfide. Six reports are summarized in Table 6 (in
chronological order). Three of these focused on the extensive studies done on
the Coleville oil field in Canada during two experimental injections [13, 51, 71,
88]. The microbial community in the Coleville oil field was extensively
characterized using the RSGP method, and the produced water was the source of
the well-studied NR-SOB, Thiomicrospira strain CVO and Arcobacter strain
FWKO B. Results from nitrate addition to two oil fields in the North Sea have
also been reported [11, 75] (Table 6). These include an 8-month study [11] and a
333
long-term application, with data reported after 32 months of operation [75].

Numbers of planktonic NRB were monitored in the first five studies listed in
Table 6, and numbers of sessile NRB were reported in the last study given in
Table 6.
Three common observations were evident from the field studies
summarized in Table 6. First, NRB were present in each of the oil field waters
studied. Thus, no intentional inoculation of NRB was required to stimulate the
beneficial activities of these bacteria. Second, nitrate injection stimulated the
NRB and, in reports in which NRB were enumerated, their numbers increased
100- to 60,000-fold during the monitoring times. Third, nitrate injection
controlled sulfide production. Each of these observations was completely
predicable from laboratory studies summarized in Tables 3, 4, and 5.
6.7. U.S. Patents

The ability to stop sulfide production in oil fields, or to remove sulfide
from sour waters and petroleum are essential in petroleum recovery and
processing. The inhibition of sulfate reduction decreases corrosion and other
problems associated with SRB and provides huge cost saving to the oil field
operators. Therefore, it is not surprising that several patents have been issued for
the use of nitrate or NRB for sulfide removal or control. Table 7 lists some of
the U.S. patents dealing with these processes.
Patent no. 4,879,240 uses a mutant strain of Thiobacillus denitrificans that
is tolerant to elevated concentrations of sulfide and glutaraldehyde (presumably
strain F) to control sulfide in environments such as oil field injection waters,
reservoirs, and waste treatment of materials that contain SRB. A sulfide-tolerant
strain of Thiobacillus denitrificans is the microbial component of patent no.
4,880,542 used to remove H2S from sour waters originating from petroleum
production, anaerobic sewage digestion or other industries. These autotrophic
bacteria are co-immobilized with CaCO3 in alginate beads and placed in a
column, through which the wastewater is pumped. Nitrate or O2 can serve as the
terminal electron acceptor.
The activities of heterotrophic denitrifying bacteria are stimulated by
supplementing oil field waters (or other sulfide-containing waters) with nitrate
and an organic compound, such as acetate (patent nos. 5,405,531 and 5,750,392;
Table 7). This allows the NRB to out-compete the SRB for organic substrates. In
addition, these patents include the addition of molybdate to further inhibit SRB.
The use of the autotrophic NR-SOB Thiomicrospira (formerly
Campylobacter sp.) strain CVO and Arcobacter strain FWKO B for the removal
of sulfide from oil field brines is covered by patent nos. 5,686,293 and
5,789,236 (Table 7). The uses include aboveground treatment of sour waters or
injection of these NR-SOB into subterranean formations. The waters are
supplemented with nitrate and phosphate.
334
Table 6
Field studies and operations using nitrate to control sulfide production.
Ref. Summary
12 Ammonium nitrate (45 T) was injected into a souring oil field at the Southeast Vassar Verta
Sand Unit in Oklahoma, U.S.A. At the time of injection, no nitrate was detected in three
adjacent production wells. Forty-five days after injection, nitrate was detected at these wells,
and the sulfide concentrations were reduced by 40 to 60%.
71 In 1994, a solution of NH4NO3 and NaH2PO4 was injected into three wells in the Coleville field
in Saskatchewan, Canada. Prior to treatment, the produced waters from these wells contained
between 52 and 160 mg sulfide L"1. After injection, there were shut-in periods of between 24
and 70 h before pumping resumed. The sulfide concentrations dropped by as much as 98% of
the initial concentrations, with ranges between 40% and 60% being sustained for several hours.
The numbers of NRB increased by 100- to 10,000-fold.
88 In 1996, a solution of NH4NO3 and NaH2PO4 was injected into two injection wells in the
& Coleville field for 50 d. Two producer wells were monitored for 90 d after the injection began.
13 After 10 d, the sulfide in the producers decreased by as much as 50 to 60% of the initial
concentrations of 60 and 40 mg L"1. The cumulative sulfide removal from the two producers
were estimated to be 50 and 70 kg over the 90-d test period. The numbers of NRB increased at
least 1,000-fold during the time of nitrate injection.
51 Samples were taken from the Coleville field in 1996. These were taken 8 d before and 20, 55,
and 82 d after the injection of a solution of NH4NO3 and NaH2PO4 began. RSGP analyses,
using 47 DNA standards, showed that strain CVO became the dominant community member
immediately after injection. The abundance of CVO decreased within 30 d after completion of
nitrate injection.
11 Studies were done in the Skjold oil field in the North Sea in 2000. Three injection strategies
were used. In each case, the highest nitrate concentrations were used at the beginning of the
treatment, then the concentration was decreased. First, nitrate (4.5 to 1.7 mM) was injected into
one well for 1 month; second, nitrate (3.8 to 1.8 mM) was injected into this well plus another
well for 2 months; third, nitrate (4.4 mM to a mean of 2.8 mM) was injected into all of the
other wells for 3 months. Only one of the monitored production wells showed marked
reduction in H2S. This well was in the highly fractured zone of the reservoir, and nitrate
reached it within 24 h of the start of injection. The amount of H2S in the produced gas dropped
from 240 ppm to between 30 to 60 ppm. After nitrate addition, the numbers of mesophilic NRB
and NR-SOB increased about 10,000- and 1,000-fold, respectively.
75 Data were presented after 32 months of adding nitrate to water injected from the Veslefrikk
platform in the North Sea. Glutaraldehyde injection was stopped in January 1999, and replaced
by continuous 0.25 mM nitrate injection. Microbial counts in biofilms were monitored and
corrosion was measured by weight loss from C-steel biocoupons. After 32 months, the
numbers of SRB decreased 20,000-fold and after 18 months, the number of NRB increased
60,000-fold. Most of the NRB were heterotrophic facultative anaerobes. Sulfate-reducing
activity (measured using 35S-sulfate) decrease 50-fold. Prior to nitrate treatment, the corrosion
rate was 0.7 mm y"1. This fell to 0.02 mm y"1 after 4 months of nitrate injection.
335
Table 7
Examples of United States patents for the control of sulfide through the application of NRB.
Patent no. Inventors and Title

year
4,879,240 Sublette et al. Microbial control of hydrogen sulfide production by sulfate

1989 reducing bacteria
4,880,542 Sublette Biofilter for the treatment of sour water

1989
5,405,531 Hitzman et al. Method for reducing the amount of and preventing the
1995 formation of hydrogen sulfide in an aqueous system
5,686,293 Jenneman et al. Sulfide-oxidizing bacteria

1997
5,750,392 Hitzman et al. Composition for reducing the amount of and preventing the
1998 formation of hydrogen sulfide in an aqueous system,
particularly in an aqueous system for oil field applications
5,789,236 Jenneman Process of using sulfide-oxidizing bacteria

1998
6.8. Economics and advantages of using nitrate to control sulfide

production
Based on the trial injections at the Coleville oil field in Canada, Jenneman
et al. [88] did a cost analysis for sulfide removal using different chemicals. They
injected ammonium nitrate (cost US$0.31 kg"1) and monosodium phosphate
(cost US$2.57 kg"1) to stimulate NRB in the reservoir. The combined cost of
these chemicals was determined to be between US$0.76 and $1.19 kg"1 H2S
removed. They compared this cost to reported costs for sulfide removal from
wastewaters using hydrogen peroxide or sodium hypochlorite. With hydrogen
peroxide, the estimated cost was between US$4.40 and $17.60 kg"1 H2S
removed, and with sodium hypochlorite the estimated cost was between
US$3.96 and $13.20 kg"1 H2S removed. With data from one well, Jenneman et
al. [88] estimated the cost of using ammonium nitrate and monosodium
phosphate to be $0,018 barrel"1, or $1.80 (100 barrels)"1, of produced water
treated.
Herbert [92] compared the costs of using nitrate with those of using the
biocides glutaraldehyde or tetrakishydroxymethylphosphonium sulfate (THPS)
336
for offshore oil fields. The costs did not include the cost of transporting the
chemicals. The estimated prices per litre of the chemicals were: US$0.25 for
nitrate (as a 40% solution of CaNO3), $2.50 for glutaraldehyde (as a 50%
solution), and $4.00 for THPS (as a 50% solution). Although the cost of nitrate
was lower, the solution was continuously injected at a dose of 60 mg L"1. In
contrast, the two biocides were injected for 1 h, twice per week at a dose of
500 mg L"1. Based on treating 200,000 barrels of produced water per d, the
yearly costs for chemicals were US$575,000 for nitrate, $345,00 for
glutaraldehyde, and $500,000 for THPS. Per 100 barrel of water treated, these
costs become US$0.79, and $0.47, and $0.68, respectively.
From these two cost analyses, the use of nitrate for sulfide control is
competitive with other chemicals. The cost of treating 100 barrels of water
calculated from the data given by Jenneman et al. [88] is higher than that
reported by Herbert [92], because Jenneman et al. [88] also injected
monosodium phosphate, which is 8 times as expensive as the ammonium nitrate.
Herbert [92] used only calcium nitrate. The need to add a phosphate source to
stimulate NRB would have to be evaluated for each oil field.
Besides the cost, other factors must be considered when choosing
chemicals for controlling sulfide in produced waters. Most notably, workers
safety and potential environmental impact of spilled chemical must be
considered. Nitrate salts are far less toxic than the biocides commonly used in
oil fields, and therefore its use presents few safety issues for oil field workers.
Spilled biocides have negative affects on the environment. In contrast, nitrate is
widely used as an agricultural fertilizer, so spills on land present no major
problem. Nitrate is listed as a substance that poses little or no risk to the marine
environment [75]. However, caution must be used to avoid contamination of
fresh surface waters or potable ground waters with nitrate (or any biocide).
The use of nitrate to control microbially-produced sulfide in oil fields is a

proven biotechnology that is grossly under-used by the petroleum industry. Its
effectiveness has been demonstrated in many laboratory investigations and in
some field studies. The microbiology is adequately well-understood, although it
is not clear whether heterotrophic or autotrophic NRB play the more important
role. This may vary from oil field to oil field. Nonetheless, from the results in
the literature, nitrate amendment (and in some cases phosphate or organic acid
amendment) stimulates NRB in the oil field waters, and there appears to be little
need to add an inoculum of NRB.
Nitrate has replaced biocides in some of the oil fields in the North Sea,
and the results have been very positive. Besides controlling sulfide levels, there
is also preliminary evidence that corrosion rates are reduced [75]. In addition,
337
there are plans to use nitrate in the Gulf of Mexico when sea water injection
begins in the near future (Stephen Maxwell, Commercial Microbiology Inc.,
personal communication). In contrast, there is little or no use of nitrate in land-
based souring oil fields in North America. It is now very clear that land-based
oil field operators should seriously consider using this proven biotechnology to
control, and possibly eliminate, microbially-induced souring and the problems
associated with H2S formation.
REFERENCES
[I] G.B. Farquhar, Corros. Prev. Contr., 45(2) (1998) 51.

[2] G.E. Jermeman, D. Gevertz and M. Wright, In Proceedings of the Third International
Petroleum Environmental Conference, Vol. II, Albuquerque, NM, 1996, pp. 693-704.
[3] D.O. Hitzman and G.T. Sperl, Paper SPE 27752, presented at the 9th Symposium on
Improved Oil Recovery Tulsa, Oklahoma, USA, April 17-20, 1994.
[4] B.P. Tissot and D.H. Welte, Petroleum Formation and Occurrence 2nd edition,
Springer-Verlag, Berlin, 1984.
[5] R.C. Selley, Elements of Petroleum Geology, 2nd edition, Academic Press, NY, USA,
1998,296.
[6] H.G. Machel, Sediment. Geol, 140 (2001) 143.
[7] H.G. Machel and J.M. Foght, In R.E. Riding and S.M. Awramic (eds.), Microbial
Sediments, Springer-Verlag, Berlin, 2000, pp. 105-120.
[8] B.K. Manzano, M.G. Fowler and H.G. Machel, Org. Geochem., 27 (1997) 507.
[9] WJ. Cochrane, P.S. Jones, P.F. Sanders, D.M. Holt and M.J. Mosley, Presented at the
SPE European Petroleum Conference, London UK, October 1988, SPE paper 18368,
Society of Petroleum Engineers, Richardson, Texas, USA.
[10] L.C. Frazer and J.D. Boiling, Presented at the International Arctic Technology
Conference, Anchorage Alaska, May, 1991, SPE paper 22105, Society of Petroleum
Engineers, Richardson, Texas, USA.
[II] J. Larsen, Paper 02025 Proceedings of the NACE Expo 2002 Annual Conference and
Exposition, Denver, Colorado, USA, April 7-11, 2002.
[12] M.J. Mclneraey, K.L. Sublette, V.K. Bhupathiraju, J.D. Coates and R.M. Knapp, In E.T.
Premuzic and A. Woodhead. (eds.), Microbial Enhancement of Oil Recovery-Recent
Advances, Elsevier Science Publishing, BV. Amsterdam, 1993, pp. 363-371.
[13] G.E. Jenneman, P.D. Moffitt, G.A. Bala and R.H. Webb, SPE Prod. Facil., 63 (1999)
219.
[14] I. Davidova, M.S. Hicks, P.M. Fedorak and J.M. Suflita, J. Ind. Microbiol. Biotechnol.,
27 (2001) 80.
[15] R.E. Eckford and P.M. Fedorak, J. Ind. Microbiol. Biotechnol., 29 (2002) 243.
[ 16] American Petroleum Institute, Recommended Practices for Oil and Gas Producing and
Gas Processing Plant Operations Involving Hydrogen Sulfide, 2nd edition, API
Recommended Practice 55, Washington DC, USA, 1995.
[17] P.G. Stecher, Hydrogen Sulfide Removal Processes, Noyes Data Corp., New Jersey,
USA, 1972, p. 1.
[18] R. Cord-Ruwisch, W. Kleinitz and F. Widdel, J. Petrol. Techno!., 39 (1987) 97.
338
[19] R.N. Tuttle and R.D. Kane (eds.), H2S Corrosion in Oil and Gas Production - A
Compilation of Classic Papers, National Association of Corrosion Engineers, Houston,
Texas, USA, 1981. pp. 1018-1038.
[20] A. Ikeda and M. Kowaka, Chem. Econ. Eng. Rev., 10 (1978) 12.
[21] T.A. Bertness, G.V. Chilingarian and M. Al-Basson, In G.V. Chilingarian, J.O.
Robertson, Jr. and S. Kumar (eds.), Surface Operations in Petroleum Production II,
Elsevier Science Publishing Co. Inc., New York, USA, 1989, pp. 283-317.
[22] I.B. Beech, In G. Bitton (ed.), Encyclopedia of Environmental Microbiology, Wiley and
Sons, New York, USA, 2002, pp. 465-475.
[23] W.P. Iverson and G.J. Olson, In R.M. Atlas (ed.), Petroleum Microbiology, Macmillan
Publishing Company, New York, USA, 1984, pp. 633-683.
[24] F.A. Giuliano (ed.), Introduction to Gas and Oil Technology, 3rd edition, Prentice Hall,
Inc., Englewood Cliffs, New Jersey, USA, 1989.
[25] A.G. Collins and C.C. Wright, In E.C. Donaldson, G.V. Chilingarian and T.F. Yen.
(eds.), Enhanced Oil Recovery I: Fundamentals and Analysis, Elsevier Science
Publishing Co. Inc., New York, USA, 1985, pp. 151-221.
[26] C.C. Wright and G.V. Chilingarian, In G.V. Chilingarian, J.O. Robertson, Jr. and S.
Kumar (eds.), Surface Operations in Petroleum Production II, Elsevier Science
Publishing Co., Inc. New York, USA, 1989, pp. 319-372.
[27] S.C. Rose, J.F. Buckwalter and R.J. Woodhall (eds.), The Design Engineering Aspects
of Waterflooding, Society of Petroleum Engineers, Richardson, Texas, USA, 1989.
[28] T.R. Jack, In F.T. Premuzic and A. Woodhead (eds.), Microbial Enhancement of Oil
Recovery - Recent Advances, Elsevier Science Publishing, BV. Amsterdam, 1993, pp.
7-16.
[29] J.M. Akagi, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp.,
New York, USA, 1995, pp. 89-112.
[30] J.R. Postgate, The Sulphate-Reducing Bacteria, Cambridge University Press,
Cambridge, UK. 1979.
[31] J.R. Postgate, The Sulphate-Reducing Bacteria, 2nd edition, Cambridge University
Press, Cambridge, UK, 1984.
[32] J.M. Odom and R. Singleton, Jr. (eds.), The Sulfate-Reducing Bacteria: Contemporary
Perspectives, Springer-Verlag, New York, USA, 1992.
[33] L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp., New York,
USA, 1995.
[34] H.F. Castro, N.H. Williams and A. Orgam, FEMS Microbiol. Ecol., 31 (2000) 1.
[35] H. Cypionka, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp.,
New York, USA, 1995, pp. 151-184.
[36] N. Pfennig, F. Widdel and H.G. Triiper, In M.P. Starr, H. Stolp, H.G. Triiper, A. Balows
and H.G. Schegel (eds.), The Prokaryotes, A Handbook of Habitats, Isolation, and
Identification of Bacteria, Vol. 1, Springer-Verlag, Berlin, 1981, pp. 926-947.
[37] L.L. Barton and F.A. Tomei, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum
Publishing Corp., New York, USA, 1995, pp. 1-32.
[38] T.A. Hansen, In J.M. Odom and R. Singleton, Jr. (eds.), The Sulfate-Reducing Bacteria:
Contemporary Perspectives, Springer-Verlag, New York, USA, 1993, pp. 21-40.
[39] F. Widdel and R. Rabus, Curr. Opin. Biotechnol., 12 (2001) 259.
[40] J. Heider, A.M. Spormann, H.R. Beller and F. Widdel, FEMS Microbiol. Rev., 22
(1999)459.
[41] D. White, The Physiology and Biochemistry of Prokaryotes, Oxford University Press
Inc., New York, USA, 1995.
339
[42] A. Azadpour, L.R. Brown and A.A. Vadie, J. Ind. Microbiol., 16 (1996) 263.
[43] G.D. Faugue, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum Publishing Corp.,
New York, USA, 1995, pp. 217-242.
[44] D.E. Brink, I. Vance and D.C. White, Appl. Microbiol. Biotechnol., 42 (1994) 469.
[45] J.T. Rosnes, T. Torsvik and T. Lien, Appl. Environ. Microbiol., 57 (1991) 2302.
[46] M. Magot, B. Ollivier and B.K.C. Patel, Antonie van Leeuwenhoek, 77 (2000) 103.
[47] J. Boivin, Mater. Perform., 34(2) (1995) 65.
[48] T.R. Jack and D.W.S Westlake, In L.L. Barton (ed.), Sulfate-Reducing Bacteria, Plenum
Publishing Corp., New York, USA, 1995, pp. 265-292.
[49] S. Myhr, B.-L. Lilleb0, E. Sunde, J. Breeder, and T. Torsvik, Appl. Microbiol.
Biotechnol., 58 (2002) 400.
[50] J. Tiedje, In A.J.B. Zehnder (ed.), Biology of Anaerobic Microorganisms, Wiley, New
York, USA, 1988, pp. 179-244.
[51] A.J. Telang, S. Ebert, J.M. Foght, D.W.S. Westlake, G.E. Jenneman, D. Gevertz and G.
Voordouw, Appl. Environ. Microbiol., 63 (1997) 1785.
[52] S. Myhr and T. Torsvik, Int. J. Syst. Bacteriol., 50 (2000) 1611-1619.
[53] R.G.L. McCready, W.D. Gould, and R.W. Barendregt, Can. J. Microbiol., 29 (1983)
231.
[54] G.J. Mitchell, J.G. Jones and J.A. Cole, Arch. Microbiol., 144 (1986) 35.
[55] H.-J. Seitz and H. Cypionka, Arch. Microbiol., 146 (1986) 63.
[56] T. Dalsgaard and F. Bak, Appl. Environ. Microbiol., 60 (1994) 291.
[57] K.L. Sublette and M.E. Woolsey, Biotechnol. Bioeng, 34 (1989) 565.
[58] K.L. Sublette, MJ. Mclnerney, A.D. Montgomery and V. Bhupathiraju, In C.N. Alpers
and D. W. Blowes (eds.), Environmental Geochemistry of Sulfide Oxidation, American
Chemical Society, Washington, DC, USA, 1994, pp. 68-78.
[59] M.J. Mclnerney, V.K. Bhupathiraju and K.L. Sublette, J. Ind. Microbiol., 11 (1992) 53.
[60] M.J. Mclnerney, N.Q. Wofford and K.L. Sublette, Appl. Biochem. Biotechnol., 57/58
(1996) 933.
[61] K.L. Sublette, D.E. Morse and K.T. Raterman, Presented at the 68th Annual Technical
Conference and Exhibition of the Society of Petroleum Engineers Houston, Texas, SPE
paper 26396, Society of Petroleum Engineers, Richardson, Texas, USA, October 3-6,
1993.
[62] D. Gevertz, A.J. Telang, G. Voordouw and G.E. Jenneman, Appl. Environ. Microbiol.,
66(2000)2491.
[63] F.A. Rainey, D.P. Kelly, E. Stackebrandt, J. Burghardt, A. Hiraishi, Y. Katayama and
A.P. Wood, Int. J. Syst. Bacteriol., 49 (1999) 645.
[64] W. Ludwig, G. Mittenhuber and C.G. Friedrich, Int. J. Syst. Bacteriol., 43 (1993) 363.
[65] L.A. Robertson and J.G. Kuenen, J. Gen. Microbiol., 129 (1983) 2847.
[66] W.W. Carothers and Y.K. Kharaka, Am. Assoc. Petrol. Geol. Bull, 62 (1978) 2441.
[67] T. Barth, Appl. Geochem, 6 (1991) 1.
[68] E.A. Greene, C. Hubert, M.. Nemati, G.E. Jenneman and G. Voordouw, Environ.
Microbiol, 5 (2003) 607.
[69] P.A. Loka Bharatm, S. Nair and D. Chandramohan. J. Mar. Biotechnol, 5 (1997) 172.
[70] J.P. Adkins, L.A. Cornell and R.S. Tanner, Geomicrobiol. J, 10 (1992) 87.
[71] D. Gevertz , G.E. Jennemen, S. Zimmerman and J. Stevens, In R. Bryant (ed.),
Proceedings of the Fifth International Conference on Microbial Enhanced Oil Recovery
and Related Biotechnology for Solving Environmental Problems, Richardson, Texas,
USA, 1995, pp. 295-309.
[72] A.J. Telang, G.E. Jenneman and G. Voordouw, Can. J. Microbiol, 45 (1999) 905.
340
[73] G. Voordouw, J.K. Voordouw, R.R. Karkhoff-Schweizer, P.M. Fedorak and D.W.S.
Westlake, Appl. Environ. Microbiol., 57 (1991) 3070.
[74] R.E. Eckford and P.M. Fedorak, J. Ind. Microbiol. BiotechnoL, 29 (2002) 83.
[75] T. Thorstenson, G. Bodtker, B.-L.P Lillebo, T. Torsvik, E. Sunde and J. Beeder, Paper
02033 Proceedings of the NACE Expo 2002 Annual Conference and Exposition,
Denver, Colorado, USA, April 7-11, 2002.
[76] E.G. Beauchamp, J.T. Trevors and J.W. Paul, Adv. Soil Sci., 10 (1989) 113.
[77] R.K. Thauer, K. Jungermann and K. Decker, Bacteriol. Rev., 41 (1977) 100.
[78] G.E. Jenneman, M.J. Mclnerney and R.M, Knapp, Appl. Environ. Microbiol., 51 (1986)
1205.
[79] G.E. Jenneman, A.D. Montgomery and M.J. Mclnerney, Appl. Environ. Microbiol., 51
(1986)776.
[80] H.-E. Jacob, In J.R.. Norris and D.W. Ribbons (eds.), Methods in Microbiology, Vol. 2,
Academic Press Inc., New York, USA, 1970, 93-123.
[81] S. Moura, C. Bursakov and J.J.G. Moura, Anaerobes, 3 (1997) 279.
[82] W.T. Carpenter, Water Works Sewerage, 79 (1932) 175.
[83] W.A. Lawrance, Sewage Ind. Wastes, 22 (1950) 820.
[84] R.A. Poduska and B.D. Anderson, J. Water Pollut. Control Fed., 53 (1981) 299.
[85] K.L. Londry and J.M. Suflita, J. Ind. Microbiol. BiotechnoL, 22 (1999) 582.
[86] M.A. Reinsel, J.T. Sears, P.S. Stewart and M.J. Mclnerney, J. Ind. Microbiol., 17 (1996)
128.
[87] M. Wright, G.E. Jenneman and D. Gevertz, Proceedings of the 4th International
Petroleum Environmental Conference: Environmental Issues and Solutions in
Exploration, Production and Refining. San Antonio, Texas, USA, (on CD-ROM), 1997.
[88] G.E. Jenneman, P.D. Moffitt, G.A. Bala and R.H. Webb, Paper SPE 38768, presented at
the 1997 SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA,
October 5-8, 1997.
[89] M. Nemati, G.E. Jennenman and G. Voordouw, BiotechnoL Prog., 17 (2001) 852.
[90] R.M. Lacatena, S. Lunetto, A. Robertiello, M. Marzorati and F. de Ferra, Presented at
2nd International Conference on Petroleum Biotechnology, Mexico City, November 5-7,
2003.
[91] M. Nemati, G.E. Jennenman and G. Voordouw, BiotechnoL Bioeng., 74 (2001) 424.
[92] B. Herbert, Lecture presented at Reservoir Microbiology Forum 9: Use of Nitrates to
Control Bacterial Problems, September 9, 2003, London, UK.
Chapter 12
Regulation of toluene catabolic pathways and toluene

efflux pump expression in bacteria of the genus
Pseudomonas
J.L. Ramos, E. Duque, M.T. Gallegos, A. Segura and S. Marques
Estacion Experimental del Zaidin, CSIC, C / Profesor Albareda 1, 18008

Granada, Spain
1. TOLUENE EXTRUSION AND DEGRADATION PATHWAYS

INFLUENCE SURVIVAL IN PSEUDOMONADS
Aromatic hydrocarbons have been present in the environment for millions of

years since they are the product of the natural pyrolysis of organic material [1]
and are widely distributed in natural environments. One ring aromatic
compounds such as benzene, xylenes, ethylbenzene and toluene, which have a
logPow (logarithm of its partition coefficient in «-octanol and water) between 2.5
and 3.5, are toxic for microorganisms and other living cells because they
partition preferentially in the cytoplasmic membrane, disorganizing its structure
and impairing vital functions [2]. The toxicity of these compounds depends not
only on the inherent toxicity of the solvent but also on the intrinsic tolerance of
the species and strains. Because living organisms have been in contact with
these chemicals through long evolutionary periods of time, it is not surprising
that microbes have developed the capability to degrade them. Many of the
aromatic compound-degrading organisms are bacteria that belong to the
Pseudomonadaceae. All Pseudomonas strains that use toluene as a carbon
source have a series of mechanisms that allow them to cope with the stress
imposed by toluene itself. Nonetheless, most Pseudomonas strains are highly
sensitive to aromatic hydrocarbons such as toluene (logPow 2.5), styrene (logPow
3.6) orp-xylene (logPow 3.2); however, independent laboratories have isolated
Pseudomonas putida strains tolerant to these toxic compounds [3-7] .
A common theme in toluene tolerance in Pseudomonas is the change in
cis/trans isomerization of unsaturated fatty acids [8]. The increase in trans
isomers (which are directly synthesized from the cis isomers with no shift in the
342
position of the double bond) causes rigidification of the membrane to counteract

the increase in membrane fluidity caused by the organic solvent [9-12]. These
changes probably represent a first response that allows the cells to prepare for
the de novo biosynthesis of other components involved in tolerance towards
organic solvents. One such response was thought to be the metabolism of
organic solvents, but some of the tolerant strains are not able to degrade toluene.
For example, P. putida S12 is a toluene-tolerant strain that cannot degrade it
[13], and a P. putida DOT-TIE mutant deficient in the tod (toluene-
dioxygenase) degradation pathway is as tolerant as the wild type to a sudden
toluene shock [14]. These two observations suggest that degradation of the toxic
compound is not a key factor in solvent tolerance.
All the efflux pumps for organic solvents identified so far in gram-
negative bacteria belong to the Resistance-Nodulation-Cell Division (RND)
family. The functioning of these efflux pumps seems to be coupled to the proton
motive-force via the TonB system, although the intimate mechanism of energy
transfer remains elusive [15, 16]. Bacterial RND efflux pumps work together
with a membrane fusion protein (MFP) and an outer membrane protein (OMP).
These three components form a structure that expand both the inner and outer
membranes [17-19]. The efflux pump transporter AcrB of the E. coli RND
multidrug efflux system AcrAB-TolC was recently crystalized. The AcrB
component is as a trimer with a 50-A transmembrane region and a 70-A part
located in the periplasm that is thought to be involved in substrate recognition
[18, 20]. The crystal structure of the outer membrane protein TolC -which forms
a trimeric channel that penetrates the periplasm and contacts the efflux pump
transporter has also been reported [17]. Finally, a lipoprotein anchored to the
inner membrane which expands into the periplasmic space may serve as a
bracket for the other two components [19, 21]. This structural organization
allows substrates to be extruded into the external medium by passing the
periplasmic space [17, 18, 22].
In spite of its toxicity and thermodynamic stability, toluene can be
degraded by many microbes using a common strategy to weaken the aromatic
ring prior to its cleavage: the introduction of two hydroxyl groups that
destabilize the chemically stable resonant structure. However, bacteria have
developed different molecular mechanisms to produce this dihydroxylated
compound. So far, five aerobic pathways have been described for the bacterial
degradation of toluene, all of them leading to catechol, methylcatechols or
protocatechuate: the so-called TOL, TOD, TMO, TOM, and TBU pathways.
However, only three of them, the TOL, TOD and TMO pathways, have been
found in Pseudomonas species. The enzymes that carry out the first reaction, i.e.
the direct insertion of one or two oxygen atoms in the toluene molecule, largely
determine the pathway that is followed for degradation. The key steps involved
in the pathways are briefly described below, and they are summarized in Fig. 1.
343
Fig. 1. Pathways for the aerobic degradation of toluene.
1.1. Toluene degradation pathways

The TOL pathway, coded by the archetypical plasmid pWWO [23, 24], is
a very well characterized pathway for toluene degradation from a biochemical
and genetic point of view. This pathway is composed of two segments, an upper
and a lower pathway. Through the upper pathway, the methyl group of toluene
is sequentially oxidized to render benzoate (Fig. 1). The first enzyme of this
upper pathway is a toluene monooxygenase that oxidizes the methyl group of
toluene to yield benzyl alcohol. Subsequent oxidation of the side chain is
accomplished in two steps: first benzyl alcohol dehydrogenase renders
benzaldehyde, which is further oxidized to benzoate by a benzaldehyde
dehydrogenase. It is of interest to note that the enzymes of the upper pathway
also accept as substrates the corresponding compounds substituted with a methyl
group at the meta and/or para position and ethyl group at the meta position.
These compounds are, therefore, oxidized to 3-, 4-methylbenzoate, 3,4-
dimethylbenzoate and 3-ethylbenzoate, respectively.
The aromatic carboxylic acids are then further metabolized through the
meto-cleavage pathway, in which the benzoate, alkylbenzoate(s) is(are) then
oxidized and decarboxylated to produce the corresponding catechol(s) (Fig. 1).
344
These compounds undergo meta fission to yield 2-hydroxymuconic acid

semialdehyde or the corresponding alkyl derivatives. Metabolism of the
semialdehydes occurs via a branched pathway that rejoins at a common
intermediate, 2-oxopent-4-enoate. The semialdehyde produced from w-toluate
is hydrolyzed, whereas the semialdehyde from benzoate and p-methylbenzoate
is metabolized via the oxalocrotonate branch, which involves at least three
enzymatic steps [25]. 2-Oxopent-4-enoate is further converted in 2-oxo-4-
hydroxypentonate, which eventually renders Krebs cycle intermediates. For a
more detailed description of the characteristics of the pathway enzymes, the
reader is referred to earlier reviews and original papers [24, 26-30].
A second degradation pathway found in Pseudomonas is the so-called
TOD pathway, which was first described in P. putida strain Fl [31]. In this
pathway, probably the best known in terms of the biochemistry involved, the
first step is carried out by a three-component enzyme complex, the toluene 2,3-
dioxygenase (TOD), which renders czs-toluene dihydrodiol (Fig. 1). This
compound then undergoes dehydrogenation to yield 3-methylcatechol.
Ethylbenzene is also a substrate of this enzyme being converted into 3-
ethylcatechol. The alkylcatechols are then the substrates for ring fission in the
meta position in a set of reactions similar to those described for the TOL meta-
cleavage pathway.
Finally, a third pathway was described in P. mendocina KR1, where the
first step in toluene metabolism is carried out by the toluene-4-monooxygenase
(TMO), which hydroxylates the aromatic ring in the para position to render p-
cresol (Fig. 1 and [32, 33]). In the subsequent steps the methyl group is
transformed by a methyl hydrolase, first rendering the alcohol and then the
aldehyde derivative, which is finally oxidized to /?-hydroxybenzoate by a
dehydrogenase. The ring is further oxidized to 3,4-dihydroxybenzoate, which is
the substrate for ring cleavage in the ortho position to enter the P-ketoadipate
pathway.
As mentioned above, two additional pathways for toluene degradation
with different initial reactions have been described in strains that were originally
considered Pseudomonas: the so-called toluene-3-monooxygenase pathway
(TBU) of Ralstonia picketti, where toluene was proposed to be first oxidized to
ra-cresol and then to 3-methylcatechol [34]. However, recent work in Tom
Wood's laboratory has suggested that toluene-3-monooxygenase indeed
functions as a hydroxilating enzyme at the para position, and that 90% of
toluene is oxidized to p-cresol, which is subsequently oxidized to yield 4-
methylcatechol (Fig. 1 and [35]). The second pathway, not present in
Pseudomonas, is the one known as the toluene 2-monooxygenase pathway
(TOM) of Burkholderia cepacia where the first oxygen is inserted in the ortho
position to render o-cresol, which is then oxidized to 3-methylcatechol (Fig. 1).
In addition, another pathway for the degradation of toluene was recently
345
described in P. stutzeri OX1, a strain able to degrade o-xylene through an initial

step that involves two successive monooxygenations of the aromatic ring carried
out by the same enzyme, toluene o-xylene monooxygenase (ToMO) [36-38].
This enzyme is interesting because of its broad substrate specificity and its
relaxed regioselectivity, which make it able to hydroxylate more than one
position of an aromatic substrate [39].
A common feature to all toluene pathways from different bacteria is that
the genes involved in the different reactions are organized as operons, which are
either independent for the different segments of the pathways (i.e., the TOL
pathway) or transcribed as a single unit (i.e. the tod operon). In all cases, these
pathways are under the control of regulatory mechanisms, which are ultimately
modulated by toluene or intermediate substrates. Below we review the
regulatory networks controlling the expression of the different pathways below
and the mechanisms developed to regulate toluene efflux pumps.
2. REGULATION OF THE TOL PATHWAY

In P. putida mt-2, the genetic information that determines growth on toluene,
xylenes and other alkyl derivatives is encoded by the xyl operons of the TOL
plasmid pWWO [24]. The xyl genes are organized in four transcriptional units:
the upper and the meta operons and the xylS and xylR genes (Fig. 2). The upper
operon xylUWCAMBN codes for the enzymes necessary for the oxidation of
toluene to benzoate, whereas the meta-operon xylXYZLTEGFJQKIHencodes the
enzymes for the oxidation of benzoate, the ensuing ring cleavage and the
degradation to TCA cycle intermediates. The xylS and xylR genes, which are
transcribed divergently, are located close to the meta operon 3' end, and their
products regulate the expression of the meta and upper catabolic operons,
respectively. This pathway is no doubt the most extensively characterized
regulatory system among the aromatic degradation pathways, and, as a whole,
the TOL regulatory network can be considered a paradigm of integrated
transcriptional regulation in prokaryotes: there are two o54-dependent promoters,
each with unique features, one regulator belonging to the NtrC, one to the AraC
family of regulators, a o32/o38-dependent promoter, and several o70-dependent
promoters, all of them under superimposed global control.
2.1. Overview of the regulatory network

A scheme of the regulatory network that operates in the TOL pathway is
presented in Fig. 2. The model summarizes experimental evidence collected
since the 1970s in different laboratories to explain the expression of the pathway
enzymes in the presence of toluene, benzoate or their derivatives. Two different
regulatory circuits operate, depending on the nature of the aromatic compound
present in the culture medium [40]. When cells are growing in the absence of
346
any aromatic substrate, the xylS gene is expressed at low levels from the a70-
dependent promoter Ps2, ensuring the presence of basal levels of XylS protein.
This protein as such is not able to activate transcription. When a substrate of the
meta pathway, e.g. 3-methylbenzoate, is present in the growth medium, the XylS
protein interacts with it and becomes active to promote transcription from the
Pm promoter, which controls expression of the meta pathway. Expression from
Pm requires RNA polymerase with either a32 in the early exponential phase or
a38 thereafter. The XylR protein, which regulates its own transcription from two
o70-dependent promoters, is synthesized in sufficient amounts under all growth
conditions. When a substrate of the upper pathway, e.g. toluene, is present in the
culture medium, the binding of this effector to the protein triggers a series of
molecular events that result in the activation of transcription from two o5 -
dependent promoters: Psl for the xylS gene, and Pu, which drives expression of
the upper pathway. This latter activation requires the integration host factor
(IHF). As a consequence of Psl activation, the XylS protein is overproduced,
and even in the absence of a meta pathway effector, transcription from Pm
occurs. The current knowledge of the molecular biology of each step on the
regulatory pathway is reviewed in detail below.
Fig. 2. The TOL pathway regulatory network. Elliptical boxes indicate the inactive form of
the regulatory proteins. Shaded square boxes indicate the active form of the regulatory
proteins. Lines represent the connections between regulatory proteins and promoters, where
(+) is activation of transcription and (-) is inhibition of transcription; GR, global regulation.
The dotted line indicates transcription activation of overproduced XylS in the absence of
effector. The sigma factor(s) involved in transcription initiation are indicated above each
promoter. Aromatic substrates of the pathways that act as effectors of the regulatory proteins
are indicated. The regulatory circuits are explained in the text.
347
2.2. Regulation of the meta pathway

Two main players are responsible for the expression of the meta pathway:
the tandem Pm promoter/RNA polymerase, and the tandem XylS/effector
molecule. Extensive genetic data have been obtained to determine the
architecture of Pm promoter and the mechanisms through which the regulator
becomes active to promote transcription from Pm. As stated above, basal levels
of the XylS protein are guaranteed under any growth conditions by the activity
of the constitutive o70-dependent promoter Ps2 (Fig. 2).
2.2.1. The XylS regulator

XylS belongs to the AraC/XylS family of transcriptional regulators, which
includes at least 284 different proteins [41-44]. Members of this family present
two domains: a 100 amino acid conserved domain involved in DNA binding (the
C-terminal domain in most of the proteins of the family, including XylS), and a
nonconserved domain (the N-terminal domain in XylS) involved in effector
binding and dimerization.
Interactions between XylS and its effector have been studied by analyzing
the ability of the protein to activate transcription in the presence of a wide range
of substituted benzoates, as well as by selecting XylS mutants with altered
effector specificity [16, 43, 45, 46]. Recognition of ring substituents strongly
depends on the position and nature of the chemical substituent, with meta being
the most permissive position in the aromatic ring (-CH3, -CH2H5, and -OCH3
groups and F, Cl, Br and I atoms are permissible substituents), whereas positions
ortho and para pose some restricitions to substituents (-CH3 and F and Cl atoms
are allowed but not -C2H5 and I atoms) [45]. Although disubstitutions involving
positions o- and m-, and m- and p-, are permissible, other combinations are
usually non permissible, which suggests that interactions between the effector
and the regulator are nonsymmetrical. Ramos et al. [48, 49] and Michan et al.
[46] isolated and sequenced a series of mutant regulators able to recognize
substituted benzoate effectors that are not recognized by the wild-type regulator.
Key residues clustered in two noncontiguous segments in the N-teminus end of
XylS. Mutations were found to be clustered at positions 37-45, 88-92, 151-155,
and around residues 256 and 288. These finding suggest that the recognition
pocket for XylS effectors may be composed of two or more noncontiguous
segments of its primary sequence. Arg-41 seems to be a critical residue for
interaction(s) with effectors, as changes in this position result in many different
phenotypes. For example, XylSArg41Gly is a mutant regulator whose ability to
recognize o- and p-methylbenzoate was lost, although it retained its capacity to
be activated by w-methylbenzoate. Substitution of Arg41 by Leu resulted in a
mutant that was unable to respond to benzoate effectors [46]. Therefore,
348
transition from the inactive to the active form may be mediated by effector
binding. Recent studies further pinpointed residues Asp 137 and His 153 as
crucial for interactions with the effector molecule [50]. In addition to
influencing effector specificity, these two residues were shown to contact
specific residues in the RNA polymerase a subunit carboxy terminal domain (a-
CTD) [50] .
XylS mutants such as XylSArg41Cys, XylSPro37Gly XylSGly44Ser,
XylSSer229Ile, XylSAsp274Val, and XylSAsp274Glu mediated transcription
from Pm in the absence of effectors [46, 47]. These results support the
hypothesis that XylS exists in vivo in a dynamic equilibrium between an inactive
and an active form, with respect to transcriptional stimulation. Within the
family, some regulators such as MarA are present in solution as monomers,
whereas most of the members of the family are found as dimers [51-54]. XylS is
likely active as a dimer and in vivo and in vitro assays have shown that Leu 193
and Leul94 in XylS play a crucial role in dimerization [55].
It is predicted that the DNA binding domain of XylS consists of seven a-
helix units which fold to assemble two helix-turn-helix (HTH) motifs that
interact with two neighboring major grooves on one face of the target DNA.
Involvement of the XylS C-terminal domain in DNA binding was first predicted
after the finding of mutations in this domain that rendered mutant regulators able
to promote high transcription levels in the absence of effectors [47, 48].
Mutation analysis of the predicted conserved positions of the HTH motifs of
XylS showed that the most conserved positions in the family seem to be
essential to preserve the structure of this domain [56]. Deletion of the 209 N-
terminal residues of XylS rendered a C-terminal domain-protein able to bind Pm
promoter and, when overproduced, able to activate transcription in vivo to levels
similar to those in the wild type protein. However, activity was clearly reduced
when the C-terminal fragment was synthesized at physiological levels. As
expected, the truncated protein was not responsive to effector-mediated control
[57].
2.2.2. The Pm promoter

XylS-mediated transcription activation from Pm requires a DNA fragment
extending to position -70 upstream from the transcription start site. The DNA in
this region exhibits a 40° bend centered between positions -41 and -46 [58]. The
XylS binding site in the Pm promoter was first defined through site-directed
mutagenesis [59-63] and further confirmed by in vitro and in vivo footprint
assays [60, 64, 65]. The XylS binding site in Pm consists of two directed repeats
(5'-TGCAN6GGNTA-3') spanning positions -34 to -68, and overlapping the
RNA polymerase biding site by 1 bp [58, 60]. This overlap with the RNA
polymerase binding site is also observed in several other members of the family
[66-68].
349
In vivo transcription from Pm is mediated by two different RNA

polymerases, depending on the growth phase. In the early exponential phase,
RNA polymerase with a32 is necessary for transcription, whereas in late
exponential growth phase and in stationary phases, a38 is required [65, 69, 70].
Despite the alternation in RNA polymerase, the same transcription start site is
detected along the growth curve, suggesting that the same promoter is used by
both forms of RNA polymerase. Transcription of o32-dependent promoters
requires the stabilization of this sigma factor, which takes place through a series
of events known as heat-shock response [71]. In the TOL pathway regulatory
network, aromatic effectors are required not only because of their direct role in
XylS activation to promote transcription, but also to trigger the heat-shock
response and provide the appropriate RNA polymerase for transcription in the
exponential phase [70]. The direct involvement of the two sigma subunits in Pm
transcription was further supported by the finding that a mutant Pm promoter
with an altered XylS binding site, combined with the mutant regulator
XylSGly44Ser, was able to overcome the requirement of a38 for transcription in
the stationary phase [69]. Moreover, the mutants XylSAspl37Glu and
XylSHisl53Gln were able to stimulate transcription from Pm in the absence of
a38 [50].
2.3. Regulation of the upper pathway

In the presence of toluene or a substrate of the upper pathway, P. putida
ensures the coordinated expression of the two catabolic operons, so that the
aromatic compound is totally degraded to TCA intermediates. The key regulator
in this process is XylR, which is responsible for the coordinated expression of
the o54-dependent promoters Pu and Psl. Pu drives transcription of the upper
pathway and Psl increases the synthesis of XylS, responsible for meta pathway
expression (Fig. 2).
2.3.1. The XylR regulator

XylR protein belongs to the NtrC family of enhancer-binding proteins
(EBP) [72-74]. It contains the four distinctive domains of this family: i) An N-
terminal A-domain responsible for signal reception, i.e., interaction with the
effector molecule (see below), ii) the A-domain is linked to the central domain,
called domain C by the short B-domain (Q-linker), iii) the C-domain is involved
in ATP binding and hydrolysis, and plays a major role in the isomerization of
the o54-dependent promoters from close to open complexes, and iv) the C-
terminal D-domain contains the HTH motif for DNA binding.
XylR is activated by aromatic compounds with a wide variety of
substitutions such as alkyl groups of different length or oxidized intermediates
of the toluene methyl group, such as benzyl alcohol, benzaldehyde and
derivatives [75, 76]. Early evidence indicated that the A-domain was the signal
350
receptor from the environment and the direct sensor of the aromatic molecule.
This was surmised from the ability of the protein to activate transcription from
the Pu promoter in the presence of a wide range of toluene derivatives, and by
experiments with XylR mutants with altered effector specificity [75, 76, 77, 78].
These data, obtained in the heterologous host E. coli, led to the conclusion that
XylR was directly activated via interaction with the effector. XylR is closely
related to the DmpR regulator for phenol degradation in Pseudomonas sp.
CF600, which recognizes phenol and derivatives, but not toluene, as an effector
[79]. Further evidence for the direct interaction of the A-domain of these
proteins with the effector molecule came from the construction of a chimeric
protein in which the receptor domain of DmpR was replaced by the
corresponding domain of XylR, resulting in a hybrid regulator that responded to
toluene for activation of the Vo promoter of the phenol degradation pathway
[80]. DNA shuffling assays to create hybrid A-domains between DmpR and
XylR confirmed that the residues 110 to 186 of both proteins were responsible
for the effector profile of these regulators [80].
The A-domain operates as an intramolecular repressor of the central
activating domain of the protein [81, 82]. In fact, a XylR derivative in which the
A domain has been deleted is able to activate Pu in the absence of an aromatic
effector. The truncated derivative of XylR depleted of the A domain and
therefore unable to respond to effector-dependent modulation showed intrinsic
ATP binding and hydrolysis activity, located in the central activation domain
(C-domain). This activity was stimulated by the presence of a DNA fragment
containing the native XylR binding site in Pu (UAS) [83]. Furthermore, binding
of ATP to this truncated protein alone was able to induce conformational
changes in the protein. Initially, a cyclic model to explain XylR activation of Pu
was proposed by Perez-Martin and de Lorenzo [83], according to which ATP
binding to the XylR central domain led to multimerization of the regulator
bound to its UAS in Pu, followed by ATP hydrolysis. This in turn triggered a54-
dependent transcription initiation in Pu, allowing the system to return to its
initial disassembled state [83]. Recently, Shingler and co-workers studied the
analogous regulator DmpR, and suggested an alternative mechanism to explain
effector-dependent activation of a54-dependent promoters. According to their
model, DmpR dimers are activated after binding of the effector molecule to the
A domain, followed by a conformational change that allows ATP binding to the
central domain and oligomerization to a hexameric conformation, probably
required to promote transcription initiation. Finally, ATP hydrolysis leads to
dissociation of the hexameric structure and dissociation of the effector [83].
2.3.2. The Pu promoter

Pu promoter belongs to the class of promoters dependent on the
alternative sigma factor o54 (Fig. 2), and shows the typical architectural
351
organization typical of the promoters of this class. A -12/-24 sequence is

responsible for the recognition of o54-RNAP [84, 85], and upstream activator
sequences (UAS) ensure XylR binding between positions -120 and -175 [86,
87]. An IHF binding site between these two elements (positions -52 to -79) [87]
is required to bring the regulator into contact with the RNA polymerase by
looping out the intermediate sequence. Direct evidence that IHF causes Pu to
form an open bend was obtained by atomic force microscopy, where an angle of
123° was measured between the UAS and the RNA polymerase binding site
[88]. However, Bertoni et al. [89] recently found that a second upstream
element, reminiscent of the so-called a-CTD-binding UP elements of o70-
dependent promoters, was important in Pu recognition by a5 -RNAP, and that
IHF-binding played an additional role in Pu. This role consisted of a5 -RNAP
recruitment to its promoter, determined by the correct positioning of the UP-like
element with respect to the -12/-24 binding site after IHF-dependent DNA
bending [39]. That IHF-mediated o54-RNAP recruitment to Pu was reproduced
in vitro with the XylR regulator acting from solution, i.e., in the absence of UAS
[90]. These authors have shown that RNA polymerase binding is an important
rate-limiting step in Pu activation, that could become crucial when the enzyme is
present at a low concentration [90].
2.4. Expression of the regulatory genes

The level of the XylR and XylS proteins is finely modulated in vivo. This
fine regulation takes place in the 300-bp intergenic region between the xylR and
the xylS genes, which contains four promoters: the two a70-dependent tandem
promoters of xylR, Prl and Pr2, divergent from the two that drive transcription
of xylS, the a54-dependent promoter Psl and the a70-dependent promoter Ps2.
The binding sites for the different regulatory proteins in this short region totally
or partially overlap; thus XylR UASs in Psl partly cover the two RNA
polymerase binding sites of Prl and Pr2. In addition, two sequences with
different affinity for IHF are found which overlap the Psl -12/-24 RNA
polymerase binding site and one of the UAS [91-93]. As a consequence, the
levels of expression normally observed in the wild-type strain for each promoter
are far below maximum values, suggesting the involvement of repressive
element(s) in the maintenance of appropriate levels of expression.
In fact, as expected from the promoter architecture described above, XylR
strongly represses its own synthesis [75, 89, 94-97]. Activation of the Psl
promoter and autoregulation of XylR expression seem to be the consequence of
the binding of XylR to the UASs for Psl that overlap the Prl and Pr2 promoters.
This is in agreement with the finding that XylR is consistently bound to target
sequences [75]. Thus, xylR promoters may be subjected to two levels of
repression depending on the mechanism of XylR activation discussed above: an
ATP-independent level resulting from non-cooperative interaction of
352
nonactivated XylR with the Psl UAS [94], and an ATP-dependent repression
level resulting from the cooperative oligomerization of activated XylR at the
UAS in Psl [89]. As soon as the protein is activated and the UAS is strongly
bound by the regulator, XylR expression is minimized, thus limiting the period
of time during which the Ps 1 and Pu promoters of the TOL plasmid are in an
activated state [89].
The role of IHF in Psl expression deserves special attention. Analysis of
Psl activity in isogenic IHF-plus and minus backgrounds showed that in the
presence of toluene, the highest levels of expression were achieved in the
absence of IHF [97]. This may reflect a better access of either XylR to its
binding site or of o54-RNA polymerase to the -12/-24 region of Psl, or both. On
the other hand, it may be the consequence of structural hindrance, as the DNA
bending induced by IHF bound to two sites may give rise to a highly ordered
structure that restricts the access of regulatory proteins to the corresponding
promoters. The high level of expression from Psl in the IHF-minus background
in the presence of effectors contrasts with the diminished expression from the
TOL plasmid Pu promoter for the upper pathway in an IHF-deficient
background. The most noticeable difference between the two promoters is the
position of the IHF binding site, which in Pu lies between the UASs and the -
12/-24 box. In addition to affecting Psl expression, the close proximity of the
regulatory sequences in the intergenic region results is a high expression level
from Ps2 in the absence of a54, i.e., when RNA polymerase is unable to bind to
the Psl promoter [97].
In general, the physiological consequence of this organization is that in
the absence of any effector in the culture medium, Ps2, Prl and Pr2 promoters
are slightly repressed. In the presence of toluene, activation of Psl causes a
stronger repression of both xylR promoters. As a result, the level of XylR
decreases at approximately 30 monomers per cell [98], which are apparently
sufficient to promote high expression of both xylS and the upper pathway. Under
these conditions the XylS protein is overproduced, which allows induction of
expression from the Pm promoter even in the absence of meta pathway
substrates. Therefore in the presence of toluene or a substituted derivative, both
the upper and the meta pathways are coordinately expressed to optimize total
degradation of the aromatic (Fig. 2).
2.5. Integration in the bacterial metabolism

The expression of the TOL pathways is tightly regulated according to the
carbon sources available for growth [99-104]. The regulation is exerted mainly
at the level of the two o54-dependent promoters Pu and Psl, and was first
observed as a delay in the induction of expression from these promoters when
cells were induced in a rich complex medium [102, 103]. Because both
promoters were silent in this medium during rapid exponential growth, and
353
expression appeared only at the end of the exponential phase, this behavior was
named exponential silencing [105]. However, it is worth noting that exponential
silencing is only observed in rich medium; in defined minimal media with
succinate (for example) as a carbon source, expression of both Psl and Pu is
observed immediately after induction [102, 103, 106]. Growth rate as a
determinant of Pu and Psl expression was ruled out through a series of
continuous culture experiments that compared different growth rates controlled
by different limiting substrates. The results led to the conclusion that repressive
conditions correlated with a high energy status of the cells [99, 100]. In other
words, in all conditions tested where excess carbon was available, the system
was repressed. However, when oxygen was the growth-limiting substrate, a
situation where carbon was also present in excess, a certain degree of
derepression was observed although activity never reached maximum values
(Fig. 3).
Fig. 3. Integration of cell signals that lead to modulation of the expression of the Pm and Pu
promoters.
354
In minimal medium batch cultures amended with casamino acids, carbon

sources such as glucose, gluconate or a-ketoglutarate expression from Pu was
inhibited. In general this phenomenon reflects the preferential and sequential use
of the different carbon sources present in a mixture; hence it could be considered
a typical case of catabolite repression. However, catabolite repression in
Pseudomonas seems to be exerted through mechanisms that differ greatly from
the classical CRP-dependent phenomenon observed in E. coli. Similar
physiological regulations of other catabolic operons has been observed in
Pseudomonas [79, 107].
The molecular basis for the observed repression of Pu and Ps 1 expression
remains unknown. Several alternatives have been envisaged and the current
picture is compatible with the partial involvement of different systems in the
global regulation response. Originally, the phenomenon known as exponential
silencing was shown to be due neither to a late activation of XylR by the
aromatic effector nor to changes in the intracellular levels of IHF during growth
[105]. However, recent findings obtained with in vivo UV laser footprint
technology have shown that IHF occupancy of its target site in Pu increases
upon entry into the stationary phase, in parallel with an increase in IHF
concentration in the cell. Therefore, this could explain the increase in Pu activity
with growth phase in batch cultures. Nevertheless, these results do not preclude
the integration of physiological repressive signals through additional
mechanisms [108].
The a54 factor of RNA polymerase has also been considered a possible
target of global regulation. Overproduction of the a54 factor allowed Pu to
partially overcome exponential silencing, although not carbon source-dependent
repression [105]. Because a54 protein levels remained approximately constant
during growth under physiological conditions [109], exponential silencing of Pu
may be caused ultimately by changes in the activity of the sigma factor itself.
The ATP-dependent physiological protease FtsH, a member of the so-called
AAA family of ATPases responsible for the stability of various transcription
factors such as a32 [110], has been shown to play a key role in Pu expression.
FtsH is required for XylR-mediated Pu transcription in a process that is not
related to XylR or IHF, but which is rather exerted through a mechanism that
involves the loss of a54 activity (Fig. 3). In fact, overproduction of a54 restored
about 60% of Pu activity in the absence of FtsH [111]. Furthermore, the
overproduction of FtsH partially relieved exponential silencing of Pu expression.
In this connection, the target of FtsH activity seems to be an additional factor
that downregulates a54 post-translationally, or that hinders the contacts of o54-
RNAP with the promoter or with the activator. Interestingly, E. coli FtsH levels
are controlled in response to physiological signals and its proteolytic function is
stimulated by the proton-motive force [112].
355
Finally, exponential silencing and carbon source-dependent repression can

be distinguished at a genetic level. The rpoN gene, which codes for the o54
sigma factor required for Pu activity, is the first gene of an operon found in
gram-negative bacteria that includes 4 additional ORFs. Two genes of this
cluster, ptsN and ptsO, code for two proteins, IIANtr and NPr respectively that
show similarity to phosphotransferases belonging to the phosphoenol
pyruvate: sugar phosphotransferase system (PTS) family. To understand the
putative role of the ORFs in this cluster in the global control of Pu, knock-out
mutants were generated and analyzed. A mutant in the ptsN gene (which
encodes IIANtr) relieves C source inhibition, but not the exponential silencing of
Pu [113]. The ptsO gene together with ptsN operates in Pu regulation, where
phosphorylation of the pteOencoded protein NPr is necessary for the normal
response of Pu to glucose [114, 115]. NPr probably modulates IIANtr activity,
promoting its dephosphorylation. This increases the concentration of
unphosphorylated IIANtr, and as a result inhibition of Pu disappears.
Interestingly, a site-directed ptsN mutant in the conserved phospho-acceptor
His-68 residue made Pu unresponsive to the presence or absence of glucose,
thus supporting the notion that phosphorylation of IIANtr mediates the C source
inhibition of the promoter [116].
The observations reported above suggest that the global regulation of
TOL pathway expression responds to several complex mechanisms that act at
the level of o54-dependent promoters (Fig. 3).
3. REGULATION OF THE TOD PATHWAY
Pseudomonas putida strains Fl and DOT-TIE use toluene and ethylbenzene as

the sole carbon source [6, 117, 118]. The catabolic genes for the complete
conversion of toluene/ethylbenzene to TCA cycle intermediates are clustered in
a single unit, the tod operon, as todXFClC2BADEGIH [14, 119-123] and these
genes are coordinately induced by toluene/ethylbenzene (Fig. 4) [12, 124].
Wang and coworkers [121] and Mosqueda and coworkers [14] identified a
single promoter upstream from the todX gene, whose -10 and -35 regions
showed homology with P. putida o70-dependent promoters [125]. Expression of
the tod catabolic operon is regulated by todST gene products, which are located
as a separate transcriptional unit downstream of todH, the last gene of the
operon. Translational coupling between todS and todT ensures the balanced
transcription of both genes [119]. TodS and TodT proteins belong to the family
of two-component signal transduction systems. The regulation mechanism of
two-component control systems is based on a histidine-aspartate phosphorelay
circuit working between the two components. One of them is a sensor that
autophosphorylates in response to an external signal, and the other one is the so-
called response regulator, which receives the phosphate from the former and
356
activates transcription. TodS, the 108-kDa sensor protein, is a member of the

hybrid class of histidine kinases and possesses multiple protein domains. The N
terminus of TodS contains a motif characteristic of the basic region leucine
zirjper (bZIP), that consists of a region with several basic residues which
probably contact DNA, and an adjacent region containing a heptad repeat of
leucine, the leucine zipper. Indirect evidence suggests that in TodS, the leucine
zipper mediates dimerization, which is required for DNA binding [119]. A
duplicated histidine kinase motif, each element of which is characterized by five
short sequence blocks that are highly conserved, flanks a receiver domain of the
response regulator located at the center of the protein adjacent to one set of a
PAS domain, known as a signal sensor and found in various redox, light and
hydrocarbon sensor proteins [126, 127]. The todT gene encodes a 206-residue
protein with an estimated mass of 23 kDa. Analysis of the amino acid sequence
of TodT shows significant similarity with response regulators of two-component
signal transduction systems [128, 129]. TodT consists of an N-terminal receiver
domain to accept the phosphoryl group from TodS, and a helix-turn-helix (HTH)
DNA-binding domain. The TodT protein was shown to specifically bind to the
todX promoter region at a 6-bp inverted repeat located 105 bp upstream from the
transcription start site, and known as the todbox [119].
Fig. 4. Organization of the tod genes and its regulatory circuit. Top. The tod genes are
organized in two operons expressed from sigma-70 dependent promoters upstream from todX
and todS. The TodS protein (O) is synthesized in an active form that in the presence of
toluene phosphorylates TodT ( • ) , which functions as the activator of the todX...
357
This two-component signal transduction system positively regulates the

tod operon [119]. TodS is predicted to function as a hybrid kinase that uses ATP
to autophosphorylate a specific histidine residue in response to toluene. TodT as
a response regulator probably receives the phosphate at a conserved aspartate
(Asp-56) and then mediates transcription activation at the todX promoter.
Although biochemical evidence supporting this model is consistent [119], in
vivo studies confirming the role of each member of the tandem are scarce.
Furthermore, the physiological behavior of this system under different growth
conditions and the precise mechanism through which the presence of toluene
triggers changes in tod gene expression in vivo are not completely understood.
On the other hand, the role of TodS as a sensor for directly detecting inducers
has not been clearly demonstrated, nor has the role of these proteins in self-
regulation been clarified. Expression from the todX promoter occurred in
response to toluene, ethylbenzene, styrene, xylenes and other aromatic
hydrocarbons, although the greatest level of expression was obtained with
toluene. Expression from the todS gene was constitutive regardless of which
aromatic was tested [14].
It is interesting to note that both TodS and TodT proteins are required for
chemotaxis to aromatic hydrocarbons in P. putida Fl [130]. This observation
indicates that both catabolism and chemotaxis are coordinately regulated at the
transcriptional level.
4. REGULATION OF THE TOLUENE-4-MONOOXYGENASE

PATHWAY
Genes involved in toluene degradation in Pseudomonas mendocina KR1 are

organized in an unusual manner: the five proteins that make up the
multicomponent enzyme toluene-4-monooxygenase, which carries out the
primary oxidation of toluene to />-cresol (Fig. 5), are coded by the
tmoXABCDEF gene cluster [33, 131], where tmoX is homologous to the outer
membrane protein that codes todX from P. putida Fl [132]. Complete toluene
oxidation to TCA cycle intermediates requires another operon, pcuCAXB for p-
cresol oxidation to ^-hydroxybenzoate, and the pobA gene, which presents two
alleles, pobAl and pobA2, for the further transformation of this compound into
protocatechuate. The pcuCAXB and pobA genes are not clustered with the tmo
genes on the P. mendocina chromosome. These operons and the pobA genes are
also independently regulated [133]. The regulatory elements involved in the
toluene degradation pathway in Pseudomonas mendocina KR1 are summarized
in Fig. 5.
358
Fig. 5. Regulation of toluene degradation in P. mendocina KR1. The upper part of the figure
summarizes the oxidation of toluene to protocatechuate and the genes needed for each
reaction. The lower part shows a scheme of the cluster of tmolpculpob genes in this strain.
tmoABCDEF, toluene-4-monooxygenase genes; c, putative cytochrome c gene; pcuRCAXB,
/?-cresol utilization genes; pobRl and pobAl, regulator and p-hydroxybenzoate hydroxylase,
respectively; tmoST, two-component signal transduction system; p-HBOH, />-hydroxybenzyl
alcohol; />-HBHO, p-hydroxybenzyl aldehyde; p-HBA, p-hydroxybenzoate; PCA,
protocatechuate. Reproduced with permission from [132].
4.1. Regulation of tmo operon expression

The transcription initiation point from the tmo operon has been mapped
and the sequence upstream has revealed strong identity with the promoter of the
tod operon of P. putida Fl, including an inverted repeat located at position -100
and an almost identical P. putida F1 tod box. This suggested the involvement of
a regulatory system similar to TodS-TodT for the transcriptional control of the
P. mendocina toluene degradation pathway. In fact, a novel two-component
signal transduction system was recently described in P. mendocina KR1 [132].
Transcription from the VtmoX promoter, which directs the expression of the
tmoXABCDEF gene operon, is induced in the presence of toluene or/>-cresol by
a two-component system made up of TmoS and TmoT, which are 83% and 85%
identical, respectively, to the TodS and TodT proteins described above.
Furthermore, transcription from P,mox and P,O<K can be mediated by TmoS-TmoT
or TodS-TodT in the presence of toluene, revealing cross-regulation between
these two catabolic pathways. In the absence of the two-component regulatory
system, no expression of the tmo operon is observed, thus confirming the role of
359
tmoST as primary regulators. Surprisingly, in P. mendocina the tmoST genes are

not located close to their target operon tmoXABCDE, as is the case in P. putida
Fl, but are found approximately 700 bp downstream from the pobA 1 gene forp-
hydroxybenzoate hydroxylase, in a different region in the chromosome. It is
interesting to note that truncated sequences showing homology to transposases
are found downstream from tmoST and upstream from the tmo operon (Ramos-
Gonzalez, M.I., unpublished). In fact, early mobilization experiments found
good linkage between the pcu and pob operons together with their
corresponding regulatory genes, but no linkage between this cluster and the
tmoXABCDE operon [133]. Sequencing of the tmoST surrounding region
recently confirmed this distribution [132].
4.2. Regulation of the lower segments of the toluene degradation pathway

In P. mendocina KR1, the further degradation of />-cresol to TCA cycle
intermediates requires the catabolic operons pcuCAXB, pobA, and the gene
necessary for ring opening and subsequent degradation.
The pcuCAXB operon is required to transform the/?-cresol produced from
toluene by toluene monooxygenase into /?-hydroxybenzoate. The pcuCAXB
operon is regulated by the divergently transcribed pcuR gene, which belongs to
the NtrC family. Expression analysis of the pcuCAXB operon using a reporter
gene fused to the promoter showed that only substrates of the pathway, such as
p-cresol, p-hydroxybenzyl alcohol or p-hydroxybenzyl aldehyde, were effectors
of PcuR. Neither toluene nor />-hydroxybenzoate was able to induce expression
of the pathway [134]. The pobA gene codes for/>-hydroxybenzoate hydroxylase,
which converts /?-hydroxybenzoate into protocatechuate, the substrate for ring
fission [135, 136]. It has been shown that in this strain, expression of the pobA 1
gene is under the control of the divergently transcribed pobRl gene. However,
no data are available on the specific molecular mechanisms involved in this
process.
5. OVERVIEW OF THE REGULATION OF THE TBU AND TOM

PATHWAYS IN RALSTONIA PICKETTII AND BURKHOLDERIA
CEPACIA.
Ralstonia pickettii PK01 (formerly Pseudomonas pickettii PK01) is able to grow

on toluene, phenol, and benzene as the sole carbon and energy source [34]. The
pathway responsible for the degradation of these compounds to TCA cycle
intermediates is called the toluene-3 -monooxygenase pathway (also tbu pathway
for toluene benzene utilization) (Fig. 1). The genes that encode enzymes for the
pathway are grouped in three operons. The tbuAlUBVA2C and tbuT operon
encodes the initial toluene-p-monooxygenase and the transcriptional activator
TbuT [137]. The tbuD operon encodes phenol/cresol hydroxylase [138, 139],
360
and the tbuWEFGKIHJ operon encodes enzymes of the /weta-cleavage pathway

for the conversion of catechol and methylcatechols to tricarboxylic acid cycle
intermediates [140]. In addition, tbuX codes for a putative facilitator of toluene
entry into the cell [141], and is located immediately downstream from tbuT. In
turn, TbuT, a member of NtrC family of transcriptional activators, controls
transcription of each of these operons in response to aromatic effector
compounds [137]. TbuT is activated by aromatic effectors (toluene, benzene and
ethylbenzene) and trichloroethylene. Expression of tbuT auto-regulated, as
suggested by the finding that it is linked to the expression of the tbuAlUBVA2C
operon by read-through transcription of tbuT from the toluene-/>-
monooxygenase promoter. As a result, transcription of tbuT is low when the
toluene-p-monooxygenase operon is not induced and high when expression of
tbuAlUBVA2C is induced by toluene. Thus, the toluene-/>-monooxygenase
promoter drives the cascade expression of both the toluene-p-monooxygenase
operon and tbuT, resulting in a positive feedback circuit [137]. Upstream from
the o54-dependent toluene-p-monooxygenase promoter (P/&«AI), a DNA region
with dyad symmetry may serve as the TbuT-binding site [137, 142]. Two
additional regulatory genes, TbuS and TbuR, are located upstream from tbuD. In
the absence of effectors, TbuS represses transcription of tbu WEFGKIHJ. The
current view of the regulatory mechanism suggests that in the presence of the
effectors phenol or w-cresol, these compounds interact with TbuS and the
effector-TbuS complex acts as a transcription activator of tbuWEFGKIHJ. In
addition, the effectors that interact with TbuR form complexes able to activate
transcription oftbuD [34].
Burkholderia cepacia G4, formerly Pseudomonas cepacia, degrades
toluene through a unique initial step that involves toluene-o?t/j0-monooxygenase
(TOM) (Fig. 1). This strain also contains a catechol 2,3-dioxygenase for the
meta-cleavage of methylcatechol. Because of its unique substrate specificity, the
biochemistry of the TOM enzyme has been studied extensively. However, little
is known about the genetics and regulation of the pathway. The genes
responsible for this pathway are located in a 70-100 kb megaplasmid present in
this strain. It has been shown that expression of the pathway is constitutive
[143].
Burkholderia sp. strain JS150 contains a plasmid that carries the genes
encoding for a toluene-2-monooxygenase clustered in the operon tbmABCDEF
and its NtrC-like regulator coding gene tmbR. Additionally, a toluene-4-
monooxygenase activity was assigned to an independent region of the same
plasmid, which was shown to be regulated by TbmR [144]. Finally, cross-
activation between toluene-3-monooxygenase and toluene-2-monoxygenase by
regulators TbuT and TbmR has been reported [142].
361
6. TRANSCRIPTIONAL REGULATION OF THE ORGANIC SOLVENT

EFFLUX PUMPS IN Pseudomonasputida
In a recent study, several Pseudomonas putida strains were analyzed with regard
to toluene tolerance [145]. Three of these strains have been classified as highly
resistant (P. putida DOT-TIE, P. putida S12 and P. putida MTB6). In P. putida
DOT-TIE, three efflux pumps are involved in solvent tolerance: TtgABC [146],
TtgDEF [147] and TtgGHI [148]. The same three efflux pump operons are
present in the P. putida MTB6 chromosome although their participation in
organic solvent extrusion has not been studied in detail. Pseudomonas putida
S12 contains two of these efflux pumps encoded by the arpABC genes (98%
identical to ttgABQ [149], and the srpABC (99% identical to ttgGHI), although
only one of these efflux pumps, SrpABC, has been involved in solvent tolerance
in the S12 strain [151]. P. putida Fl has two efflux pumps ttgABC and sepABC
[152] and is more tolerant to toluene that P. putida KT2440, which only has the
ttgABC pump, but it is more sensitive than DOT-TIE.
In P. putida DOT-TIE, solvent tolerance is an inducible process, as
growth of P. putida DOT-TIE in the presence of toluene supplied in the gas
phase has a clear effect on cell survival: the sudden addition of 0.3% (vol/vol)
toluene to P. putida DOT-TIE pre-grown with toluene in the gas phase resulted
in survival of almost 100% of the initial cell number, whereas only 0.01% of the
cells pre-grown in the absence of toluene tolerated exposure to this aromatic
hydrocarbon (Fig. 6) [146, 153]. The three efflux pumps in this strain should
therefore work together in this strain to achieve the maximal level of solvent
tolerance, and they are probably tightly regulated in order to produce an optimal
response to solvent stress.
Most of the regulatory genes that encode for proteins involved in control
of the expression of the efflux pumps belonging to the RND family are located
adjacent to the structural genes of the pump, divergently transcribed from the
efflux pump operon [154].
6.1. Regulation of the ttgABC efflux pump operon

The TtgABC efflux pump was the first efflux pump identified in P. putida
DOT-TIE as involved in solvent tolerance. Physiological experiments done with
a ttgB knockout mutant suggested that this efflux pump was involved in the so-
called intrinsic tolerance. This mutant (P. putida DOT-TIE-18) did not
withstand the sudden toluene shock (0.3% vol/vol) at all, and only a small but
significant fraction (about 1 out of 105 cells) survived if pre-exposed to low
toluene concentrations [146]. On the basis of this observation the existence of
other(s) efflux pump(s) involved in toluene extrusion was postulated. The fact
that in P. putida DOT-TIE-18 cultures, no survival at all was observed after
sudden toluene shock compared with the 0.01% cell survival observed in the
362
wild-type, pointed out to a high basal expression of the TtgABC pump

responsible for this noninduced intrinsic resistance. This hypothesis was
confirmed after studying the expression levels of the ttgABC operon. In fact, the
operon is transcribed at relatively high levels in the presence and also in the
absence of toluene in accordance with its physiological behavior [155, 156].
Interestingly, the analysis of the TtgABC efflux pump susbstrate range
revealed that it not only extruded different organic solvents such as toluene,
styrene or p-xylene, but also different antibiotics such as ampicillin,
carbenicillin, tetracycline, nalidixic acid and chloramphenicol [146, 155, 157].
Gene fusion to lacZ and primer extension assays showed that some
substrates of the TtgABC efflux pump, such as chloramphenicol or tetracycline,
did increase (to a different extent) the expression of the pump operon, whereas
others (nalidixic acid, streptomycin or carbenicillin) did not [156]. Therefore,
the contribution of this solvent efflux pump to antibiotic resistance is probably a
consequence of the broad substrate specificity of both the pump itself and the
transcriptional regulator.
Upstream of the ttgABC operon, there is an open reading frame (named
ttgR) encoding a protein that shows 50-60% sequence similarity with a number
of transcriptional repressors such as AcrR (repressor of the acrAB operon in
Escherichia coli) or MtrR (regulator of the MtrCDE efflux pump in Neisseria
gonorrhoeae [158, 159]). All these proteins belong to the TetR family of
transcriptional regulators [160]. TtgR is a repressor of the TtgABC efflux pump:
a ttgR knockout mutant (P. putida DOT-TIE-13) exhibited an increased
expression (6-fold higher than the wild-type) of the ttgABC operon. However,
this increase in the efflux pump transcription levels did not lead to a higher
survival rate of the culture when shocked with 0.3% (vol/vol) toluene, but did
increase resistance towards different antibiotics such as chloramphenicol,
carbenicillin and tetracycline [155].
ttgR gene expression was very similar to that observed for its cognate
efflux pump operon: both are induced by chloramphenicol and tetracycline but
not by toluene [155, 156]. In the TtgR-deficient mutant strain the basal activity
of the ttgR promoter was eight-fold higher than the wild-type background,
indicating that TtgR also down regulates its own transcription.
The transcription initiation points of ttgA and ttgR have been mapped. The
-10 region, although not the -35 region, of both the ttgA and ttgR promoters
exhibits a certain degree of similarity to promoters recognized by sigma-70
[155]. The location of the start sites indicates that the divergent promoter
regions fully overlap (Fig. 6). This overlap could explain some of the great
similarity in the expression pattern.
363
Fig. 6. Organization of the ttgABC (A), ttgDEF (B) and ttgGHI (C) operons and their
respective regulatory genes. The regulatory regions of each gene cluster are zoomed. TtgR
(A), TtgT (B) and TtgV (C) DNA binding regions, deduced from DNAsel footprinting, are
shadowed. Putative palindromic (arrows) or symetric (bold and underlined) recognition sites
for each repressor are indicated. The +1 and the direction of transcription are marked with
small triangles for each promoter (except for ttgT one, which distance from
indicated). The -10 and the -35 positions of each promoter are also shown.
364
A recombinant and functional His-tagged TtgR protein has been

overexpressed, purified and used in several in vitro experiments carried out to
elucidate its regulatory role in ttgABC and ttgR expression. Gel shift assays
showed that TtgR binds specifically to a DNA fragment corresponding to the
ttgR-ttgABC intergenic region. Teran and coworkers [156] demonstrated that the
addition of chloramphenicol or tetracycline to the binding reaction led to a
gradual dissociation of TtgR from the ttgR-ttgABC intergenic region, suggesting
that TtgR is able to bind these two antibiotics, which triggers its release from the
promoter regions with subsequent enhanced expression. DNAse I footprint
assays allowed the identification of the TtgR operator within the ttgA-ttgR
intergenic region: a 36-bp DNA segment that includes the -10 and -35 region of
the ttgABC promoter and the -10 of the ttgR promoter. The TtgR binding site
deduced from DNAse I footprint, revealed a particularly long inverted repeat (28
bp) comprising two 12-bp half sites separated by 4 bp [156]. This indicates that
probably each half site would accept two monomers of TtgR, as reported for the
TtgR homolog QacR (repressor of the qacA multidrug pump gene of
Staphylococcus aureus), whose 3D structure bound to its operator was resolved
[161].
The mepABC efflux pump of a toluene-resistant variant of P. putida
KT2442 has also been implicated in solvent and antibiotic resistance, and its
sequence is practically identical to those of the TtgABC and ArpABC efflux
pumps. The corresponding regulatory protein is probably encoded by mepR,
although its function as a regulator has not been investigated yet [162].
Although TtgABC-like efflux pumps are widespread in different
Pseudomonas putida strains (P. putida MTB5, P. putida KT2440, P. putida
SMO116, among others) with different levels of toluene tolerance [145]. The
role of these efflux pumps in solvent or antibiotic resistance and their regulation
remains unknown.
6.2. Regulation of the ttgDEF efflux pump operon

As described above, solvent tolerance studies in a ttgB knockout mutant
of P. putida DOT-TIE (P. putida DOT-TIE-18) suggested the existence of
other(s) inducible solvent efflux pump(s). By sequencing downstream from the
toluene dioxygenase (tod) operon of P. putida DOT-TIE, Mosqueda and Ramos
[147] identified three open reading frames (ttgDEF) that encode for the three
components of an efflux pump which shares homology with other efflux pumps
of Pseudomonas. The transporter, named TtgE, shares 59% identity with the
previously described TtgB and 75% identity with TtgH. The contribution of this
efflux pump to solvent tolerance was studied in a ttgD knockout mutant (P.
putida DOT-T1E-1). A culture of this mutant strain showed a survival rate
identical to the wild type when shocked with 0.3% (vol/vol) toluene. However if
the cells were pre-induced with toluene in the gas phase, the survival rate of the
365
mutant was 100 times lower than in the wild type. Expression studies of the
ttgDEF operon at the transcriptional level revealed that this pump is not
expressed during growth under normal laboratory conditions, and demonstrated
its inducible character in the presence of organic solvents (toluene or styrene)
(Table 1) [147].
The wild type multiple antibiotic resistance was not affected in a TtgDEF-
deficient strain; moreover, no increase in antibiotic resistance was obtained by
pre-inducing the culture with toluene [147], suggesting that the substrate
specificity of this pump is limited to organic solvents. There was also no
induction of the ttgDEF operon in the presence of several antibiotics in the
culture media (Teran et al., unpublished).
Upstream from the ttgDEF operon and divergently transcribed, there is an
open reading frame whose product shares homology with several members of
the IclR family of transcriptional regulators. This gene, called ttgT, encodes for
a protein 70% identical to the SrpS-negative regulator of SrpABC solvent efflux
pump of P. putida S12 (see below).
A ttgT knockout mutant showed a small increase in ttgDEF expression
under non-inducing conditions, suggesting its involvement in the negative
regulation of this operon. The fact that in this mutant strain there was still a
strong induction of the ttgDEF expression in the presence of organic solvents
suggested that TtgT is not the only protein involved in the induction of this
operon by organic solvents (Teran et al., unpublished).
Differently from ttgR, ttgT gene expression remained unaltered regardless
of the organic solvent present in the growth medium, which suggested that
expression from ttgDEF and ttgT promoters was not coordinated. Moreover, in
the TtgT-deficient mutant, the activity of the ttgT promoter was similar to that of
the wild-type, indicating that TtgT does not regulate its own transcription (Teran
et al., unpublished).
Gel shift experiments showed that TtgT was able to specifically bind a
DNA fragment containing the ttgT-ttgDEF intergenic region (Teran et al.,
unpublished). DNAse I footprint assays revealed a single binding site along the
ttgT-ttgDEF intergenic region which covers only the ttgDEF promoter region (-
37 to +5 from the transcription start point) and not the ttgT one, consistent with
the in vivo expression studies described above. Therefore TtgT is directly
involved in ttgDEF operon repression, probably by competing with the RNA
polymerase for access to the efflux pump promoter. Analysis of the operator
sequence does not reveal the presence of a clear single inverted repeat. Recent
results of our laboratory suggests the induction of this efflux pump in a ttgT-
deficient background by organic solvents is mediated regulator by the TtgV
regulator.
366
6.3. Regulation of the ttgGHI efflux pump operon

The third efflux pump involved in solvent tolerance in P. putida DOT-
TIE is called TtgGHI. A knockout mutant in which the TtgGHI efflux pump of
P. putida DOT-TIE is not functional (P. putida DOT-T1E-PS28) is not able to
survive a sudden 0.3% (vol/vol) toluene shock regardless of the growth
conditions [163]. In contrast, the TtgABC and TtgDEF knockout mutants were
still able to survive the toluene shocks to a different extent. The extreme toluene
sensitivity of the P. putida DOT-T1E-PS28 mutant strain (ttgH::Q.Sm) under
induced or noninduced conditions suggested that this efflux pump is involved in
intrinsic as well as inducible resistance to organic solvents. The ttgGHI operon
is expressed from a single promoter PG2 at a certain basal level in the absence of
solvents, and its expression increases several-fold in the presence of aromatic
hydrocarbons such as toluene and styrene, aliphatic alcohols such as 1-octanol,
but not in the presence of antibiotics [163, 164].
In P. putida DOT-TIE, two genes ttgV and ttgW were identified upstream
from the ttgGHI operon. They are transcribed divergently from the efflux pump
operon. TtgV showed an overall 50%-60% similarity with a number of
transcriptional regulators belonging to the IclR family, whereas ttgW is probably
a pseudogene and the protein encoded seems not to be functional.
A TtgV knockout mutant was constructed and characterized [163]. This
strain showed increased resistance toward toluene shocks under noninduced
conditions when compared with the wild type. The fraction of cells that survived
the sudden addition of 0.3% (vol/vol) toluene was the same (107 cells) under
induced and noninduced conditions. Analysis of the expression of the ttgGHI
operon in this genetic background showed that the level of transcription
increased 4-fold in the absence of toluene. Taken together, these data clearly
indicated that TtgV is a repressor that prevents expression of the ttgGHI operon.
The transcription initiation point of the ttgVW operon was mapped in cells
growing in the absence and in the presence of toluene. The operon was
transcribed from a single promoter regardless of the growth conditions, but the
level of expression in the presence of toluene was 3- to 4-fold higher than in the
absence of the aromatic hydrocarbon. The ttgVW operon was also shown to be
induced by several organic solvents but not by antibiotics. The ttgVW operon
showed a pattern of inducibility similar to that of ttgGHI, probably because both
promoters are regulated in the same way. Sequence analysis of the promoter
region showed that the -10 and -35 boxes of the VG2 overlap with the -35 and -
10 boxes of the ftgFpromoter (Fig. 6). P-galactosidase assays carried out with a
transcriptional fusion of VtlgV promoter to lacZ in a TtgV-deficient background
showed that expression was about 3-fold higher than in the wild type strain, also
in the absence of toluene, indicating that TtgV negatively controls its own
expression. Together, these results suggested that the TtgV protein is a repressor
of its own synthesis as well as of ttgGHI operon expression [163]. It is
367
interesting to note that in TtgV-deficient background expression from ttgDEF

operon promoter increased 3-fold suggesting the involvement of this regulator in
the control of this operon.
Sequence analysis of PHgG and VttgV showed that both promoters
overlapped. The TtgV protein has been overexpressed and purified with an N-
terminal histidine tag. In vitro gel mobility shift assays demonstrated the specific
binding of TtgV to a 210-bp DNA fragment comprising the ttgG and ttgV
intergenic region. The DNAse I-protected region extended a 40-bp covering the
-10/-35 regions of the ttgG promoter and the divergently oriented ttgV promoter
[163]. To gain insight into the mechanism of regulation of ttgGHI transcription
by TtgV, in vitro transcription experiments were carried out using the purified
protein and the ttgV-ttgG intergenic region on a supercoiled plasmid. When the
TtgV protein and the plasmid were incubated before the addition of RNA-
polymerase, ttgGHI transcription was completely repressed. However, when
TtgV was added after the formation of the RNA-polymerase-«gGif/ promoter
open complex, the repression level became negligible [164]. These findings
support the idea that TtgV binding to the intergenic region blocks the entry of
RNA-polymerase to transcribe both operons.
In vitro transcription assays were carried out in the absence and in the
presence of increasing concentrations of 1-hexanol, a known inducer of ttgGHI
operon. Addition of the inducer to the transcription reaction in the presence of
TtgV led to transcriptional levels similar to those observed in the absence of
TtgV repressor, resulting in VttgG expression. This suggests that 1-hexanol
decreased TtgV binding to the intergenic region, as it has been demonstrated by
gel shift assays [164].
Acknowledgments
Work in our laboratory was supported by grants of the European
Commission (QLK3-CT-1999-00041, QLK3-CT-2001-00435 and QLK3-CT-
2000-0170) and a grant from the Human Science Foundation (RGY0021/2001).
We thank C. Lorente for reading the manuscript and improving the language.
REFERENCES
[1] S. Dagley, Adv. Microb. Physiol, 6 (1971) 1.

[2] J. Sikkema, J.A.M. de Bont, and B. Poolman, Microbiol. Rev. 59 (1995) 201.
[3] D.L. Cruden, J.H. Wolfram , R.D. Rogers and D.T. Gibson , Appl. Environ. Microbiol.
58 (1992) 2723.
[4] A. Inoue and K. Horikoshi, Nature 338 (1989) 264.
[5] K. Kim, S. Lee, K. Lee, and D. Lim, J. Bacteriol. 180 (1998) 3692.
[6] J.L. Ramos, E. Duque, MJ. Huertas and A. Haidour, J. Bacteriol. 177 (1995) 3911.
[7] F.J. Weber, S. Isken and J.A.M. de Bont, Microbiology 140 (1994) 2013.
[8] H. Keweloh and H.J. Heipieper, Lipids 31 (1996) 129.
368
[9] HJ. Heipieper and J.A.M. de Bont, Appl. Environ. Microbiol. 60 (1994) 4440.
[10] HJ. Heipieper, G.Meulenbeld, Q. van Oirschot and J.A.M. de Bont, Appl Environ
Microbiol. 62(1996)2773.
[11] B. Loffeld and H. Keweloh , Lipids31 (1996)811.
[12] A. von Wallbrun, N.H.Richnow,, Neumann, G., Meinhardt, F., and Heipieper, H.J.,
(2003), J. BacterioL, 185 1730.
[13] F.J. Weber and J.A.M. de Bont, Biochim. Biophys. Acta 1286 (1996) 225.
[14] G. Mosqueda, M. I. Ramos-Gonzalez and J. L. Ramos, Gene, 232 (1999) 69.
[15] P. Godoy, M.I. Ramos-Gonzalez and J.L. Ramos, J. BacterioL 18 (2001) 5285.
[16] Q. Zhao, X.Z. Li, A. Mistry, R. Srikumar, L. Zhang, O. Lomovskaya and K. Poole,
Antimicrob. Agents. Chemother. 42 (1998) 2225.
[17] V.Koronakis, A. Sharfl, E. Koronakis, B. Luisi and C. Hughes, Nature 405 (2000) 914.
[18] S. Murakami, R. Nakashima, E. Yamashita and A. Yamaguchi, Nature 419 (2002) 587.
[19] H. I. Zgurskaya and H. Nikaido J. Mol. Biol. 285 (1999) 409.
[20] E.W. Yu, G. McDermott, H.I. Zgurskaya, H. Nikaido and D.E. Jr. Koshland, Science
300(2003) 976.
[21] H.I. Zgurskaya and H. Nikaido, J. BacterioL 182 (2000) 4264.
[22] C.A. Elkins and H. Nikaido, Drug Resist. Update 6 (2003) 9.
[23] S. J. Assinder and P.A. Williams, Adv. Microb. Physiol., 31 (1990) 1.
[24] P.A. Williams, R.M. Jones and G. Zylstra, (2004), Genomics of catabolic plasmids. In
Pseudomonas, vol. I, Chapter 6, pp. 165-196, Ed. J.L. Ramos, Kluwer
Academic/Plenum Publishers.
[25] S. Harayama, N. Mermod, M. Rekik, P.R. Lehrbach and K.N. Timmis, J. BacterioL,
169 (1987) 558.
[26] S. Harayama and K.N. Timmis (1989), Catabolism of aromatic hydrocarbons by
Pseudomonas. In Genetics of Bacterial Diversity, ed. D. Hopwood, K. Chater, 151-174.
New York Academic.
[27] C. Nakai, H. Kagamiyama and M. Nozaki, J. Biol. Chem., 258 (1983) 2923.
[28] A. Polissi and S. Harayam, EMBO J. 12 (1993) 3339.
[29] J.P. Shaw and S. Harayama, Eur. J. Biochem., 191 (1990) 705.
[30] M. Suzuki, T. Hayakawa, J.P. Shaw, M. Rekik and S. Harayama, J. BacterioL 173
(1991) 1690.
[31] D. T. Gibson, J. R. Koch and R. E. Kallio, Biochemistry, 7 (1968) 2653.
[32] G. M. Whited and D. T. Gibson, J. BacterioL, 173 (1991) 3010.
[33] K. M. Yen, M. R. Karl, L. M. Blatt, M. J. Simon, R. B. Winter, P. R. Fausset, H. S. Lu,
A.A. Harcourt and K. K. Chen, J. BacterioL, 173 (1991) 5315.
[34] R. H. Olsen, J. J. Kukor and B. Kaphammer, J. BacterioL, 176 (1994) 3749.
[35] A. Fishman, Y. Tao and T.K. Wood, J. BacterioL, (2004) in press.
[36] G. Baggi, P. Barbieri, E. Galli and S. Tollari, Appl. Environ. Microbiol., 53 (1987)
2129.
[37] G. Bertoni, F. Bolognese, E. Galli and P. Barbieri, Appl. Environ. Microbiol., 62
(1996)3704.
[38] G. Bertoni, N. Fujita, A. Ishihama and V. de Lorenzo, EMBO J., 17 (1998) 5120.
[39] G. Bertoni, S. Marques and V. de Lorenzo, Mol. Microbiol., 27 (1998) 651.
[40] J.L. Ramos, E. Duque, J.J. Rodriguez-Herva, P. Godoy, A. Haidour and J.L. Ramos,
J. Biol. Chem. 272 (1997) 3887.
[41] EganS.M., (2002) J. BacterioL, 184 5529.
[42] M. T. Gallegos, R. Schleif, A. Bairoch, K. Hofmann and J. L. Ramos, Microbiol. Mol.
Biol. Rev., 61 (1997) 393.
369
[43] J. L. Ramos, C. Michan, F. Rojo, D. Dwyer and K. Timmis, J. Mol. Biol., 211 (1990)
373.
[44] R.Tobes and J. L Ramos, Nucleic Acids Res., 30 (2002) 318.
[45] J. L. Ramos, A. Stolz, W. Reineke and K. N. Timmis, Proc. Natl. Acad. Sci. USA, 83
(1986) 8467.
[46] C. Michan, L. Zhou, M. T. Gallegos, K. N. Timmis and J. L. Ramos, J. Biol. Chem.,
267 (1992) 22897.
[47] L. M. Zhou, K. N. Timmis and J. L. Ramos, J. Bacteriol., 172 (1990) 3707.
[48] J. L. Ramos, F. Rojo, L. Zhou and K. N. Timmis, Nucleic Acids Res., 18 (1990) 2149.
[49] J.L. Ramos, A. Wasserfallen, K. Rose and K.N. Timmis, Science 235 (1987) 593.
[50] R. Ruiz and J. L. Ramos, J. Biol. Chem, 277 (2002) 7282.
[51] N. LaRonde-LeBlanc and C. Wolberger, Biochemistry, 39 (2000) 11593.
[52] C. A. Poore, C. Coker, J.D. Dattelbaum and H. L. Mobley, J. Bacteriol, 183 (2001)
4526.
[53] S. Rhee, R. G. Martin, J. L. Rosner and D. R. Davies, Proc. Natl. Acad. Sci. USA, 95
(1998) 10413.
[54] S. M. Soisson, B. MacDougall-Shackleton, R. Schleif and C. Wolberger, Science, 276
(1997)421.
[55] R. Ruiz, S. Marques and J. L. Ramos, J. Bacteriol, 185 (2003) 3036.
[56] M. Manzanera, S. Marques and J. L. Ramos, FEBS Lett, 476 (2000) 312.
[57] N. Kaldalu, U. Toots, V. de Lorenzo and M. Ustav, J. Bacteriol, 182 (2000) 1118.
[58] M. M. Gonzalez-Perez, S. Marques, P. Dominguez-Cuevas and J. L. Ramos, FEBS
Lett, 519(2002) 117.
[59] M. T. Gallegos, S. Marques and J. L. Ramos, J. Bacteriol, 178 (1996) 6427.
[60] M. M. Gonzalez-Perez, J. L. Ramos, M. T. Gallegos and S. Marques, J. Biol. Chem,
274(1999)2286.
[61] B. Kessler, V. de Lorenzo and K. Timmis, J. Mol. Biol, 230 (1993) 699.
[62] B. Kessler, M. Herrero, K. N. Timmis and V. de Lorenzo, J. Bacteriol, 176 (1994)
3171.
[63] B. Kessler, S. Marques, T. Kohler, J. L. Ramos, K. N. Timmis and V. de Lorenzo, J.
Bacteriol, 176(1994)5578.
[64] N. Kaldalu, T. Mandel and M. Ustav, Mol. Microbiol, 20 (1996)569.
[65] K. Miura, S. Inouye and A. Nakazawa, Mol. Gen. Genet, 259 (1998) 72.
[66] I. Artsimovitch, K. Murakami, A. Ishihama and M. M. Howe, J. Biol. Chem, 271
(1996) 32343.
[67] S. Busby and R. H. Ebright, Mol. Microbiol, 23 (1997) 853.
[68] A. Dhiman and R. Schleif, J. Bacteriol, 182 (2000) 5076.
[69] S. Marques, M. T. Gallegos and J. L. Ramos, Mol. Microbiol, 18 (1995) 851.
[70] S. Marques, M. Manzanera, M. M. Gonzalez-Perez, M. T. Gallegos and J. L Ramos,
Mol. Microbiol, 31 (1999) 1105.
[71] T. Yura and K. Nakahigashi, Curr. Opin. Microbiol, 2 (1999) 153.
[72] S. Inouye, A. Nakazawa and T. Nakazawa, Gene, 66 (1988)301.
[73] S. Kustu, E. Santero, J. Keener, D. Popham and D. Weiss, Microbiol. Rev, 53 (1989)
367.
[74] A. K. North, K. E. Klose, K. M. Stedman and S. Kustu, J. Bacteriol, 175 (1993) 4267.
[75] M. A. Abril, C. Michan, K. N. Timmis and J. L. Ramos, J. Bacteriol, 171 (1989)
6782.
[76] A. Delgado and J. L. Ramos, J. Biol. Chem, 269 (1994) 8059.
[77] A. Delgado, R. Salto, S. Marques and J. L. Ramos, J. Biol. Chem, 270 (1995) 5144.
370
[78] J. Garmendia, D. Devos, A. Valencia and V. de Lorenzo, Mol. Microbiol., 42 (2001)

47.
[79] V. Shingler,, (2004), Transcriptional regulation and catabolic strategies of phenol
degradative pathways. In Pseudomonas, vol. II, Chapter 16, pp. 451-478, Ed. J.L.
Ramos, Kluwer Academic/Plenum Publishers.
[80] E. Skarfstad, E. O'Neill, J. Garmendia and V. Shingler, J. Bacteriol, 182 (2000) 3008.
[81] S. Fernandez, V. de Lorenzo and J. Perez-Martin, Mol. Microbiol, 16 (1995) 205.
[82] J. Perez-Martin and V. de Lorenzo, J. Mol. Biol, 258 (1996) 575.
[83] J. Perez-MartinandV.de Lorenzo, Cell, 86(1996)331.
[84] R. Dixon, Mol. Gen. Genet, 203 (1986) 129.
[85] S. Inouye, A. Nakazawa and T. Nakazawa, Gene, 29 (1984) 323.
[86] M. A. Abril, M. Buck and J. L. Ramos, J. Biol. Chem, 266 (1991) 15832.
[87] V. de Lorenzo, M. Herrero, M. Metzke and K. N. Timmis, EMBO J, 10 (1991) 1159.
[88] G. H. Seong, E. Kobatake, K. Miura, A. Nakazawa and M. Aizawa, Biochem. Biophys.
Res. Commun, 291 (2002) 361.
[89] G. Bertoni, M. Martino, E. Galli and P. Barbieri, Appl. Environ. Microbiol, 64 (1998)
3626.
[90] M. Carmona, V. de Lorenzo and G. Bertoni, J. Biol. Chem, 274 (1999) 33790.
[91] A. Holtel, M. A. Abril, S. Marques, K. N. Timmis and J. L. Ramos, Mol. Microbiol, 4
(1990) 1551.
[92] A. Holtel, D. Goldenberg, H. Giladi, A. B. Oppenheim and K. N. Timmis, J.
Bacteriol, 177(1995)3312.
[93] A. Holtel, K. Timmis and J. L. Ramos, Nucleic Acids Res, 20 (1992) 1755.
[94] G. Bertoni, J. Perez-Martin and V. de Lorenzo, Mol. Microbiol, 23 (1997) 1221.
[95] M. Gomada, S. Inouye, H. Imaishi, A. Nakazawa and T. Nakazawa, Mol. Gen. Genet,
233(1992)419.
[96] S. Inouye, A. Nakazawa and T. Nakazawa, J. Bacteriol, 163 (1985) 863.
[97] S. Marques, M. T. Gallegos, M. Manzanera, A. Holtel, J. Bacteriol, 180 (1998) 2889.
[98] S. Fraile, F. Roncal, L. A. Fernandez and V. de Lorenzo, J. Bacteriol, 183 (2001) 5571.
[99] W. A. Duetz, S. Marques, C. de Jong, J. L. Ramos and J. G. van Andel, J. Bacteriol,
176(1994)2354.
[100] W.A. Duetz, S. Marques, B. Wind, J.L. Ramos and J.G. Van Andel, Appl. Environ.
Microbiol, 62(1996)601.
[101] A. Holtel, S. Marques, I. Mohler, U. J akubzik and K. N. Timmis, J. Bacteriol, 176
(1994) 1773.
[102] N. Hugouvieux-Cotte-Pattat, T. Kohler, M. Rekik and S. Harayama, J. Bacteriol, 172
(1990)6651.
[103] S. Marques, A. Holtel, K. N. Timmis and J. L. Ramos, J. Bacteriol, 176 (1994) 2517.
[104] M. J. Worsey, and P. A. Williams, J Bacteriol. 124 (1975) 7.
[105] I. Cases, V. de Lorenzo and J. Perez-Martin, Mol. Microbiol, 19 (1996) 7.
[106] M. Carmona, M. J. Rodriguez, O. Martinez-Costa and V. de Lorenzo, J. Bacteriol, 182
(2000)4711.
[107] F. Rojo and M.A. Dinamarca, (2004), Catabolite repression and physiological control.
In Pseudomonas, vol. II, Chapter 13, pp. 365-388, Ed. J.L. Ramos, Kluwer
[108] M. Vails, M. Buckle and V. de Lorenzo, J. Biol. Chem, 277 (2002) 2169.
[109] P. Jurado, L. A. Fernandez and V. de Lorenzo, J. Bacteriol, 185 (2003) 3379.
[110] T. Tomoyasu, K. Yamanaka, K. Murata, T. Suzaki, P. Bouloc, A. Kato, H. Niki, S.
Hiraga and T. Ogura, J. Bacteriol, 175 (1993) 1352.
371
[111] M. Carmona and V. de Lorenzo, Mol. Microbiol, 31 (1999) 261.

[112] Y. Akiyama, Proc. Natl. Acad. Sci. USA, 99 (2002) 8066.
[113] I. Cases, and V. de Lorenzo, J. Bacteriol., 182 (2000) 956.
[114] I. Cases, J. A. Lopez, J. P. Albar and V. de Lorenzo, J. Bacteriol., 183 (2001) 1032.
[115] I. Cases, F. Velazquez and V. de Lorenzo, J. Bacteriol., 183 (2001) 5128.
[116] I. Cases, J. Perez-Martin and V. de Lorenzo, J. Biol. Chem., 274 (1999) 15562.
[117] B. A. Finette, V. Subramanian and D. T. Gibson, J. Bacteriol., 160 (1984) 1003.
[118] D. T. Gibson, G. H. Zylstra and S. Chauhan, (1990), Biotransformations catalyzed by
toluene dioxygenase from Pseudomonasputida Fl, p. 121-132. In S. Silver, A.M.
[119] P. C. Lau, Y. Wang, A. Patel, D. Labbe, H. Bergeron, R. Brousseau, Y. Konishi and M.
Rawlings, Proc. Natl. Acad. Sci. USA, 94 (1997) 1453.
[120] F. M. Menn, G. J. Zylstra and D. T. Gibson, Gene, 104 (1991) 91.
[121] Y. Wang, M. Rawlings, D. T. Gibson, D. Labbe, H. Bergeron, R. Brousseau and P. C.
Lau Mol. Gen. Genet., 246 (1995) 570.
[122] G. J. Zylstra and D. T. Gibson, J. Biol. Chem., 264 (1989) 14940.
[123] G. J. Zylstra, W. R. McCombie, D. T. Gibson and B. A. Finette, Appl. Environ.
Microbiol., 54 (1988)1498.
[124] B. A. Finette and D. T. Gibson, Biocatalyst 2 (1988) 29.
[125] P. Dominguez-Cuevas, and S. Marques, (2004), Compiling sigma-70 dependent
promoters. In Pseudomonas, vol. II, Chapter 11, pp. 319-344. Ed. J.L. Ramos, Kluwer
[126] B. L. Taylor and I. B. Zhulin, Microbiol. Mol. Biol. Rev., 63 (1999) 479.
[127] I. B. Zhulin, B. L. Taylor and R. Dixon, Trends Biochem. Sci., 22 (1997) 331.
[128] J. A. Hoch and T. J. Silhavy, (1995), Two-component signal transduction. American
Society for Microbiology, Washington DC.
[129] J. Reizer and M. H. J. Saier, Curr. Opin. Struct. Biol., 7 (1997) 407.
[130] R. E. Parales, J. L. Ditty and C. S. Harwood, Appl. Environ. Microbiol., 66 (2000)
4098.
[131] K. M. Yen and M. R. Karl, J. Bacteriol., 174 (1992) 7253.
[132] M. I. Ramos-Gonzalez, M. Olson, A. A. Gatenby, G. Mosqueda, M. Manzanera, M. J.
Campos, S. Vichez and J. L. Ramos, J. Bacteriol., 184 (2002) 7062.
[133] A. Wright and R. H. Olsen, Appl. Environ. Microbiol., 60 (1994) 235.
[134] A. Ben-Bassat, M. Cattermole, A.A. Gatenby, K.J. Gibson, M.I. Ramos-Gonzalez, J.L.
Ramos and S. Sariaslani, (2003), Method for the production ofp-hydroxybenzoate in
species of Pseudomonas and Agrobacterium. US Patent 6,586,229
[135] B. Entsch, Y. Nan, K. Weaich and K. F. Scott, Gene, 71 (1988) 279.
[136] C. M. Wong, M. J. Dilworth, and A. R. Glenn, Microbiology, 140 (1994) 2775.
[137] A. M. Byrne and R. H. Olsen, J. Bacteriol., 178 (1996) 6327.
[138] J. J. Kukor and R. H. Olsen, J. Bacteriol., 172 (1990) 4624.
[141] H. Y. Kahng, A. M. Byrne, R. H. Olsen and J. J. Kukor, J. Bacteriol., 182 (2000) 1232.
[142] J. G. Leahy, G. R. Johnson and R. H. Olsen, Appl. Environ. Microbiol., 63 (1997) 3736.
[143] M. S. Shields, M. J. Reagin, R. R. Gerger, R. Campbell and C. Somerville, Appl.
[144] G.R. Johnson and R.H. Olsen, Appl. Environ. Microbiol., 63 (1997) 4047.
[145] A. Segura, A. Rojas, A. Hurtado, M.J. Huertas and J.L. Ramos, Extremophiles 7
(2003)371.
[146] J.L. Ramos, E. Duque, P. Godoy and A. Segura, J. Bacteriol. 180 (1998) 3323.
372
[147] G. Mosqueda and J.L. Ramos, J. Bacteriol. 182 (2000) 937.

[148] A. Rojas, E. Duque, G. Mosqueda, G. Golden, A. Hurtado, J.L. Ramos and A. Segura,
J. Bacteriol., 183(2001)3967.
[149] J. Kieboom and J.A.M. de Bont, Microbiology 147 (2001) 43.
[150] J. Kieboom, J.J. Dennis, J.A.M. de Bont and G.J. Zylstra, J. Biol. Chem. 273 (1998)
85.
[151] J. Kieboom, J.J. Dennis, G.J. Zylstra and J.A.M. de Bont, J. Bacteriol. 180 (1998)
6769.
[152] P. Phoenix, A. Keane, A. Patel, H. Bergeron, S. Ghoshal and P.C.K. Lau, Env.
Microbiol. 5 (2003) 1309.
[153] J. L. Ramos, S. Marques and K. N. Timmis, Annu. Rev. Microbiol., 51 (1997) 341.
[154] A. Segura, E. Duque, G. Mosqueda, J.L. Ramos and F. Junker, Environ. Microbiol. 1
(1999)191.
[155] E. Duque, A. Segura, G. Mosqueda and J.L. Ramos, Mol. Microbiol. 39 (2001) 1100.
[156] W. Teran, A. Felipe, A. Segura, A. Rojas, J.L. Ramos and M.T. Gallegos, Antimicrob.
Agents Chemother. (2003), in press.
[157] J.L. Ramos, E. Duque, M.T. Gallegos, P. Godoy, M.I. Ramos-Gonzalez, A. Rojas, W.
Teran and A. Segura, Ann. Rev. Microbiol. 56 (2002) 743.
[158] D. Ma, M. Alberti, C. Lynch, H. Nikaido and J.E. Hearst, Mol. Microbiol. 19 (1996)
101.
[159] W. Pan and B.G. Spratt, Mol. Microbiol. 11 (1994) 769.
[160] W. Hillen and C. Berens, Annu. Rev. Microbiol. 48 (1994) 345.
[161] M.A. Schumacher, M.C. Miller, S. Grkovic, M.H. Brown, R.A. Skurray and R.G.
Brennan, EMBO J. 21 (2002) 1210.
[162] F. Fukumori, H. Hirayama, H. Takami, A. Inoue and K. Horikoshi, Extremophiles 2
(1998)395.
[163] A. Rojas, A. Segura, M.E. Guazzaroni, W. Teran, A. Hurtado, M.T. Gallegos and J.L.
Ramos, J. Bacteriol., 185 (2003) 4755.
[164] M.E. Guazzaroni, W. Teran, X. Zhang, M.T. Gallegos, and J.L. Ramos, J. Bacteriol.,
(2004) in press.
Chapter 13
Bacterial hydrocarbon biosynthesis revisited

B. Valderrama
Departamento de Ingenieria Celular y Biocatalisis, Universidad Nacional

Autonoma de Mexico, AP 510-3, Cuernavaca, Morelos, 62250, Mexico.
1. INTRODUCTION
One of the greatest challenges faced by the modern world is the dissociation
from the heavy dependency of the energy technologies upon the chemical bonds
of hydrocarbons. The imminent exhaustion of conventional oil sources, ranging
from a pessimistic ultimate recovery volume of 0.6 trillions of barrels to a highly
optimistic volume of 3.9 trillions of barrels [4], results in a stringent requirement
for the development of alternative technologies. It is important to note that the
world is not to run out of hydrocarbons, given the substantial amount of low-
grade, hard-to-extract supplies such as the Canadian tar sands or the abundant
heavy oil reservoirs in Venezuela and Mexico. Nevertheless, exploiting these
reservoirs is likely to be much more expensive financially, energetically,
politically and especially environmentally.
Biotechnology has greatly impacted modern industry, from the now
conventional production of goods by the use of fermentations to the novel
synthesis of valuable fine chemicals using enzymes [5, 6, 7, 8]. Notwithstanding
its enormous potential, the incorporation of biotechnological tools into the oil
industry has faltered [9]. In particular, the search of alternative hydrocarbon
sources through biotechnological media has not been assessed. Here, I compile
information regarding the biological production of hydrocarbons by bacteria and
explore its potential, not only as an environmentally-friendly fuel supply, but
also as a renewable source for basic petrochemicals.
374
2. HYDROCARBON BIOSYNTHESIS IS A COMMON TRAIT AMONG

BACTERIA
The ability of organisms to synthesize hydrocarbons has been observed in all

phyla [10]. Whereas alkanes are mainly involved in epicuticular wax
biosynthesis in plants [11], in insects, their roles are more diverse, ranging from
waterproofing of the cuticle to participation in sexual behaviors as aphrodisiac
pheromones [12]. Furthermore, many marine animals, from invertebrates to
whales, contain appreciable amounts of hydrocarbons as component of waxes,
which appear to have a variety of functions, from serving as energy source to
insulation, buoyancy and even echo-location [13]. The accumulation of non-
volatile hydrocarbons by microorganisms has been shown to occur in
microalgae [14, 15, 16, 3], in bacteria [17, 18, 19] and in yeast [20, 21].
Originally, the ability of microorganisms to produce hydrocarbons was
studied as part of the biogenic hypothesis for oil reservoir formation. This
hypothesis was actively investigated between 1930 and 1960 by C.E. ZoBell,
who proposed a significant bacterial role in the origin of petroleum [22, 23]. In
1950 he suggested that microbial modification of organic remains in sediments
could contribute precursors for oil formation by lowering their oxygen and
nitrogen content and increasing their carbon and hydrogen content, and by the
direct production of methane and other higher hydrocarbons [24, 25]. ZoBell
also suggested that hydrogenation of unsaturated fatty acids and their subsequent
decarboxylation might have contributed to petroleum formation. Although a
bacterial role in the initial processing of organic matter from which petroleum is
derived has become generally accepted, there is no experimental evidence that
bacteria were directly responsible for hydrocarbon production in significant
quantities. Although ZoBell later abandoned the idea of direct microbial
formation of hydrocarbons in large amounts from organic matter, he described
several cases where cultures of marine bacteria presented substantial capacity
for aliphatic hydrocarbon biosynthesis [26, 27, 23].
After the development of more powerful analytical procedures, the
subject of bacterial hydrocarbon production resurged between 1960 and 1970 [1,
28, 19, 18]. Unequivocal evidence of hydrocarbon accumulation was observed
in all the bacterial species tested (Table 1), including photosynthetic bacteria as
well as in non-photosynthetic bacteria. The composition of bacterial
hydrocarbons was complex, with length ranging from Ci5 up to C36, and
including n-alkanes, alkenes, and branched hydrocarbons. In particular, non-
photosynthetic bacteria accumulate long-chain n-alkanes (C27-C29), whereas n-
alkanes with shorter chains (C|7-C2o) are more abundant in photosynthetic
bacteria [29]. Photosynthetic bacteria, as well as anaerobic non-photosynthetic
bacteria, are characterized by the presence of isoprenic units of pristane and
phytane [30].
375
Table 1
Taxonomical distribution of eubacterial and archaeal species able to
synthesize and accumulate hydrocarbons.
Phyllum Class Genus Reference
Proteobacteria a Proteobacteria Rhodopseudomonas sphaeroides [33,30]

Rhodospirillum rubrum [33,30]
y Proteobacteria Chromatium [29]
Escherichia coli [33,30]
Rhodimicrobium vannielii [30]
Vibrio furn issii [17]
Serratia marinorubrum [23]
Vibrio marinus [34]
Vibrio ponticus [23]
Vibrio furnissii [17]
8/E Proteobacteria Desulfovibrio desulfuricans [35]
Desulfovibrio Essex [30]
Desulfovibrio Hildenborough [30]
Firmicute Clostridia Clostridium acidiurici [30]

Clostridium tetanomorphum [30]
Sarcinaflava [18]
Sarcina lutea [18,28]
Sarcina subflava [18]
Bacilli Staphylococcus sp. [18]
Bacillus sp. [36]
Actinobacteria Actinobacteria Micrococcus lysodeikticus [18,33,30]

Micrococcus sp. [36]
Mycobacterium sp. [36]
Corynebacterium sp. [36]
Arthrobacter sp. [36]
Cyanobacteria Nostococales Nostoc muscorum [30]

Nostoc sp. [36]
Oscillatoriales Phormidium luridum [30]
Chlorobi Chlorobia Chlorobium [30]
Euryarchaeota Thermoplasmata Thermoplasma sp. [32]

Methanomicrobia Methanosarcina barkeri [32]
Halobacteria Halobacterium cutirubrum [32]
Crenarchaeota Thermoprotei Sulfolobus sp. [31]
This ability is not restricted to eubacteria. Some archeal species from the
genus Sulfolobus, Thermoplasma, Methanosarcina and Halobacterium, have
been demonstrated able to synthesize and accumulate hydrocarbons such as
squalene and other acyclic isoprenoids (C20-C25). [31, 32]. Furthermore,
individual species that produce hydrocarbons as major components have been
isolated from mesophilic, thermophilic, psycrophilic, acidophilic, alkalinophilic,
376
and halophilic environments under aerobic or anaerobic, autotrophic or

heterotrophic conditions. The environmental distribution of hydrocarbon
producers follows no discernible pattern that can be used as a guide for finding
prolific hydrocarbon producers.
3. BIOSYNTHETIC PATHWAYS
Non-isoprenoid biological hydrocarbons were presumed to derive from fatty

acids. Currently, there are two known pathways for the conversion of fatty acids
to straight-chain hydrocarbons. The best known of them is the elongation-
decarboxylation process (Fig.lA). In this case, a fatty acid precursor, such as
oleic acid, is elongated by the continuous addition of a C2 unit derived from
malonyl-CoA. The hydrocarbon produced is then cappedoff through a
decarboxylation reaction when it reaches the designated length. The second
mechanism involves the "head-to-head" condensation of two fatty acids (Fig.
IB). In this path, one of the acid derivatives is specifically decarboxylated
following the condensation step, while the total carbon chain of the other is
incorporated into the hydrocarbon [3].
The commitment step in these pathways is the decarboxylation reaction. It
has been well documented that CO2 elimination from carboxylic acids requires
high energy and therefore has to be activated by a P-substituent able to stabilize
the negative charge generated by CO2 release. Accordingly, it has generally been
thought that activated fatty acid derivatives are the intermediates in the
decarboxylation leading to hydrocarbons.
Fig. 1 Metabolic pathways for aliphatic hydrocarbon biosynthesis (Modified from [3] and
[1]). LCFA - Long Chain Fatty Acids
377
Fatty acyl-CoA reductase activity has been identified in a variety of other

organisms, from bacteria to animals [40, 41, 42, 38]. The gene encoding this
enzyme has been recently cloned from the y-proteobacteria Acinetobacter
calcoaceticus and Photobacterium leiognathi, as well as from seeds of jojoba
(Simmondsia chinensis) [42, 43, 44]. As can be seen in Fig. 2, all of them harbor
the residues conforming the catalytic triad observed in related dehydrogenases
[45]. Interestingly, each one of these reference sequences represents independent
groups, on the basis of sequence similarity (see Fig. 3). Groups I and II are
rather selective, being all their members either plantsor bacteria, respectively.
Group III has the sequence from Acinetobacter calcoaceticus as sole member.
This sequence is similar to oxido-reductases with different substrate specificities
from various sources.
The aldehydes generated from fatty acid reduction in B. braunii are
further reduced to hydrocarbons (alkanes). The initial observation that the
resulting alkane had one less carbon than the aldehyde leaded to the proposal of
a decarbonylation as the final step. Such an activity would yield one molecule of
alkane and one molecule of CO as products. Two plant aldehyde decarbonylases
(from pea and from B. braunii) have been studied in some detail [46, 47]. They
are integral membrane proteins with the pea decarbonylase suggested to be
located in the cuticular cell membrane and the algal decarbonylase in the
microsomal membranes. Both use highly hydrophobic fatty aldehydes as
substrate and need metal ions for their function. B. braunii decarbonylase is a
cobalt-porphyrin enzyme able to convert a fatty aldehyde to hydrocarbon and
CO without requiring any other cofactor under anaerobic conditions [48]. The
partially purified decarbonylase from pea is merely known to depend on metal
ions, probably copper, the activity being severely inhibited in the presence of
metal ion chelators [47]. Unfortunately, none of their genes have been cloned to
date.
Genetic approaches produced mutants of Arabidopsis that have altered
surface composition, including a decreased amount of hydrocarbons. One of
these mutants, cerl, was postulated to be located in a decarbonylase because it is
proportionally deficient only in alkanes (and alkane-derived metabolites) and
accumulated fatty aldehydes [11]. The cerl gene was cloned and shown to
bedeposits actively expressed in stem and in fruit tissue, which corresponded to
the main of waxes. Interestingly, cerl affected pollen fertility. The CER1
protein has been related, on the basis of sequence similarity, to C-5 sterol
desaturase and C-4 methyl oxidase, involved in cholesterol biosynthesis [49,50].
All these enzymes are integral membrane proteins containing many conserved
histidine residues.
378
Fig. 2. Multiple alignment of fatty acyl CoA reductases from the proteobacteria Acinetobacter
calcoaceticus [43] and Photobacterium leiognathi [44] and from the plant Simmondsia
chinensis (jojoba) [42]. Residues conforming the predicted catalytic triad are highlighted.
The n-heptane tissue-specific biosynthetic pathway in Pinus jeffreyi

proceeds through the polymerization of acetate via a tipical fatty acid synthase
reaction sequence yielding a C8 thioester, which subsequently undergoes a two-
electron reduction to generate a free thiol molecule and octanal, the latter
undergoes the direct loss of Ci to generate n-heptane [2] (Fig.4).
Aside from the obvious generality of the aforementioned pathway,
alternative reactions for alkane biosynthesis have been revealed recently. In
insects, whereas aldehyde was the immediate precursor of alkanes, the enzyme
involved in the deearbonylation step was proposed to be a P450 enzyme which
required the presence of molecular oxygen and NADPH or NADH (less
effectively) with the production of CO2 instead of CO [51]. In archaea, the
379
synthesis and cleavage of acetyl-CoA is catalyzed by the acetyl-CoA

decarbonylase synthase (ACDS) complex, a completely different enzyme
composed of five different subunits, [52, 53]. There is no information available
regarding an equivalent acyl-CoA decarbonylation reaction in bacteria.
3.1. Bacterial biosynthesis

In contrast to the heterogeneity of plant synthesized hydrocarbons [3], a
very large proportion of bacterial hydrocarbons are less dispersed in length and
structure. The best known example comes from the study of Sarcina lutea,
whose most abundant hydrocarbon presented a branch methyl on both ends of
the molecule, a double bond near the center, and contained a number of carbon
atoms equal to one less than two times the average number of carbon atoms in
the most abundant fatty acid [54] (Fig. 5). These structural characteristics, as
well as the distribution of the carbon chains of iso-leucine, valine and acetate in
the fatty acids and hydrocarbons synthesized in vivo, are consistent with a head-
to-head condensation of fatty acids mechanism [1, 54, 55]. A plausible precursor
for this mechanism would be multiple methyl-branched fatty acyl-CoA
molecules produced by the methyl-malonate driven elongation of fatty acid
molecules, as has been observed in mycobacteria [56]. In the head-to-head
condensation mechanism, one molecule of fatty acid undergoes decarboxylation.
When S. lutea cells are grown with low acetate, approximately 70% of the
incorporated palmitate molecules are decarboxylated, in contrast, when acetate
was included in the growth medium, palmitate was incorporated without
undergoing further decarboxylation, suggesting high specificity [57].
Fig. 3. Deduced Fatty Acyl CoA reductases are organized in three different groups on
the basis of sequence similarity. Sequence identification numbers in parenthesis.
380
Fig. 4. Proposed pathway for formation of n-heptane from acetate in Pinusjeffreyi.

Adapted from [2]
The in vitro incorporation of palmitate into hydrocarbons requires the

addition of CoA, Mg++, ATP, NADPH and pyridoxal or pyridoxamine
phosphate [58]. The requirement for the first three cofactors was consistent with
the participation of acyl-CoA and this was confirmed by showing that in the
absence of added CoA, palmitoyl-CoA was over 20 times better a precursor than
the free acid. Under these conditions, palmitate was sometimes decarboxylated
when added to the system. The specificity was originated from the source of
palmitate, whether it was free or esterified. Approximately 30% of the free acid
was decarboxylated while the methyl ester derivative was essentially 100%
decarboxylated. Also, the CoA derivative of palmitate was extensively
decarboxylated. Furthermore, the direct incorporation of palmitate required
piridoxamine as a cofactor instead of CoA [58]. Although it is clear that
biosynthesis occurs by a head-to-head condensation mechanism, the
intermediates of the pathway have not yet been fully elucidated.
Fig. 5 Relationship between isoleucine and acetate, the anteiso C15 fatty acid and the
C29 hydrocarbon with anteiso branch methyls in both ends. Modified from [1].
381
All this information leads to the idea that ability of bacteria to synthesize
hydrocarbons is widespread and that it probably occurs by more than one
mechanism which are significantly different compared to those described in
plants.
4. DOWNSTREAM PROCESSING
The isolation and refinement of bacterial hydrocarbons has not been approached
at production scale. Nevertheless, there is abundant information regarding the
operations developed for a similar procedure with B. braunii. The process of
extracting hydrocarbons from these cells can be thought of as consisting of three
major operations. The first is that of harvesting the cells from the growth
medium. This involves the concentration or flocculation of cells from the liquid
where it is grown. This operation can be achieved through a variety of means
that include filtration, mechanical centrifugation or concentration, gravitational
concentration or chemical flocculation. The most efficient method for large-
scale hydrocarbon recovery is chemical flocculation. For efficient extraction, the
cells must be concentrated to a semi-dry paste.
The second step is that of the actual physical extraction of the
hydrocarbon fuel from the cells. Under suitable conditions, up to 70% of the
total hydrocarbon content can be released by 30 min of contact with solvents.
The selected solvent should be immiscible with water, with a density
significantly different than water, should be non-toxic and reusable. In view of
these considerations, hexane appears to be the solvent of choice [59]. Growth
and hydrocarbon production are not affected by repeated extraction with hexane.
In fact, a higher content of hydrocarbons has been observed in hexane-treated
biomass relative to controls. Nevertheless, recovery yields are influenced by the
physiological status of the culture. Scale up of the extraction can be difficult,
given the algae propensity to aggregate. Extensive clumping shields a large
fraction of the biomass from exposure to solvent. Alternative methods aimed to
increase the oil extraction yield have been explored. Recovery yields are
markedly increased relative to freely suspended controls when cells immobilized
by adsorption in polyurethane foam were continuously extracted with hexane
[60]. Supercritical fluid extraction is another technology that has been applied
[61].
The third operation would be the collection and concentration of the
hydrocarbon product. Although biosynthetic hydrocarbons can be directly used
in internal combustion engines after extraction with hexane, its performance is
improved by further modification. Cellular hydrocarbons can be converted to
gasoline (60 to 70%), light cycle oil (10 to 15%), heavy cycle oil (2 to 8%) and
coke (5 to 10%) after catalytic cracking [62]. The yield of gasoline obtained by
382
this method is comparable to the yield obtained from petroleum. Additionally,

the gasoline produced has a sufficiently high octane number for direct use in
automobiles. In principle, all or some of these operations might be directly
applied for the extraction and upgrading of bacterial hydrocarbons.
5. FUTURE PROSPECTS
One of the main features impelling the study of microalgae as an alternative

source of hydrocarbons lies in their ability to fix carbon dioxide through
photosynthesis. The possibility of reducing the atmospheric carbon load by
direct recycling into fuels is appealing [63]. Despite all efforts, the production of
algal hyodrocarbons is not competitive with petroleum derived fuels, mainly due
to the slow growth rate of microalgae, the low extraction yields and the high
viscosity of the cultures. The alternative approach of cloning the algal
hydrocarbon synthesis genes into other microorganisms has proven to be
difficult.
Nevertheless, the necessity for an alternative source of hydrocarbons, not
only as fuels, but also for the fine-chemicals industry is still there. As presented
in this document, the ability to synthesize hydrocarbons is widely distributed
among eubacteria. The biosynthetic pathways seem to be completely different
compared to those observed in plants, involving a novel set of enzymatic
activities. At the present time, two different strategies might be suggested in
order to increase the accumulation of aliphatic hydrocarbons in bacteria. A
direct one, involving the identification and cloning of the genes encoding the
relevant activities involved, aimed at their subsequent expression in a suitable
host might result in an important advance. However, this strategy might not be
as straightforward as appears. It is well understood that the arbitrary
modification of the cellular carbon fluxes might severely impair cell viability
and performance. In this case, the natural deviation of the cellular carbon pool
into a reserve compound (hydrocarbons) might not be gratuitous but the result of
a finely tuned metabolic network. The deeper understanding of the physiology
of the process might eventually provide the knowledge basis for the rational
design of an imbalanced metabolism yielding the desired accumulation of
hydrocarbons without compromising cell viability. The current availability of
more powerful analytical as well as genetic tools enables us to face this
challenge.
Aknowledgements
The author thanks Shirley Ainsworth for assistance during the
bibliographical investigation. This work was supported by PEMEX grant 138.
383
REFERENCES
[I] P.W. Albro and J.C. Dittmer, Lipids, 5 (1970) 320.
[2] T.J. Savage, M.K. Hristova and R. Croteau, Plant PhysioL, 111 (1996) 1263.
[3] P. Metzger, C. Largeau and E. Casadevall in W. Herz, G.W. Kirby, W. Steglich and Ch.
Tamm (eds.) Progress in the Chemistry of Organic Natural Products, Springer-Verlag,
New York (1991).
[4] C. Hall, P. Tharakan, J. Hallock, C. Cleveland and M. Jefferson, Nature, 426 (2003)
318.
[5] A.L. Demain, Biotechnol. Adv., 18 (2000) 499.
[6] J.D. Rozzell, Bioorg. Med. Chem., 7 (1999) 2253.
[7] W. Gerhartz (eds.), Enzymes in industry: production and applications, VCH, Germany,
(1990).
[8] A. Liese and M.V. Filho, Curr. Opin. Biotechnol., 10 (1999) 595.
[9] R. Vazquez-Duhalt, E. Torres, B. Valderrama and S. Le Borgne, Energy Fuels, 16
(2002) 1239.
[10] Chemistry and Biochemistry of Natural Waxes, Elsevier, Amsterdam, (1976).
II1] M.G. Aarts, C.J. Keijzer, WJ. Stiekema and A. Pereira, Plant Cell, 7 (1995) 2115.
[12] M. Ashburner (eds.) Drosophila. A laboratory handbook, Cold Spring Harbor
Laboratory Press, Cold Spring Harbor (2003).
[13] RJ. Hamilton (eds.), Waxes: Chemistry, molecular biology and functions, The Oily
Press, Dundee, (2003).
[14] T. Rezanka, J. Zahradnik and M. Podojil, Folia Microbiol, 27 (1982) 450.
[15] R. Vazquez-Duhalt, Current Topics in Phytochemistry, 14 (1995) 69.
[16] A. Banerjee, R. Sharma, Y. Chisti and U.C. Banerjee, Crit. Rev. Biotechnol., 22 (2002)
245.
[17] M.-O. Park, M. Tanabe, K. Hirata and K. Miyamoto, Appl. Microbiol. Biotechnol., 56
(2001) 448.
[18] T.G. Tornabene, S.J. Morrison and W.E. Kloos, Lipids, 5 (1970) 929.
[19] E.G. Dediukhina and V.K. Eroshin, Usp. Sovrem. Biol, 76 (1973) 351.
[20] J. Baraud, A. Maurice and C. Napias, Bull. Soc. Chim. Biol., 52 (1970) 421.
[21] E.G. Dediujina, V.P. Zhelifonova, L.V. Andreev and B.K. Eroshin, Prikl. Biokhim.
Mikrobiol, 9(1973)813.
[22] H.L. Ehrlich in C.R. Bell, M. Brylinsky and P. Johnson-Green (eds.) Microbial
Biosystems: New Frontiers, Atlantic Canada Society for Microbial Ecology, Halifax,
Canada (1999).
[23] R.W. Stone and C.E. ZoBell, Ind. Eng. Chem., 44 (1952) 2564.
[24] C.E. ZoBell, World Oil, 130 (1950) 128.
[25] C.E. ZoBell, J. Sediment Petrol., 22 (1952) 42.
[26] G.J. Jankowski and C.E. ZoBell, J. Bacteriol., 47 (1944) 447.
[27] C.E. ZoBell, Science, 102 (1945) 364.
[28] P.W. Albro and C.K. Huston, J. Bacteriol., (1964)
[29] J.G. Jones and B.V. Young, Arch. Mikrobiol., 70 (1970) 82.
[30] J. Han and M. Calvin, Proc. Natl. Acad. Sci. U. S. A., 64 (1969) 436.
[31] T.G. Tornabene, J. Mol. Evol, 11 (1978) 253.
[32] T.G. Tornabene, T.A. Langworthy, G. Holzer and J. Oro, J. Mol. Evol., 13 (1979) 73.
[33] J. Han, E.D. McCarthy, W. Van Hoeven, M. Calvin and W.H. Bradley, Proc. Natl.
Acad. Sci. U. S. A., 59 (1968) 29.
384
[34] J. Oro, T.G. Tornabene, D.W. Nooner and E. Gelpi, J. Bacteriol, 93 (1967) 1811.
[35] J.B. Davis, Chem. Geol., 3 (1968) 155.
[36] J.G. Jones, J. Gen. MicrobioL, 59 (1969) 145.
[37] J. Templier, C. Largeau and E. Casadevall, Phytochemistry, 23 (1984) 1017.
[38] X. Wang and P.E. Kolattukudy, FEBS Lett., 370 (1995) 15.
[39] J. Vioque, T. Sirakova and P.E. Kolattukudy, J. Phycol., 35 (1999) 121.
[40] X. Wang and P.E. Kolattukudy, Biochem. Biophys. Res. Commun., 208 (1995) 210.
[41] J. Vioque and P.E. Kolattukudy, Arch. Biochem. Biophys., 340 (1997) 64.
[42] K.D. Lardizabal, J.G. Metz, T. Sakamoto, W.C. Hutton, M.R. Pollar and M.W. Lassner,
Plant Physiol, 122 (2000) 645.
[43] S. Reiser and C. Sommerville, J. Bacteriol., 179 (1997) 2969.
[44] J.W. Lin, Y.F. Chao and S.F. Weng, Biochem. Biophys. Res. Commun., 191 (1993)
314.
[45] J. Benach, S. Atrian, R. Gonzalez-Duarte and R. Ladenstein, J. Mol. Biol., 282 (1998)
383.
[46] M.W. Dennis and P.E. Kolattukudy, Arch. Biochem. Biophys., 287 (1991) 268.
[47] F. Schneider-Belhaddad and P.E. Kolattukudy, Arch. Biochem. Biophys., 377 (2000)
341.
[48] M. Dennis and P.E. Kolattukudy, Proc. Natl. Acad. Sci. U. S. A., 89 (1992) 5306.
[49] M. Bard, D.A. Bruner, C.A. Pierson, N.D. Less, B. Biermann, L. Frye, C. Koegel and R.
Barbuch, Proc. Natl. Acad. Sci. U. S. A., 93 (1996) 186.
[50] B.A. Arthington, L.G. Bennett, P.L. Skatrud, C.J. Guynn, R.J. Barbuch, CD. Ulbright
and M. Bard, Gene, 102 (1991) 39.
[51] J.R. Reed, P. Hernandez, G.J. Blomquist, R. Feyereisen and R.C. Reitz, Insect Biochem.
Mol. Biol., 26(1996)267.
[52] S. Gencic and D.A. Grahame, J. Biol. Chem., 278 (2003) 6101.
[53] E. Kocsis, M. Kessel, E. DeMoll and D.A. Grahame, J. Struct. Biol., 128 (1999) 165.
[54] P.W. Albro, J. Bacteriol., 108 (1971) 213.
[55] P.W. Albro, T.D. Meehan and J.C. Dittmer, Biochemistry, 9 (1970) 1893.
[56] S. Kikuchi, D.L. Rainwater and P.E. Kolattukudy, Arch. Biochem. Biophys., 295 (1992)
318.
[57] P.W. Albro and J.C. Dittmer, Biochemistry, 8 (1969) 1913.
[58] P.W. Albro and J.C. Dittmer, Biochemistry, 8 (1969) 3317.
[59] J. Frenz, C. Largeau, E. Casadevall, F. Kollerup and A.J. Daugulis, Biotehnol. Bioeng.,
34(1989)755.
[60] J. Frenz, C. Largeau and E. Casadevall, Enzyme Microb. Technol., 11 (1989) 717.
[61] R.L. Mendes, J.P. Coelho, H.L. Fernandes, IJ. Marrucho, J.M.S. Cabral, J.M. Novais
and A.F. Palavra, J. Chem. Technol. BiotechnoL, 62 (1995) 53.
[62] H. Kitazato, S. Asaoka and H. Iwamoto, Sekiyu Gakkaishi, 32 (1989) 28.
[63] P. Pedroni, J. Davison, H. Beckert, P. Bergman and J. Benemann, J. Energy Environ.
Res., 1 (2001) 136.
© 2004 Elsevier B .V. All rights reserved. 3 85
Chapter 14
The microbial diversity of deep subsurface oil reservoirs

N.-K. Birkeland
Department of Biology, University of Bergen, Box 7800, N-5020 Bergen,

Norway
1. THE DISCOVERY OF MICROBIAL LIFE IN DEEP OIL WELLS
"Souring" of oil reservoirs by the formation of hydrogen sulfide has been a

problem since the beginning of commercial oil production (see chapter 11).
Whether the sulfide is a result of abiotic chemical processes or due to
microbiological activities has been debated for many decades. The first
indications of an active role of sulfidogenic bacteria in this process were
presented already in 1926 based on the observation that sulfate-reducing bacteria
were widespread in oil-well production waters [1], Thermophilic sulfate-
reducing bacteria recovered from water produced from a North Sea oil-well was
in 1991 found to be able to survive, multiply and actively produce sulfide under
simulated reservoir conditions at temperatures and pressure up to 80°C and
4,500 psi, respectively [2, 3], thus demonstrating that sulfide formation can be
caused by microorganisms even under the extreme physical conditions found in
deep and hot petroleum reservoirs. Hyperthermophilic organisms able to grow at
even higher temperatures were soon to be recovered from deep oil wells in
Alaska [4] and the North Sea [5]. The question whether these bacteria were
contaminants that had been introduced to the oil wells during drilling or through
the repressurization by water injection into the oil-bearing strata, or whether
they belonged to an indigenous community of subsurface microbes remained an
open question.
During the last decade, however, our perception of this has changed with
the recovery from numerous oil wells of a large number of anaerobic
microorganisms representing a wide range of different metabolic types,
including sulfate- and iron-reducing bacteria, fermentative bacteria and
methanogenic Archaea. A number of different thermophilic and hyperthermo-
philic organisms have now been recovered from both terrestrial and offshore
386
wells that have never been water-flooded [6, 7], strongly indicating an
indigenous origin. The presence of an indigenous microbial community is
further supported by the isolation of novel species that never has been recovered
from any other sources, and by the physiological characteristics of some isolates
indicating a close adaptation to the respective in situ reservoir conditions.
Recovery of closely related strains from remote oil fields [7-9] also supports the
existence of a widespread microbial biosphere in oil-bearing strata. However,
the problems associated with recovery of biological samples from oil wells are
extensive. Sampling from wellheads is the only way of collecting samples from
petroleum reservoirs, and the possible sources of contamination are numerous. It
is furthermore possible that exogenous mesophilic bacteria can propagate in top
facilities of the oil field installations. Occasionally, aerobic and microaerophilic
bacteria are recovered from produced oil-well water, but available chemical data
suggest that oxygen is absent in oil reservoirs, and these isolates should thus not
be considered as being truly indigenous to deep oil wells. In addition to SRB
and fermentative bacteria, Voordouw et al. [10] detected several aero- and
microaerophilic bacteria in a 600 m deep water flooded oil reservoir in Canada.
Nitrate-reducing bacteria were recovered from a similar shallow oil field [11]. It
is postulated that oxygen and nitrate is able to reach these shallow oil-bearing
formations through diffusion or convection from surface layers, giving support
to a limited community of bacteria respiring with nitrate or oxygen [10, 11].
The number of bacterial cells in water produced from oil reservoirs is
highly variable. Total bacterial counts demonstrated the presence of more than
106 cells per ml in water from a non-water flooded reservoir in California [6],
and from sulfide-rich production water from a German water-flooded petroleum
reservoir up to 6.3 x 106 colony-forming units of sulfate-reducing bacteria per
ml has been obtained [12]. Although only a few bacteria per ml have been
detected in water produced from some oil wells, these results show that the
bacterial density can be significant. In the present chapter our current knowledge
of these microorganisms is reviewed.
2. SULFATE-REDUCING BACTERIA AND ARCHAEA (SRB)
Sulfate-reducing prokaryotes constitute a diverse physiological group of sulfide-

producing microorganisms able to carry out anaerobic respiration with sulfate as
a terminal electron acceptor. They are widespread in anaerobic environments
were sulfate is available. Typically, they oxidize organic acids either to acetate,
or by complete oxidation of acetate or other acids to CO2, but autotrophic
species using H2 and CO2 as energy and carbon-sources are also common. A
wide range of organic acids, e.g. acetate, propionate, butyrate, pentanoate and
hexanoate, at concentrations up to 20mM has been found in oil reservoirs [13,
14]. On the other hand, the concentration of sulfate seems to be rather low in
387
non-water-flooded reservoirs, and the abundance of SRB is therefore probably

sulfate-limited.
There are few reports on the quantification of SRB in produced water, but
the few estimates that have been made show a considerable variation. As
mentioned above, up to 6.3 x 106 colony-forming units of SRB per ml was
obtained from wellhead water from a German sulfide-rich reservoir [11].
Acetate was used as a substrate. In a more recent survey, up to 4.5 x 104 SRB
per ml were counted in wellhead water from low-temperature reservoirs using
most probable number (MPN) technique and with lactate + acetate as substrates
[15]. The number of SRB decreased with increasing in situ temperature, and
from wells with an in situ temperature of 85° no SRB could be detected. These
results are comparable to the results reported by Nazina et al. [16] and Rozanova
and Nazina [17], who detected only very low populations of SRB in high-
temperature reservoirs. A low number of SRB have also been detected in
reservoirs of high salinity. In a study of sulfate reducers in a high-temperature
oil field in the North Sea up to 2 x 104 thermophilic or hyperthermophilic SRB
were detected directly in the produced water using fluorescent antibody
technique with conjugated antibody directed against three specific SRB groups
[18]. No correlation between the duration of seawater injection and the number
of SRB in the water was observed, indicating that thermophilic and
hyperthermophilic SRB are indigenous to this oil field.
The salinity of oil-well water is very variable and is an important factor
for the in situ microbial activity and diversity. Oil-well brines with salinity
above 20% have been found, but most oil well formation waters have a
moderate salinity (<6%). No extremely halophilic SRB has ever been recovered
from oil wells, the most halophilic SRB being Desulfovibrio vietnamensis [19]
and Desulfotomaculum halophilum [20], which both grow optimally at a salinity
around 5%. Voordouw et al. [21] detected different communities of SRB in
enrichments from oil wells with low and high salinity, demonstrating that
salinity is an important discriminating factor.
2.1. Mesophilic SRB

Mesophilic bacteria generally have growth optima in the 25-40°C
temperature range. Mesophilic SRB have frequently been isolated from oil field
production waters (Table 1). These organisms are believed proliferate in the top
facilities of the petroleum installations. They also seem to proliferate in shallow
low-temperature oil wells, but can also be found in deeper wells that have been
subject to long-term water flooding. However, these organisms cannot be
regarded as indigenous components of the deep oil well microbial community.
Most of the mesophilic isolates are Gram-negative bacteria belonging to the
delta subdivision of proteobacteria. They include species from the genera
Desulfobacter, Desulfobacterium, Desulfobulbus, Desulfomicrobium and
388
Desulfovibrio. One Gram-positive mesophilic SRB has been recovered,

Desulfotomaculum halophilum. Except for Desulfobacter vibrioformis and
Desulfobacterium cetonicum, which oxidize their they substrates completely to
CO2, they oxidize organic acids to the level of acetate. D. vibrioformis is
restricted to oxidation of acetate only, while the other species can oxidize a
range of substrates. Most of them can oxidize lactate and pyruvate, which are
common substrates for SRB. Desulfovibrio gabonensis and Desulfovibrio
vietnamensis can use a wide range of substrates, including formate, malate,
fumarate and ethanol. D. cetonicum has a unusual substrate range, as it, in
addition to common substrates like buturate, lactate, pyruvate, alcohols and fatty
acids can oxidize ketones, benzoic acid, toluene and p/m-cresol. This property
makes D. cetonicum an interesting organism from a bioremediation perspective.
Many of the isolates can also utilize H2 as energy source when grown in
presence of a carbon source such as formate and acetate, but only
Desulfomicrobium apsheronum is able to grow with H2 autotrophically using
CO2 as the sole carbon source.
2.2. Thermophilic SRB

Thermophilic SRB isolated from oil well production waters also include
members of the delta subdivision of proteobacteria and the Gram-positive genus
Desulfotomaculum, but in addition, members of the genus Thermodesulfo-
bacterium, a deeply branching lineage of thermophilic Gram-negative bacteria,
have been recovered (Table 1). The most frequently isolated thermophilic SRB
belong to the Gram-positive spore-forming genus, Desulfotomaculum.
Desulfotomaculum nigrificans, isolated from Western Siberia in 1978, is a rather
restricted organism oxidizing lactate and ethanol incompletely.
Desulfotomaculum kuznetsovii was originally isolated from an underground
thermal water system [22], and has later been recovered from wellheads in the
Paris Basin [23]. D. kuznetsovii is nutritionally a very versatile organism. It is a
methylotrophic bacterium able to oxidize methanol in addition to substrates like
H2, formate, acetate, aliphatic fatty acids, ethanol, lactate, fumarate and malate.
It oxidizes its substrates completely to CO2. One novel species,
Desulfotomaculum thermocisternum, has been recovered from a North Sea oil-
well. It has a more restricted substrate range than D. kuznetsovii, but shares its
ability to grow on lactate, alcohols and aliphatic carboxylic acids. D.
thermocisternum was isolated from produced oil reservoir water obtained before
breakthrough of injected seawater, suggesting that it is a true indigenous oil-
field organism.
Thermodesulfobacterium thermophilum, originally described in 1974, was
the first thermophilic organism to be isolated from water produced from an oil
well. It was first described as Desulfovibrio thermophilus, but later renamed to
Thermodesulfobacterium mobile before its final name, T. thermophilum, was
389
validated [24]. The genus Thermodesulfobacterium is the third deepest-

branching phylogenetic lineage in the bacterial domain. Thermodesulfo-
bacterium commune was originally isolated from a geothermal area in
Yellowstone National Park [25], but later it has been recovered from a
continental non-water flooded reservoir in the Paris Basin [6]. T. thermophilum
has been recovered also from the Paris Basin and the North Sea [26]. This
widespread occurrence indicates a cosmopolitan subterranean distribution of
these species which obviously must be indigenous to oil reservoirs. Both species
can utilize H2, formate, lactate and pyruvate as energy sources. The organic
acids are oxidized incompletely to CO2.
Two thermophilic SRB belonging to novel genera of the delta subdivision
of proteobacteria, Desulfacium infernum and Thermodesulforhabdus norvegicus
have been isolated from North Sea oil wells [27, 28]. Both species can utilize a
range of organic acids, including acetate, which they oxidize completely to CO2.
Table 1
Sulfate-reducing prokaryotes recovered from oil field production waters
Species Temperature Location of Complete Ref.

optimum [°C] oil field oxidizer
Archaeoglobus fulgidus strain 7324 76 North Sea + 5
'Archaeoglobus lithotrophicusM nd North Sea nd 4
Archaeoglobus profundus nd North Sea + 4
Desulfacinum infernum 60 North Sea + 27
Desulfobacter vibrioformis 33 North Sea + 73
Desulfobacterium cetonicum 30-35 Russia + 74
Desulfobulbus rhabdoformis 31 North Sea - 75
Desulfomicrobium sp. 25-35 North Sea - 76
Desulfomicrobium apsheronum 25-30 Apsheron - 77
Desulfotomaculum spp. 65 North Sea - 2
Desulfotomaculum halophilum 35 Paris Basin, France - 20
Desulfotomaculum kuznetsovii 60-65 Russia + 22
Desulfotomaculum nigrificans 60 Russia - 78
Desulfotomaculum thermocisternum 62 North Sea - 79
Desulfovibrio gabonensis 30 Gabon, WestAfrica - 15
offshore
Desulfovibrio longus 35 Paris Basin, France - 80
Desulfovibrio vietnamensis 37 Vietnam, offshore - 19
Thermodesulfobacterium 65 Caspian Sea - 26,
thermophilum North Sea 81
Thermodesulfobacterium commune 70 Paris Basin, France - 6
Thermodesulforhabdus norvegicus 60 North Sea + 28
a
Not yet validly described.
390
2.3. Hyperthermophilic SRB

Hyperthermophilic microbes grow at temperatures above 80°C and most
of them belong to the domain Archaea. Until recently, Archaeoglobus was the
only known archaeal genus able to carry out dissimilatory sulfate reduction, but
an additional archaeal sulfate reducer, Caldivirga maquilingensis, isolated from
a hot spring in the Philippines, has now been described [29]. Only species
belonging to the genus Archaeoglobus have so far been found in oil-field
production waters. The type species of the genus, Archaeoglobus fulgidus (type
strain VC-16), was first isolated from a shallow hydrothermal system in the
Mediterranean [30, 31]. In 1994, A. fulgidus strain 7324 was recovered from an
oil well in the Norwegian sector of the North Sea [5]. A. fulgidus has also been
recovered from the East Shetland Basin of the North Sea along with its relatives,
Archaeoglobus profundus and 'Archaeoglobus lithotrophicus' [4]. A. fulgidus
strain 7324 has a lower optimal growth temperature (76°C) than the type strain,
VC-16 [83°C], and unlike VC-16 it cannot grow autotropohically with H2 and
CO2. Both strains can grow on lactate and pyruvate. In contrast to VC-16, strain
7324 is able to grow on starch [32]. A. profundus is a obligate mixotrophic
species growing only on a mixture of H2 and an organic carbon source (e.g.
acetate) [33]. A. litotrophicus can grow autotrophically on H2 and CO2 [5], but
has not yet been validly described.
3. METHANOGENIC ARCHAEA
Biological generation of methane is limited to a group of strict anaerobic

archaeal organisms, the methanogens, which form methane as a product of
anaerobic respiration. They are widespread in nature, and have been found in
most natural anoxic environments. Their most frequently used energy source is
hydrogen, which is usually oxidized with CO2 as electron acceptor. These
hydrogenotrophic methanogens are also autotrophs, using CO2 as carbon source
by assimilation of carbon via the acetyl-CoA pathway. Other possible energy
sources include one-carbon compounds like formate, methanol and
methylamines. Methanol is converted to methane, carbon dioxide and water. A
few methanogens can utilize acetate, which is converted to methane and CO2
through an acetoclastic reaction. Ethanol and propanol can also be used by some
species. Some commonly used methanogenic reactions and their standard
changes in free energies are given in Table 2.
Biological methane formation in oil-bearing strata has been well
documented [34, 35, 36, 37], but few methanogens able to grow under in situ
conditions have been isolated. Although hyperthermophilic methanogens are
frequently isolated from geothermal environments like hot springs and
hydrothermal vents, hyperthermophilic methanogens have never been isolated
from oil wells. Methanobacterium thermoalcaliphilum, which grows optimally
391
at 65°C and has an upper growth-limit at 80°C, has been recovered from oil-
fields in Tataria and Western Siberia [38]. Two other thermophilic methanogens
have been recovered from oil wells, which both share with M.
thermoalcaliphilum, the ability to use hydrogen as energy source and CO2 as
carbon source (Table 3). Although positive enrichments of acetate-utilizing
methanogens at 60°C have been obtained, it has not yet been possible to obtain
pure cultures of acetoclastic thermophilic methanogens from oil wells [37, 38,
39]. A plausible reason is that thermophilic methane production from acetate in
these environments might be a result of interspecies hydrogen transfer [37].
Several mesophilic methanogens have been isolated, including
hydrogenotrophic types as well as strains utilizing methylamines and acetate
(Table 3).
4. FERMENTATIVE BACTERIA AND ARCHAEA
Fermentative organisms are able to utilize organic substances such as

carbohydrates and peptides for growth producing organic acids, ammonium and
hydrogen as fermentation products. In contrast to SRB and methanogens, these
organisms do not use any external electron acceptor in their energy-yielding
reactions, thus maintaining an internal red-ox balance. Occasionally,
fermentative microbes can transfer excess reduction power to sulfur compounds
like thiosulfate or elemental sulfur, which both are reduced to sulfide, thereby
improving growth rate and substrate utilization. However, the sulfur compounds
only serve as electron "sinks", and the sulfur-reducing reactions do not appear to
be linked to any energy-conserving mechanism. This sulfur-reducing ability can
be an important step in the geochemical cycling of sulfur in these anaerobic
thermal environments. Fermentative bacteria from a great variety of
phylogenetic lineages have been isolated from oil reservoirs, especially during
the recent years (Table 4).
Table 2
Examples of methanogenic reaction and their standard free energy changes.
Reaction AG 0 ' [KJ/mol CH 4 ]

4 H 2 + CO 2 -> CH 4 + 2H 2 O -135.6
4 Methanol - • 3CH 4 + CO 2 + 2H 2 O -104.9
4 Methylamine + 2H 2 O -> 3CH 4 + CO 2 + 4NH 4 + -75.0
Acetate -> CH 4 + CO 2 -31.0
Table 3
Methanogens recovered from oil reservoirs
Species Temperature Location Substrates used References

optimum [°C] [methane produced from]
Methanobacterium bryantii 37 Tatarstan and Western H2 38
Siberia
Methanobacterium ivanovii 45 Tatarstan H2 34,82
Methanobacterium 60 California H2 83
thermoaggregans
Methanobacterium 65 Tatarstan and Western H2 38
thermoalcaliphilum Siberia
Methanobacterium 60 Tatarstan H2 36
thermoautrophicum
'Methanocalculus 38 France H2, formate 84
halotolerans'
Methanococcus 60 North Sea H2 39
thermolithotrophicus
Methanohalophilus euhalobius 28-37 Western Siberia methylamines 85
''Methanoplanus petrolearius' 37 Gulf of Guinea H2 + CO2, formate, 2- 86
propanol + CO2
Methanosarcina mazei 37 Tatarstan Acetate, methylamines 87
Methanosarcina siciliae 40 Gulf of Mexico methylamines 88
393
4.1. Order Thertnotogales

Six novel species belonging to the genera Thermotoga, Petrotoga and
Thermosipho, which all belong to the order Thermotogales have been isolated
from oil reservoirs and described the last 4 years [40, 41, 42, 43]. Members of a
related genus, Geotoga, were isolated earlier [44]. Bacteria belonging to this
order are rod-shaped organisms possessing a characteristic outer sheath-like
structure called a 'toga' [45] and represent one of the deepest phylogenetic
branches in the bacterial line of evolutionary descent. Most members are
thermophiles with optimal growth between 50 and 70°C, but some species of the
genus Thermotoga are hyperthermophiles. In total, 13 different species of this
order have been recovered from sources around the world, including oil
reservoirs in France, Western Siberia, Japan, Africa, USA, the Gulf of Mexico
and the North Sea. Most of these species have so far only been recovered from
oil reservoir. The widespread distribution and uniqueness of these bacteria is
clearly in support of their cosmopolitan nature as subterranean indigenous
bacteria belonging to a natural microbial community in oil-bearing strata. Their
temperature growth ranges also correlate well with the in situ temperature of the
reservoirs. Nutritionally, these organisms are very versatile heterotrophs,
fermenting a variety of organic substrates ranging from mono- and
disaccharides, polysaccharides and protein hydrolysates.
Most of the oil-well isolates belonging to the Thermotoga genus can
ferment a large number of carbohydrates and grow also on peptide substrates
like peptone and bio-Trypticase. Small amounts of yeast extract is usually
required for growth on carbohydrates. Thermotoga subterranea, however, seems
to be restricted to growth on complex media such as peptone and yeast extract
and cannot grow on defined carbon sources. Thermosipho geolei is nutritionally
very similar to the Thermotoga members, but is rather limited with regard to
carbohydrate utilization, as it was found to only grow on glucose when
carbohydrates were tested as substrates [42]. While members of the Thermotoga
and Thermosipho genera have been isolated also from deep-sea hydrothermal
vents, members from the Petrotoga and Geotoga genera have so far only been
recovered from petroleum reservoirs. It is possible that these genera are unique
to these special microbial habitats. An interesting characteristic of
Thermotogales members is the widespread ability to degrade and utilize
polymeric substrates such as starch, maltodextrin, xylan and peptides. Even a
cellulolytic bacterium has been recovered (T. petrophila). Xylanases from
thermophilic microbes has received much interest lately as it has a considerable
biotechnological potential for the paper pulping industry. The common capacity
of these organisms to utilize a wide range of polymeric substances for growth
indicates a saprophytic life style and a role as consumers in this subsurface
microbial ecosystem.
394
Table 4
Fermentative bacteria and Archaea recovered from oil reservoirs
Species Optimal Location of oil Reduction of sulfur Ref.

Temp. reservoir compounds
S° S2O3"
Acetoanaerobium romashkovii 37 Western Siberia Nd nd 59
Anaerobaculum thermoterrenum 55 Utah + + 56
Dethiosulfovibrio peptidovorans 42 Congo, offshore + + 58
Fusibacter paucivorans 37 Congo, offshore + + 89
Geotoga petraea 50 Oklahoma/Texas + nd 44
Geotoga subterranea 45 Oklahoma/Texas + nd 44
Haloanaerobium acetoethylicum 34 Gulf of Mexico Nd nd 48
Haloanaerobium congolense 42 Congo, offshore + + 49
Haloanaerobium kushneri 35-40 Oklahoma - -? 51
Haloanaerobium salsuginis 40 Oklahoma Nd nd 50
Petrotoga mexicana 55 Gulf of Mexico + + 40
Petrotoga miotherma 55 Oklahoma/Texas + nd 44
Petrotoga mobilis 58-60 North Sea + + 90
Petrotoga olearia 55 Western Siberia + - 41
Petrotoga sibirica 55 Western Siberia + - 41
Spirochaeta smaragdinae 37 Congo, offshore + + 53
Thermoanaerobacter Nd Western Siberia Nd Nd 68
acetoethylicus
Thermoanaerobacter subterraneus 65 France 55
Thermoanaerobacter brockii 55-60 France + + 54
Thermococcus sp. 85 Niigata, Japan Nd Nd 46
Thermococcus sibericus 78 Western Siberia + nd 47
Thermosipho geolei 70 Western Siberia + - 42
Thermotoga elfii 66 Africa - + 95
Thermotoga hypogea 70 Cameroon - + 97
Thermotoga maritima Ml2597 nd Western Siberia Nd Nd 68
Thermotoga naphthophila 80 Niigata, Japan + [weak +] 43
Thermotoga petrophila 80 Niigata, Japan + + 43
Thermotoga subterranea 70 Paris Basin, - + 98
France
4.2. Archaea
Hyperthermophilic fermentative Archaea belonging to the genus
Thermococcus were first isolated from Japanese oil reservoirs in 2000 [46].
Although the isolates were not described at the species level, they were
nutritionally very similar to other thermococci, growing on proteinaceous
substrates, yeast extract and amino acids. Although the in situ reservoir
temperature ranged from 50 to 58°C, the optimal temperature of the isolates was
above 80°C [43, 46]. The number of hyperthermophilic cocci that were present
395
in produced water from the oil wells was estimated to be up to 4.6 x 104
cells/ml. The organisms were not able to grow in produced water due to lack of
required nutrients, but under starved conditions at 50°C the viable cell count was
stable for 200 days, indicating that they have developed an amazing ability to
survive prolonged periods under starved conditions. This feature is probably
important for the continued existence in a hot subterranean oil reservoir where
the supply of nutrients is limited [43]. A novel thermococcal species,
Thermococcus sibiricus, has been isolated from an oil reservoir in Western
Siberia [47], and a novel species has also recently been recovered from a North
Sea oil well (Birkeland, unpublished). This indicates that the high-temperature
oil reservoir biosphere is also inhabited by indigenous hyperthermophilic
Archaea. There is also evidence for their presence in other oil wells [4, 7, 8].
4.3. Halophilic bacteria

Some oil reservoirs contain highly saline brines with salinity above 20%.
Anaerobic fermentative bacteria that can grow at this high salinity have been
isolated from such oil wells in Africa, the Gulf of Mexico and USA [48, 49, 50,
51]. These bacteria belong to the genus Haloanaerobium, one of a few genera of
anaerobic fermentative bacteria that are adapted to high-salt conditions. These
organisms have frequently been isolated from bottom sediments of hypersaline
lakes and lagoons. They are typically saccharolytic bacteria, fermenting a range
of carbohydrates. Only mesophilic members, with growth optima between 34
and 42°C, have so far been recovered from oil reservoirs, but related
thermophilic bacteria have been isolated from other sources. In contrast to other
halophilic bacteria, which accumulate organic osmotic solutes in order to
maintain an osmotic balance between the surrounding medium and the
cytoplasm, the members of the Haloanaerobium genus accumulate Na+, K+ and
Cl as compatible solutes [48, 52]. Accumulation of inorganic ions is a property
they share with extremely halophilic Archaea, which accumulate up to 3M KC1.
Another unusual feature of the Haloanaerobium genus is the nature of their cell-
wall structure as compared to their phylogenetic position. They stain Gram-
negative when subjected to the Gram-staining procedure, but electron
micrographs show the presence of a typical Gram-negative cell wall. However,
based on the sequence of the 16S rRNA they form a deep-branching cluster
within the phylum of Gram-positive bacteria. This is taken as evidence that
certain descendants of the ancestors of the Gram-positive bacteria maintained
their Gram-negative cell wall structure, which is also the case with certain other
strict anaerobic relatives.
A moderately halophilic spirochete, Spirochaeta smaragdinae, growing
optimally at a salinity of 5% has been isolated from an offshore oil well in
Congo [53]. This bacterium is nutritionally very versatile, fermenting
396
carbohydrates, glycerol, fumarate, peptides and yeast extract, and is the only
spirochete so far isolated from the deep subsurface. Evidence for the presence of
a closely related spirochete in North Sea oil wells has, however, been obtained
using direct molecular techniques (Birkeland, unpublished).
4.4. Other mesophilic and thermophilic fermentative bacteria

Thermoanaerobacter subterraneus and Thermoanaerobacter brockii
subsp. lactiethylicus are Gram-positive thermophilic carbohydrate-fermenting
bacteria isolated from French oil wells [54, 55]. They grow optimally at 65 and
55-60°C, respectively. T. brockii forms endospores. Spores have not been
observed in cultures of T. subterraneus, but because this organism can survive
autoclaving for 45 minutes, the presence of heat-resistant forms has been
suggested [55]. Anaerobaculum thermoterrenum is a Gram-positive bacterium
isolated from an oil well in Utah [56]. It defines a novel moderately
thermophilic genus, Anaerobaculum, phylogenetically related to Thermo-
anaerobacter. It is nutritionally versatile, growing on a wide range of
carbohydrates including cellulose, as well as peptone and organic acids like
citrate. It is able to utilize crotonate as electron acceptor, reducing it to butyrate
[57]. Although it groups within the phylum of Gram-positive bacteria, A.
thermoterrenum has a Gram-negative cell wall, a feature it shares with the
halophilic genus Haloanaerobium. A mesophilic thiosulfate-reducing bacterium
termed Dethiosulfovibrio peptidovorans, which can only utilize peptides and
amino acids for growth, has been isolated from a corroding offshore oil well in
Congo [58]. The isolate was shown to cause a strongly enhanced corroding
activity of steel in the presence of thiosulfate, indicating that apart from SRB,
thiosulfate-reducing bacteria can contribute to this process. Phylogenetically,
this organism groups within Gram-positive bacteria of the clostridial group, but
shares with Anaerobaculum a multilayered cell wall ultrastructure typical of
Gram-negative bacteria. Together with Anaerobaculum and a few other small
genera, Dethiosulfovibrio spp. form a separate phylogenetic cluster in the
Clostridium group of Gram-positive bacteria. Another Gram-positive bacterium,
Fusibacter paucivorans, isolated from an African oil well, defines a novel genus
within the Clostridium phylum. F. paucivorans is a mesophilic and halotolerant
Gram-positive bacterium fermenting a limited number of carbohydrates. Spores
have never been observed.
Acetoanaerobium romashkovii is a homoacetogenic mesophilic bacterium
isolated from a Siberian oil field in 1992 [59]. Homoacetogenic bacteria is a
group of anaerobes that can use CO2 as an electron sink and reduce it to acetate
as a fermentation product via the carbon monoxide dehydrogenase pathway.
Hydrogen can be used as energy source, but as is the case for A. romashkovii,
various one-carbon compounds, amino acids and sugars can also be utilized.
397
Although the members of the genus Acetoanaerobium stain Gram-negative, they

possess a Gram-positive cell-wall architecture [60].
5. IRON REDUCERS
It has been suggested that reduction of iron is an ancient and widespread

mechanism for anaerobic respiration [61]. The contribution of Fe(III) reduction
to the cycling of iron and organic matter in various anaerobic environments,
including the subsurface [62, 63], is now well known, and geochemical evidence
suggest that Fe(III) was the first electron acceptor of global significance during
the early evolution of microbial energy metabolism [64]. Mesophilic Fe(III)-
reducing bacteria have been detected in oil field fluids [65, 66], some of which
have been identified as Shewanella putrefaciens (formerly Alteromonas
putrefaciens) [65, 66]. S. putrefaciens can use hydrogen and formate as electron
donors during iron respiration, and can also reduce sulphur and sulphite. A
moderately thermophilic Fe(III)-reducing bacterium, Deferribacter
thermophilus, has been isolated from a North Sea oil well [67]. In addition to
iron, it can use manganese and nitrate as electron acceptors, and use
proteinacous substrates, hydrogen and organic acids as energy sources. Most
iron reducers that have been tested can reduce manganese in addition to iron.
In a survey of the iron-reducing capability of thermophilic and
hyperthermophilic isolates from oil reservoirs in Western Siberia, 8 of nine
strains were found to reduce Fe(III) using pepton or hydrogen as energy source
[68]. These iron reducers included 5 strains belonging to the thermophilic
bacterial genera Thermotoga and Thermoanaerobacter, and 3 hyperthermophilic
archaeal strains belonging to genus Thermococcus. The isolates had not been
subjected to an enrichment step in Fe(III)-containing media. In the same
investigation it was shown that the major part of 25 samples taken from these oil
reservoirs were positive for Fe(III) reduction in peptone or hydrogen
enrichments. These results suggest that iron reduction is a common feature of
thermophilic and hyperthermophilic microorganisms in deep subsurface
petroleum reservoirs. This is further supported by the observation that the
sulphate-reducing archaeon A. fulgidus and the methanogen M.
thermolithotrophicus, which both have been found in hot water produced from
North Sea oil wells, also can reduce Fe(III) [62]. Whereas Thermotoga maritima
was previously considered to possess only a fermentative metabolism, it was
found later to grow respiratory in the presence of Fe(III), coupling Fe(III)-
reduction with energy conservation [62]. The function of iron reduction in the
energy metabolism of these bacteria needs further investigations.
398
6. CULTURE-INDEPENDENT APPROACHES
Direct analyses of uncultured natural microbial communities based on

fluorescence microscopy using fluorescent antibodies (FA) or oligonucleotide
probes directed against specific bacterial groups, and amplification and analyses
of genes from DNA extracted from environmental samples have contributed
significantly to an improved understanding of the structural complexity of
natural microbial communities during the last decade. The development of the
polymerase chain reaction (PCR), automated DNA sequencing and DNA
microchip technologies has provided efficient tools for culture-independent
analyses of microbial diversity.
Genus specific antibodies directed against the hyperthermophilic
Archaeoglobus, and the thermophilic genera Desulfotomaculum and
Thermodesulforhabdus have been used for analyzing the distribution of SRB in
produced oil reservoir waters sampled at different dates and from different wells
in the Gullfaks field in the North Sea [18]. Archaeoglobus and
Thermodesulforhabdus strains were detected in 4 of 16 samples, but in most
samples the numbers were below the detection limit. The number of cells varied
from 400 to 2 x 104 per ml. Desulfotomaculum strains were only detected in one
of the wells. In 3 wells only one of the three types could be detected. This
investigation demonstrated that the distribution of these SRB in the Gullfaks
field is subject to strong spatial and temporal variations.
Oligonucleotide microchips containing specific 16S rRNA probes
targeting selected microbial groups encompassing key genera of thermophilic
bacteria and Archaea were used for probing the diversity in water samples from
the Samotlor high-temperature oil reservoir in Western Siberia [37]. The results
confirmed the presence of organisms identified by culture-based methods, but
organisms that had not been identified by culture-dependent methods were also
detected. These organisms included representatives of the aerobic genus
Thermus and the microaerophilic Aquifwales group, as well as anaerobes
belonging to the genera Desulfurobacterium and Thermovibrio. None of these
groups have previously been detected in oil reservoirs. Orphan et al. [7] made
16S rDNA libraries from total DNA from water produced from high-
temperature petroleum reservoirs in California, using either universal or archaeal
primer sets, 83 unique clones were identified from the universal library, and
sequence analysis revealed that the majority of the clones grouped within the
bacterial domain, with only 8.8% of the library affiliated with the domain
Archaea. The dominating bacterial phylotypes were close relatives of Gram-
positive fermentative genera Acidaminococcus and Thermoanaerobacter, and to
the halophilic proteobacterium Halomonas. The genus Acidaminococcus has
never been isolated from petroleum reservoirs. Clones related to aerobic bacteria
were also identified. The archaeal library was dominated by methanogen-like
399
clones, with a lower percentage of clones belonging to the fermentative

Thermococcales. Interestingly, archaeal sequence types related to the
acetoclastic Methanosarcinales were identified, suggesting the presence of
acetate-utilizing methanogens. The molecular analyses demonstrated a much
higher diversity than indicated by cultivation-dependent methods. The same
strategy was also used by Voordouw et al. [69] with samples from shallow low-
temperature oil field in Canada. A variety of Gram-negative SRB were detected,
but a limited number of clones representing fermentative organisms were
obtained. The results also indicated the presence of microaerophilic types.
Molecular approaches have also been used for assessing the diversity in
primary enrichments. Cloning and sequencing of 16S rDNA have been used for
bacterial diversity analysis of SRB enrichments from North Sea oil field samples
[70]. The library was dominated by Gram-positive SRB (Desulfotomaculum)
and fermenters belonging to the Gram-positive genera Thermoanaerobacter and
Clostridium. In addition, two clones, which probably represent a novel
undescribed genus of deeply branching Gram-positive bacteria were obtained.
Reverse sample genome probing (RSGP) is a microbial community
fingerprinting technique that have been used successfully for both qualitative
and quantitative analysis of SRB enrichments [21, 71, 72]. Using different
carbon sources for enrichment of mesophilic SRB, 34 different types of SRB
were detected by RSGP [21], giving a glimpse of the enormous diversity of SRB
in low-temperature oil fields.
7. CONCLUSIONS
Several lines of evidence indicate the existence of an indigenous deep

subsurface microbial community in petroleum reservoirs; a) a large variety of
prokaryotic groups have been isolated from produced waters around the world,
b) many of the isolates belong to unique species or genera that have not been
recovered from any other habitats, c) highly similar microorganisms have been
isolated from geographically remote oil fields, d) several isolates are adapted to
growth under the extreme in situ reservoir conditions, and e) using cultivation-
independent approaches, molecular analyses have directly verified the presence
of previously cultivated strains in produced water and have demonstrated the
existence of a highly diverse community. A wide range of chemolitoautho-
trophic and organotrophic types have been isolated from deep oil wells;
hydrogenotrophic methanogens and sulfate-reducers, heterotrophic
methanogens, fermentative bacteria and Archaea, heterotrophic sulfate-reducers,
iron-reducing microbes, and saprophytes. Organisms with temperature optima
from 25 to 85°C have been recovered, and optimal salinity ranges from fresh-
water conditions to more than 10% salinity. This enormous physiological
diversity suggests that this microbial community constitutes a complex
400
ecosystem with an active biogeochemical cycling of carbon and minerals.

Further characterization of this subsurface biosphere in order to understand the
ecological role and significance of these organisms is a major
geomicrobiological challenge.
REFERENCES
[I] E.S. Bastin, Science, 63 (1926) 21.
[2] J.T. Rosnes, T. Torsvik and T. Lien, Appl. Environ. Microbiol., 57 (1991) 2302.
[3] J.T. Rosnes, A. Graue and T. Lien, SPE Production Engineering, May (1991) 217.
[4] K.O. Stetter, R. Huber, E. Blochl, M. Kurr, R.D. Eden, M. Fielder, H. Cash and I.
Vance, Nature, 365 (1993) 743.
[5] J. Beeder, R.K. Nilsen, J.T. Rosnes, T. Torsvik and T. Lien, Appl. Environ. Microbiol.,
60(1994)1227.
[6] S. L'Haridon, A.L. Reysenbach, P. Glenat, D. Prieur and P. Jeanthon, Nature, 377
(1995)223.
[7] V.J. Orphan, L.T. Taylor, D. Hafenbradl and E.F. Delong, Appl. Environ. Microbiol., 66
(2000) 700.
[8] G.S. Grassia, K.M. McLean, P. Glenat, J. Bauld and A.J. Sheehy, FEMS Microbiol.
Ecol.,21(1996)47.
[9] M. Magot, Nature, 379 1996) 681.
Gevertz, Appl. Environ. Microbiol., 62 (1996) 1623.
[II] D. Gevertz, A.J. Telang, G. Voordouw and G.E. Jenneman, Appl. Environ. Microbiol.,
66(2000)2491.
[12] R. Cord-Ruwisch, W. Kleinitz and F. Widdel, Erdol-Erdgas-Kohle, 102 (1986) 281.
[13] T. Barth, Appl. Geochem., 6 (1991) 1.
[14] T. Barth and M. Riis, Org. Geochem., 19 (1992) 455.
[15] C. Tardy-Jacquenod, P. Caumette, R. Matheron, C. Lanau, O. Arnauld and M. Magot,
Can. J. Microbiol., 42 (1996) 259.
[16] T.N. Nazina, E.P. Rozanova and S.I. Kuznetsov, Geomicrobiol. J., 4 (1985) 103.
[17] E.P. Rozanova and T.N. Nazina, Microbiology (Engl. Tr.) 48 (1979) 907.
[18] R.K. Nilsen, J. Beeder, T. Thostenson and T. Torsvik, Appl. Environ. Microbiol., 62
(1996) 1793.
[19] P.N. Dang, T.C.H. Dang, T.H. Lai and H. Stan-Lotter, Anaerobe, 2 (1996) 385.
[20] C. Tardy-Jacquenod, M. Magot, B.K.C. Patel, R. Matheron and P. Caumette, hit. J. Syst.
Bacteriol. 48 (1998) 333.
[21] G. Voordouw, J.K. Voordouw, J.K. Jack, J. Foght, P.M. Fedorak andD.W. Westlake,
[22] T.N. Nazina, A.E. Ivanova, L.P. Kanchaveli and E.P. Rozanova, Microbiology [Engl.
Tr.], 57(1988)823.
[23] M. Magot, B. Ollivier and B.K. Patel, Antonie Van Leeuwenhoek, 77 (2000) 103.
[24] H.G. Triiper, Int. J. Syst. Evol. Microbiol., 53 (2003), 927.
[25] J.G. Zeikus, M.A. Dawson, T.E. Thompson, K. Ingvorsen and E.C. Hatchikian, J. Gen.
Microbiol., 129 (1983) 1159.
[26] B. Christensen, T. Torsvik and T. Lien, Appl. Environ. Microbiol. 58 (1992) 1244.
401
[27] G.N. Rees, G.S. Grassia, AJ. Sheehy, P.P. Dwivedi and B.K.C. Patel, Int. J. Syst.
Bacteriol., 45 (1995) 85.
[28] J. Beeder, T. Torsvik and T. Lien, Arch. Microbiol, 164 (1995) 331.
[29] T. Itoh, K.-i. Suzuki, P.C. Sanchez and T. Nakase, Int. J. Syst. Bacteriol., 49 (1999)
1157.
[30] K.O. Stetter, G. Laurer, M. Thomm and A. Neuner, Science, 236 (1987) 822.
[31] K.O. Stetter, Syst. Appl. Microbiol., 10 (1988) 172.
[32] A. Labes and P. Schonheit, Arch. Microbiol., 176 [2001) 329.
[33] S. Burggraf, H.W. Jannasch, B. Nicolaus and K.O. Stetter, Syst. Appl. Microbiol., 13
(1990)24.
[34] LA. Borzenkov, S.S. Belyaev, Y.M. Miller, LA. Davidova and M.V. Ivanov,
Microbiology (Engl. Tr.), 66 (1997) 104.
[35] S.S. Belyaev and M.V. Ivanov, Ecol. Bull, 35 (1983) 273.
[36] M.V. Ivanov, S.S. Belyaev, A.M. Zyakun, V. Bondars and K. Laurinivicius,
Geokhimiya 11 (1983) 1647.
[37] E.A. Bonch-Osmolovskaya, M.L. Miroshnichenko, A.V. Lebedinsky, N. A. Chernyh,
T.N. Nazina, V.S. Ivoilov, S.S. Belyaev, E.S. Boulygina, Y.P. Lysov, A.N. Perov, A.D.
Mirzabekov , H. Hippe, E. Stackebrandt, S. L'Haridon and C. Jeanthon, Appl. Environ.
Microbiol., 69 (2003) 6143.
[38] LA. Davydova-Charakhch'yan, V.G. Kuznetsova, L.L. Mityushina and S.S. Belyaev,
Microbiology [Engl. Tr.], 61 (1992) 299.
[39] R.K. Nilsen and T. Torsvik, Appl. Environ. Microbiol., 62 (1996) 728.
[40] E. Miranda-Tello, M.L. Fardeau, P. Thomas, F. Ramirez, L. Casalot, J.L. Cayol, J.L.
Garcia and B. Ollivier, Int. J. Syst. Evol. Microbiol., 54 (2004) 169.
[41] S. L'Haridon, M.L. Miroshnichenko, H. Hippe, M.L. Fardeau, E.A. Bonch-
Osmolovskaya, E. Stackebrandt and C. Jeanthon, Int. J. Syst. Evol. Microbiol. 52 (2002)
1715.
[42] S. L'Haridon, M.L. Miroshnichenko, H. Hippe, M.L. Fardeau, E. Bonch-Osmolovskaya,
E. Stackebrandt and C. Jeanthon, Int. J. Syst. Evol. Microbiol., 51 (2001) 1327.
[43] Y. Takahata, M. Nishijima, T. Hoaki and T. Maruyama, Int. J. Syst. Evol. Microbiol., 51
(2001)1901.
[44] M.E. Davey, W.A. Wood, R. Key, K. Nakamura and D.A. Stahl, System. Appl.
Microbiol. 16(1993)191.
[45] R. Huber and K.O. Stetter, In: A. Balows, H.G. Triiper, M. Dworkin, W. Harder and
K.H. Schleifer (eds.), The prokaryotes, 2nd edn., Springer, Berlin Heidelberg New York,
(1992) 3809.
[46] Y. Takahata, M. Nishijima, T. Hoaki and T. Maruyama, Appl. Environ. Microbiol., 66
(2000) 73.
[47] M.L. Miroshnichenko, H Hippe, E. Stackebrandt, N.A. Kostrikina, N.A. Chernyh, C.
Jeanthon, T.N. Nazina, S.S. Belyaev, E.A. Bonch- Osmolovskaya, Extremophiles, 5
(2001) 85.
[48] S. Rengpipat, T.A. Langworthy and J.G. Zeikus, Syst. Appl. Microbiol., 11 (1988) 28.
[49] G. Ravot, M. Magot, B. Ollivier, B.K.C. Patel, E. Ageron, P.A.D. Grimont, P. Thomas
and J.L. Garcia, FEMS Microbiol. Lett., 147 (1997) 81.
[50] V.K. Bhupatiraju, A. Oren, P.K. Sharma, R.S. Tanner, C.R. Woese and MJ. Mclnerney,
Int. J. Syst. Bacteriol., 44 (1994) 565.
[51] V.K. Bhupatiraju, and M.J. Mclnerney, C.R. Woese and R.S. Tanner, Int. J. Syst.
Bacteriol., 49(1999)953.
[52] A. Oren, Can. J. Microbiol., 32 (1986) 4.
402
[53] M. Magot, M.L. Fardeau, O. Arnauld, C. Lanau, B. Ollivier, P. Thomas and B.K.C
Patel, FEMS Microbiol. Lett., 155 (1997) 185.
[54] J.L. Cayol, B. Ollivier, B.K.C. Patel, G. Ravot, M. Magot, E. Ageron, P.A.D. Grimont
and J.L. Garcia, Int. J. Syst. Bacteriol. 45 (1995) 783.
[55] M.L. Fardeau, M. Magot, B.K. Patel, P. Thomas, J.L. Garcia, B. Ollivier, Int. J. Syst.
Evol. Microbiol., 50 (2000) 2141.
[56] G.N. Rees, B.K.C. Patel, G.S. Grassia and A.J. Sheehy, Int. J. Syst. Bacteriol. 47 (1997)
150.
[57] R.J. Menes and L. Muxi, Int. J. Syst. Evol. Microbiol., 52 (2002) 157.
[58] M. Magot, G. Ravot, X. Campaignolle, B. Ollivier, B.K.C. Patel, M.L.Fardeau, P.
Thomas, J.L. Crolet and J.L. Garcia, Int. J. Syst. Bacteriol. 47 (1997) 818.
[59] LA. Davydova-Charakhch'yan, A.N. Mileeva, L.L. Mityushina and S.S. Belyaev,
Microbiology (Engl. Tr.), 61 (1992) 306.
[60] R. Sleat, R.A. Mah and R. Robinson, Int. J. Syst. Bacteriol., 35 (1985) 10.
[61] M. Vargas, K. Kashefi, E.L. Blunt-Harris and D.R. Lovley, Nature, 395 (1998), 65.
[62] D.R. Boone, Y. Liu, Z.J. Zhao, D.L. Balkwill, G.R. Drake, T.O. Stevens, H.C. Aldrich,
Int. J. Syst. Bacteriol., 45 (1995) 441.
[63] S.V. Liu, J. Zhou, C. Zhang, D.R. Cole, M. Gajdarziska-Josifovska and T.L. Phelps,
Science, 277(1997)1106.
[64] A.G. Cairns-Smith, A.J. Hall and M.J. Russell, Orig. Life Evol. Biosph., 22 (1992) 161.
[65] K.M. Semple and D.W.S Westlake, Can. J. Microbiol., 33 (1987) 366.
[66] T.N. Nazina, A.E. Ivanova, O.V. Golubeva, R.R. Ibatulin, S.S. Belyaev and M.V.
Ivanov, Microbiology (Engl. Tr.), 64 (1995) 203.
[67] A.C. Greene, B.K.C. Patel and A.J. Sheehy, Int. J. Syst. Bacteriol., 47 (1997) 505.
[68] A.I. Slobodkin, C. Jeanthon, S. L'Haridon, T. Nazina, M. Miroshnichenko, E. Bonch-
Osmolovskaya, Curr. Microbiol., 39 (1999) 99.
Gevertz, Appl. Environ. Microbiol., 62 (1996) 1623.
[70] J.-Y. Leu, C.P. McGovern-Traa, A.J.R. Porter and W.A. Hamilton, Anaerobe, 4 (1998)
165.
[71] G. Voordouw, J.K. Voordouw, R.R. Karkhoffschweizer, P.M. Fedorak and D.W.S.
Westlake, Appl. Environ. Microbiol., 57 (1991) 3070.
[72] G. Voordouw, Y. Shen, C.S. Harrington, A.J. Telang, T.R. Jack, D.W.S. Westlake,
[73] T. Lien and J. Beeder, Int. J. Syst. Bacteriol., 47 (1997) 1124.
[74] A.S Galushko and E.P. Rozanova, Microbiology (Engl. Tr.), 60 (1991) 102.
[75] T. Lien, M. Madsen, I.H. Steen and K. Gjerdevik, Int. J. Syst. Bacteriol., 48 (1998) 469.
[76] J.-Y. Leu, C.P. McGovern-Traa, A.J. Porter, W.A. Hamilton, Lett. Appl. Microbiol., 29
(1999)246.
[77] E.P. Rozanova, T.N. Nazina and A.S. Galushko, Microbiology (Engl. Tr.) 57 (1988)
634.
[78] T.N. Nazina and E.P. Rozanova, Microbiology [Engl. Tr.], 47 (1978) 142.
[79] R.K. Nilsen, T. Torsvik and T. Lien, Int. J. Syst. Bacteriol., 46 (1996) 397.
[80] M. Magot, P. Caumette, J.M. Desperrier, R. Matheron, C. Dauga, F. Grimont and L.
Carreau, Int. J. Syst. Microbiol., 42 (1992) 398.
[81] E.P. Rozanova and A.I. Khudyakova, Microbiology [Engl. Tr.] 43 (1974) 1069.
[82] S.S. Belyaev, R. Wolkin, W.R Kenealy, M.J. De Niro, S. Epstein and J.G. Zeikus, Appl.
[83] T.K. Ng, P.J. Weimer and L.J. Gawel, Geomicrobiol. J., 7 (1989) 185.
403
[84] B. Ollivier, M.L. Fardeau, J.L. Cayol, M. Magot, B.K.C. Patel, G. Prensier and J.L.
Garcia, Int. J. Syst. Bacteriol., 48 (1998) 821.
[85] A.Y. Obraztsova, O.V. Shipin, L.V. Bezrukova and S.A. Belyaev, Microbiology [Engl.
Tr.] 56 (1988) 523.
[86] B. Ollivier, J.L. Cayol, B.K.C. Patel, M. Magot, M.L. Fardeau and J.L. Garcia, FEMS
Microbiol. Lett., 147 (1997) 51.
[87] A.Y. Obraztsova, V.E. Tsyban, K.S. Laurina Vichus, L.V. Bezrukova andS.S. Belyaev,
Microbiology [Engl. Tr.], 56 (1987) 807.
[88] S. Ni and D.R. Boone, Int. J. Syst. Bacteriol. 41 (1991) 410.
[89] G. Ravot, M. Magot, M.-L. Fardeau, B.K.C. Patel, P. Thomas, J.-L. Garcia and B.
Ollivier, Int. J. Syst. Bacteriol. 49 (1999) 1141.
[90] T. Lien, M. Madsen, F.A. Rainey and N.-K. Birkeland, Int. J. Syst. Bacteriol., 48 (1998)
1007.
[95] G. Ravot, M. Magot, M.L. Fardeau, B.K.C. Patel, G. Prensier, A. Egan, J.L. Garcia and
B. Ollivier, Int. J. Syst. Bacteriol., 45 (1995) 308.
[97] M.L. Fardeau, B. Ollivier, B.K.C. Patel, M. Magot, P. Thomas, A. Rimbault, F.
Rocchiccioli and J.L. Garcia, Int. J. Syst. Bacteriol. 47 (1997) 1013.
[98] C. Jeanthon C, A.L. Reysenbach, S. L'Haridon, A. Gambacorta, N.R. Pace, P. Glenat
and D. Prieur, Arch. Microbiol., 164 (1995) 91.
Chapter 15
Biotechnological approach for development of microbial

enhanced oil recovery technique
K. Fujiwara3, Y. Sugaib, N. Yazawac, K. Ohnoc, C.X. Hong" and H.
Enomotoe
a
Chugai Technos Co. Ltd., 9-20 Yokogawa-Shinmachi Nisi-ku Hiroshima City
733-0013,Japan
b
Akita University Venture Business Laboratory, 1-1 Tegatagakuen-cho Akita
City ,010-8502, Japan
technology Research Center, Japan National Oil Corporation, 1-2-2 Hamada,
Mihama-ku, Chiba 261-0025, Japan
d
PetroChina Company Limited, Jilin Oilfield Company, Jilin province, China
e
Department of Geoscience and Technology, Graduate School of Environmental
Studies, Tohoku University, Aramaki, Aoba-ku, Sendai 980-0845, Japan
1. INTRODUCTION
1.1. Microbial processes for oil recovery

Limited opportunities for discovering major new oil accumulations have
focused attention on processes which improve petroleum recovery and prolong
the life of existing wells. The microbial processes for oil recovery are classified
into three basic applications: well bore clean up, well stimulation and enhanced
waterfloods. Well bore clean up is normally carried out when paraffin and scale
are deposited on the well bore. Well stimulation and waterfloods enhancement,
namely, microbial enhanced oil recovery (MEOR), are conducted when the
target reservoir and oil production experience the following conditions:
formation damage, pore damage, high water production, poor displacement
efficiency, and/or poor sweep efficiency.
The authors have focused on MEOR because MEOR is one of the
techniques expected to be both economically feasible and environmentally
friendly, while also considerably increasing oil production.
406
1.2. Development of MEOR technique

Since Beckman proposed in 1926 [1] that bacterial metabolites assist in
the release and transport of oil in geological structures, MEOR processes
including well stimulation and waterflood enhancement have been attempted in
numerous oil fields throughout the world [2]. Fig. 1 shows activity of MEOR
that have the potential to enhance oil recovery.
Fig. 1. Mechanisms and effects expected for MEOR

407
According to previous research [3-4], reservoir conditions necessary for

MEOR processes are as follows: lower reservoir temperatures (below 70 °C),
higher permeability (above 50-75 md), higher porosity (above 20%), lower total
salinity (below 15%), appropriate pH ranges (4 to 9) and lower oil viscosity
(above 5-50mPa.s).
1.3. Types of MEOR processes

There are three variations on the MEOR process. The first involves the
injection of both microbes and nutrients. The microbes used here are selected for
their ability to make products, such as gases, surfactants, polymers and biomass,
that have the potential to increase oil recovery. This process often uses molasses,
an inexpensive by-product produced in the sugar refining process, as the
nutrients supply. The second variation involves simply the injection of nutrients
in order to utilize indigenous microbes within the reservoir. These nutrients
generally consist of molasses and/or plant fertilizers. The third process uses
microbes which can utilize hydrocarbon.
When applying these MEOR processes to a given reservoir, it is critical
that the optimum process is chosen based on observed reservoir characteristics,
such as geology and indigenous microbes. For example, the existence of useful
indigenous microbes in the target reservoir is necessary for successful
application of the first and second process variations described above.
1.4. Current stage of MEOR

MEOR has not been commonly accepted by the petroleum industry and,
for many, the question still remains as to whether MEOR can really be used in
increasing oil recovery. One industry concern is that many of the processes
simply "do not work." That is, MEOR is thought to be either based on
unsupported theories or on isolated laboratory work, so that when MEOR is
subjected to actual field tests it failed to generate substantial volumes of
incremental oil production. Another reason for the dismissal of MEOR is that
most people in the petroleum industry do not fully understand it. Part of this lack
of understanding stems from the fact that many people do not realize that
MEOR is a multiplicity of technologies, not a single process.
1.5. Current obstacles and breakthrough points

The study of MEOR is entering a turning point and there are technical
breakthrough points for overcoming the main obstacles to MEOR use. Current
obstacles include the following: first, the need for a collection of solid data
demonstrating the success of MEOR. Second, fundamental technologies in
MEOR need to be more fully developed and tested. Breakthrough points for
MEOR are listed below.
408
(1) Breakthrough Point 1:

Most reports on field trials have been poorly documented for scientific
acceptance. For example, many studies have demonstrated a lack of
scientific knowledge about the fundamental microbiology related to
MEOR and a lack of adequate control experiments. This is likely due to
the fact that research into the fundamental microbiology related to
MEOR and control experiments are added expenses and will not
contribute directly to an increase in oil recovery. However, getting the
data to scientifically document the success of MEOR technology is
needed if MEOR is to become a respected tool of the oil industry.

A technique for transplanting microbes into a formation has not yet been
well established.

So far, the metabolic function of target microbes in the reservoir has not
been scientifically demonstrated.
1.6. Objectives of the present study

The objectives of this research were twofold: the collection of valuable
data proving MEOR's effectiveness, and the development of fundamental
technologies in MEOR. These objectives were established for a collaborative
research project between Technology Research Center (TRC) of Japan National
Oil Corporation and PetroChina Jilin Oilfield Company, run from 1996 to 2002
[5-21]. The strategies of this research are as follows:
(1) First step: development of biotechnological tools for estimating the

behavior of microbes in the reservoir.
(2) Second step: understanding of the MEOR test field. That is, the
collection of scientific data about the fundamental microbiology and
reservoir related to the MEOR field trial.
(3) Third step: development of the following techniques for use in

assessing the effectiveness of MEOR:
a) Technique for transplanting microbes into the formation.
b) Technique for demonstrating microbial metabolic function in the
reservoir when needed.
409
2. TEST FIELD
The test field, Fuyu oilfield, is located in the northeastern area of China (Fig. 2
A and B). Oil production in this area began in 1973, and waterflooding was
instituted in 1983. Current production is conducted using sucker rod pumping.
Fig.2. Location of the test field

(A) Location of Fuyu oilfield (Jilin province in China)
(B) Wells map at the test field in Fuyu oilfield
410
Table 1
Reservoir data for test area
Block East 24-23 East 24-26
Reservoir Areas [km2] 0.562 0.219
Res. Depth [m] 300-450 320-450
Res. Temperature [°C] 28.0 28.0
Res. Thickness [m] 15.2 (net) 14.8 (net)

Permeability [md] 240 240
Porosity [%] 27 27
Current Pressure [kg cm"2] 28.8 18.6
Water Cut
(1995end) [%] 73.8 69.7
* Present water cut of most wells is more than 90%.

* There are heterogeneous fracture zones between each
injection well and production well.
The reservoir data for test area is shown in Table 1. The target reservoir is
sandstone and its depth is from 300 to 450 m. The temperature of the reservoir is
approximately 30 °C. The average permeability is approximately 240 md and
porosity is 27%. The present water cut of most wells averages more than 90%
and there are heterogeneous fracture zones between each injection well and
production well.
3. THE COLLECTION OF SCIENTIFIC KNOWLEDGE OF THE

FUNDAMENTAL MICROBIOLOGY RELATED TO MEOR
3.1. Development of investigation technique of microbes related to MEOR.

Fig. 3 shows the analytical protocol of microbes related to MEOR. The
authors developed a biotechnological tool, a combination of plating and
PCR-RFLP analysis [11] to estimate the behavior of microbes which are able to
propagate in the reservoir using molasses. Those microbes unable to propagate
using molasses were not relevant to our study and therefore did not need to be
analyzed. The RFLP profile described in the present study refers to the profile of
411
Restriction Fragments Length Polymorphism based on the 16S rDNA sequence

of bacteria. This PCR-RFLP analysis is able to discriminate microbes by
comparing their RFLP profiles. Matching of the RFLP profiles of three
restriction endonucleases also improved microbe identification. This method
consists of the following stages:
1) Microbe extractions are incubated on the molasses agar plate.
2) 16S rDNA of each type of colony is individually amplified by PCR.
3) Amplified 16S rDNA are treated by restriction endonucleases (Hhal, Mspl,
AM)
4) RFLP profiles obtained by electrophoresis of the digested 16S rDNA are
compared with each other.
5) The microbes are classified based on their RFLP profiles.
The PCR-RFLP analysis includes two major contrived conditions. One is the
primer setting and PCR condition. Highly conserved regions of 16S rDNA are
used as universal primer; amplifications of 16S rDNA of all microbes are
accomplished using only one condition. Another is the selection of the three
restriction endonucleases Mspl, Hhal and Alul. Through computer simulation,
these endonucleases are shown to be particularly effective in generating many
restriction fragments in various microbes, mainly resulting in moderate-size
fragments of 100 to 1000 bp that are easily distinguished.
3.2. Investigation of microbes inhabiting the reservoir rock which have the
ability to propagate using molasses.
To obtain closed reservoir core samples, a #J15 well and #J16 well were
drilled in the test area [11] (see Fig. 2(b)). A total of 8 and 4 samples of the
reservoir rock were collected from, respectively, the #J15 and #J16 wells (see
Fig. 4). Whole cores obtained from each layer were cut off using a hatchet;
samples of reservoir rock were then collected by scratching the center of the
whole core's cross section with a small sterilized pickax. Within two days,
microbes inhabiting these samples were analyzed using the combination of
plating and PCR-RFLP analysis described above.
Within the #J15 reservoir rock, we observed and isolated a total of 177
species having the ability to propagate on molasses. Of these, 59 were
determined to have unique RFLP profiles based on RFLP analysis. A total of 87
species having the ability to propagate on molasses were isolated from the
r e s e r v o i r r o c k of # J 1 6 ; of t h e s e , 47 w e r e d e t e r m i n e d to
have unique RFLP profiles based on RFLP analysis. Moreover, homology
analysis (phylogenetic relationships) based on the RFLP profiles demonstrated
that almost all of the microbes were different from the general soil bacteria, and
some types of yeasts were detected in the high permeability zones and their
surroundings.
412
Fig. 3. Analytical protocol of microbes related to the MEOR
In the comparison of RFLP profile of microbes from #J15 with those from
#J16, 15 to 20% of RFLP profiles from #J16 matched those of #J15, and 20 to
25% of RFLP profiles from #J16 are closely related species (that is, two RFLP
profiles were exact matches) to microbes from #J15. Almost all microbes of
these microbes were detected at 102to 105 cfu g"1, and some grew to more than
108 cfu ml"1 using molasses.
These results indicate that aerobes and facultative anaerobes isolated from
the reservoir rock are trapped in high permeability zones and highly saturated
water zones, such as fractures created by hydraulic fracturing operations. The
aerobes and facultative anaerobes have also been carried into the reservoir from
the surface by water flooding operations, and have accumulated over a long
period of time. During these operations, the injection water, including these
microbes, is apt to enter into the high permeability zones. Therefore, in the
development of MEOR techniques, we must consider that microbes injected into
the reservoir will need to co-exist with these indigenous microbes. It is
necessary to monitor the indigenous microbes that make use of molasses,
particularly those microbes' potential to suppress the growth of in situ microbes
injected into the reservoir.
413
Fig.4. Sampling location of reservoir rock
4. ESTIMATION OF BEHAVIOR OF IN SITU MICROFLORA AT THE

MOLASSES INJECTION TEST USING HUFF & PUFF PROCESS
Four producing wells were selected for this experiment and the Huff and Puff
molasses injection process shown in Fig. 5 [12]. The Huff and Puff process
encompassed several steps. From the surface facilities, a 10% molasses solution
was shipped in sterilized tank trucks and transported to the well site. At the well
site, 200 kl per well of molasses solution was injected into the reservoir at an
injection rate of 30 kl h"1. After injection, these wells were shut in for 20 days
before production resumed.
414
Fig. 5. Injection process of molasses solution by Huff & Puff injection method
Table 2 shows the microbes detected inhabiting the ground water,

molasses, and reservoir brine before the injection test. In the ground water,
seven species were distinguished by their RFLP profile, and theconcentration of
viable cells of each species was 10 to 103 cfu ml"1. Of these, three species grew
to more than 107 cfu ml"1 using 4% molasses.
On the other hand, in the molasses and reservoir brine, the number of
species distinguished by their RFLP profile was two to eight, and the
concentration of viable cells of each species was 10 to 106 cfu ml"1. Of these,
almost no species grew to more than 107 cfu ml"1 using molasses.
415
Table 2
The number of species distinguished by RFLP profile
(in ground water, molasses, reservoir brine)
The number of The number of species
species detected grown more than 107 cfu ml"1
in the samples using 4% molasses.
Ground water 7 (10-103) 3
Molasses 5 (102-103) 1
Reservoir brine from
#26-231 2 (10-104) 0
#24-24, 8 (102-103) 1
#20-28 4 (104-105) 0
#T-59 6 (102-103) 0
( ) ; Concentration of viable cells [cfu ml"1]
Table 3 shows the microbes detected in the injected fluid before injection.
These injected fluids were collected from tank trucks at each well site. Within
the four injection fluid samples, the number of species distinguished by their
RFLP profile was three to five, and the concentration of viable cells of each
species was 104 to 107 cfu ml"'. Of these, almost all species grew to more than
107 cfu ml"1 using molasses. Moreover, based on their RFLP profiles, almost all
microbes detected in the injected fluid matched microbes which were isolated
from the ground water.
Table 4 shows microbes detected in the production water after the
injection test. In samples obtained from the four wells, three predominant
species are distinguished by their RFLP profiles. These species, in viable cell
concentrations of 103 to 107 cfu ml"1 were also detected in the 5 to 13 samples of
production water collected daily throughout the 20-day test period. Moreover,
based on their RFLP profiles, these microbes matched microbes which were
isolated from the ground water.
Fig. 6 (A, B, C and D) shows the production history of the producing
wells in which the molasses solution was injected. It is apparent from these data
that molasses injection into the target reservoir did not result in increased oil
recovery.
416
The number of The number of species

species detected grown more than 107 cfu ml"1
in the samples using 4% molasses.
Injected fluid for
#26-23, 4 (104-107) 4
#24-24, 4 (104-106) 4
#20-28 3 (106-107) 3
#T-59 5 (106-107) 5
Analysis of the collected data indicates the following:

(l)Microbes inhabiting the ground water thrive in the presence of molasses
and may threaten the growth of other microbes injected into the reservoir.
(2) Based on their RFLP profiles, those microbes are "Enterobacteriaceae" or
that of a closely related species.
(3) An injection of molasses alone is not expected to increase oil recovery in
our target reservoir.
(4) Microbes related to the molasses injection (such as those inhabiting the
ground water, molasses, reservoir brine and reservoir rock) do not cause an
increase in oil recovery when provided with molasses.
Table 4
Predominant species in production water
distinguished by RFLP profile
(In production water)
DCTB n Frequency of
RFLP profile /* . , . .
detection (times)
A 5 (104-106)
B 8 (103-105)
C 13 (103-107)
417
Fig. 6. Result of oil production at the molasses injection tests

• : Oil production • : Total liquid o : Water cut
Considering these results, it is clear that an increase in oil recovery will

require the use of microbes selected for reservoir characteristics such as the
development history of the reservoir, indigenous microbes, and geological
features.
418
5. DEVELOPMENT OF TECHNIQUES VERIFYING MEOR

EFFECTIVENESS
In the initial stages of the experiment, oil field conditions were investigated in
detail in order to determine what in situ metabolic processes can be supported in
the reservoir. It was assumed that selective plugging of highly permeable zones
would be effective for this reservoir, because there are many horizontal fractures
near the production wells, caused by the hydraulic fracturing operations.
Successful selective plugging field operations have been reported previously
[22-24], and this methodology is regarded as an effective MEOR process.
Fig. 7 shows the plugging mechanism in highly permeable zones. In
conventional water flooding, water injected into the reservoir flows
predominantly into large channels for a long period of time. When microbes are
injected, they also enter primarily the large channels, growing and producing
insoluble polymer in these places. As a result, insoluble polymer, including
microbial cell mass, selectively plugs high permeability zones, and injection
water is diverted from the large channels into previously un-swept areas of the
reservoir.
Fig.7. Plugging mechanism of high permeable zones

419
5.1. Screening of microbes for injection in the reservoir

5.1.1. Screening of microbes [16]
The essential parameters of microbes used for "selective plugging" are
shown below. The microbes must:
(1) Produce an insoluble polymer using relatively inexpensive molasses.
(2) Propagate under both aerobic and anaerobic conditions.
(3) Form biofilm at the surface of reservoir rock.
(4)Propagate and produce insoluble polymer under the reservoir's unique
conditions.
Fig. 8 shows the screening protocol of microbes used for MEOR. This
protocol consists of two stages:
(1) Microbe extractions from reservoir samples (such as reservoir core and
brine) are incubated on the molasses agar plate.
(2) Each type of colony is incubated individually in the molasses liquid
medium, and candidate microbes are selected visually based on their
ability to produce the insoluble polymer.
A strain CJF-002, which demonstrated the potential described above, was
screened from reservoir rock in Fuyu oilfield. Based on 16S rDNA sequences, it
was identified as a strain belonging to the Enterobacter species.
Fig. 8. Screening of microbes for injection in the reservoir

420
Fig. 9. Growth and production of insoluble polymer of strain CJF-002 (After 1 day
incubation).
Fig. 9 shows the growth and production of insoluble polymer from the
strain CJF-002. It is apparent from these data that one of the notable features of
CJF-002 is its ability to grow and produce insoluble polymer when fed molasses.
Moreover, the cells of the strain CJF-002 are small enough to pass through the
pore throat of average sandstone.
Fig. 10 shows the visual image (A) and SEM images (B and C) of
insoluble polymer. Cellulase can degrade this insoluble polymer.
Results of sugar composition (Table 5) and methylation analysis (Table 6)
also indicated that the insoluble polymer produced by strain CJF-002 is a
cellulose derivative.
5.1.2. Development of the monitoring technique of viable strain CJF-002

propagating in the reservoir
Conventional culture-based bacteriological methods for detecting
microbes in environment samples depend on their recovery from each sample
and therefore on their culture conditions. However, these methods are not
suitable for MEOR because their lower selectivity can not distinguish target
microbes from the various microbes inhabiting the reservoir fluid. These
methods also require several days to produce results. The process of MEOR
421
with flooding is a massive undertaking, involving the injection of a culture broth

of microbes and other nutrients over a long period. Therefore, the technique
for monitoring the strain CJF-002 must be not only effectively discriminating,
but also rapid and simple.
Fig. 10. Visual image and SEM images of insoluble polymer by strain CJF-002
(A);Visual image, (B and C);SEM images (After 1 day incubation)
Table 5.
Sugar composition of insoluble polymer
Sugar
Detected substance composition
Glucose 97.31
Mannose 1.32
Arabinose 0.42
Galactose 0.95
Uronic acid 0.0
*Grown with 4% molasses

422
Table 6.
Result of methylation analysis of insoluble polymer
Peak Retention partially methylated Binding Peak

ratio
No. time (MS) sugar alcohol location area
1,5diO-acethyl-2,3,4,6-tetra-O- nonreduced
1 19'09" 2840 1.00
methylglucitol end Glc
1,4,5-tri-O-acethyl-2,3,6-tri-O-
2 20' 50" ^4Glc 88917 29.22
methylglucitol
l,5,6-tri-O-acethyl-2,3,4-tri-O-
3 20' 57" -+6Glc 310 0.10
methylglucitol
l,3,4,5-tetra-O-acethyl-2,6-di-
4 21' 27" ^3,4Glc 706 0.22
O-methylglucitol
l,2,4,5-tetra-O-acethyl-3,6-di-
5 21'38" —2,4Glc 1572 0.48
O-methylglucitol
1,4,5,6-tetra-O-acethyl-2,3-di-
6 22' 01" ^4,6Glc 1549 0.48
O-methylglucitol
The authors developed a combination of plating and Direct-PCR analysis

to estimate the behavior of strain CJF-002 in the reservoir. The Direct-PCR
methodologies, which permit the rapid direct detection of a specific DNA
sequence in a target microbe, have been studied for use in detecting a specific
microbe in environmental samples including soil, foods and water [25-28].
These methodologies have the potential to identify a given target microbe
because detection of DNA fragments indicates the presence of that microbe.
Hence, it may be possible to apply these methodologies effectively in
understanding the behavior of strain CJF-002 during the MEOR process.
The PCR primer was designed using intergenetic spacer regions located
between the 16S and 23S rDNA, which consist of highly species-specific
sequences. Previous research has demonstrated that these intergenetic sequences
are more available than 16S rDNA sequences when identifying bacteria by
direct PCR method [29-31]. The sequences of these primers are
5'-AGGCCTACCAAATTTCAGCT-3' (CJF-2F, forward), and
5'-GAGACTCGCAGAACAGTTCG-3' (CJF-2R, reverse).
Experimental specificity testing of the PCR primers was also performed
on pure cultures of bacterial strains. Bacterial genomic DNA was extracted from
each bacterium using an InstaGene matrix (Bio-Rad lab.) containing sterile
423
distilled water (SDW) and Chelex 100 resin, as per the manufacturer's
instructions. PCR amplification was performed under the following conditions.
PCR mixtures contained 2 ul of 10x PCR buffer (500 mM KC1, 100 mM
Tris-HCl [pH 8.3], 15 mM MgCl2), 2 ul of 25 mM MgCl2, 2 ul of a
deoxynucleotide triphosphate (dNTP) mixture (concentration of each dNTP, 2.5
mM), 10 pM of each primer, 5 ul of the extracted DNA sample and 0.4 U of Taq
DNA Polymerase (Takara Shuzo Co., Ltd., Kyoto, Japan), in a total volume of
20 ul.
After the solution was overlaid with 30 ul of mineral oil (Chill-out 14
Liquid Wax, MJ Research Inc., Watertown, MA), the PCR program was initiated
with a preincubation at 94°C for 30 s. The amplification profile is 94°C for 45s,
58°C for 50 s, and 72°C for 60 s. PCR products were electrophoresed in a 1.5 %
agarose gel and visualized by UV transillumination after being stained in
ethidium bromide solution (5 ug ml"1).
A primer annealing temperature close to the theoretical primer melting
point, that is, 58°C, allowed amplification of a single 280 bp product only in
strain CJF-002 (see Table 7), according to the direct PCR protocol. The band
size of the amplificates matched the expected size. These results demonstrated
that the 16S-23S spacer sequence of strain CJF-002 is sufficiently
species-specific for the derivation of PCR primers used to identify strain
CJF-002.
5.1.3. Biofilm formation test

A biofilm formation test was performed as follows. Sliced reservoir rock
was set vertically inside a bottle filled with the 4% molasses solution,
synthetic brine, and strain CJF-002 (see Fig. 11). The components of the
synthetic brine were established previously based on the components of
reservoir brine obtained from the test field (see Table 8). The molasses medium
with strain CJF-002 was then stirred 1.7cm per second and incubated at 30 °C.
The medium predominantly affected one face of the rock during the test period.
After one day, the surface of the sliced reservoir rock was covered with
biofilm consisting of the insoluble polymer produced by strain CJF-002 (see Fig.
12). These results suggest that the insoluble polymer produced by strain
CJF-002 will adhere to the surface of reservoir rock in the target reservoir.
424
Table 7.
Summary of PCR amplification for various species by spacer primers
Species Strain No. Size of amplificats (bp)
CJF-002 280
Acinetobacter calcoacelis IFO 12552
Aeromonas hydrophila IFO 13286
Alcaligenes faecalis IFO 14479
Azotobacter vinelandii IFO 12018
Bacillus subtilis IFO 3134
Rhodococcus erythropolis IFO 12320
Staphylococcus aureus IFO 12732
Pseudomonas fluorescens IFO 14160
Enterobacter cloacae IFO 13535
Citrobacter freundii IFO 13546
Escherichia coli IFO 13898
Erwinia carotovora IFO 14082
Klebsiella pneumoniae IFO 13541
Clostridium acetobutylicum IFO 13948
Clostridium butyricum IFO 13949
Fig. 11. Biofilm formation test by strain CJF-002

425
Teble 8.
The components of reservoir brine
Synthetic brine (1000ml)
NaCl : 1210 (mg)

KC13 23
NaHCO 2820
CaCl2 140
MgCl2 53
FeCl3 2
KH2PO4 10
NaHSO4 3
5.2. Investigation of availability of strain CJF-002 for the environments of

given reservoir.
Investigation into the availability of strain CJF-002 was carried out in
order to determine whether the strain CJF-002 will survive and perform
the desired metabolic functions in a given reservoir.
A controllable factor in the reservoir, the effect of molasses concentration
was studied first. Results showed that growth of the strain CJF-002 started in the
presence of more than 0.1% molasses, and that production of insoluble polymer
accelerated with 1% molasses.
The effects of factors difficult to control in the real reservoir were then
investigated. Growth of strain CJF-002 and production of insoluble polymer
were observed at more than 15°C and at a pH higher than 5.4. Production of
insoluble polymer also accelerated when NaHCO3 was present in the reservoir
brine. Moreover, strain CJF-002 grew and produced insoluble polymer in
co-existence with microbes inhabiting the reservoir brine, ground water,
molasses and injection fluid.
Table 9-(A) shows the results of competitive culture tests with indigenous
microbes in the reservoir brine, and Table 9-(B) is with microbes in the injection
water. Notably, in co-existence with the indigenous microbes in the reservoir
brine, CJF-002's potential for propagation and survival are exceedingly high
when compared with strains A and B. The strain CJF-002 was detected at
approximately 108 cfu ml"1 until day 10 and detected at more than 104 cfu ml"1
by the 20th day. Thus it appeared that strain CJF-002 would have the ability to
grow and survive in the environment of our target reservoir.
426
Table 9.
Result of competitive vulture test
(A) with indigenous microbes in the reservoir brine
Screened Incubation period
place initial lday 3 days 5 days lOdays 20days
Strain A Japan x 1O s xlO6 xlO6 xlO 5 • OlO3 • OlO3
Strain B Japan xlO6 xlO7 xlO4 DOlO 3 •OlO 3 NT
CJF-002 China xio 6 xlO8 xlO 8 xlO8 xlO 8 xlO4
(B) with microbes in the injection water

Screened Incubation period
place initial lday 3 days 5 days lOdays 20days
Strain A Japan xlO6 xlO8 xlO 8 xlO8 xlO8 • OlO3

Strain B Japan xlO5 xlO7 • OlO3 •OlO 3 •OlO 3 • OlO3
CJF-002 China xlO6 xlO9 xlO8 xlO8 xlO8 • OlO3
The survivability of strain CJF-002 also depends on its initial

concentration. More than 0 . 1 % of molasses and 105 cfu ml"1 of strain CJF-002
are vital for growth, insoluble polymer production and overcoming other
microbes.
Finally, the effect of uncontrollable factors in the reservoir was
investigated. Growth of strain CJF-002 and the production of insoluble polymer
were found to be unrelated to the presence of oxygen. In the presence of oil,
however, the rate of insoluble polymer production decreased approximately 20%.
That rate also decreased to approximately 1/3 of the original production under
the reservoir pressure (30 atm). Growth of strain CJF-002 and the production
of insoluble polymer were not affected by the micro culture environment.
427
Fig. 12. Result of biofilm formation test by strain CJF-002
5.3. Evaluation of microbial profile modification effect

An experiment was performed to evaluate the effect of microbial profile
modification at the laboratory [19]. Two sets of porous media were prepared (see
Fig. 13): one high-perm sand pack, and one low-perm berea sandstone core. The
two cores were first saturated with the synthetic brine described above. The
absolute permeability of both cores was then measured by injecting brine
through the pump. Permeability of the sand pack was determined to be 11,000
md, and the sandstone core permeability was found to be 900 md. Strain
CJF-002 and 5% molasses were then injected.
When the waterflooding was started after a 5-day shut in, core
permeability was almost unchanged, while the permeability of the sand pack fell
drastically (see Fig. 14). Also, prior to the injection of the strain CJF-002 and
molasses, the fluid ratio, (the production rate of high permeability sand pack to
the low permeability core), was 16, but it decreased to only 1.5 after the shut-in.
This decrease indicates that the strain CJF-002 selectively plugs high
permeability zones. In other words, strain CJF-002 has the ability to modify
profiles.
428
Fig. 13. Laboratory experiment to confirm an ability of profile modification
5.4. Huff and Puff tests using production wells

Huff and Puff tests using strain CJF-002 were performed at six producing
wells in order to confirm the survivability of strain CJF-002 and the production
of insoluble polymer [16-17]. Wells selected for the Huff and Puff tests featured
a relatively high water cut (more than 70%) and high water production.
Culture conditions of strain CJF-002 and the process of injection into the
reservoir were as follows. In the laboratory, strain CJF-002 was incubated using
a 1% molasses medium containing 0.01% of cellulase at 100 ml and 10L, in due
order. The cellulase was added to the medium to prevent the production of
insoluble polymer during the propagation of the strain CJF-002. At the surface
facility, the culture scale increased to 1000 1. The 400 1 of final culture broth was
mixed with water or molasses solution (final cone. 0.6 - 8%), then injected into
the oil reservoir through six producing wells as described below:
(1) 10 kl of CJF-002 culture broth & 80 kl of molasses were injected
separately, followed by an injection of 20 kl of driving water.
(2) 10 kl of CJF-002 culture broth & 80 kl of molasses were injected
simultaneously, followed by an injection of 20 kl of driving water.
After a 10-day shut-in period, concentrations of strain CJF-002 in production
water drawn from producing wells were measured using the following method.
100 ul of each sample was placed on nutrient broth agar plates (Difco Co., Ltd.,
Detroit, MI); these plates were then incubated at 30°C under aerophilic
conditions for two days. Concentration of the CJF-002-like colonies was
measured by the colony forming units (CFU) method. Some of those colonies
were selected from the agar plates to undergo the direct PCR analyses described
429
above, in order to demonstrate that the colonies distinguished by their

characteristics were, in fact, strain CJF-002. The total number of non- CJF-002
microbes was also evaluated, using the plating method described above. All data
from these experiments are presented in Table 10.
The strain CJF-002 was detected at all six of the wells and a definite
increase in oil production was observed at four wells. The results of oil
production and monitoring of microbes at the 22-264 well are illustrated in Fig.
15 and 16, respectively. After shut-in, oil production increased remarkably and
strain CJF-002 was detected in the production water at a relatively high
concentration. It is apparent from these data that the strain CJF-002 has the
ability to survive in the given reservoir and injection of strain CJF-002 into the
target reservoir contributes to increased oil recovery.
In the Huff and Puff tests, the concept of highly-permeable-zone plugging
by insoluble polymer is as follows. If there are highly-permeable zones in the
reservoir, injected fluid, including the strain CJF-002 and molasses, enters
primarily into these zones. Insoluble polymer is produced selectively in the areas
which contain the mixture of strain CJF-002 and molasses. A stream line of
injected water is then modified, and residual crude oil in low-permeable zones
(unswept oil) is newly recovered.
Fig. 14. Result of laboratory waterflooding experiment.

• Permeability (Sand Pack) o Permeability (Core)
D
x Producing fluid ratio Injection pressure
430
Teble 10.
Result of Huff & Puff Test by strain CJF-002
Before injection Condition of injection After injection

Well*1)'2) Water R t Molasses Injection system Microbe Oil WC*5)
3) 4)
cut cone. (CJF-002and in Prod.* Prod.* r 0/1
[%] [t day"1] [%] molasses) [%] L J
°
1. #22-27 99 0.2 1.9-3.0 Separately 83.3
2. #24-23 4 97-98 0.1-0.2 0.6-1.5 Separately 85.7
3.#22-26 4 98 0.2-0.3 3.3-5.7 Separately 88.9 ?9 ?9
4. #26-231 98 0.2 4.2 Separately 57.1
5. #26-254 97 0.2-0.3 4.6-7.4 Simultaneously 62.5
6. #24-24 j 99 o.l 4.4-8.0 Simultaneously 85.7
*1) 1.-4. 10m3 of CJF-002 culture broth & 80m3 of molasses were injected separately.
*2) 5. and 6. 10m3 of CJF-002 culture broth & 80m3 of molasses were injected
simultaneously.
*3) The ratio of samples with CJF-002 to samples of produced water analyzed.
*4) Oil production; Increase , Same level
*5) Water cut; Decrease , Same level
Fig. 15. Result of oil production at well 22-264

• Oil production A Water cut
431
Fig. 16. Result of microbial monitoring at well 22-264
• : Viable cell number before Injection

O: Total viable cell number after resuming production
• : CJF-002 viable cell number after resuming production
5.5. Continuous injection tests using two injection wells

These tests were conducted to demonstrate the fundamental breakthroughs
in MEOR technology [19-21]. Fig. 17 shows the location of the test area; there
are 10 production wells and two injection wells. Production history of the test
area is shown in Fig. 18. An increase in the monthly oil rate had not been
observed for approximately 20 years.
Fig. 19 shows a schematic of the surface facility where microbial
treatment took place in the first trial. The microbial incubation facility is located
3 km away from the well site. The schedule of microbial treatments in the first
trial is shown in Fig. 20. In these treatments, strain CJF-002 solution and
molasses were injected separately to prevent the produced insoluble polymer
from plugging the perforation holes in injection wells.
The strain CJF-002 was incubated as it was for the Huff and Puff tests
described above. The 400 1 of final culture broth was mixed continuously with
50 kl of water for one day, then injected into the reservoir through two injecting
wells at a rate of 25 kl per day per well. The concentration of strain CJF-002 at
the injection pump was approximately 106 cfu ml"1. The strain CJF-002 was
injected for two weeks, followed by a one-week injection of 0.1% molasses
solution. In addition, 1, 5 and 20% molasses solutions were injected subsequent
to the strain CJF-002 injections. After the strain CJF-002 and molasses
injections, water injection was continued as usual.
432
Fig. 17. Wells map at continuous injection test aria
Fig. 18. Production history of East 24-26 Block

Monthly oil rate, Monthly water rate
433
Fig. 20. Schedule of microbial treatments (1st trial).

434
5.5.1. First Trial (2000)

Throughout this experiment, the concentration of strain CJF-002 in both
the injection fluid obtained from the manifold and the production water drawn
from producing wells were measured by the plating and Direct-PCR analysis
described above. The total number of non-CJF-002 microbes were also
evaluated using this plating method.
The microbial concentrations in injected water taken from the manifold
are shown in Fig. 21. At the first injection of strain CJF-002 and molasses, strain
CJF-002 and non-CJF-002 microbes are at similar concentration levels of 105
and 106 cfu ml"'. The competitiveness of strain CJF-002 under similar conditions
had been confirmed in our previous laboratory study, however, at the second
injection, the concentration of other microbes had become over 100 times higher
than that of the strain CJF-002. By the third and final injection, non-CJF-002
microbes were completely dominant. Consequently, the strain CJF-002 was
detected in only 60% of producing wells in this test area. In such conditions, the
competitiveness of the strain CJF-002 is considered to be inferior to other
microbes
In addition, an increase in oil recovery, which should have been observed,
was not confirmed in the first trial. These results indicate that it is crucial to
inject the strain CJF-002 at high concentrations and keep its predominance in the
injected fluid.
The microbial monitoring of production water from well #J18 showed the
first detection of strain CJF-002 was observed within four days, though this
production well is located 60 m away from the injecting well. The strain
CJF-002 detected in the production water was known to have come from the
strain injected at the injection wells, because the strain CJF-002 was not
detected in the pure molasses injection test described above. From this data, it
appears that the strain CJF-002 injected into the reservoir flows at a high speed
in the highly permeable regions.
The detection period of the strain CJF-002 after the molasses injection is also a
remarkable point. After the 0.1% molasses injection, the strain CJF-002 became
undetectable in a week, while after a high molasses injection, a relatively high
concentration of strain CJF-002 was detectable for a month (see Fig. 22). These
results indicate that the strain CJF-002 has survivability in the reservoir
environments when a desirable concentration of molasses, namely, more than
5%, is present.
435
5.5.2. Second Trial (2001)

Based on the results of first trial, the following modifications to the
injection method were determined:
(1) Inject strain CJF-002 and molasses simultaneously to increase the chance
of interaction between the CJF-002 and the molasses.
(2) Inject 10% molasses to maximize the amount of insoluble polymer.
(3)For one week preceding the molasses injection, inject only the strain
CJF-002, so as to dominate the reservoir with strain CJF-002.
(4)Reduce the other microbes in the pipeline by injecting the molasses closer
to the well site.
Fig. 23 shows a schematic of the surface facility used for microbial
treatment in the second trial. The strain CJF-002 was injected through an
injection pump at the microbial incubation facility, whereas the 10% molasses
solution was injected from a pump near the injection wells. The schedule of
microbial treatments in the second trial is shown in Fig. 24. The strain CJF-002
was injected along with injection water for one week.
Fig. 21. Results of microbial monitoring at manifold (1st trial)

• CJF-002 o Other microbes
436
Fig. 22. Result of microbial monitoring of production water taken

from well J-18 (2nd trial)
• CJF-002 o other microbes
Fig. 23. Microbial treatment facility of 2nd trial.

437
Fig. 24. Schedule of microbial treatments (2nd trial).
A 10% molasses solution was then injected along with strain CJF-002.
After two months of CJF-002 and molasses injections, water injection was
continued as usual. The strain CJF-002 and other microbes were measured as in
the first trial, described above.
The microbial concentrations in injection water taken from the manifold
are shown in Fig. 25. The strain CJF-002 and non-CJF-002 microbial
concentrations were at almost same levels during injection of the strain CJF-002
and molasses. Theoretical concentration, that is, approximately 105 cfu ml" of
strain CJF-002 was routinely detected in the injection water. In other words,
results of this microbial monitoring indicate that the strain CJF-002 can increase
in cell number and produce the insoluble polymer in the reservoir.
Figure 26 shows the results of the microbial monitoring of production
water from well #J18. Notably, the concentration of strain CJF-002 detected in
the production water was relatively high, approximately 103 to 106cfu ml"1, for
20 days following the initial injection. Consequently, the strain CJF-002 was
detected in all producing wells in this test area. Parts of insoluble polymer which
may have come off the reservoir rock were also detected in all producing wells
by HPLC analysis combined with cellulase degradation. This result proves that
the strain CJF-002 has the ability to produce insoluble polymer in porous media
in the reservoir. Results also show that the concentrations of the injected
CJF-002 and molasses during the second trial are enough to sustain the strain
CJF-002's competitiveness and survivability.
The oil production of all wells in the test area is shown in Fig. 27.
Eventually, the oil production increased by more than two times for at least one
year, and incremental oil production reached 3,392 tons [approximately 24,521
bbls] after microbial injection. Total water cut also fell from 88% to 65%.
Notably, a dramatic increase in oil production was observed approximately 20
days after beginning injections. Even after the final CJF-002 and molasses
injection, the great improvements in oil production and water cut continued, and
the oil recovery rate after one year was still doubled. These results indicate that
438
the increased oil production is related to the high concentration of the strain
CJF-002 which was detected at the manifold in the 20 days following the initial
injection.
The authors believe that the principal cause of the sustained increase in oil
recovery over the following year is the stability of the insoluble polymer in the
reservoir. Degradation by strain CJF-002 and the other microbes inhabiting the
reservoir is one factor that may affect the stability of the insoluble polymer.
Some microbes producing cellulase, such as Clostridium sp., are generally
well-known, and some researchers have reported that the microbes belonging to
these species inhabit reservoirs. In our preliminary experiment, however, we
confirmed that the insoluble polymer is not degradable by any microbes
inhabiting the target reservoir or by the strain CJF-002 (data not shown).
Another factor influencing insoluble polymer stability is the possible
absorption of insoluble polymer into reservoir rocks. Considering the results of
the biofilm formation test described above, it is presumed that the insoluble
polymer produced by strain CJF-002 will adhere to the surface of reservoir rock
in the target reservoir.
Fig. 25. Results of microbial monitoring at manifold (2nd trial)

• CJF-002 A Other microbes
439
Fig. 26. Result of microbial monitoring at producing wellJ-18

(2nd trial)
• CJF-002 o Other microbes A Insoluble polymer
Fig. 27. Behavior of total oil production at all test area

(2nd trial)
A
Total liquid production • Oli production
m
Water cut
440
Figure 28 shows the carbon number of production oil from well #24-254'
after microbial treatment. The carbon number of production oil shifted to a low
molecular weight which can be recovered easily. This result indicates that the
production oil after microbial treatment was produced from previously unswept
zones.
In addition, the effectiveness of the microbial treatment was evaluated
through a comparison of tracer tests using NH4SCN solution. After the first and
second microbial treatment, 30,000 ppm of NH4SCN solution was injected into
the reservoir through two injecting wells. The cumulative quantity of that tracer
was approximately 500 kg. In well #22-27, tracer could not be detected before
the treatment, but could be detected after the treatment (see Fig. 29 (A)). In
contrast, in well #26-25, tracer could be detected before, but not after the
treatment (see Fig. 29 (B)). Furthermore, in the other wells, changes in tracer
peak or time of detection were observed. These results show that the microbial
treatment modified the sweep pattern of injection water.
The results described above prove that the insoluble polymer increases the
volumetric sweep efficiency by diverting injection water from the most highly
permeable zones to previously unswept oil-bearing zones. Moreover,
considering the results of the second trial, injection of strain CJF-002 and
molasses for 20 days when the oil production rate decreases may be effective in
continuing to increase oil recovery.
441
Fig. 29. Result of tracer test at well 22-27 and well 26-25 (2nd trial)
O Before microbial treatment A After microbial treatment
Based on the present study, future development of the "selective

plugging" technique is suggested as follows. The exploration of optimum
conditions in field tests, such as the injection concentrations and the injecting
periods of strain CJF-002 and molasses, should be continued based on
biotechnological monitoring data. This research will contribute to improving the
effectiveness of the plug, reliable plugging of the target location in the reservoir,
and higher efficiency and durability of the selective plugging effect.
A technique utilizing multiple diverse microbes should also be developed,
in addition to using biotechnology to screen highly-useful microbes.
442
5.5.3. Economic feasibility o/MEOR

Additional running costs for two months in the second trial were as
follows: 236 tons of molasses (US$27,000), 40 tons of diesel oil (US$15,500)
and the labor costs of two to three people. Additional facilities in the second trial
included a molasses tank (5 kl), a microbial incubation tank (1 kl), and injection
pumps for the molasses and strain CJF-002. In contrast, additional income over
one year of the second trial was US$490,240 (US$20 per bbl).
Previous studies report that operational costs of the MEOR process range
from $2 to $4 per incremental barrel of oil. Based on the present results, the
running costs (aside from labor costs and facility costs) for increasing oil
recovery is 1.7 US$ per bbl. These results indicate that MEOR increases oil
recovery in an economically attractive manner.
6. CONCLUSION
6.1. Conclusions of all the present studies

Based on the results of laboratory experiments and MEOR field tests, the
following conclusions were drawn:
1) Biotechnological tools for estimating the behavior of indigenous microbes in

the reservoir (PCR-RFLP method), as well as injected microbes (Direct-PCR
method), were successfully developed.
2) Biotechnological tools have improved our understanding of micro flora in the

MEOR test field.
a) Microbes isolated from reservoir rock have been carried into the target
reservoir through the development of the oilfield.
b) Fracture zones and their surroundings are susceptible to injected water,
and various microbes included in the injected water are apt to propagate at those
zones.
c) Using microbes selected for reservoir characteristics (such as the
development history of the reservoir, microbes inhabiting the reservoir
environment, and geological features) is necessary for increasing oil recovery.
3) Transplanting microbes into the high permeability zones of the reservoir and
demonstrating the resulting insoluble polymer production are best accomplished
with operations which reflect and respond to data monitoring the behavior of the
injected microbe.
4) MEOR is expected to be an economically feasible technique.

443
5) It is clear that the following points are very important for development of the
MEOR technique:
a) Understanding of microbes related to MEOR (including microbes
inhabiting the ground water, molasses and reservoir environments, in addition to
the injecting microbe).
b) Monitoring of behavior of these microbes at field trials.
c) Designing of field operations which reflect those monitoring data.
d) Establishing techniques for transplanting microbes and demonstrating
microbial metabolic function in the reservoir.
In the present study, valuable data to demonstrate the MEOR effect were
successfully collected and the results obtained from this research support the
theory that MEOR can effectively increase oil recovery. Though the results are
seen in only a few cases yet at most, the authors believe that this is an example
of a successful application of MEOR to a broad range of reservoirs.
6.2. Contribution of biotechnology to the development of the MEOR

technique.
In the examination of microbes related to MEOR processes, PCR-RFLP
analysis of the 16S rDNA of microbes was useful for investigating microbes
inhabiting the ground water, molasses, reservoir brine and reservoir rock. This
method is also useful for confirming a state of sterilization in equipment, such as
incubation tanks, tank trucks, and fluid lines, during field operations.
As demonstrated in this study, the Direct-PCR method, with primers
including sequences in the ribosomal spacer region between the 16S and 23S
rDNA, is a powerful tool for screening the injection microbe. The screening
involves several tests, such as competitive culture tests, on multiple diverse
microbes. The Direct-PCR method is also useful for monitoring injection
microbes in the incubation tanks, tank trucks, manifolds and well sites (injection
wells and production wells) during field operations.
Therefore, biotechnologies contribute to promoting MEOR because the
MEOR process must be designed based on the features of microbes and
geological characteristics in each oil field or reservoir.
Acknowledgements
We would like to thank Japan National Oil Corporation, PetroChina
Company Limited Jilin Oilfield Company and Chugai Technos. Company
Limited for the permission to present this paper. We also thank Tohoku
University and KRI Inc. for their constant support in microbial analysis at Fuyu
oilfield. We are also very grateful to fellows for their continuous assistance
during this collaborative research project.
444
REFERENCES
[I] J. W. Beckman, Ind. Eng. Chem., November, 10 (1926) 3.

[2] I. Lazar, and E. C. Donaldson (eds.), Microbial Enhancement of Oil Recovery - Recent
Advances, MEOR Field trials carried out over the world during the last 35 years,
Elsevier, 1991
[3] J. B. Clark, D. M. Munnecke, G. EJanneman, Dev. Ind. Microbiol., 22 (1981) 695.
[4] R. S. Bryant, U. S. DOE Report, NIPER-478, August, DE91002208 (1990).
[5] H. Yonebayashi, M. Taguchi, K. Fujiwara, S. Yoshida, H. Enomoto, J. Jap. Associ. Petrol.
Technol, 61 (1996) 485.
[6] H. Yonebayasi, H. Enomoto, T. Chida and K. Fujiwara, Proceeding of 17th Workshop of
the International Energy Agency, Sydney Australia, September, 1996
[7] H. Yonebayasi, K. Ono, H. Enomoto, T. Chida, C-X. Hong and K. Fujiwara, Society of
Petroleum Engineers 38070 (1997).
[8] H. Yonebayasi, H. Enomoto, K. Fujiwara, T. Chida and C-X. Hong, Laboratory R & D
leads to MEOR Field Pilot in Fuyu-oilfield, Chaina, Proceeding of 9th European
Symposium on Improved Oil Recovery, The Hague, October 20-22, 1997
[9] K. Ono, S. Maezumi, H. K. Sarma, H. Enomoto, C-X. Hong, S-C. Zhou and K. Fujiwara,
Society of Petroleum Engineers 54328 (1999)
[10] S. Maezumi, H. K. Sarma, N. Yazawa, S-C. Zhou, K. Fujiwara, H. Enomoto and C-X.
Hong, Proceeding of 20th Workshop of the International Energy Agency,
Enghien-les-Bains (Paris), France, September 22-24, 1999
[II] K. Fujiwara, S. Tanaka, M. Ohtsuka, N. Ichimura, H. Yonebayashi, C. X. Hong and H.
Enomoto, Sekiyu Gakkaishi (J. Jpn. Petrol. Inst.), 42 (1999) 342.
[12] K. Fujiwara, S. Tanaka, M. Ohtsuka, K. Nakaya, S. Maezumi, N. Yazawa, C. X. Hong, T.
Chida and H. Enomoto, Sekiyu Gakkaishi (J. Jpn. Petrol. Inst.), 43 (2000) 274.
[13] K. Fujiwara, S. Tanaka, M. Ohtsuka, H. Yonebayashi and H. Enomoto, Sekiyu
Gakkaishi (J. Jpn. Petrol. Inst.), 43 (2000) 43.
[14] K. Fujiwara, Proceeding of International Symposium on Research and Education in the
21st Century, Sendai, Japan, August 18-25, 2000
[15] H. Enomoto, K. Fujiwara, H. Yonebayashi Sekiyu Gakkaishi (J. Jpn. Petrol. Inst.), 43
(2000)91.
[16] K. Nagase, S. T. Zhang, H. Asami, N. Yazawa, K. Fujiwara, H. Enomoto, C. X. Hong
and C. X. Liang, Society of Petroleum Engineers 68720 (2001)
and C. X. Liang, Proceeding of 22th Workshop of the International Energy Agency,
Poland, September 10-14, 2001
[18] K. Fujiwara, Cellulose Commun., 8 (2001) 127.
and C. X. Liang, Society of Petroleum Engineer 75238 (2002)
[20] K. Nagase, S. T. Zhang, H. Asami, N. Yazawa, K. Fujiwara, H. Enomoto and C. X.
Hong, J. Jap. Associ. Petrol. Technol., 68 (2003) 271.
445
[21] K. Fujiwara and H. Enomoto, Proceeding of 2nd International Conference of Petroleum

Biotechnology, Mexico City, Mexico, November 5-7, 2003
[24] A. A. Valie, J. O. Stephens and L. R. Brown, Society of Petroleum Engineers 35448
(1996)
[22] L. R. Brown, Society of Petroleum Engineers 59306 (2000)
[23] G. E. Jenneman, R. E. Lappan and R. H. Webb, Society of Petroleum Engineers 59307
(2000)
[25] A. A. Khan, R. A. Jones and C. E. Cerniglia, J. Ind. Microbiol. Biotechnol, 20 (1988)
90.
[26] Y. Tasi, and B. H. Olsen, Appl. Environ. Microbiol., 57 (1991) 1070.
[27] R. J. Steffan, and R. M. Atlas, Appl. Environ. Microbiol., 54 (1992) 2185.
[28] C. R. Kuske, K. L.Banton,, D. L. Adorada, P. C. Stark, K. K.Hill, and P. J.Jackson, Appl.
[29] C. D. Smart, B. Schneider, C. L. Blomquist, L. J. Guerra, N. A. Harrison, U. Ahrens, K.
H. Lorenz, E. Seemuller and B. C. Kirkpatrick, Appl. Envir. Microbiol., 62 (1996) 2988.
[30] N. P. Rijpens, G. Jannes, M. V. Asbroeck, R. Rossau and L. M. Herman, Appl. Envir.
Microbiol., 62(1996) 1683.
[31] J. Chun, A.Huq and R. R. Colwell, Appl. Envir. Microbiol., 65 (1999) 2202.
Chapter 16
Phytoremediation of hydrocarbon-contaminated soils:

principles and applications
R. Kamath, J. A. Rentz, J. L. Schnoor and P. J. J. Alvarez
Department of Civil and Environmental Engineering, Seamans Center,

1. INTRODUCTION
1.1. Common Target Contaminants
Total petroleum hydrocarbons (TPH) comprise a diverse mixture of
hydrocarbons that occur at petrochemical sites and storage areas, waste disposal
pits, refineries and oil spill sites. TPHs are considered persistent hazardous
pollutants, and include compounds that can bioconcentrate and bioaccumulate in
food chains [1], are acutely toxic [2], and some such as benzene [3] and
benzo[a]pyrene are recognized mutagens and carcinogens [4]. Since this group
includes chemicals that have physical and chemical characteristics that vary over
orders of magnitude, TPHs are divided into two categories (Fig. 1). Gasoline
range organics (GRO) corresponds to small chain alkanes (C6-Ci0) with low
boiling point (60°-170 C) such as isopentane, 2,3-dimethyl butane, rc-butane and
fl-pentane, and volatile aromatic compounds such as the monoaromatic
hydrocarbons benzene, toluene, ethylbenzene, and xylenes (BTEX). Diesel
range organics (DRO) includes longer chain alkanes (Cio-C4O) and hydrophobic
chemicals such as polycyclic aromatic hydrocarbons (PAH).
Whereas most of these contaminants do have natural sources,
concentration and release of contaminants through anthropogenic activities has
led to significant contamination of soil and groundwater. The extent of
petroleum hydrocarbon contamination throughout the United States is reflected
by the large number of Superfund sites and Leaking Underground Storage Tanks
(LUST) sites that contain these contaminants (Fig. 2 and 3). These sites often
contain high concentrations of contamination. However, individual
contaminants behave differently. Some contaminants such as BTEX compounds
are highly mobile in the environment, while others such as PAHs tend to bind
448
strongly to soil particles near the source or remain entrapped within an organic
phase.
ii) Polycyclic Aromatic Hydrocarbons (PAH)
Fig. lb. Examples of common Diesel Range Orgamcs (DRO).

449
Since hydrocarbon spills at different sites represent different mixtures, it

is very difficult to find a single, efficient method of cleanup. Current treatment
techniques usually involve excavation and ex situ treatment of the source
material and the contaminated soils. However, residual contamination often
exceeds regulatory limits by a relatively small margin, and occurs over extensive
areas [5]. The large volume of soil affected precludes ex-situ treatment due to
economical constraints and requires the use of relatively inexpensive
remediation schemes, such as phytoremediation.
Fig. 2. United States Superfund sites containing petroleum hydrocarbon contamination for
FY1982 to FY1999 (834 total projects, [6]).
Fig. 3. Total United States underground storage tank corrective actions (FY 1992 to FY 2003,
[7]).
450
Research and application of phytoremediation for treatment of petroleum

hydrocarbon contamination over the past fifteen years has provided much useful
information that can be used to design effective remediation systems and drive
further improvement and innovation. This chapter will attempt to provide a
strong foundation for understanding phytoremediation of petroleum
hydrocarbon contaminated sites from principles to practice.
1.2. General Scope of Phytoremediation

Phytoremediation is a biological technology process that utilizes natural
plant processes to enhance degradation and removal of contaminants in
contaminated soil or groundwater. Broadly, phytoremediation can be cost-
effective for:
a) Large sites with shallow residual-levels of contamination by organic,
nutrient, or metal pollutants, where contamination does not pose an imminent
danger and only "polishing treatment" is required; and
b) Where vegetation is used as a final cap and closure of the site [8].
Advantages of using phytoremediation include cost effectiveness,

aesthetic advantages, and long-term applicability (Table 1). Furthermore, the use
of phytoremediation as a secondary or polishing in situ treatment step minimizes
land disturbance and eliminates transportation and liability costs associated with
offsite treatment and disposal. Increasing public and regulatory acceptance are
likely to extend the use of phytoremediation beyond current applications.
2. PHYTOREMEDIATION MECHANISMS
Phytoremediation utilizes physical, chemical, and biological processes to

remove, degrade, transform, or stabilize contaminants within soil and
groundwater. Hydraulic control, uptake, transformation, volatilization, and
rhizodegradation are important processes used during phytoremediation (Fig. 4)
and are discussed below.
2.1. Hydraulic Control
Phytoremediation applications can be designed to capture contaminated
groundwater plumes to prevent off-site migration and/or decrease downward
migration of contaminants, as illustrated in Fig. 5. Trees and grasses act as a
solar "pump" removing water from soils and aquifers through transpiration.
Contaminant plume capture relies on the formation of a cone of
depression within an aquifer due to uptake of water by plants and subsequent
transpiration.
451
Table 1
Advantages and disadvantages of phytoremediation over traditional technologies such as
pump and treat of contaminated groundwater and soil excavation and above-ground treatment.
Advantages Disadvantages
Relatively low cost Longer remediation times
Easily implemented and maintained Climate dependent
Several mechanisms for removal Effects to food web might be unknown
Environmentally friendly Ultimate contaminant fates might be unknown
Aesthetically pleasing Results are variable
Reduces landfilled wastes
Harvestable plant material
Figure 4. Schematic of different mechanisms of contaminant removal by plants [8].

452
The key to forming a successful barrier against plume migration is for

trees to be rooted into a shallow water table aquifer. Phreatophytes, deep-rooted
plants including hybrid poplars and willows are most often used for hydraulic
control. When planted densely (more than 600 trees per acre), poplars and
willows usually reach optimum working conditions after 3-4 years during
canopy closure when almost all the direct sunlight is intercepted.
The application of phytoremediation requires that the bottom of the
aquifer be confined by materials of low hydraulic conductivity such as clay,
shale, or rock (hydraulic conductivity < 10"6 cm/s) and does not "leak" water
vertically down to another unit. However, plume capture is not limited to
shallow aquifers, as poplar trees planted in well casings have been used to tap
water tables at a depth of 10-m [10].
Fig. 5. Plan view of trees planted on a line (similar to an interdiction well field) to capture a
shallow groundwater plume (Modified after [9]).
Downward migration of contaminants due to percolation of rainwater can

also be controlled with phytoremediation. Within the upper region of an aquifer,
grasses with dense, fibrous root systems are used to transpire water and limit
percolation of contaminants through the vadose zone and to intercept rainwater
that may discourage tree root penetration through the water table.
2.2. Uptake, translocation, and transformation
Moderately hydrophobic (log KQW = 1.0 to 3.0) hydrocarbons, including
BTEX, can be removed from soil and groundwater through direct plant uptake.
The transpiration stream concentration factor (TSCF), an indirect measure of
uptake efficiency, has been used to adequately predict whether contaminants
will be taken up by plants (Fig. 6). Briggs [11] proposed a bell-shaped
relationship between TSCF and contaminant hydrophobicity, indicated by the
453
logarithmic of the octanol-water partitioning coefficient (log Kow). This

relationship was developed for pesticide uptake by barley plants, and is given by
equation (1) below. Burken and Schnoor [12] adapted this equation to describe
the uptake of a wide variety of organic contaminants (including BTEX) by
hybrid poplar trees. This relationship is represented by equation (2) and is
depicted in Figure 6.
TSCF =0.784 exp {-(log K™ - 1.78)2/ 2.44} (1)
TSCF = 0.756 exp{-(log K™ - 2.50)2 / 2.58} (2)
The bell-shaped curve shown in Fig. 6 reflects poor plant uptake of

hydrophilic compounds (log Kow < 1), which have little affinity for root
membranes; high uptake efficiency of moderately hydrophobic hydrocarbons
such as BTEX (1.5 < low Kow < 3.5); and poor uptake of hydrophobic
hydrocarbons such as PAHs (log Kow > 4), which strongly sorb to soil and are
therefore, not bioavailable.
The rate of contaminant removal has been found to be a function of
uptake efficiency (e.g., TSCF), transpiration rate, and the contaminant
concentration in soil water, as discussed in section 5.1. Uptake efficiency varies
with plant species, age, health, and physico-chemical properties of the root zone.
Transpiration rate also varies dramatically and depends on the plant type, leaf
area, nutrients, soil moisture, temperature, wind conditions, and relative
humidity.
Once the organic xenobiotic enters the plant system, it is partitioned to
different plant parts through translocation. Unlike microbial species that
metabolize organic contaminants to carbon dioxide and water, plants use
detoxification mechanisms that transform parent chemicals to non-phytotoxic
metabolites. The detoxification mechanism within plants is often described
using the "green liver" concept [13, 14]. Once a contaminant enters the plant,
any number of reactions within the following series may occur.
• Phase I - Conversion
• Phase II - Conjugation
• Phase III - Compartmentation
454
Fig. 6. Estimated transpiration stream concentration factors (TSCF) for BTEX using Eq. 2.
Conversion reactions include oxidations, reductions, or hydrolysis that the

plant uses to begin detoxification. Conjugation reactions chemically link the
Phase I products to glutathione, sugars, or amino acids and thus, the plant alters
the solubility and toxicity of the contaminant. Once conjugated, xenobiotics can
be removed as waste or compartmentalized. During compartmentation,
chemicals are conjugated and segregated into vacuoles or bound to the cell wall
material (hemicellulose or lignin). Phase III conjugates are often described as
"bound residues" because chemical extraction methods do not recover the
original contaminants.
Trichloroethylene (TCE), which is not a hydrocarbon but is one of the
more studied volatile organic compounds, has been shown to degrade to
trichloroethanol, trichloroacetic acid, and dichloroacetic acid in hybrid poplars
[15]. However, overall mass balances have been poor, indicating that other
processes or further transformations that result in bound residues may be
occurring [16]. Whereas Burken and Schnoor (1996) demonstrated that BTEX
compounds translocate to the leaves, not much is known about the fate of BTEX
compounds or other hydrocarbons in plants [17].
In general, the ultimate fate of phytotransformed contaminants with
respect to C-cycling between a plant and its environment remains unclear.
Concern centers on whether transformed contaminants will pose a threat to
human or ecological health. Products of conversion reactions could be more
455
toxic than the parent contaminants when consumed by animals or potentially

leached to the environment from fallen leaves [18]. Release of contaminants
from conjugated complexes or compartmentalization could occur in the gut of a
worm, snail, or butterfly [8]. This raises the potential of re-introducing the
pollutant into the food chain. Therefore, a thorough understanding of pathways
and end products of enzymatic processes within a plant is required if
phytoremediation is to be applied successfully and accepted widely.
2.3. Phytovolatilization
The natural ability of a plant to volatilize a contaminant that has been
taken up through its roots can be exploited as a natural air-stripping pump
system. Phytovolatilization is most applicable to those contaminants that are
treated by conventional air-stripping i.e., contaminants with a Henry's constant
KH > 10 atm m3 waterm"3 air, such as BTEX, TCE, vinyl chloride and carbon
tetrachloride. Chemicals with KH < 10 atm m3 waterm"3 air such as phenol and
PCP are not suitable for the air-stripping mechanism because of their relatively
low volatility.
Volatile pollutants diffuse from the plant into the atmosphere through
open stomata in leaves. Radial diffusion through stem tissues has also been
reported [19-21]. For example, methyl-tert-butyl ether (MTBE) can escape
through leaves, stems, and the bark to the atmosphere [22-23]. Tree core
samples of hybrid poplars exposed to TCE also showed radial diffusion from the
stem [24] rather than transpiration from leaves [24, 25] as the main dissipation
mechanism. Generally, the concentration of VOCs in the xylem decreases with
increasing distance from the roots [24].
Once released into the atmosphere, compounds with double-bonds such as
TCE and perchloroethylene (PCE) could be rapidly oxidized in the atmosphere
by hydroxyl radicals. However, under certain circumstances (e.g., poor air
circulation) phytovolatilization may not provide a terminal solution. For
example, MTBE is long lived in the atmosphere and can pose a risk to shallow
groundwater during precipitation [26]. In such cases, simple mass balance
models can be utilized to determine if phytovolatilization poses a significant risk
to humans and/or the environment [20, 24, 27]. Nevertheless, the rate of release
of VOCs from plant tissues is generally small relative to other emissions [27].
Thus, phytovolatilization is a potentially viable remediation strategy for many
volatile organic chemicals.
2.4. Rhizodegradation
Microbial degradation in the rhizosphere might be the most significant
mechanism for removal of diesel range organics in vegetated contaminated soils
[28-34]. This occurs because contaminants such as PAHs are highly
hydrophobic and their sorption to soil decreases their bioavailability for plant
uptake and phytotransformation. Briggs (1982) first demonstrated that the
456
lipophilicity of a pesticide determines its fate in a barley plant [11]. High Kow
values (an indicator of hydrophobicity) corresponded to a greater possibility that
the compound would be retained in the roots (Eq. 3). Burken and Schnoor
(1998) published similar results for the sorption of a wide range of organic
contaminants to roots of hybrid poplar plants grown hydroponically (Eq. 4) [12].
log (RCF - 0.82) = 0.77 log Kow -1.52 (3)
log (RCF - 3.0) = 0.65 log Kow -1.57 (4)
Where the Root Concentration Factor (RCF) (L/kg dry roots) is the ratio
of organic chemical sorbed on the root (mg/kg of fresh root tissue) to that in
hydroponic solution (mg/L). This equilibrium partitioning coefficient has
generally proved to be a good indicator of whether a plant retains a contaminant
in the root, which increases the probability of microbial degradation (not
withstanding significant bioavailability limitations). However, a few exceptions
exist such as phenol and aniline, which bind irreversibly to the root (especially
aniline) and are chemically transformed. They are not appreciably desorbed
because they are covalently bound as metabolic products in plant tissue [35].
Fig. 7. Estimated Root concentration factors (RCF) for PAHs using Eq. 4.
457
Figure 7 uses Eq. 4 to estimate RCF values for a few common PAHs. The
hydrophobic (high sorption) characteristics of PAHs and other DRO compounds
result in high retention in the root zone. Fortunately, the rhizosphere of most
plants promotes a wealth of microorganisms that can contribute significantly to
the degradation of petroleum hydrocarbons during phytoremediation. Thus,
though a plant may not directly act upon these contaminants, a plant can
influence the microbial community within its root zone to a great extent.
Potential rhizosphere interactions that may be important for
phytoremediation of petroleum hydrocarbons include:
1. Prolific microbial growth

2. Repression/induction of catabolic enzymes
3. Co-oxidation of contaminants
4. Changes in bioavailability
5. Chemotaxis of competent strains
Deposition of plant-derived carbon sources through root exudation, and/or

root turnover provides rhizosphere bacteria with numerous organic substrates
[36]. Rhizodeposition can account for release of 7 to 27 percent of the total
carbon fixed during plant photosynthesis [37] and varies between plants.
Commonly reported estimates are between 10 - 100 mg-C g-root material"1 [38]
of which root exudation is reported to range between 0.4 - 27.7 mg-C g-root
material"1 [39-41]. The composition and quantity of root-derived material
released into the rhizosphere varies depending on the season [42], the age of
plant [42] and the health of the plant [43] but generally contains sugars (15 -
65% total organic carbon), organic acids (9 - 33% total organic carbon), amino
acids (2 - 31% total organic carbon) [34,39-40] and phenolics (0.3-4 mg-Cg-
root material"1) [42-44]. Plant stress and age generally increase rhizodeposition.
The availability of simple organic carbon sources that can be used for
growth promotes rhizosphere microbial populations which have been reported to
be 4- to 100- fold greater than that observed in surrounding bulk soils [33, 45-
48]. Selection of competent microorganisms during phytoremediation has been
hypothesized. Miya and Firestone [28] observed greater percentages of
phenanthrene degrading bacteria in rhizosphere soil than bulk soils and
suggested the rhizosphere selected for PAH degraders. Siciliano et al. (2003)
observed a higher frequency of catabolic genes in tall fescue rhizosphere than in
bulk soil [49], suggesting that gene transfer or another mechanism of selection
exists in the rhizosphere. However, the presence of contaminants in these
experimental systems likely provided a strong selective pressure for competent
strains [50]. Investigation of competent degraders within the rhizosphere of
uncontaminated soil has not been reported; such studies are needed to provide
conclusive evidence for selection of specific degraders through plant influence.
458
Induction of microbial aromatic degradation has also been hypothesized

due to the deposition of phenolic compounds that are structurally analogous to
known inducers of enzymes responsible for degradation of aromatic
contaminants [51-52]. Gilbert and Crowley demonstrated induction of
polychlorinated biphenyl (PCB) degradation in Arthrobacter sp. strain BIB, a
gram-positive organism, using spearmint products and identified /-carvone as
the compound responsible [52]. Interestingly, /-carvone was not a growth
substrate tor Arthrobacter sp. strain BIB, and it inhibited growth of the bacteria
on fructose. Induction of PAH degrading enzymes by plant root products has not
been demonstrated in the literature.
In a screening test of inducers of naphthalene dioxygenases potentially
released by plants [53], none were detectable in root extracts at concentrations
required for catabolic gene induction. Furthermore, Kamath et al., and Rentz et
al. observed inhibition of catabolic enzyme activity on a per cell basis following
exposure to environmentally relevant concentrations of plant root products
(exudates and turnover) [53-54]. This was attributed to the presence of organic
acids, carbohydrates, and amino acids, known repressors of aromatic catabolism
within soil bacteria. However, both studies concluded that proliferation of
competent genotypes through growth could compensate for the interference that
labile substrates exert on the expression of PAH catabolic genes. Currently, little
information concerning the expression of other catabolic enzymes during
petroleum hydrocarbon phytoremediation is available.
Several researchers have suggested that co-oxidation of high molecular
weight (HMW) PAH within the rhizosphere [37,47-48] is an important
mechanism for phytoremediation. Generally, HMW PAHs do not serve as
carbon and energy source for microbial populations during degradation. The use
of plants as a method to "inject" growth substrates to contaminated soil could
overcome this limitation to degradation [28]. Soil experiments with plants and
root exudates (pyrene, 4-rings) have shown degradation of HMW PAH and co-
oxidation was implied. However, oxidation or metabolism of HMW PAH has
not been demonstrated using a well-defined system. Co-oxidation and
cometabolism is likely an important process within the rhizosphere with the
availability of a wide array of growth substrates, although no studies have
assessed the importance of this mechanism compared to other processes.
The bioavailabilitiy of hydrophobic contaminants may also be altered
with the root zone environment. Exudation of organic acids could promote
contaminant desorption from soil and solublization, but re-sorption to roots [55]
may compete with microbial utilization. For carcinogenic and highly
hydrophobic benzo[a]pyrene, sorption to roots could prove to be an acceptable
end-point with respect to human and environmental risk. However, no studies
have assessed the potential of this attenuation mechanism.
459
Chemotaxis of competent bacteria towards the rhizosphere may also

enhance rhizoremediation. Ortega-Calvo et al. demonstrated chemotaxis of
PAH-degrading rhizosphere bacteria towards root exudates [56]. Interestingly,
these bacteria were also attracted to naphthalene and phenanthrene, but repelled
by anthracene and pyrene.
4.5. Summary of mechanisms
The different mechanisms discussed above could be utilized for the
remediation of a wide variety of contaminants (Table 2). Phytoremediation
could therefore be applied for the remediation of numerous contaminated sites.
However, not much is known about contaminant fate and transformation
pathways, including the identity of metabolites. Little data also exists on
contaminant removal rates and efficiencies directly attributable to plants under
field conditions. Therefore, further research is required before a tree can be
designed as an engineered reactor system and optimized for efficiency at the
field-scale.
3. PILOT STUDIES
While numerous studies have been carried out at the lab-scale, very little has
been published about field scale implementation of phytoremediation.
Nedunuri et al. [57] investigated total petroleum hydrocarbon (TPH) removal at
several field sites contaminated with crude oil, diesel fuel, or petroleum refinery
wastes, at initial TPH concentrations of 1,700 to 16,000 mg/kg. Plant growth
varied by species, but the presence of some species led to greater TPH
disappearance than with other species or in unvegetated soil. At a crude oil-
contaminated field site near the Gulf of Mexico, an annual rye-soybean rotation
plot and a St. Augustine grass-cowpea rotation plot had significantly (P < 0.05)
greater TPH disappearance than did sorghum-sudan grass or unvegetated
control plots, at 21 months. At a diesel fuel-contaminated Craney Island field
site in Norfolk, Virginia, the fescue plot had significantly (P < 0.10) greater
TPH removal than did an unvegetated plot. At a refinery waste site, statistical
analyses were not presented due to the short time since establishment of the
plots, but Nedunuri et al. reported that qualitatively, the vegetated plots had
greater TPH removal than the unvegetated control plots. After investigating the
potential to use phytoremediation at a site contaminated with hydrocarbons, the
Alabama Department of Environmental Management granted a site, which
involved about 1500 cubic yards of soil of which 70% of the baseline samples
contained over 100 ppm of total petroleum hydrocarbon (TPH). After 1 year of
vegetative cover, approximately 83% of the samples contained less than 10-
ppmTPH[58].
460
Table 2
Potential clean-up mechanisms during phytoremediation of hydrocarbon-
contaminated sites based on physical properties of the target pollutants such as
octanol-water partitioning coefficient (KoW) and Henry's dimensionless constant (KH).
Potential Removal
Contaminants Sources KoW* KH* Mechanisms
Gasoline Range Organics (GRO)
Refineries, LUST, . , { K H n m Hydraulic Control
Fuel spills Pnytovolatihzation
Gasoline TITCT ,„ , ln-4 Hydraulic Control
Oxygenates Phytovolatilization
Diesel Range Organics (DRO)
Coal-gasification,
TIATT Fpetroleum distillation, ^1A4 „ -s „, . ,. ..
PAH . >10 < 2 x l 1A
O Rhizoremediation
wood preservation,
waste disposal
4. FIELD SCALE CONSIDERATIONS
Design of a phytoremediation system varies according to the contaminant/s, the

conditions at the site, the level of clean up required, and the plant/s that are used.
Nevertheless, it is possible to specify a few design considerations that are a part
of most phytoremediation efforts. These include:
• Site Treatability
• Plant selection and planting density
• Irrigation, agronomic inputs and maintenance
• Cost Estimation
• Mathematical Modeling
• Clean-up time required
• Analysis of failure modes
4.1. Site Treatability
4.1.1. Source Removal

For phytoremediation to succeed, it is very important to physically
remove the source of contamination (e.g., excavation of highly-contaminated
soil and/or extraction of free phase). The presence of a continuous source can be
detrimental to the health of the plants and can extend the life of the
phytoremediation project indefinitely.
461
4.1.2. Depth of Contamination

Phytoremediation is most effective at sites with shallow (i.e., root
accessible) contaminated soils where contaminants can be treated in the
rhizosphere and/or by plant uptake. Roots of phreatophytic trees can be expected
to grow at least 3 meters into a soil profile, and it is possible to encourage
rooting to a depth of 5 meters or more using the tree-in-a-well concept [10]. On
the other hand, roots of some grasses (alfalfa, switchgrass, tall fescue) can reach
soil depths of only 0.25-0.4 m. Buffelgrass roots to a depth of 0.75 m but has
been observed to have dense rooting pattern within 0.3 m from the topsoil layer.
Hawaiian plants, Milo and Kou which were used to remediate saline soils
contaminated with TPHs, rooted to a depth of more than 1.5 m by growing
through the brackish water table into a zone of concentrated contaminants [59].
Optimizing irrigation patterns can also facilitate biodegradation of
contaminants by creating an "expanded rhizosphere" due to translocation of
organic root exudates and inorganic nutrients to relatively deep soil layers.
Phytoremediation can therefore influence soils to the depth where irrigation
water reaches, even if the roots are sparse in the contamination zone.
4.1.3. Soil composition and quality

Soil quality is another important factor for determining successful
germination, growth and health of plants. Heavily contaminated soils have a
tendency towards poor physical conditioning which is unsuitable for vigorous
growth of vegetation and rhizosphere bacteria. It is therefore critical to use
amendments to improve the quality of soil before planting. Common limitations
are poor moisture-holding capacity, insufficient aeration, low permeability and
nutrient deficiencies. Agronomic soil analysis and preliminary greenhouse or
pilot scale experiments can help identify these constraints. For example, nutrient
analysis of contaminated soils from a site at the Unocal Bulk Storage Terminal
at Superior, Wisconsin [54] indicated general deficiencies in nitrogen,
phosphorus, potassium, and zinc. To decrease the soil pH, an addition of sulfur
was also recommended. This information was subsequently used in greenhouse
treatability studies, from which a formula of 50 lb/ac phosphorus, 225 lb/ac zinc,
and 50 lb/ac potassium was identified as optimum for growth of native grasses.
Organic amendments such as aged manure, sewage sludge, compost,
straw, or mulch can be used to increase the water-holding capacity of a
contaminated soil. Soil pH can be increased and decreased by the addition of
lime and sulphur respectively.
4.1.4. Weather
Phytoremediation might be best suited for tropical countries where plant
growth occurs all year round. In temperate climates, the active contribution of
phytoremediation is restricted to the growing period only. Winter operations
462
may pose problems for phytoremediation when deciduous vegetation loses its
leaves, transformation and uptake cease, and soil water is no longer transpired.
However, a combination of grasses can be used to prolong the growing period.
4.2. Plant Selection Criteria
Plants should be selected according to the needs of the application, the
contaminants of concern and their potential to thrive on contaminated soil.
Design requirements should include the use of native plants to avoid
introduction of invasive species. Apart from this, vegetation should be fast
growing, hardy, easy to plant and maintain. The main aim is to ensure that roots
expand throughout the entire contaminated zone. In temperate climates with
shallow contaminated aquifers, phreatophytes, such as Populus sp. (hybrid
poplar, cottonwood, aspen) and Salix sp. (willow) are often selected because of
fast growth, deep rooting ability down to the surface of groundwater, large
transpiration rates, and the fact that they are native throughout most of the
country. Among tropical plants tested for use in Pacific Islands, three coastal
trees, kou {Cordia subcordata), milo (Thespesia populnea), and kiawe (Prosopis
pallida) and the native shrub beach naupaka (Scaevola sericd) tolerated field
conditions and facilitated clean-up of soils contaminated with diesel fuel [59].
Grasses are often planted in tandem with trees at sites with organic
contaminants as the primary remediation method. They provide a tremendous
amount of fine roots in the surface soil, which is effective at binding and
transforming hydrophobic contaminants such as TPH, BTEX, and PAHs.
Grasses are often planted between rows of trees to provide for soil stabilization
and protection against wind-blown dust that can move contaminants off-site.
Legumes such as alfalfa (Medicago sativa), alsike clover (Trifolium hybridum ),
and peas (can be used to restore nitrogen to poor soils. Fescue (Vulpia myuros),
rye (Elymus sp.), clover {Trifolium sp.) and reed canary grass (Phalaris
arundinacea) have been used successfully at several sites, especially
petrochemical wastes. Once harvested, the grasses can be disposed off as
compost or burned.
Plant tolerance to high contaminant concentrations is also a very
important factor to keep in mind. The phytotoxicity of petroleum hydrocarbons
is a function of the specific contaminant composition, its concentration, and the
plant species used. Major adverse effects typically include reduced germination
and growth if contaminant concentrations are sufficiently high. In general, TPH
values of 15 percent or greater can result in significant reductions in plant
growth and in some cases mortality. Compared with uncontaminated soil, soils
with 2% TPH reduced alfalfa yields by 32 percent [61]. Production of biomass
by ryegrass was reduced 46 percent at a soil concentration of 0.5 percent (5000
mg/kg) hydrocarbons [47]. It was found that plants pre-grown in clean soil and
subsequently transplanted to the contaminated soil grew nearly as well as the
463
control, showing that toxicity was associated with germination and/or early plant
growth. Similarly, poor rooting of ryegrass compared to legumes appeared to
adversely affect the removal of TPH from Gulf War-contaminated soils [62].
Also, although the germination of sunflower seeds and beans was greater than
that of maize, vegetative growth was greater for maize than beans,
demonstrating that germination and later plant growth may be affected
differently [63].
Aged spills tend to be much less phytotoxic than fresh ones, possibly
because of the lower bioavailability of toxic compounds in the aged spills.
However, the speciation of petroleum hydrocarbons is also very important in
determining phytotoxicity. A fuel oil with 30 percent aromatics resulted in LC50
germination (oil concentration lethal to 50 percent of test plants) values of
7 percent (70,000 mg/kg) for sunflower seeds. The volatile fraction can prove
most toxic to plants. Aromatic volatile petroleum hydrocarbons such as benzene
have been used as herbicides in the past years, illustrating their phytotoxicity
when applied to plant leaves [64]. In contrast, no phytotoxic effects were
observed in hybrid poplar trees exposed to a simulated groundwater containing a
mixture of VOCs including BTEX, chlorinated aliphatics, and alcohols at a total
concentration of 169mg/L [65]. Reduction of the volatile fraction may be
accomplished through management, such as by tillage of the soil. If initial
efforts at plant establishment at a site fail, replanting the area may ultimately
lead to success as concentrations or bioavailability of the more phytotoxic
components decline.
Solution-phase concentrations of hydrocarbons are also important,
particularly for aquifer remediation applications of phytoremediation. Additional
components with phytotoxic effects include various unsaturated hydrocarbons
and acidic hydrocarbons such as alicyclics with carboxylic acid groups
(naphthenic acids) [64].
A screening test and knowledge from the literature of plant attributes is
essential for selection of plants. Most experts recommend a mixture of grasses
or legumes to address surface soils contaminated with petroleum hydrocarbons.
However, design engineers should work in interdisciplinary teams that include a
botanist and/or agricultural specialist to identify and select plants that will grow
well at the site. Preliminary greenhouse studies should also be used to identify
plants that can thrive and enhance transformation of contaminants of concern to
non-toxic or less toxic products.
The U.S. Department of Agriculture also provides two databases on plants
(http://Plant-Materials.nrcs.usda.gov/ and http://plants.usda.gov/). For
information specifically pertaining to plants used for phytoremediation of
petroleum hydrocarbons, the Phytopet database compiled by the Department of
Soil Science, University of Saskatchewan in co-operation with Environment
Canada is available at http://www.phytopet.usask.ca.
464
4.2.1. Time scale of clean-up

Degradation of organics may be limited by mass transfer, i.e., desorption
and mass transport of chemicals from soil particles to the aqueous phase may
become the rate determining step. Therefore, phytoremediation may require
more time (see Section 4) to achieve clean-up standards than other more costly
alternatives such as excavation or ex-situ treatment, especially for hydrophobic
pollutants that are tightly bound to soil particles. In many cases,
phytoremediation may serve as a final "polishing step" to close sites after more
aggressive clean-up technologies have been used to treat the hot spots.
4.2.2. Plant Density

Planting density depends on the application. Louis Licht, Ecolotree® Inc.,
(http://www.ecolotree.com), pioneered the use of hybrid poplar trees as riparian
zone buffer strips, landfill caps, and at hazardous waste sites. For hybrid poplar
trees, 1000-2000 trees per acre are typically planted with a conventional tree
planter at 12-18 inches depth or in trenched rows 1-6 ft deep. The poplars are
planted simply as "sticks", long cuttings that will root and grow rapidly in the
first season. Several phreatophytes in the Salix family, such as willow and
cottonwood, can be planted in a similar manner. Poplars have the ability to root
along the entire buried depth. If a row conformation is used, the trees may be
spaced with 2 ft between trees and 10 ft between rows. Hardwood trees and
evergreens may require a lower planting density initially.
Projects using hydraulic control are most effective at canopy closure,
when transpiration is maximized (within 5-6 years). Theoretically, this can be
determined based on the amount of energy received from the sun and that
required to evaporate water. For mid-latitudes during the growing season, the
earth receives an average 30 million Joules per square meter per day (30 x 106 J
m"2 d"1) of solar insolation. It takes about 2.5 x 106 Joules to evaporate one liter
of water. Thus, it is thermodynamically possible to evaporate 12 L m"2 d"1. But
no plant is 100% efficient, and energy is required to lift the water from the
groundwater to the atmosphere with friction. Typical crops, like corn, can
evapotranspire about 4-5 L m"2 d"1 during their growth period. Poplars can
perform about 30% more efficiently than corn if they are rooted in the
groundwater table, but they actively transpire only about 4-6 months of the year
(due to seasonal changes), depending on the geographic location. Thus, the best
that can be expected from a phytoremediation effort where the trees have
canopied and are rooted in the groundwater table is 4.5 L m" d"1 x 1.3 x 6/12 x
365 days per year x (lnrVlOOO L) = 1.07 m/yr, which is approximately one
million gallons per acre per year. Typically, evapotranspiration rates range from
about 0.4-1.0 million gallons per acre per year for a good phytoremediation
effort using phreatophyte trees rooted into shallow groundwater.
465
A high initial planting density assures a significant amount of

evapotranspiration in the first year which is normally desirable, but the trees will
naturally thin themselves by competition to 600-800 trees per acre over the first
six years. If desirable, hybrid poplars can be harvested on a six-year rotation and
sold for fuel wood or pulp and paper, and the trees will grow back from the cut-
stump (coppicing trait). The dense, deep root system stays in place to sustain
growth for the next year. The lifetime of hybrid poplars such as Populus
deltoides x nigra DN-34 (Imperial Carolina) is on the order of 30 years which is
usually sufficient as the design life of the project.
4.3. Agronomic Inputs
4.3.1. Irrigation
Results suggest that irrigation can enhance bioremediation of certain
diesel components. For terrestrial phytoremediation applications, it is often
desirable to include irrigation costs on the order of 10-20 inches of water per
year, in the design. Spray irrigation is less efficient than drip irrigation as it
encourages the growth of weeds that compete for nutrients with plants and
hinder their delivery to the contaminated zone. Irrigation of the plants is
especially important during the start of the project. However, after the first year,
hydrologic modeling can beused to estimate the rate of percolation to
groundwater under irrigation conditions. Over time, irrigation can be withdrawn
from the site, provided the area receives sufficient rainfall to sustain the plants.
4.3.2. Fertilizer Requirements

Contaminated soils are usually deficient in macro- and micro-nutrients
(Table 3) necessary for establishing healthy vigorously growing plants and
stimulating microbial contaminant degradation.
Nitrogen fertilization of motor oil-contaminated soils was found to
increase the growth of corn and reduce what appeared to be nitrogen-deficient
yellowing of the leaves [66]. The source of nutrients also appeared to affect the
germination and growth of plants. Organic sources of nitrogen are better than
inorganic sources. This is probably because organic nitrogen sources provide a
slow release source of nitrogen, and also help to improve soil structure and soil-
water relationships for plant growth. It was found that poultry manure increased
the growth of corn in a soil containing 3 percent weight per volume crude oil
more than an inorganic fertilizer containing nitrogen, phosphorus, and potassium
[67]. The addition of sawdust alone improved germination by decreasing oil
contact with seeds, but accentuated the adverse effect of the oil on later growth,
apparently by further widening the carbon-to-nitrogen ratio [67].
466
Table 3
Macro- and Micro-nutrients required for healthy plant growth.
Macronutrientsa (-100 ppm) Micronutrientsb (~1 ppm)

Nitrogen (N) Iron (Fe)
Phosphorus (P) Boron (B)
Potassium (K) Zinc (Zn)
Magnesium (Mg) Copper (Cu)
Calcium (Ca) Manganese (Mn)
Sulfur (S) Molybdenum (Mo)
http://extension.oregonstate.edu/mg/botany/table3.html
b
http://extension.oregonstate.edu/mg/botany/table4.html
With respect to TPH degradation, nutrient addition during

phytoremediation has yielded mixed results. Hutchinson et al. observed better
degradation of TPH using grasses with N/P amendments than without inorganic
amendments [68]. Joner et al. reported improved degradation of 3 and 4 ringed
PAHs with the addition of N/P, but diminished degradation of 5 and 6 ringed
PAHs [69]. Finally, Palmroth et al. observed no improved degradation of diesel
fuel with nutrient amendments during phytoremediation with pine, poplar, or
grasses [70].
Microbial bioremediation of TPH contaminants with nutrient addition
also produced widely varying results. Diesel fuel degradation was stimulated
with the addition of N/P using cold region soils [71] and P amendments
stimulated creosote degradation [72]. Breedveld and Sparrevik observed
improved degradation of 4 ringed PAHs with N/P addition, but no increased
degradation of 3 ringed PAHs [73]. However, Graham et al. assessed an array of
N/P amendments for hexadecane biodegradation and suggested amendments
above stoichiometric requirements can lead to diminished rates of degradation
[74]. This potentially occurs because addition of excessive nitrogen additions
results in an increase in soil salinity and this increases the osmotic stress and
suppresses the activity of hydrocarbon-degrading organisms [71]. Carmichael
and Pfander observed slower degradation of 3 and 4 ringed PAHs with N
addition and no effects for P addition [75]. Johnson and Scow reported similar
results indicating N/P addition inhibited or did not change phenanthrene
degradation (3 ringed PAH) [76]. Their results showed that soil with initial low
concentrations of N or P is more likely to show decreased degradation with N/P
addition. Many PAH-degrading organisms are adapted to low nutrient
conditions and activity may decrease with the addition of soil amendments.
Thus, addition of nutrients should be considered on a site-by-site basis
and a balance should be considered between biodegradation and plant growth.
Application of amendments exclusively for plant growth may result in
diminished contaminant degradation, the ultimate goal of phytoremediation.
467
4.3.3. Oxygen requirements

Soil oxygen is required for optimal aerobic microbial degradation of
petroleum hydrocarbon contaminants. Similar to nutrient deficiencies, oxygen
depletion is caused by natural microbial respiration of contaminants. Within
phytoremediation, plants may be a net positive or negative oxygen source [77].
Plants may improve soil oxygen through two mechanisms. First, specially
adapted plants use aerenchyma, channels of reduced air resistance, to transport
oxygen to the root zone, enhancing aerobic biological degradation [37, 78];
although there are no reports of aerenchyma within hybrid poplars, the subject
of this report. Second, soil dewatering and fracturing increases soil porosity,
allowing increased diffusion of atmospheric oxygen [6].
Plant roots can also be a net oxygen sink within petroleum-contaminated
soils. Rentz et al. [79], observed stimulation of hybrid poplar growth and
increased poplar root density with the addition of Oxygen Release Compound®
(ORC) when plants were grown in petroleum smear zone soils (high
biochemical oxygen demand). Flux of oxygen into soil by plants could be offset
by root turnover and root exudation that provides microbial populations with
simple carbon sources that could deplete soil oxygen when metabolized [80].
Furthermore, plant roots are known to require oxygen [81]. For soils with a high
biochemical oxygen demand, oxygen addition may be required to promote plant
growth and stimulate microbial degradation.
Passive methods of oxygen delivery are suggested to keep costs of
phytoremediation low and include the following. Perforated aeration tubes,
placed next to cuttings, can supply oxygen to roots along a vertical axis [82].
Perforated ADS tubing, placed at depth prior backfilling the planting trench
provides oxygen on a horizontal plane. Gravel used to backfill planting trenchs
allows permeation of oxygen on vertical and horizontal axis. Finally, the use of
solid peroxides (e.g. Oxygen Release Compound®) can provide oxygen to soils
when in contact with water [83].
4.4. Cost
Phytoremediation is usually less costly than competing alternatives such
as soil excavation, pump-and-treat, soil washing, or enhanced extraction. Apart
from costs incurred during installation of vegetation at the site, a field-scale
phytoremediation project involves expenditure on design, site preparation,
reporting, monitoring, and operation and maintenance. It would be prudent to
include preliminary greenhouse experiments along with agronomic soil testing
during the design phase to ensure vigorous plant growth at the field-site.
Mathematical modeling may be necessary to demonstrate the effectiveness of
the technology to regulatory agencies (See section 6).
468
Including all these costs, the start-up cost for phytoremediation at $10,000
- 25,000/ acre is still considerably less expensive than other competing
technologies (Table 4). However, since phytoremediation usually requires five
or more years, it is very important to make sure that funding for operation and
maintenance is available during the life of the project.
4.5. Operation and Maintenance Issues
Operation and maintenance (O & M) is vital to ensure vigorous growth of
plants. Some of the major problems in the field have been weeds, killing frosts
or drought, insect or disease infestation, beaver or deer browse, and damage by
voles. It has been estimated that at least 30 percent of the plants may need to be
replanted in the second or third year. Phreatophytic trees are also a source of
concern since there is a potential for the expanding roots to enter and restrict
flow of subsurface drains and sewers and break power and communication
cables and small pipelines. Further, mowing, pruning, harvesting, monitoring
vegetation for contaminants, irrigation and fertilizer costs should be included in
the initial estimated costs. Jordahl, et al. (2002) provides a good summary of key
siting and O&M issues that occur during the life of a field-scale project [85].
Table 4.
Five-Year Cost Comparison of Phytoremediation by Hybrid Poplar Trees versus
Conventional Pump and Treat [84]
1. Phytotransformation
Design and Implementation $ 50,000
Monitoring Equipment
Capital 10,000
Installation 10,000
Replacement 5,000
5-Year Monitoring
Travel and administration 50,000
Data collection 50,000
Reports (annual) 25,000
Sample analysis 50,000
TOTAL $ 250,000
2. Pump and Treat (3 wells and Reverse Osmosis System)
Equipment $ 100,000
Consulting 25,000
Installation/Construction 100,000
5-Year Costs
Maintenance 105,000
Operation (electricity) 50,000
Waste disposal 180,000
Waste disposal liability 100,000
TOTAL $ 660,000
469
5. Mathematical Modeling
5.1. Groundwater Capture and Transpiration
One must understand where the water is moving at a site in order to
estimate contaminant fate and transport. For applications involving groundwater
remediation, a simple capture-zone calculation [86] can be used to estimate
whether the phytoremediation "pump" can be effective at intercepting and
extracting the plume of contaminants. Trees can be grouped for consideration as
average withdrawal points. The goal of such a phytoremediation effort is to
create a water table depression where contaminants will flow to the vegetation
for uptake and treatment or volatilization. It is important to realize that organic
contaminants are not taken-up at the same concentrations that are present in the
soil or groundwater. Rather, there is a transpiration stream concentration factor
(a fractional efficiency of uptake) that accounts for the partial uptake of
contaminant (due to membrane barriers at the root surface).
U = (TSCF) (T) (C) (5)
where:
U = uptake rate of contaminant, mg/day
TSCF = transpiration stream concentration factor, dimensionless
T = transpiration rate of vegetation, L/day
C = aqueous phase concentration in soil- or ground-water, mg/L
A method for estimating the Transpiration Stream Concentration Factor

(TSCF) for eq. (5) was given by eq. (1) and (2).
If the contaminant plume is not completely taken-up by the vegetation,
the plume that remains could be evapoconcentrated; i.e., the mass of
contaminant in the plume will be less due to uptake by vegetation, but the
concentration remaining will actually be greater due to preferential uptake of
water over the contaminants. This is a potential concern for phytoremediation of
groundwater plumes or created wetlands, where a relatively hydrophilic
contaminant can be concentrated on the downstream side of the phytosystem.
Mature phreatophyte trees (poplar, willow, cottonwood, aspen, ash, alder,
eucalyptus, mesquite, bald cypress, birch and river cedar) typically can transpire
3-5 acre-ft of water per year (36-60 inches of water per year). This is equivalent
to about 600-1000 gallons of water per tree per year for a mature species planted
at 1500 trees per acre. Transpiration rates in the first two years would be
somewhat less, about 200 gallons per tree per year, and hardwood trees would
transpire about half the water of a phreatophyte. Two meters of water per year is
a practical maximum for transpiration in a system with complete canopy
coverage (a theoretical maximum would be 4 m/yr based on the solar energy
470
supplied at 40°N on a clear day that is required to evaporate water). If

evapotranspiration of the system exceeds precipitation, it is possible to capture
water that is moving vertically through soil. Areas that receive precipitation in
the wintertime (dormant season for deciduous trees) must be modeled to
determine if the soil will be sufficiently dry to hold water for the next spring's
growth period. The Corps of Engineers Hydrologic Evaluation of Landfill
Performance (HELP) model (Vicksburg, Mississippi) and other codes have been
used to estimate vertical water movement and percolation to groundwater.
5.2. Contaminant Uptake and Clean-up Time

From equation (5) above, it is possible to estimate the uptake rate of the
contaminant/s. First order kinetics can be assumed as an approximation for
clean-up time. The uptake rate should be divided by the mass of contaminant
remaining in the soil:
k = U/Mo (6)
where:
k = first order rate constant for uptake, yr"1
U = contaminant uptake rate, kg/yr
Mo = mass of contaminant initially, kg
Then, an estimate for mass remaining at any time is expressed by equation (7)
below.
M = Moe"kt (7)
where:
M = mass remaining, kg
t = time, yr
Solving for the time required to achieve clean up of a known action level:
t = -(lnM/M 0 )/k(8)
where:
t = time required for clean-up to action level, yr
M = mass allowed at action level, kg
Mo = initial mass of contaminant, kg
471
Equations (5-8) can be applied to most sites where soil clean-up

regulations are known for metals or organic contaminants.
5.3. Rhizodegradation
The Root Concentration Factor, which was previously described (eq. (3)
and eq. (4)) is defined as the ratio of the contaminant in roots to the
concentration dissolved in soil water (fig/kg root per |ag/L). It is important in
estimating the mass of contaminant sorbed to roots in phytoremediation systems.
While RCF is a simple indicator of whether a contaminant will be retained on
the root surface, mathematical modeling of the removal of contaminants in the
rhizosphere is highly complex. The most sophisticated rhizosphere fate model
available is the Pesticide Root zone model (PRZM) available from the EPA
(http://www.epa.gov/oppefedl/models/water/index.htm). It allows for the
estimation of the fate of pesticides in the root zone through hydrologic and
chemical transport simulation. The processes of plant uptake, surface run-off,
erosion, decay, volatilization, advection, dispersion and adsorption are
considered. However, for PAHs and other highly hydrophobic contaminants,
factors such as microbial mobility, spatial variability, plant root growth and
depth of root penetration, root turnover and rhizosphere volume are probably
more important.
Current models [87] are built on a conceptual framework in which the
soil-plant contaminant system is compartmentalized into multiple zones: the root
itself, a series of root influenced zones (the rhizosphere), a decaying root zone
and a non-root-influenced zone (the bulk soil). The essence of the system
conceptualization is that each of the modeled zones is treated as a variable
volume, uniformly mixed continuous reactor. The change in each zone's volume
over time is determined from a pair of forcing functions that describe the
specific growth and senescence rates of the plant system. Thus, as the new roots
penetrate the soil and the associated microbial community is established, bulk
soil will be transformed into rhizosphere soil. Similarly, as root senescence
occurs, the root and rhizosphere volume will be converted into decaying root
zone that ultimately returns, through humification, to bulk soil. Different growth
and senescence functions can be used to simulate various grass species growth
and biomass production patterns throughout an annual cycle. The model also
includes the idea that the rhizosphere bacteria will face a gradient of influencing
factors as the distance from roots increases.
The model is useful to identify important variables from those with only
minor effects, and to extrapolate results for one geographic region to another,
based on the patterns of interaction between physical and biological factors.
However, it does not take into account the effect of temperature and availability
of nutrients such as nitrogen, phosphorus, soil oxygen, moisture and pH on
472
degradation rates. Also, it cannot simulate growth of multiple plant species that
might be used in field-scale applications.
6. REGULATORY ISSUES
Compliance with regulatory concerns is a critical factor when considering

remediation of a site. State and federal acceptance of the technology has been
slow but is the product of input by the Interstate Technology and Regulatory
Cooperation Work Group (ITRC), the Superfund Innovative Technology
Evaluation (SITE) program and the Research Technologies Demonstration
Forum (RTDF) program of EPA. The Phytotechnologies Work Team, a part of
the ITRC (www.itrcweb.org), published a Decision Tree (1999) and a Guidance
Document (2001) as a first approximation for whether phytoremediation is
suitable for a particular site. The latter guidance document in combination with
the USEPA document titled "Introduction to Phytoremediation " (EPA 600-R-
99-107) should be useful in guiding industrial site managers.
Apart from the ITRC, the SITE program and RTDF were also designed to
evaluate the potential of phytoremediation for field-scale purposes.
Phytoremediation has been the subject of six SITE investigations and over 25
field trials by RTDF (http://www.rtdf.org). SITE is a formal program established
by EPA's Office of Solid Waste and Emergency Response (OSWER) and the
Office of Research and Development (ORD) in response to the Superfund
Amendments and Reauthorization Act of 1986 (SARA). Consultants are
responsible for operating the innovative system on site and are expected to pay
the costs of the demonstration, together with site owners. EPA is responsible for
project planning, sampling and analysis, quality assurance and quality control,
preparing reports, disseminating information, and transporting and disposing of
treated waste materials.
Under Superfund laws, EPA (1998) [88] lists nine criteria for
consideration:
1. Overall protection of human health and the environment
2. Compliance with Applicable and Relevant and Appropriate Requirements
3. Long-term effectiveness and permanence
4. Reduction of contaminant toxicity, mobility, or volume
5. Short-term effectiveness (including the length of time needed to
implement the technology and associated risks to workers, residents, and the
environment during that time)
6. Implementability (including availability of goods and services)
7. Cost including capital, operation and maintenance, and monitoring
8. State (and federal) acceptance of the technology and its evaluation of its
performance
473
9. Community acceptance which is addressed in the Record of Decision

(ROD) Amendment (including responsiveness summary that presents public
comments and responses to those comments)
Of these, phytoremediation addresses concerns about aesthetics, cost, ease

of implementation and community acceptance. Phytoremediation also has an
advantage over constructed remedies in the long run. Unlike other remediation
technologies, the efficiency of phytoremediation increases with time until the
system reaches its maximum during canopy closure. Further, since it is possible
to monitor the effect of phytoremediation in mitigating vertical percolation of
contaminants as well as soil erosion, it fulfills the criteria required by Risk
Based Corrective Action (RBCA) as well as Monitored Natural Attenuation
(MNA). For most other actions including Voluntary Programs, it is usually
sufficient to show that the cover is lush and growing and that phytoremediation
meets routine (quarterly to annual) groundwater monitoring requirements.
There are certain regulatory limitations to applying phytoremediation to a
site. Phytoremediation is passive technology. Meeting cleanup goals might be
difficult and could require 10 years or more without a guarantee of reaching
specific performance standards. Furthermore, if phytoremediation is to be used
in conjunction with Monitored Natural Attenuation (MNA), it is necessary to
demonstrate that the plume (or contaminated zone) is stable or shrinking and
that it is not causing unacceptable risk to humans or the environment. In
addition, proof that the contaminants are not in danger of moving off the site,
and knowledge of the mechanism of degradation (metabolites, pathways,
products) and/or immobilization/sequestration is required. The following is a list
of environmental monitoring requirements that are often appropriate for
phytoremediation efforts.
• Tree survival rates and replacement requirements.

• Plant (leaf area index) or root densities and replacement requirements.
• Levels of contaminants and/or metabolites measured in leaves or grasses.
Quarterly groundwater monitoring for applicable or relevant and appropriate
requirements (ARAR).
• Sap flow or evapotranspiration estimates to calculate volume of water
treated.
• Soil gas measurements and oxygen profiles with soil depth to demonstrate
aerobic degradation of aromatic constituents or gradual improvement.
• Soil corings to demonstrate that treatment is occurring at the site
(heterogeneity makes this monitoring requirement imprecise and sometimes
misleading)
474
Nevertheless, the fate of contaminants taken up by the plant or

transformed in the rhizosphere is not well-understood and it can sometimes
prove difficult to show that the technology reduces toxicity of the contaminants,
prevents cross-media transfer of pollutants and/or reduces risks to human and
ecological receptors. Furthermore, since the distribution and composition of
contaminants in field-scale projects is very heterogeneous, it is almost
impossible to prove that phytoremediation enhances the rate of contaminant
removal at field sites.
In summary, long-term monitoring and evaluation of phytoremediation
technology is still needed to demonstrate efficacy, to further define suitable
plants and applications, and to gain acceptance from regulatory agencies.
7. EMERGING ETHICAL ISSUES, OPPORTUNITIES AND

CHALLENGES
One emerging issue requiring consideration is the use of plants that could be
genetically modified to exhibit beneficial traits for phytoremediation, such as
increased water uptake for hydraulic control, drought and pest tolerance, and
increased enzyme activity for faster and more complete phytotransformation of
organic contaminants. A similar potential innovation is the inoculation of the
rhizosphere with genetically modified organisms (GMOs) that overexpress
catabolic enzymes for enhanced rhizoremediation.
The use of (microbial and/or plant) GMOs represents a research frontier
with broad implications. The potential benefits of using GMOs are significant,
and extend beyond improved contaminant removal efficiency and lower O&M
costs. For example, GMO's might facilitate coupling phytoremediation with the
production of marketable non-food (cash) crops that could be used for energy
production (e.g., biomass production for fuel wood, biodiesel, or fuel ethanol) or
raw materials for commercial products (e.g., pulp for paper or feedstock for
cosmetic or pharmaceutical industries). Nevertheless, although GMOs have been
extensively used in agriculture, little research has been conducted to assess their
long-term life cycle impacts, including the consequences of increased genetic
drift across species on biodiversity and biological community structure. This
gives rise to much speculation and polarization regarding the consequences of in
vitro genetic manipulation, which represents a significant political barrier to the
use of GMOs in phytoremediation. Furthermore, the need for GMOs may be
questionable for many projects, considering that indigenous species often
perform adequately and that we have not tapped the full potential of wild species
due to our limited understanding of various phytoremediation mechanisms,
including the regulation of enzyme systems that degrade pollutants.
In summary, the potential benefits and risks associated with the use of
genetic manipulation suggest that we need to be very cautious of GMOs, but not
475
necessarily rule out their application in phytoremediation yet. Additional

scientific input will hopefully contribute to dissipating myths, discern the
benefits and consequences of using GMOs, and ensure their safe use when their
application is justified.
8. CONCLUSIONS
Over the past 15 years, phytoremediation has developed into a more acceptable
technology for the remediation of soils and groundwater polluted with residual
concentrations of petroleum hydrocarbons. However, regulators as well as
consumers are still wary about the efficiency, predictability and applications of
the technique. The ITRC guidelines and decision tree has supported the use of
phytoremediation for most field-scale applications. Yet, at this point there is an
urgent need for strong evidence supporting the potential of phytoremediation in
protecting human as well as ecological receptors from exposure to contaminants,
using rigorous methods of risk analysis. For direct application to field projects, it
would be desirable if more protocols for designing preliminary greenhouse
experiments reflecting field-environments and cheap innovative methods of
encouraging growth of healthy plants were published. Research examining the
long-term fate of contaminants in the environment would be particularly
relevant. Also important is the difficult task of evaluating acceptable endpoints
(e.g., humification) using standard ecological toxicity or bioavailability assays
that might support phytoremediation.
Albeit, phytoremediation is an emerging technology that is based on
sound ecological engineering principles. Phytoremediation is a practical and
cost-effective approach with aesthetical and atmospheric-carbon-sequestration
ancillary benefits, and is particularly attractive for rural areas with residual and
shallow contamination. Phytoremediation also holds great potential to manage a
wide variety of environmental pollution problems, including the cleanup of soils
and groundwater contaminated with hydrocarbons and other hazardous
substances, the attenuation of pollutants dispersing through the environment in
agricultural drainage, landfill leachates, and other forms of surface runoff or
sub-surface migration, and the assimilation of industrial wastewater effluents to
support efforts to move towards a zero-discharge policy from industrial facilities
(e.g., refineries). Although phytoremediation is not a panacea that would be
universally applicable, it is rapidly achieving pedagogical maturity and it has
already earned an important place in the menu of alternatives from where we
select solutions for our environmental pollution problems.
476
REFERENCES
[I] E. McElroy, J. W. Farrington, and J. M. Teal, In: U. Varanasi (ed.) Metabolism of
Polycyclic Aromatic Hydrocarbons in the Aquatic Environment, CRC Press Inc., Boca
Raton, Florida, 1989.
[2] M. A. Heitkamp, and C. E. Cerniglia, Appl. Environ. Microbiol., 54 (1988) 1612.
[3] International Agency for Research on Cancer (IARC), Monographs on the Evaluation of
the Carcinogenic Risk of Some Industrial Chemicals to Humans, Lyon, France, 2000.
[4] K. Mortelmans, S. Harworth, T. Lawlor, W. Speck, B. Tainer and, and E. Zeiger,
Environ. Mutagen, 8 (1986) 1.
[5] National Research Council, Report of the National Research Council Committee on
Groundwater Cleanup Alternatives, National Academy Press, Washington, DC, 1994.
[6] Environmental Protection Agency (EPA), Treatment Technologies for Site Cleanup:
Annual Status Report (Tenth Edition), 2001.
[7] Office of Solid Waste and Emergency Response (OS WER), FY 2003 Semi-Annual
(Mid-Year) Activity Report, 2003.
[8] J. L. Schnoor, L. A. Licht, S. C. McCutcheon, N. L. Wolfe, and L. H. Camera, Environ.
Sci. Technol, 29 (1995) 318A.
[9] P. A. Domenico and F. W. Schwartz, Physical and Chemical Hydrogeology, Wiley,
New York, 1998.
[10] E. G. Gatliff, Remediation, No. 8 (1994) 343-352.
[II] G. G. Briggs, R. H. Bromilow, and A. A. Evans, Pesticide Sci., 13 (1982) 495.
[12] J. G. Burken and J. L. Schnoor, Environ. Sci. Technol., 32 (1998) 3379.
[13] H. Sandermann Jr., Pharmacogenetics, 4 (1994) 225.
[14] H. Ohkawa, H. Imaishi, N. Shiota, T. Yamada, and H. Inui, (eds. G. T. Brooks and T. R.
Roberts). In: Pesticide Chemistry and Bioscience-The Food Environment Challenge,
Special Publication 233, The Royal Society of Chemistry, Cambridge, UK, 1999, pp.
259-264.
[15] L. A. Newman, S. E. Strand, N. Choe, J. Duffy, G. Ekuan, M. Ruszaj, B. B. Shurtleff, J.
Wilmoth, P. Heilman, and M. P. Gordon, Environ. Sci. Technol., 31 (1997) 1062.
[16] W. E. Schnabel, A. C. Dietz, J. G. Burken, J. L. Schnoor, and P. J. J. Alvarez, Wat. Res.,
31 (1997)816.
[17] J. G. Burken, J. L. Schnoor, J. Environ. Eng. ASCE, 122 (1996) 958.
[18] J.M. Yoon, B.T. Oh, C.L. Just and J.L. Schnoor, Environ. Sci. Technol. 36 (2002) 4649.
[19] Q. Zhang, L. C. Davis, and L. E. Erickson, Environ. Sci. Technol., 35 (2001) 725.
[20] M. Narayanan, L. E. Erickson, and L. C. Davis, Environ. Progress, 18 (1999) 231.
[21] L. C. Davis, D. Lupher, J. Hu, and L. E. Erickson, (eds. L. E. Erickson and M. M.
Rankin), Proceedings of the 1999 Conference on Hazardous Waste Research, Kansas
State University, Manhattan, Kansas, (1999) pp. 219-223,
http://www.engg.ksu.edu/HSRC.
[22] M. S. Hong, W. F. Farmayan, I. J. Dortch, C. Y. Chiang, S. K. McMillan, and J. L.
Schnoor, Environ. Sci. Technol., 35 (2001) 1231.
[23] S. Trapp, J. C. McFarlane (eds.), Plant Contamination: Modeling and Simulation of
Organic Chemical Processes, Lewis Publishers, Boca Raton, Florida, 1994.
[24] X. Ma and J. G. Burken, Environ. Sci. Technol., 37 (2003) 2534.
[25] L. A. Newman, S. L. Doty, K. L. Gery, P. E. Heilman, I. Muiznieks, T. Q. Shang, S. T.
Siemieniec, S. E. Strand, X. P. Wang, A. M. Wilson, and M. P. Gordon, J. Soil
Contamination, 7 (1998) 531.
477
[26] J. F. Pankow, N. R. Thompson, R. L. Johnson, A. L. Baehr, and J. S. Zogorski, Environ.

Sci. Technol., 31(1997)2821.
[27] E. W. Aitchison, S. L. Kelley, P. J. J Alvarez, and J. L. Schnoor, Wat. Environ. Res., 72
(2000)313.
[28] W. Aprill, and R. C. Sims, Chemosphere, 20 (1990) 253.
[29] M. K. Banks, E. Lee, A. P. Schwab, J. Environ. Qual, 28 (1999) 294.
[30] P. Binet, J. M. Portal, and C. Leyval, Soil Biol. Biochem., 32 (2000) 2011.
[31] H. H Liste and M. Alexander, Chemosphere, 40 (2000) 11.
[32] K. A. Reilley, M. K. Banks, and A. P. Schwab, J. Environ. Qual., 25 (1996) 212.
[33] R. K. Miya and M. K. Firestone, J. Env. Qual, 29 (2000) 584.
[34] R. K. Miya and M. K. Firestone, J. Env. Qual. 30 (2001) 1191.
[35] S. Lang, M. S. Thesis, University of Iowa 1998.
[36] M. B. Leigh, J. S. Fletcher, X.O. Fu, and F. J. Schmitz, Environ. Sci. Technol, 36
(2002) 1579.
[37] J.F. Shimp, J.C. Tracy, L.C. Davis, W. Huang, L.E. Erickson, and J.L. Schnoor, Crit.
Rev. Environ. Sci. Technol, 23 (1993) 41.
[38] J. M. Whipps and J. M. Lynch, Ann. Proc. Phytochem. Soc, 26 (1985) 59.
[39] W. Hiitsch, J. Augustin, and W. Merbach, J. Plant Nutr. Soil Sci, 165 (2002) 397.
[40] Kraffczyk, G. Trolldenier, and H. Beringer, Soil Biol. Biochem. 16 (1984) 315.
[41] J. A. Trofymow, D. C. Coleman, and C. Cambardella, Plant Soil, 97 (1987) 333.
[42] R. S. Hegde and J. S. Fletcher, Chemosphere, 32 (1996) 2471.
[43] Z. Rengel, Plant Soil, 245 (2002) 59.
[44] H. Wu, T. Haig, J. Pratley, D. Lemerle, and M. An, J. Agric. Food Chem, 48 (2000)
5321.
[45] H. Chaineau, J. L. Morel, and J. Oudot, J. Environ. Qual, 29 (2000) 569.
[46] J. L. Jordahl, L. Foster, J. L. Schnoor, and P. J. J. Alvarez, Environ. Toxicol. Chem, 16
(1997)1318.
[47] T. Gunther, U. Dornberger, and W. Fritsche, Chemosphere, 33 (1996) 203.
[48] T. D. Nichols, D.C. Wolf, H.B. Rogers, C. A. Beyrouty, and C. M. Reynolds, Wat. Air
Soil Poll, 95 (1997) 165.
[49] S. D. Siciliano, J. J. Germida, K. Banks, and C. W. Greer. Appl. Environ. Microbiol, 69
(2003) 483.
[50] S. D. Siciliano, N. Fortin, A. Mihoc, G. Wisse, S. Labelle, D. Beaumier, D. Ouellette, R.
Roy, L. G. Whyte, M. K. Banks, P. Schwab, K. Lee, and C. W. Greer, Appl. Environ.
Microbiol, 67 (2001) 2469.
[51] J. S. Fletcher and R. S. Hegde, Chemosphere, 31 (1995) 3009.
[52] S. Gilbert and D. E. Crowley, Appl. Environ. Microbiol, 63 (1997) 1933.
[53] R. Kamath, J. L. Schnoor, and P. J. J. Alvarez, Environ. Sci. Technol, (In Press, 2004)
[54] J. A. Rentz, P. J. J. Alvarez, and J. L. Schnoor, Environ. Microbiol, (hi Press, 2004).
[55] A. P. Schwab, A. A. Al-Assi, and M. K. Banks, J. Environ. Qual, 27 (1998) 220.
[56] J. J. Ortega-Calvo, A. I. Marchenko, A. V. Vorobyov, and R. V. Borovick. FEMS
Microbiol. Ecol, 44 (2003) 373.
[57] K. V. Nedunuri, R. S. Gouindaraju, M. K. Banks, A. P. Schwab, and Z. Chen, J.
Environ. Eng, 126 (2000) 483.
[58] D. Hecht and G. Badiane, New Internationalist, June (1998) 12.
[59] U.S. Army Corps of Engineers, Agriculturally Based Bioremediation of Petroleum-
Contaminated Soils and Shallow Groundwater in Pacific Island Ecosystems, May 2003.
[60] K. Precht, M. S. Thesis, University of Iowa 2003.
478
[61] C.C. Wiltse, W. L. Rooney, Z. Chen, A. P. Schwab, and M. K. Banks, J. Environ. Qual.,
27 (1998) 169.
[62] Yateem, A., A.S. El-Nawawy, and N. Al-Awadhi,. Soil and Groundwater Cleanup,
1999, pp. 31-33.
[63] C. H. Chaineau, J. L. Morel, and J. Oudot, J. Environ. Qual., 26 (1997) 1478.
[64] Baker, J.M. 1970. Environ. Pollut., 1 (1970) 27.
[65] Ferro, J. Kennedy, R. Kjelgren, J. Rieder, and S. Perrin, Int. J. Phytoremediation, 1
(1999) 9.
[66] J. Giddens, J. Environ. Qual., 5 (1970) 179.
[67] Amadi, A. A. Dickson, and G.O. Maate, Wat. Air Soil Pollut., 66 (1993) 59.
[68] S. L. Hutchinson, N. K. Banks, and A. P. Schwab, J. Environ. Qual., 30 (2001) 395.
[69] E. J. Joner, S. C. Corgie, N. Amellal, and C. Leyval, Soil Biol. Biochem., 34 (2002)
859.
[70] M. R. T. Palmroth, J. Pichtel, and J. A. Puhakka, Bioresource Technol, 84 (2002) 221.
[71] J. L. Walworth, C. R. Woolard, and K. C. Harris, Cold Regions Sci. Technol., 37 (2003)
81.
[72] T. M. Phillips, A. G. Seech, D. Liu, H. Lee, and J. T. Trevors, Environ. Toxicology, 15
(2000) 99.
[73] D. Breeveld, and M. Sparrevik, Biodegradation, 11 (2000) 391.
[74] D. W. Graham, V. H. Smith, D. L. Cleland, and K. P. Law, Wat., Air, and Soil Poll.,
I l l (1999)1.
[75] L. M. Carmichael, and F. K. Pfaender, Biodegradation, 8 (1997) 1.
[76] C.R. Johnson and K.M. Scow, Biodegradation, 10 (1999) 43.
[77] R.W. Lee, S.A. Jones, E.L. Kuniansky, G. Harvey, B.S. Lollar, and G.F. Slater, Int. J.
Phytoremed., No. 2 (2000) 193.
[78] L. E. Erickson, M. K. Banks, L. C. Davis, A. P. Schwab, N. Muralidharan, K. Reilley,
and J.C. Tracy, Environ. Progress, No. 13 (1993) 226.
[79] J. A. Rentz, B. Chapman, P. J. J. Alvarez, and J. L. Schnoor, Int. J. Phytoremed., No. 5
(2003) 57.
[80] J. M. Lynch, The Rhizosphere, New York, Wiley, 1990.
[81] D. S. Neuman, M. Wagner, J. H. Braatne, and J. Howe, Stress Physiology - abiotic, In:
Biology of Populus and its implications for management and conservation, (eds. R. F.
Stettler, H. D. Bradshaw, Jr., P. E. Heilman, and T. M. Hinckley), NRC Research Press,
National Research Council of Canada, Ottawa, ON, 1996, pp. 423-458.
[82] Ferro, J. Chard, R. Kjelgren, B. Chard, D. Turner, and T. Montague, Int. J. Phytoremed.,
No. 3 (2001) 105.
[83] S. S. Koenigsberg and R. D. Norris, Accelerated Bioremediation Using Slow Release
Compounds, Regenesis Bioremediation Products, San Clemente, CA, 1999.
[84] E.G. Gatliff, Phytoremediation. Ground Water Monitoring Review, Winter/1996.
[85] J. L. Jordahl, M. F. Madison, H. M. E. Smesrud, and M. Q. Motte, (Eds. S. C.
McCutcheon and J. L. Schnoor), In: Phytoremediation- Degradation and Control of
Contaminants, Wiley fnterscience, New York, NY, 2002.
[86] P. A. Domenico and F. W. Schwartz, Physical and Chemical Hydrogeology, NY, John
Wiley & Sons, Inc., 1998.
[87] G. J. Thoma, T. B. Lam, and D. C. Wolf, Int. J. Phytoremed., 5 (2003) 47.
[88] Environmental Protection Agency (EPA), Phytoremediation Handbook for Site
Manager, 1998 Draft.
Chapter 17
Biological treatment of polluted air emissions
S. Revaha and R. Auriab

a
Department of Process Engineering, Universidad Autonoma Metropolitana-
Iztapalapa (UAM-I). Apdo. Postal 55-534, 09340 Mexico D.F., Mexico
b
Laboratoire IRD de Microbiologie, Universite de Provence, CESB/ESIL, Case
925, 163 Avenue de Luminy 13288, Marseille Cedex 9 France
1. METHODS OF ODOR AND VOC CONTROL

1.1. VOCs and odor definition.
Volatile organic and inorganic compounds (VOCs and VICs) are emitted
as gases from certain solids or liquids and have an impact on the health and the
environment. VOCs are organic compounds having vapor pressure exceeding
0.1 millimeters of mercury (mrnHg) at standard conditions (20 °C and 760
mmHg). Diverse VOCs have shown short- and long-term adverse health effects.
In the environment, VOCs have been identified as major contributors to
atmospheric photochemical reactions and to smog, which can cause different
damages to humans, plant, animal life and to many materials. Some common
examples of the volatile inorganic gases (VICs) are hydrogen sulfide (H2S),
sulfur dioxide (SO2) and ammonia (NH3). Odor-pollution problems are caused
by mixtures of highly volatile compounds with very low threshold detection
limit that are generally in small concentrations. Air pollution by the emission of
VOCs, VICs and odors are widespread in the petroleum industry including
extraction, refining, transport and distribution. The large volume emitted can
result in significant health-related and ecological deterioration problems.
1.2. Methods for VOC, VIC and odor control from stationary sources.
To select an appropriate control method it is essential to consider the
physical, thermodynamic and reaction properties of the pollutant. These
properties, the conditions of the stream and the extent of treatment required
480
determine the control method [1]. Figure 1 shows a classification of common

technologies applied for VOC, VIC and odor control.
Figure 2 was obtained from actual applications and shows that the initial
choice can be made based on the stream flow and the pollutant concentration [2,
3]. Further selection of the biological system should include the biodegradability
and other technical factors such as temperature, oxygen content of waste gas,
stream composition, solubility, operating schedule, presence of particles,
production of by-products, utility, and maintenance requirements. The overall
selection must include the investment and maintenance costs and the secondary
environmental impacts.
2. BIOLOGICAL METHODS
2.1. Introduction
The basis of biological air treatment systems (BATS) is the competence of
active microorganisms, including bacteria, yeast, and fungi, to transform certain
organic and inorganic pollutants into compounds with lower health and
environmental impact. These compounds result generally from oxidative
reactions and include carbon dioxide, water, nitrate, and sulfate. Since
microorganisms are unable to transform the pollutants directly in the gas phase,
the first step is the solubilization in the biologically active aqueous phase.
Microbes utilize these molecules as a source of nutrients and energy for growth,
producing more biomass, which is partly recycled. The performance of the
process is determined by the relative rates of the physical, chemical and
biological processes. Several books have been published that cover different
aspects of BATS [2,3,41.
Fig. 1. Methods for decontamination of emissions from stationary sources.

481
Fig. 2. Comparison of methods for decontamination of air-polluted emissions.
Although the basic-transfer and reaction mechanisms are the same for all
the BATS, there are different equipment configurations. These can be grouped,
as shown in Table 1, depending on the state of biomass and liquid phase in the
reactor.
2.2. Types of reactors

The principal reactors described in Table 1 are:
2.2.1. Biofllter (BF).
In biofilters (Fig. 3), the polluted air percolates through a moist packed
bed, which supports the microorganisms, that grow on its surface and crevices,
forming a biofilm. Pollutant transformation rates depends on the microbial
density and activity, its bioavailability and the environmental conditions, such as
temperature, nutrients, pH and humidity. The biofilm humidity is one of the
most critical condition to maintain a proper performance, since biological
activity is highly dependent on the water activity (aw). Drying occurs by
incoming air with low humidity and by the heat generated by the biological
reaction [5]. Increased drying rates are obtained with dry air and high
elimination capacity, therefore the air is generally pre-humidified and the
support periodically water sprayed. Biofilters have generally a high void fraction
to limit pressure drop and to reduce ventilation costs.
The supports can be natural and bioactive, inert or a mixture. The natural
bioactive supports are soil, peat, compost, bark etc. They are relatively
482
inexpensive and abundant, and have been used for many applications. They can
retain water and generally contain an initial microbial population with enough
mineral nutrients [6]. Suitable structural characteristics are obtained by mixing
with a coarser fraction, including plastics or ceramics, to prevent high pressure
drop and to limit bed compaction. The natural supports may degrade with time
and lose their structure and water-retaining capacity, inducing channeling and
performance loss [7]. In some cases, re-mixing the support with some fresh
material and nutrients allows to recover the activity [8], but eventually, it will
need to be replaced. With proper maintenance, the support can be used for
several years. As mentioned above, biofilters can use inert, natural or synthetic
supports such as activated carbon, ceramics, lava rock, polyurethane foam,
vermiculite and perlite. On one hand, these supports lack the nutrients required
to sustain the microbial activity, therefore it is necessary to intermittently add
them. On the other hand, they are not degraded and, in theory, could be
engineered to have optimal properties such as controlled head loss, porosity,
adsorptive capacity, etc. This remains an area of active research [2, 3, 4].
The high surface and low water content favor degradation of the less-
hydrophilic pollutants (Henry's constant, H >10). Empty bed retention time
(EBRT) is generally between 30 seconds and 2 minutes. Due to the type of
supports used, the height of the packed bed is generally about 0.8 to 1.2 m,
making thus necessary to have a large footprint, which may be a disadvantage
for situations where space is limited.
2.2.2. Biotricklingfilters (BT).

In BT, the polluted air (Fig. 4) flows upflow or downflow through a
packed column where liquid is continuously recirculated. The pollutant is first
solubilized in the falling liquid film and then transferred to the biofilm
developed on the support. The liquid provides moisture, nutrients, pH control to
the biofilm and allows the removal of inhibiting products and excess biomass.
Table 1.
Classification of biological reactors
Biomass Liquid phase Reactor

Fixed on a support Stationary Biofilter, BF
Fixed on a support Flowing Biotrickling filter, BT
Rotating contactors, RC
Suspended Flowing Bioscrubber, BS
Suspended or fixed Stationary Suspended growth, SR
Fixed on a membrane Flowing Membrane, MR
483
Fig. 3. Schematic representation of bio filter (BF).
Inert random supports or structured packing are used. Some examples

include plastic corrugated structured PVC sheets, Raschig or Pall rings and
saddles, lava rock and polyurethane foam [2]. To maintain low pressure drop
and reduce clogging, the supports have low porosity and low specific surface
(100 - 400 m2 m"3). EBRT are normally around 30 seconds but systems with
EBRT as low as 2 seconds have been reported for low H2S concentration [9].
Fig. 4. Schematic representation of biotrickling filter (BT)

484
2.2.3. Bioscrubbers (BS).

In bioscrubbers, the pollutant in the gas phase is first absorbed in a gas-
liquid contactor (Fig. 5). Subsequently, it is eliminated in a bioreactor and the
liquid, containing the suspended microorganisms, is returned to the contactor.
Nutrients and pH regulators can be added to maintain microbial activity and the
excess of biomass and sub products can be controlled by purging.
The gas-liquid contactors can be packed towers, venturi scrubbers or
spray towers [10]. Bioscrubbers are designed to favor mass transfer with low
pressure drop (< 3 cm H2O m"1). In the bioreactor, supplementary air is added to
favor the oxidation of the pollutant. Water retention time in the reactor is
calculated to eliminate the soluble pollutant, and the biomass concentration is
generally about 5-8 g L"1 [11] to foster high volumetric rates while reducing
clogging problems in the contactor. Bioscrubbers are used for hydrophilic
pollutants (H<1) to avoid big absorbers or large water flows [12].
2.2.4. Other configurations.

Membrane bioreactors (MB). This is an emerging technology derived
from the development of new porous materials [4, 12, 13]. The pollutant in the
gas phase is transferred through a membrane (hollow fibers or flat sheets) to the
biofilm, which is attached to the other side where oxygen and aqueous nutrients
are fed. In hollow fibers, the gas is usually passed through the lumen of the tube
and the biomass is on the shell side. These reactors have been used for other
waste treatment applications where the conditions of the stream excluded the
possibility of direct contact with the biomass. A distinct characteristic of the
Membrane bioreactors is the physical separation of the polluted gaseous stream
from the biomass, which allows the use of BATS in certain applications such as
indoor air or in extreme case, for spaceship air treatment.
Fig. 5. Schematic representation of bioscrubber (BS).

485
Suspended cell bioreactor. In these reactors, the polluted air is bubbled

directly in the bulk of the liquid containing suspended microorganisms. The
most common reactor [14] is an activated-sludge aerator where the sparged
polluted air is treated simultaneously with the municipal wastewater. The reactor
parameters, such as biomass concentration, air feed and sparger design, are
generally imposed by the requirements of the wastewater treatment. A review of
the characteristics of several facilities is given in Ref. 15. In other cases, the
reactors are designed to optimize mass transfer from the bubble to the bulk
liquid where biodegradation occurs and to control the conditions that promote
high microbial rates. These reactors include airlift, external loop, split cylinder,
etc.
Rotating Biological contactors (RC). In these systems, initially developed
to treat water, the polluted air flows through the headspace of a reactor
containing discs that serve as support for a biofilm and are assembled on a
rotating shaft. The shaft is slowly rotated (around 2 rpm) and the discs are
partially wetted in water containing nutrients and other additives. Air can be fed
tangentially to the disks or through perforations in a hollow shaft [16].
Intermittent wetting of the biomass favors mass transfer and biological activity.
2.3. Mechanism
BATS involve complex physical, chemical and biological interactions, [2,
3,4] which will be shortly reviewed:
2.3.1. Gas-liquid phase equilibrium of the pollutant

As the degradation of pollutants cannot occur directly in the gaseous
phase, they have to be first transferred to the biofilm or to the liquid with the
suspended cells. Gas-liquid resistance is often negligible and consequently
interfacial concentrations can be considered in equilibrium. Moreover, because
the concentration of pollutants involved in biological treatment is low, Henry's
law is generally used to describe this equilibrium. Henry's law partition
coefficient can be described as
H = C g /C, (1)
Where
H: dimensionless Henry's constant
Cg: gas phase concentration (g L"1)
Q: liquid phase concentration (g L"1)
In Table 2, some Henry's coefficients are reported for different pollutants.

For high values of Henry's coefficient (H), the pollutant is slightly soluble in
water. For example, with hexane concentration in air of 1 g m"3, corresponding
486
to those usually treated in biological processes, its concentration in water is

lower than 30 ug L"1. Henry's coefficient increases with an higher salinity and
temperature. Data is usually found for H in pure water but it may be different
when biomass is present. A strong decrease of H value has been reported for
toluene, benzene, and trichloroethylene when biomass is present in the aqueous
mixture. In biofiltration, biomass, extra-cellular polymers (EPS) and support
promote the solubilization of hydrophobic compounds [17]. A very important
aspect of BATS performance is the solubilization of oxygen required for the
oxidation of the pollutant. Oxygen limitation can be found when the pollutants
are in high concentration or have very low Henry's coefficient.
Gas- liquid phase equilibrium can be altered by reaction in the liquid. For
example, soluble H2S dissociation depends on pH according to the following
equilibrium reactions. At pH =10, sulfides are present mainly as HS" which is
very soluble and consequently the apparent Henry's coefficient is 3 orders of
magnitude lower than at pH =4.
2.3.2. Pollutant diffusion in the biofilm

Once the pollutant and the oxygen are solubilized in the liquid, diffusion
in the biolayer occurs under the influence of concentration gradients. Diffusion
is described using Fick's law:
Table 2
Henry's coefficient for some common compounds at 25 ° C
(adapted from Ref. 18).
Compound Henry's coefficient

(non-dimensional)
Methane 41.3 U)
Hexane 30.9
Oxygen 29.1
Hydrogen sulfide 0.92
Toluene 0.25
Benzene 0.22
MTBE (methyl t- butyl ether) [19] 0.026
Ethanol 0.0012
Ammonia 0.0005
(1) http://www.epa.gov/regionl/measure/Natatten.pdf
487
J=-D^l (2)
ax
where:
J: mass flux (g m"2 s"1)
D :diffusion coefficient (m2 s"1)
Cl: liquid concentration (g m"3)
x: distance in the biolayer (m)
In biofilms, diffusion coefficient may be strongly reduced by the presence

of biomass and EPS. Most of the effective diffusion coefficients in biofilms are
evaluated by either empirical or semi-empirical correlations. In terms of
concentration gradient, two typical cases of biofiltration are found:
1) No diffusion limitation exists and the biofilm is fully active. In this
case, the biological reaction is the limiting step.
2) Diffusion limitation occurs and the biofilm is not fully active, this
condition is referred as diffusion limited. In this situation, a reaction-free zone is
predicted.
2.3.3. Biodegradation of the pollutant in the biofilm

In most of BATS, pollutants are degraded by microorganisms structured
in biofilms mainly composed of microbial cells and EPS, which promote
adhesion. EPS may account up to 90 % of the total. While the biofilm is often
represented as an homogeneous mass with constant thickness, it has, in reality, a
complex structure comprising diverse biological (presence of bacteria, yeasts
and fungi, EPS,...) and physical (cavities, detachment of the biofilm, microbe-
support interactions, less-dense zones, etc.) components. Biofilm structure and
ecology are poorly known.
In BATS, organic pollutants serve as carbon and energy sources, and the
dissolved oxygen as electron acceptor. Degradation occurs not only during
growth phase, but also when the net growth rate is zero. The energy released
during degradation is used for growth and maintenance metabolism. The
pollutant uptake rate can be defined as:
(3)
where:
(a.: net cell growth (h"1)
X: biomass concentration (gxL"')
Yxs: biomass yield from the pollutant (gx g poiiutant"1)
m: maintenance coefficient (g poiiutant gx' h"1)
488
Specific growth rate (a) depends on the concentration of the limiting substrate:
(4)
where:
|imax: maximum cell growth (h"1)
Ks: half saturation constant (g L"1)
In BATS, mixed populations are generally found and their behavior can
be described by a net growth rate as:
*$=Mx (5)
BATS are open biological systems as the air is never filtered. The large
variety of microorganisms (fungi, yeast, and bacteria) come from the initial
inoculum, from the biofilter packing material (compost, peat, etc..) and from the
incoming air [2, 3, 4, 12]. The microorganisms present in the biofilters are
similar to those found in the natural ecosystems or other biological processes
such as compost or water treatment plants. Although bacteria are dominant in
biofilters, fungi are frequently observed. Recent studies showed that some fungi
[20, 21, 22, 23] can degrade toluene and hexane vapors at higher rates than
bacteria.
For a successful biological air treatment, the pollutants have to be
biodegradable. In general, organic compounds with low molecular weight,
highly soluble in water and simple-bond structures are the best candidates.
Alcohols, aldehydes, ketones, and some simple aromatics have very good
biodegradability, while phenols, aliphatic and chlorinated hydrocarbons show
moderate to slow degradation (Table 4). Ethers, like diethyl and dimethyl ethers
are generally easily degraded while MTBE (Methyl tert butyl ether) is reported
to be very recalcitrant.
Table 4
Biodegradability of pollutants [Ref. 24].
Rapidly Alcohols**, aldehydes*, ketones**, esters**, ethers*, organic

acids*, terpenes, amines*, thiols*, sulfides, ammonia
Slowly Hydrocarbons**, phenols*
Very slowly Halogenated hydrocarbons*T, polyaromatic hydrocarbons,
*Branched molecules are less biodegradable
| Removal in biofilters follows alcohols > esters > ketones > aromatics > alkanes
t Biodegradability decreases with higher number of halogens
489
2.3.4. Performance parameters

Three parameters are often used to compare the pollutant treatment
efficiency in BATS.
Inlet mass pollutant load (IMPL), is defined as the amount of pollutant
introduced into the bioreactor normalized by its empty-bed volume.
(6)
Elimination capacity (EC), is the quantity of pollutant removed per bioreactor

volume per time unit.
(7)
Removal efficiency (RE), is the fraction of the pollutant removed expressed as

percentage.
(8)
3. EXAMPLES OF TREATMENT OF VOLATILE PETROLEUM

HYDROCARBONS BY BATS
Petroleum products such as gasoline, fuel oils, and diesel fuels are among the
most important water, soil and air pollutants. They are complex mixtures of
organic compounds containing a significant volatile fraction. Hydrocarbons are
composed of four main structural classes: 1) «-alkanes (linear saturated
hydrocarbons), 2) isoalkanes (branched saturated hydrocarbons), 3)
cycloalkanes (saturated cyclic alkanes) and 4) aromatics [25]. Moreover, in the
case of gasoline, other oxygenated additives such as the ethers MTBE, ethyl t-
butyl ether (ETDE) or /-amyl methyl ether (TAME) or alcohols such as ethanol,
can be added to improve combustion and consequently air quality.
Hydrocarbons may be released into the atmosphere by evaporation during
production, transport and storage.
Table 5 presents some examples of the degradation of volatile organic
compounds from gasoline by BATS. The removal efficiency of volatile
individual compound in the gasoline varies from 5 % to 99 %. Aromatics are
generally removed more efficiently than n-alkanes. Light aliphatic compounds
(<C5), which are poorly soluble and highly volatile, are generally not removed
by more than 50 %, even at a 6-min residence time [26]. This preferential
removal is promoted by the relative high solubility of aromatics compared to n-
490
alkanes. Recent studies showed that biofilter performance was improved using
filamentous fungi [22]. An elimination capacity of toluene of 258 g m"3 h"1 (RE =
98 %) was attained. This value was higher than those obtained for bacterial
biofilters [27-28]. It is hypothesized that the aerial fungal mycelia, which are in
direct contact with the gas, give a larger superficial area and allow a direct mass
transfer of the volatile compound. Poor degradation of hydrophobic n-alkanes in
biofilter could strongly be increased using filamentous fungi as shown in Ref.
[23].
Very few reports have been published addressing MTBE biofiltration.
Except for the results obtained by Fortin et al, 1999 [29], elimination capacities
of MTBE are generally lower than 10 g m"3 h"1. MTBE is a highly soluble
compound in water but the ether bond and the ter/-butyl moiety have been
shown to limit MTBE biodegradability. Cometabolism has proved to be a good
way to increase biomass production and MTBE degradation. Microbial consortia
[30], can degrade MTBE by cometabolism with n-alkanes (hexane, pentane and
heptane) present in gasoline. Cometabolism with pentane using a single bacteria
(Pseudomonas aeruginosa) can be used to degrade MTBE in a biofilter packed
with vermiculite [31].
Biofilters used to treat gasoline from a soil vapor extraction operation
showed that higher molecular weight compounds were almost completely
removed while lower molecular weight were less degraded [32]. The
predominant compounds remaining in the outlet of the biofilter were tentatively
identified as methyl-substituted alkanes and cycloalkanes in the C6 to C9 range.
A pilot-scale biofilter system treating gasoline vapors presented total
hydrocarbon-elimination capacities of 16 g m~3 h"1 [33]. Linear alkanes and
aromatics were rapidly degraded, while branched alkanes had lower removal
efficiencies. Pilot-scale biofilter elimination capacities for hexane, isooctane and
toluene were 3.2 ghexane m"3 h"1, 3.1 g,so-octanc m"3 h"1, and 1.5 gtoiucne m"3 h"1,
respectively. Removal efficiencies for toluene were the highest and the most
stable.
4. CONCLUSIONS
BATS are among the established technologies that can be applied to control
VOC and odor emissions. For their choice, the characteristics of the stream
(flow, temperature, presence of particles, humidity, etc.), the pollutant
(composition, concentration, reactivity, solubility and biodegradability) and the
required performance have to be considered. BATS are applicable for a wide
range of volatile pollutants found in the petroleum industry and their
applications are growing continually based on scientific and technological
developments.
491
Table 5
Examples of treatment of volatile petroleum hydrocarbons by BATS
Compounds BATS Packing Cgi,,(gm-3) EC Ref.

EBRT (min) (grnV)
Efficiency
Gasoline BT Compost /pine bark 70:30 1000 ppmv 16d [33]

(v/v) 1 45
Gasoline BT Compost/perlite 140-490 ppmv [32]
2.3 89
Ethers
MTBE BF Pall rings 0.8 50 [29]
0.9 90
MTBE a BT Vermiculite 5 0.8 [31]
66 18
MTBE BT Celite ™ R-635 b 0.13 (35ppmv) 7.8 [38]
1 99
MTBEC BT Compost /pine bark 70:30 0.21 3.8 [33]
(v/v) 1 30
Aliphatics
Isopentane BT Peat moss 5 (1700 ppmv) 50 [34]
3 40
Pentane BT Vermiculite 17.4 12 [31]
35 40
Hexane BT Compost/perlite 0.7 (200ppmv) 21 [35]
50:50 (v/v) 3 99
Hexane BT Granular expanded clay 12 150 [23]
26 >80
Hexane BF Polyamide structure wire 10 80 [39]
mat 6.3 89
Cyclohexane BT Compost 0.004 (1.2 ppmv) — [36]
2 9
Aromatics
Benzene BF Fibrous sheet (cotton) 0.5 3 [37]
Q
O 5U
Xylenes BT Peat 8.2 66 [28]
1.7 23
Toluene BT Vermicul ite/Granular 5 258 [22]
activated carbon 85:15 (v/v) 1.2 98
Toluene BT Peat 1.5 25 [27]
1 31
Toluene BT Peat 6.2 165 [28]
1.3 70
a) MTBE degraded in cometabolism with pentane
b) Extruded diatomaceous earth pellet
c) MTBE as a gasoline additive
d) Methane equivalent gHC m"3 h"'
492
REFERENCES
[I] H. Rafson (ed.), Odor and VOC Control Handbook. McGraw Hill Professional. USA
1998.
[2] J.S. Devinny, M. Deshusses and T. S. Webster, Biofiltration for air pollution control.
CRC Press, Boca Raton, Fl. USA, 1999.
[3] C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas treatment. Kluwer
Academic Publishers, The Netherlands, 2001.
[4] Z. Shareefdeen and A. Singh (eds.) Biotechnology for Odour and Air Pollution,
Springer-Verlag, Germany (in press).
[5] M. Morales, S. Hernandez, T. Cornabe, S. Revah and R. Auria Environ. Sci. Technol.
37 (2003) 985.
[6] B. Cardenas-Gonzalez, S. Ergas, M. Switzenbaum and N. Phillibert, Environ. Progress
18(1999)205.
[7] J. M. Morgan-Sagastume, S. Ergas, S. Revah and A. Noyola, J. Air Waste Manage.
Assoc. 53(2003)1011.
[8] R. Auria, G. Frere, M. Morales, M. E. Acuna and S. Revah S, Biotechnol Bioeng 68
(2000) 448.
[9] D. Gabriel and M. Deshusses, PNAS 100 (2003) 6308.
[10] J. W. Van Groenestijn, In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste
gas treatment. Kluwer Academic Publishers, The Netherlands, (2001) 133.
[II] S. P. P. Ottengraf In: Biotechnology 8, Rehm HJ and Reed G (eds), VCH
Verlagsgesellschaft Weinheim, Germany, (1986) 426.
[12] J. W. Van Groenestijn and P. G. Hesselink, Biodegradation 4 (1993) 283.
[13] S. Ergas, In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas treatment.
Kluwer Academic Publishers, The Netherlands, (2001) 163.
[14] A. Bielefeldt, In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for waste gas
treatment. Kluwer Academic Publishers, The Netherlands, (2001) 215.
[15] R. P. Bowker, In H. Rafson (ed.), Odor and VOC Control Handbook. McGraw Hill
Professional. USA (1998)8.192.
[16] R. von Rohr and P. Ruediger In C. Kennes and M. C. Veiga MC (eds.), Bioreactors for
waste gas treatment. Kluwer Academic Publishers, The Netherlands, (2001) 201.
[17] B. Davison, J. Barton, T Klasson and, A. Francisco Biotechnol. Bioeng. 68, (2000) 279.
[18] T. Card, In H. Rafson (ed.), Odor and VOC Control Handbook. McGraw Hill
Professional. USA (1998)2.1.
[19] A. Fischer, M. Muller and J. Klasmeier, Chemosphere 54 (2004) 689.
[20] H. H. J. Cox, R. E. Moerman, S. Vanbaalen, W. N. Vanheiningen, H. J. Doddema and
W. Harder, Biotechnol. Bioeng. 53 (1997) 259.
493
[21] J. Woertz, K. A. Kinney, P. Mclntosh and P. J. Szaniszlo, Biotechnol. Bioeng. 75

(2001) 550.
[22] E. I. Garcia-Pena, S. Hernandez, E. Favela-Torres, R. Auria and S. Revah, 76 (2001) 61.
[23] G. Spigno, C. Pagella, M. Daria Fumi, R. Molteni, and M. De Favieri, Chem. Eng. Sci.
58 (2003) 739.
[24] S. Revah and J. M. Morgan Sagastume In Z. Shareefdeen and A. Singh (eds.)
Biotechnology for Odour and Air Pollution, Springer-Verlag, Germany (in press).
[25] R. Marchal, S. Penet, F. Solano-Serena and J. P. Vandecasteele, Oil Gas Rev. IFP. 58
(2003)441.
[26] G. Leson and B. J. Smith, J. Environ. Eng. 123, (1997) 556.
[27] M. Morales, S. Revah and R. Auria, Biotechnol. Bioeng. 60 (1998) 483.
[28] H. Jorio, K. Kiared, R. Brzezinski, A. Leroux, G. Viel and M. Heitz, J. Chem. Technol.
Biotechnol. 73(1998) 183.
[29] N. Fortin and M. Deshusses, Environ. Sci. Technol. 33 (1999) 2980.
[30] P. M. Gamier, R. Auria, C. Augur and S. Revah, J. Gen. Appl. Microbiol. 46 (2000) 79.
[31] D. Dupasquier, S. Revah and R. Auria,. Environ. Sci. Technol. 36 (2002) 247.
[32] W. F. Wright, Y. Davidova, E. D. Schroeder and D. P. Y. Chang,. Proceedings of
Conference on Biofiltration. USC Los Angeles. (1995) 18.
[33] A. Hernandez, M. Magana, B. Cardenas, S. Hernandez, S. Revah, S. Queney and R.
Auria, Proc. 94th Annual AWMA Meeting Exhibition. (2001) Paper # 1037 AE-2a.
[34] A. P. Togna and M. Singh, M Proc. 87th Annual AWMA Meeting Exhibit, (1994).
[35] E. Morgenroth, E. D. Schroeder, D. P. Y. Chang and K. M. Scow, J. Air Waste Manage.
Assoc. 46(1996)300.
[36] D. X. Li, Proc. of Conference on Biofiltration. USC Los Angeles. (1995) 1.
[37] Q. Zhou,, Y. L. Huang, D. H. Tseng, H. Shim and S. T. Yang S-T J. Chem. Technol.
Biotechnol. 73(1998)359.
[38] J. B. Eweis, E. D. Schroeder ED, D. P. Y. Chang and K. M. Scow, in G.
Wickramanayake and R. Hinchee (eds) Remediation of Chlorinated and Recalcitrant
Compounds, Physical chemical and thermal technologies, Battelle Press, Columbus
Ohio USA (1998) 341.
[39] J. W. Van Groesnetijn and M. E. Lake, hi: Arendt F, Hinsenveld M., Brink WJ van den
(eds). Kluwer, Dordrecht. (1999) 151-155.
Chapter 18
Bioremediation of marine oil spills

R. C. Prince1 and J. R. Clark2
'ExxonMobil Research & Engineering Co.

Annandale, NJ 08801
2
Fairfax, VA 22037
1. INTRODUCTION
Crude oils are the liquid fossil residues of aquatic algae, sometimes with minor
contributions from terrestrial plants, that grew in the distant past. Then as now,
we can imagine that most of this material was biodegraded and recycled on an
essentially annual timescale, but a small fraction became buried and underwent
diagenesis and catagenesis to become oil [1]. This process usually took millions
of years, and was dependent on the depth of burial and the temperature. Some
oil dates from the Precambrian (>570 million years ago), but most is rather
younger; the average age of commercially important oil is about 100 million
years, with the majority being from the Jurassic and Cretaceous (180 to 85
million years ago) [2].
Commercially important oil has migrated from its source rock to a
reservoir, and it is not unusual for these reservoirs to leak. If the leak reaches the
surface, it is known as an oil seep. Humans have used material from such seeps
for thousands of years. Early uses include hafting stone axes to their handles, as
an embalming agent, and as a medical nostrum. Genesis (11,3) says that bitumen
was used as the mortar for the Tower of Babel, and Exodus (2,3) that Moses'
basket was made waterproof with a bitumen daub. It seems likely that several
religions started around natural gas seeps, either as eternal flames or as sources
of hallucinogenic vapors [3]. But these were only very minor uses, and it is only
in the last century and a half that oil has come to play a truly central role in
modern society [4]. Terrestrial seeps were the first locations to be drilled when
oil production began in earnest in the nineteenth century, such as the 1860 Drake
well in Pennsylvania, but marine seeps were drilled by the end of the century.
496
Marine seeps are widespread, as shown in Figure 1, and are a major

source of oil into the World's oceans [5]. Even with today's enormous
commerce in oil, seeps provide about 62% of the total releases into the coastal
marine environment of North America, and 47% of the world. Such seeps must
have been occurring for millions of years, providing an important input of
degradable carbon for local ecosystems and perhaps even major fisheries [6]. A
diverse group of microorganisms exploits this natural input. Oil-degrading
microbes have been found in all marine environments where they have been
looked for, and more than seventy genera of eubacteria and archaea, and a
hundred genera of fungi, have been shown capable of degrading petroleum
hydrocarbons [7]. It is these organisms that remove oil seepage and spilled oil
from the marine and terrestrial environment, and underpin the bioremediation
strategies for dealing with spilled oil that we will describe here.
2. ANTHROPOGENIC INPUT OF OIL INTO THE WORLD'S OCEANS
Oil fuels the modern world on an enormous scale; annual consumption is of the
order of 3.5 billion US gallons (1.2 x 1010 liters) per day [8]. Much of this is
produced at sea; more than 25% of US production is from the Continental Shelf,
and 25% of Saudi Arabia's production, 80% of Nigeria's, and 100% of
Angola's, Australia's, Brazil's and Malaysia's production is offshore. This
production is associated with some oil and grease discharges into the marine
environment associated with the produced water; some 2,700 tonnes per year in
the US, and more than ten times this in the rest of the world [5]. This input is
dwarfed, however, by oil in municipal run-off from the land, and from the
standard operation of marine shipping (Figure 2). Catastrophic spills from
tankers and other ships are well known, but in fact their contribution to total oil
input into the oceans is relatively small, some 8% of the global input, 2% of
North American input. Nevertheless, since such spills are large on the local
scale, they often require environmentally appropriate and cost-effective
responses. Fortunately, despite the increasing volume of transported oil, the
amount spilled from catastrophic spills has been generally decreasing over the
last few decades [9]. The major exception to this decline was the appalling
environmental crime in the Arabian Gulf in 1991, where Iraqi forces deliberately
released more than a million tonnes (about 260 million gallons or a billion liters)
of oil into the sea near Kuwait [10]. An additional 350 million gallons (1.2 xlO9
1) were deposited in the Gulf as fallout from the smoke plumes of the > 700 oil
well fires in the Kuwait oil fields [11], making this by far the largest man-made
release of oil into the marine environment to date.
497
3. COMPOSITION OF CRUDE OIL
Crude oils are complex mixtures of hydrocarbons with significant quantities

(typically about 15%) of compounds containing additional heteroatoms such as
oxygen, sulfur or nitrogen. The hydrogen to carbon ratio of the hydrocarbons is
typically between 1.5 and 2, indicating a mixture of aliphatic (predominant
carbons are -CH 2 -, either as linear molecules, known as paraffins, or as rings
known as naphthenes) and aromatic species (principal carbons are -HC=CH- in
rings). Alkenes and alkynes, linear unsaturated molecules, are rare in crude oils,
although they can be abundant in some refined products such as gasoline. Tissot
and Welte [2] calculate the average composition of more than 525 crude oils as
58.2% saturates, 28.6% aromatics and 14.2% polar compounds, noting that the
absolute values vary widely in different oils. On average, there is rough parity
between paraffins, naphthenes and aromatic hydrocarbons in crude oils [2].
Paraffins in crude oils may start with methane, and extend to waxes with
more than seventy carbons. Their total content varies widely, from essentially
undetectable to as high as 35%, depending on source and reservoir conditions,
but they typically make up 15-20% of an undegraded crude oil. There are also
branched alkanes; especially in the C6 to C8 range, but pristane (C19H40) and
phytane (C20H42), molecular relics of the phytol chains of chlorophylls and
perhaps other biomolecules, are usually the most abundant individual branched
alkanes. Pristane is thought to be the result of initial partial degradation of
phytol in the presence of oxygen, while phytane is thought to be the result of
diagenesis in the absence of oxygen [2].
Fig. 1. Map of the major oil seeps into the World's oceans. Data taken from reference [5].
498
Fig. 2. Oil input into the World's oceans. Data taken from reference [5].
The naphthenes include parent compounds, such as cyclopentane,

cyclohexane and decalin, together with their alkylated congeners. Tissot and
Welte [2] quote the average composition of the naphthene fraction of 299 crude
oils as 54.9% one and two ring naphthenes, 20.4% tricyclic naphthenes, and
24.0% tetra and pentacyclic naphthenes. These latter molecules are amongst the
better understood molecular biomarkers in crude oils, and they are used
extensively in correlating reservoirs and source rocks [12], in assigning the
depositional environment of source rocks [12], and more recently as conserved
internal markers during biodegradation [13].
Because of the separation procedures used in the analysis of crude oils,
any molecule containing at least one aromatic ring is included in the "aromatic"
fraction, regardless of the presence of saturated rings and/or alkyl substituents.
Aromatic heterocycles containing sulfur, such as thiophenes, benzothiophenes
and dibenzothiophenes, or nitrogen, such as the indoles, carbazoles and
quinolines, also fall into the aromatic category. Alkylated aromatic species are
more abundant than their parent compounds, with mono-, di- and tri-methyl
derivatives usually being most abundant. Nevertheless, the median aromatic
structure probably has one or two methyl groups and a long-chain alkyl
substituent [14].
The polar molecules are the most difficult to characterize because they
typically cannot be analyzed by gas chromatography, the method of choice for
the molecular characterization of hydrocarbons. Petroleum polar compounds
contain heteroatoms such as nitrogen, oxygen and/or sulfur, and the category
includes the porphyrins, typically with nickel or vanadium in the tetrapyrole,
499
naphthenic acids and large molecules known as asphaltenes. Some asphaltenes

have molecular weights into the thousands and even higher, and many are
suspended in the oil rather than dissolved in it [15]. The polar fraction of the oil
contains the majority of the color centers in crude oil, and in isolation these
materials are difficult to distinguish from more recent biological residues, such
as the humic and fulvic acids [16], except with sophisticated tools such as
isotope analysis. Recent developments in electrospray mass spectrometry
promise significant progress in determining the molecular structure of these
compounds [17, 18].
Several hundred different crude oils are being produced today, and their
chemical composition and properties vary quite widely. They are typically
classified by their density. The oil industry uses a unit known as API (American
Petroleum Institute) gravity, which is defined as [142.5/(specific gravity)] -
131.5, and expressed as degrees (°). Thus water has an API gravity of 10°, and
denser fluids will have lower API gravities. For convenience, oils with API
gravities greater than 40° are said to be light oils, while those with API gravities
of less than about 17° are said to be heavy. Note that even these typically float
on water, especially seawater. Light oils have higher proportions of
hydrocarbons; heavy oils are rich in polars and asphaltenes. Viscosity is roughly
inversely proportional to API gravity, but it is also dependent on the physical
state of the polar compounds and longer alkanes in the oil, and is highly
dependent on the temperature.
Among petroleum products in commerce, crude oil is transported in the
largest volumes, both in undersea pipelines and in tankers. Refined products are
also shipped, and of course all ships contain large amounts of fuel for their own
propulsion. Refining crude oils for commercial applications starts with
distillation, and the simplest distinction of the various refined products can be
related to this process. The most volatile liquid product is aviation gasoline,
followed by automobile gasoline, jet fuels, diesel and heating oils, and then the
heavy oils that are used for fueling ships and some electrical generation. Most of
the molecules in gasoline have between four and ten carbons, most in diesel
have between nine and twenty, and heavy fuel oils typically have very few
molecules with less than fifteen carbon atoms except for some added as a diluent
to achieve the appropriate viscosity to facilitate distribution. As an aside it is
appropriate to note that fuels are valued based on their combustion properties,
and not chemical composition. Fuels with the same name may have very
different chemical compositions if they come from different refineries [19].
4. PHYSICAL FATE OF SPILLED OIL
When oil gets into the oceans from a seep, urban runoff or a spill from a
production facility, pipeline or a tanker, it becomes subject to a group of
phenomena that are usually grouped under the term "weathering" [20]. Almost
all oils float, allowing the smallest molecules to evaporate [21, 22]. These
500
molecules are either degraded photochemically [23] or are washed from the
atmosphere in rain and then biodegraded [24], likely far from the spill site.
Under particularly aggressive aeration in water, such as in surf, evaporation may
extend to molecules with >30 carbon atoms [25], but evaporation is more
usually limited to molecules with less than about 15 carbon atoms [21]. Thus
evaporation is the likely fate of most of a gasoline spill, three-quarters or more
of a diesel spill, and perhaps 20-40% of a typical crude oil. Heavy fuels, such as
the Bunker fuels used in ships, and bitumen emulsions (Orimulsion®, [26]) do
not contain a significant volatile fraction.
Two competing emulsification processes occur as water and oil mix;
water can become entrained in the oil to form an emulsion known as mousse
[27], or oil can disperse into the water column as a suspension of small droplets,
as happened during the 1993 Braer spill off the Shetland Islands [28]. Mousses
are remarkably persistent, and are thought to be the precursors of tarballs that
can last for decades [29]. As we shall discuss below, chemical dispersants that
break emulsions and stimulate the natural dispersion process are effective tools
in the oil spill response "toolkit".
Oil also interacts with small mineral particles in a process originally
termed "Clay-oil flocculation" [30], and now termed "Oil-Mineral Fines
Interactions" [31]. Like dispersion, this dramatically increases the surface area
of the oil.
Aliphatic hydrocarbons are remarkably insoluble, but small aromatics,
particularly the notorious BTEX (benzene, toluene, ethylbenzene and the
xylenes) and small polar molecules such as naphthenic acids dissolve from
floating slicks or dispersed oil, and even from oils immobilized on shoreline
sediments and particles [32]. Again, their eventual fate is biodegradation.
Aromatic hydrocarbons can be photochemically oxidized [33], converting
them to polar products that are probably polymerized species. The process is
most effective on the larger and more alkylated forms, and although such
hydrocarbons are only a minor component of crude oils [2], they have important
toxicological properties [34], and are on the USEPA list of priority pollutants
[35] and the EU list of priority substances in the field of water policy [36]. Since
light cannot penetrate very far into a dark oil slick, photooxidation has little
effect on the bulk properties of spilled oil. Nevertheless it may be important in
generating a polymerized "skin" that enhances the stability of tarballs and
"pavements" on beaches. Layers of immobile, hardened oil and sediment,
termed pavements, form when oil reaches a shoreline as a heavy, thick slick. Oil
becomes trapped in the sediment, and the oil and the sediment become saturated
with each other [37]. Oil incorporated into such pavements is effectively
preserved from weathering processes until this heavy, solidified material is
physically disrupted, so a major goal of spill clean-up operations is to prevent
the formation of pavements.
501
5. EVENTUAL FATE OF SPILLED OIL
The weathering processes described above distribute and change the oil in
various ways, but they do not actually remove oil from the environment. Only
two processes, combustion and biodegradation, actually eliminate oil by
converting it to carbon dioxide and water. Some spills do spontaneously ignite,
as happened to the Haven spill in the Mediterranean [38], and deliberate ignition
is an accepted response option in some situations, such as that of the New
Carissa off the coast of Oregon [39]. Under optimal conditions burning may
consume >90% of a spill, but there is usually only a small window of
opportunity for success [40].
Far more generally, it is biodegradation that removes oil from the
environment. As mentioned above, a diverse group of microorganisms has
evolved to degrade hydrocarbons, many able to live with hydrocarbons as their
sole source of carbon. They are probably ubiquitous, having been found in
almost all natural environments where they have been searched for, and
obviously very effective, since they have been consuming the vast majority of
the oil entering the world's oceans from natural seeps for millions of years
(600,000 tonnes, >600 million liters, per year). What separates these organisms
from other heterotrophs is their ability to transform hydrocarbons into organic
alcohols and acids that enter cellular metabolism. Under aerobic conditions the
most common microbial activation of hydrocarbons involves the addition of one
or both atoms of molecular oxygen. Alternatively, the activation may involve
the addition of hydrogen peroxide. The activation of aromatic hydrocarbons is
discussed by Foght in chapter 5, and many pathways are available in the
University of Minnesota Biocatalysis/Biodegradation Database [41] and in a
recent encyclopedia article [42]. Here it suffices to say that the vast majority of
hydrocarbons are biodegradable under aerobic conditions. Thus refined
products, such as gasoline, diesels and jet fuels, that are almost entirely
hydrocarbons, are essentially completely biodegradable. McMillen et al. [43]
examined the short-term biodegradability of 17 crude oils in soil microcosms,
and found that the API gravity was the most useful predictor of biodegradability.
At 0.5 wt% oil in soil with appropriate nutrients, moisture and aeration, more
than 61 % of the most degradable oil (API = 46 ) was lost in four weeks, while
only 10% of the least degradable oil (API =15 ) was consumed under the same
conditions. Further degradation occurred on a longer timescale, and the literature
reports biodegradation potentials as high as 97% for particularly light oils [44].
An important distinction between hydrocarbon-degrading microorganisms
and animals and plants is that many microbes degrade polycyclic aromatic
hydrocarbons to carbon dioxide, water and biomass. Animals and plants can also
activate these molecules, but they do so with enzyme systems that form stable
502
permutations of the polycyclic hydrocarbons (see chapter 3). The enzyme

systems are known as Cytochrome P450s because of their prominent absorption
band when treated with carbon monoxide [45]. These enzymes generate
epoxides that are excreted as adducts with sugars, anions, etc., but which may
alternatively intercalate and form adducts with DNA [46]. It is thus clearly
preferable that polycyclic aromatic hydrocarbons be degraded by bacteria rather
than eukaryotes, and facilitating such a preference is one of the advantages of a
successful bioremediation protocol.
Aerobic biodegradation of hydrocarbons occurs over a wide range of
environmental conditions [47]. Although no hyperthermophilic oil-degraders
have yet been found, extreme thermophiles such as Thermus and Bacillus
species degrade polycyclic aromatic hydrocarbons and long chain alkanes at 60-
70 °C [48]. Significant biodegradation occurs below 0°C [49] and extremely
halophilic oil-degrading organisms have been described [50] that degrade
hydrocarbons in the presence of several molar salt.
In the last decade it has become clear that hydrocarbons are also degraded
under anoxic anaerobic conditions. Small water-soluble aromatic compounds,
such as benzene and toluene, have been shown to undergo biodegradation under
sulfate-reducing, nitrate-reducing, perchlorate-reducing, ferric ion reducing,
humic acid-reducing and methanogenic conditions [51], and this phenomenon is
proving important in remediating terrestrial spills where these compounds have
reached groundwater [52]. Larger hydrocarbons, such as n-alkanes up to «C34H7o
[53] and two- three- and four-ring aromatic hydrocarbons [54] are also
biodegraded under anaerobic conditions. This may be important if oil spills
contaminate anaerobic environments, such as marshes, and in the degradation of
the traces of oil that become entrained in sediments in harbors.
Wherever oil is biodegraded, it is important to bear in mind the fact that
crude oil and refined products provide a rather unusual "food" for heterotrophs.
While hydrocarbons are excellent sources of carbon and energy, they do not
provide any of the other nutrients essential for life; there are no significant
amounts of biologically available nitrogen, phosphorus or other elements. Of
course most environments have at least trace amounts of these essential
nutrients, but most marine environments offer meager reserves to sustain new
growth. It is thus likely that any significant input of hydrocarbon is likely to
overwhelm the background levels of nutrients, and their availability soon limits
that biodegradation. As we shall see below, alleviating this limitation forms the
basis of the simplest forms of shoreline bioremediation.
503
6. SPILL RESPONSE
6.1. At sea:
When oil is spilled at sea, deployment of mechanical equipment designed
for containment and recovery is often a slow and inefficient, if not ineffective,
response. The rapid spreading of the oil, the slow rate at which mechanical
equipment (once deployed) can encounter spreading oil, and interference from
waves and currents often limits recovery effectiveness to less than 20% of the
oil spilled, significantly less under conditions of severe wind and weather [55,
56]. Unrecovered oil remains in the environment, and undergoes the weathering
processes described above, with the most severe environmental consequences
resulting when oil strands on shorelines [57, 55]. Beached oil increases the
likelihood of contamination for habitats and animals found in subtidal, intertidal
and supratidal areas, which include some of the most productive and diverse
portions of the marine environment.
Burning spilled oil in a contained and controlled manor, so as not to
jeopardize the bulk of remaining cargo or other response assets, can rapidly
remove bulk oil from the water surface. However, it is a logistical and
operational challenge to contain the oil, arrange and control its placement out of
the immediate area of spill response activity, and ensure sufficient oil thickness
to sustain an efficient burn [40]. Many of the logistical and physical constraints
working against efficient mechanical containment and recovery also confound
attempts to collect and burn oil on water. When the oil does burn, the unburned
residue is comprised mostly of the heavy, longer chain hydrocarbons, which are
relatively resistant to ready microbial degradation [58].
Dispersants are widely recognized by many regulatory agencies as an
effective at-sea response that provides a net environmental benefit when
compared to reliance on mechanical recovery alone (see chapter 9). Application
of chemical dispersants facilitates the breakup of the oil slick, moving oil from
the water surface into the water column as neutrally buoyant oil droplets ranging
from 1 to 100 micrometers in diameter, due to the mixing action of waves and
currents. Subsequently, this plume of oil droplets rapidly distributes throughout
the water column, mixing into lateral and deeper water masses and reducing oil
concentrations below levels of concern for marine life. The rate and
effectiveness of this process depends on the nature of the spilled oil (its API
gravity and viscosity, degree of weathering, extent of emulsification, and pour
point) and the ability of the dispersant formulation to mix with the oil.
Dispersants have been an effective aspect of oil spill response over the
past 30 years, with applications to major and smaller oil spill incidents in many
of the world's oceans (Fig. 3). From 1970 through 1998, dispersants have been
used on approximately 37% of oil spills covered in a worldwide database by the
Oil Spill Intelligence Reporter [59]. In addition to countless small-scale tests
504
that have been conducted in laboratories around the world, critical assessments
of dispersant performance have been organized by private and government
research organizations, often cooperatively, using controlled releases of large
volumes of oil and dispersant applications under real world conditions (Fig. 3).
These studies have led to modern dispersant formulations with improved
effectiveness and greater environmental safety. A range of dispersant products
are stockpiled around the world for spill response, and a few have been shown to
be effective over a broad range of oil types and environmental conditions [60].
An important environmental consideration associated with dispersant use
is assessing the environmental tradeoff between intentionally exposing water
column plants and animals to dispersed oil and the often significant effects of
unrecovered oil left to drift at sea to potentially strand on a shoreline. In most
cases, these considerations demonstrate a net environmental benefit to the use of
dispersants because the short-term, transient exposure of water column
communities has much less ecological effect than the prolonged, wide-spread
contamination of oil reaching shorelines [57, 55, 61].
The environmental risks of dispersed oil are further decreased by the
rapid degradation of the small, dispersed oil droplets moving through the water
column, compared to the persistence observed for bulk oil stranded on
shorelines and incorporated into sediments. The large surface to volume ratio
characteristic of micron-sized dispersed oil droplets provides a colonizing
substrate for oil degrading bacteria and a source of degradable hydrocarbon to
support their growth. And, because the small oil droplets are widely dispersed in
the water column, the supply of nitrogen and phosphorus nutrients needed to
support bacterial degradation of the diluted oil is sufficient to maintain a viable
degrader community in association with the oil droplet. Furthermore, laboratory
studies have shown that some dispersants can enhance the initial rate of oil
degradation due to the presence of constituents that serve as initial substrates for
nascent bacterial growth [62, 63].
Laboratory studies of the fate of dispersed oil droplets have characterized
the process by which it becomes a physical substrate for supporting a microbial
community as well as a chemical substrate to support their growth. Within 2 to
4 days, the dispersed oil droplet becomes colonized by oil degrading microbes
[63-65]. Subsequently, this can become a full floating heterotrophic community
consisting of oil, bacteria, protozoa and even nematodes. Macnaughton et al.
[65] reported that by day 16, the size of the clusters increased and sank in test
microcosms, most likely the result of reduced buoyancy due to oil
biodegradation and increased biomass associated with the droplets.
505
Fig. 3. Dispersant response to oil spills. Data taken from reference 59.
6.2. On shore:
If oil reaches shore then the first response is to collect it [66]. Oil typically
lands on only the upper portion of the intertidal zone, and on sandy beaches it
may be possible to collect the oiled sand with mechanical equipment. This was
done, for example, with the spill from the Sea Empress [67]. Particularly heavy
oils may be best picked up by hand, as in the case of the spill from the Prestige
[68]. On rocky beaches it may be possible to wash oil back into the sea where it
can be collected with skimmers, as was done following the spill from the Exxon
Valdez [69].
Once the bulk oil has been removed by physical techniques, residual oil is
eventually naturally biodegraded. Bioremediation aims to stimulate the rate of
natural biodegradation, without causing any additional adverse impact, by at
least partially alleviating whatever is limiting microbial growth. In most porous,
and therefore aerobic shorelines, the most likely limitation is biologically
available nitrogen and phosphorus, and effective bioremediation protocols have
applied various forms of fertilizers to deliver these nutrients.
Research on this topic has been going on for decades in many parts of the
world (Fig. 4; reviewed in 42, 44, 70-77). The simplest approach is to alleviate
the nutrient limitation of oil biodegradation by adding fertilizers. This was the
basis for the successful bioremediation of the spill from the Exxon Valdez [78-
506
80]. Two different fertilizers were used, an oleophilic fertilizer, Inipol EAP22®,
designed to adhere to oil and deliver nutrients at the oil-water interface [81] and
a slow-release granular agricultural product (Customblen®) that would release
nutrients over many weeks through the beach gravel. Inipol EAP22 is a
microemulsion with an external oil phase of oleic acid and trilaureth-4-
phosphate, containing an internal phase of urea in aqueous solution, co-
solubilized with butoxy-ethanol to adjust the viscosity. It contains 7.4% nitrogen
and 0.7% phosphorus by weight, and was applied with airless sprayers
transported on small shallow-draft catamarans. Customblen® is a high quality
agricultural fertilizer designed to release its nutrients over several weeks. It
consists primarily of ammonium nitrate, calcium phosphate and ammonium
phosphate, encapsulated in polymerized linseed oil. Customblen contains 28%
nitrogen and 3.5% phosphorus by weight, and was applied by workers walking
the beaches with broadcast spreaders. An extensive monitoring program
demonstrated that the fertilizer applications were successful at stimulating the
rate of biodegradation some two- to five-fold [78-80]
A quite similar approach was used on a limited scale following the spill
from the Sea Empress [82]. Much of this spill was treated with dispersants while
at sea, and most of the residue that landed on shore was collected by work
crews, but some oil landed on a relatively steep (gradient 10-12.5%) shingle and
pebble beach at Bulwell Bay. Because the beach was so steep, slow-release
fertilizer was placed in 1 m mesh bags, and secured to the beach with steel pegs.
Again, the rate of biodegradation was stimulated more than two-fold on the
fertilized part of the beach.
To our knowledge, these are the only two occasions when bioremediation
by the addition of relatively simple fertilizers was used following a spill, but
there have been field and laboratory tests all over the world that have found
similar results (see Figure 4). All sorts of fertilizers have been used, usually with
success, including soluble and slow release forms of inorganic and organic
nitrogen. Our most recent experiments were on Spitsbergen, the largest island of
Svalbard, Norway, (approximately 78° N, 17' E.) [83, 84]. Slow release and
soluble fertilizers were applied in much the same way they were in Alaska, and
they led to an approximate doubling of the rate of biodegradation, even in this
cold, Arctic environment.
A slightly more complex approach has been championed by Rosenberg
and colleagues [71, 85, 86]. In this case the fertilizer is an insoluble polymer of
urea and formaldehyde, and it is applied together with an oil-degrading bacterial
inoculum that can use this nitrogen source. The approach was apparently able to
stimulate the biodegradation of a small spill (100 tons) of a heavy crude oil on a
sandy beach between Haifa and Acre in Israel in the early 1990's [71, 86].
507
Fig. 4. Bioremediation response to oil spills. Data taken from references 70 - 80.
Others have suggested that what really limits oil biodegradation in the
environment is the absence of effective oil degrading microorganisms, and they
therefore seek to add such organisms. Most recently this has been attempted on
heavy oil spilled by the Nakhodka in the Sea of Japan [87, 88]. Assessing this
work is problematic. The published analyses of the field work rely on digital
photography of representative oiled rocks, and no detailed chemical analyses
have been presented that can be compared with what has been found in other
spills. Earlier microbial inocula did not perform well in standardized tests [89].
An important corollary to any oil spill remediation is that it should have a
net environmental benefit [90]. By aiming to stimulate natural processes,
bioremediation is likely to have minimal adverse effects if carried out carefully
and conscientiously, but there are obvious potential risks that must be evaluated.
Potential adverse impacts include the possibility that the fertilizer applications
might be acutely toxic to marine biota, might stimulate nearshore algal blooms,
might cause the production of biosurfactants that could result in increased
removal of oil from the shorelines by tidal flushing and lead to broader shoreline
impacts, or might generate toxic by-products. Careful monitoring following the
spill from the Exxon Valdez [91] and a field trial in the Arctic [92] failed to
detect any adverse environmental impact from the careful application of
fertilizers, while the rate of hydrocarbon biodegradation was stimulated two- to
five-fold.
508
7. CONCLUSIONS
Oil spill bioremediation technologies epitomize modern environmental

techniques: working with natural processes to remove spilled oil from the
environment while minimizing undesirable environmental impacts. If a floating
oil slick cannot be collected or burnt, chemical dispersants will cause the oil to
move into the water column as tiny droplets with a dramatically increased
surface area that allows rapid biodegradation. If oil reaches a shoreline and
cannot be removed physically, the careful addition of fertilizers will stimulate
oil biodegradation without adverse environmental impact. These two tools are
thus an important part of the toolkit for dealing with accidental and deliberate
releases of oil into the marine environment.
REFERENCES
[I] J.M. Hunt, Petroleum Geochemistry and Geology, 2nd edition. W.H. Freeman, New
York. 1996.
[2] B.P Tissot and D.H. Welte, Petroleum Formation and Occurrence. Springer-Verlag,
Berlin. 1984.
[3] J.Z. De Boer, J.R. Hale, and J. Chanton, New evidence of the geological origins of the
ancient Delphic oracle (Greece). Geology, 29 (2001) 707.
[4] D. Yergin, The Prize, the epic quest for oil, money and power. Simon and Schuster,
New York. 1992.
[5] National Research Council, Oil in the Sea III: Inputs, Fates and Effects, National
Academy Press, Washington DC, 2002.
[6] E.M Levy and K. Lee, Can. J. Fish. Aquat. Sci. 45 (1988) 349.
[7] R.C. Prince, Petroleum and other hydrocarbons, biodegradation of. hi Encyclopedia of
Environmental Microbiology (G. Bitton, ed.) John Wiley, New York, 2002 pp. 2402-
2416.
[8] Anonymous. Industry at a glance. World Oil 224 (2003) 75.
[9] D.S. Etkin, Historical overview of oil spills from all sources. In: Proceedings of the
1999 International Oil Spill Conference. American Petroleum Institute, Washington
DC. 1999 pp. 1097-1102.
[10] H.J. Barth, Mar. Poll. Bull. 46 (2003) 1245.
[II] A.N. AlGhadban, F. Abdali, and M.S. Massoud, Environ. International 24 (1998) 23.
[12] K. Peters and J.M. Moldowan, The Biomarker Guide; Interpreting molecular fossils in
petroleum and ancient sediments. Prentice-Hall, Englewood Cliffs, NJ. 1993.
[13] R.C. Prince, D.L. Elmendorf, J.R. Lute, C.S. Hsu, C.E. Haith, J.D. Senius, G.J. Dechert,
G.S. Douglas, and E.L. Butler, Environ. Sci. Technol. 28 (1994) 142.
[14] W.K. Robbins and C.S. Hsu, Petroleum Composition. In: Kirk-Othmer Encyclopedia of
Chemical Technology, Fourth Edition. John Wiley & Sons, New York. 1996 pp. 352-
370.
[15] Y.A. Ibrahim, M.A. Abdelhameed, T.A. Al-Sahhaf, and M.A. Fahim, Petrol. Sci.
Technol. 21 (2003) 825.
[16] J Burdon, Soil Science, 166 (2001) 752.
509
[17] K. Qian, R.P. Rodgers, C.L. Hendrickson, M.R. Emmett and A.G. Marshall, Energy &
Fuels, 15(2001)492.
[18] K. Qian, W.K. Robbins, C.A. Hughey, HJ. Cooper, R.P. Rodgers and A.G. Marshall,
Energy & Fuels, 15 (2001) 1505.
[19] A.D. Uhler, S.A. Stout, K.J. McCarthy, S. Emsbo-Mattingly, G.S. Douglas and P.W.
Beall, Contaminated Soil Sediment and Water, April/May (2002) 20.
[20] R.C. Prince, R.M. Garrett, R.E. Bare, M.J. Grossman, T. Townsend, J.M. Suflita, K.
Lee, E.H. Owens, G.A. Sergy, J.F. Braddock, J. E Lindstrom. and R.R. Lessard, Spill
Sci. Technol. Bull, 8 (2003) 145.
[21] M.F. Fingas, The evaporation of oil spills: development and implementation of new
prediction methodology. In Proceedings of the 1999 International Oil Spill Conference,
American Petroleum Institute, Washington DC, 1999 pp. 185-194.
[22] Y. Wang and Y. Huang, Appl. Geochem. 18 (2003) 1641.
[23] M.D. Hurley, O. Sokolov, T.J. Wallington, H. Takekawa, M. Karasawa, B. Klotz, I.
Barnes, and K.H. Becker, Environ. Sci. Technol. 35 (2001) 1358.
[24] K.M. Arzayus, R.M. Dickhut, and E.A. Canuel, Environ. Sci. Technol. 35 (2001) 2178.
[25] R.C. Prince, R.T. Stibrany, J. Hardenstine, G.S. Douglas and E.H. Owens, Environ. Sci.
Technol. 36 (2002) 2822.
[26] Bitor Corporation (2004) www.orinutlsionfuel.com/
[27] M.F. Fingas, B. Fieldhouse, J. Lane and J.V. Mullin, What causes the formation of
water-in-oil emulsions? In Proceedings of the 2001 International Oil Spill Conference,
American Petroleum Institute, Washington DC, 2001 pp. 109-114.
[28] R. Thomas and T. Lunel, The Braer Incident; dispersion in action. In Proceedings of the
Sixteenth Arctic Marine Oilspill Program Technical Seminar, Edmonton, Alberta, 1993
pp. 843-859.
[29] R. Goodman, Spill Sci. Technol. Bull. 8 (2003) 117.
[30] J.R. Bragg and E.H. Owens. Clay-oil flocculation as a natural cleansing process
following oil spills: Part 1-studies of shoreline sediments and residues from past spills.
In: Proceedings of the Seventeenth Arctic and Marine Oilspill Program (AMOP)
Technical Seminar. Ottawa, Canada. 1994 pp. 1-24.
[31] E.H. Owens and K. Lee, Mar. Poll. Bull. 47 (2003) 397.
[32] E. Lafargue and P. Le Thiez, Org. Geochem. 24 (1996) 1141.
[33] R.M. Garrett, I.J. Pickering, C.E. Haith, and R.C. Prince, Environ. Sci. Technol. 32
(1998)3719.
[34] International Agency for Research on Cancer. IARC Monograph on the evaluation of
the carcinogenic risk of chemicals to humans. Polynuclear aromatic hydrocarbons, Part
1, Chemical, environmental and experimental data, Vol. 32. World Health Organization,
Geneva. 1983
[35] L.H. Keith and W.A. Telliard, Environ. Sci. Technol. 13 (1979) 416.
[36] European Commission, DECISION No 2455/2001/EC OF THE EUROPEAN
PARLIAMENT AND OF THE COUNCIL of 20 November 2001 establishing the list of
priority substances in the field of water policy and amending Directive 2000/60/EC,
Official Journal of the European Union L331, 1-5, 15 December 2001.
[37] E.H. Owens, J.R. Harper, W. Robson and P.D. Boehm, Arctic 40 (1987) 109.
[38] M. Martinelli, A. Luise, E. Tromellini, T.C. Sauer, J.M. Neff and G.S. Douglas, The
M/C Haven oil spill: Environmental assessment of exposure pathways and resource
injury. Proceedings of the 1995 Oil Spill Conference, American Petroleum Institute,
Washington, D.C. 1995 pp. 679-685.
510
[39] J.J. Gallagher, H.B. Hile and J.A. Miller, The old New Carissa; a study in patience.
Proceedings of the 2001 Oil Spill Conference, American Petroleum Institute,
Washington, D.C. 2001 pp. 85-90.
[40] I. Buist, Spill Sci. Technol. Bull. 8 (1993) 341.
[41 ] The University of Minnesota Biocatalysis/Biodegradation Database (2004)
http://umbbd.ahc.umn.edu/
[42] R.C. Prince, Crude oil biodegradation In The Encyclopedia of Environmental Analysis
and Remediation, Volume 2, 1327-1342. John Wiley, New York. 1998.
[43] S.J. McMillen, A.G. Requejo, G.N. Young, P.S. Davis, P.D. Cook, J.M. Kerr, and N.R.
Gray, Bioremediation potential of crude oil spilled on soil. In: Microbial Processes for
Bioremediation (R.E. Hinchee, CM. Vogel, and FJ. Brockman, eds.). Battelle Press,
OH. 1995 pp, 91-99.
[44] R.C. Prince, Crit. Rev. Microbiol. 19 (1993) 217.
[45] C.J. Omiecinski, R.P. Remmel and V.P.. Hosagrahara, Toxicol. Sci. 48 (1999) 151.
[46] V.J. Melendez-Colon, A. Luch, A. Seidel, and W.M. Baird, Cancer Res. 59 (1999)1412.
[47] R. Margesin and F. Schinner, Appl. Microbiol. Biotechnol. 56 (2001) 650.
[48] H. Feitkenhauer, R. Muller and H. Markl, Biodegradation 14 (2003) 367.
[49] A.G. Rike, K.B. Haugen, M. Barresen, B. Engene. and P. Kolstad, Cold Regions Sci.
Technol. 37 (2003) 97.
[50] M.J. Gauthier, B. Lafay, R. Christen, L. Fernandez, M. Acquaviva, P., Bonin and J.C.
Bertrand, Int. J. System. Bacteriol. 42 (1992) 568.
[51] A.M. Spormann and F. Widdel, Biodegradation 11 (2000) 85.
[52] J.A. Cunningham, H. Rahme, G.D. Hopkins, C. Lebron, and M. Reinhard, Environ. Sci.
Technol. 35 (2000) 1663.
[53] M.E. Caldwell, R.M. Garrett, R.C. Prince and J.M. Suflita, Environ. Sci. Technol. 32
(1998)2191.
[54] J.D. Coates, J. Woodward, J. Allen, P. Philp and D.R. Lovley, Appl. Environ.
Microbiol. 63(1997)3589.
[55] A. Lewis and D. Aurand, Putting Dispersants to Work: Overcoming Obstacles. API
Technical Report IOSC-004. 1997 International Oil Spill Conference Issue Paper. API.
Washington, DC. 1997.
[56] D.S. Etkin and P. Tebeau, Assessing progress and benefits of oil spill response
technology development since Exxon Valdez. Proceedings of the 2003 International Oil
Spill Conference, American Petroleum Institute, Washington, DC, 2003 pp. 843-850.
[57] National Research Council, Using Oil Dispersants on the Sea. National Academy
Press. Washington, DC. 1989.
[58] R.M. Garrett, C.C. Guenette, C.E. Haith and R.C. Prince, Environ. Sci. Technol. 34
(2000) 1934.
[59] D.S. Etkin, Factors in the Dispersant use decision-making process: historical overview
and look to the future, Proceedings of the twenty-first Arctic and Marine Oil Spill
Program (AMOP) Technical Seminar. Environment Canada. Ottawa. 1998 pp. 281-304
[60] R.R. Lessard and G. DeMarco. Spill Sci. Technol. Bull. 6 (2000) 59.
[61] International Petroleum Industry Environmental Conservation Association (IPIECA),
Oil Spill Preparedness and Response. Volume 5: Dispersants and their Role in Oil Spill
Response. London. 2000.
[62] R. Varadaraj, M.L. Robbins, J. Bock, S. Pace and D. MacDonald, Dispersion and
biodegradation of oil spills on Water. In: Proceedings of the 1995 International Oil
Spill Conference, American Petroleum Institute, Washington, DC. 1995 pp. 101-106.
511
[63] R. Swannell and F. Daniel, Effect of dispersants on oil biodegradation under simulated
marine conditions. Proceedings of the 1999 International Oil Spill Conference,
American Petroleum Institute, Washington, DC, 1999, pp. 169-176.
[64] J.E. Lindstrom and J.F. Braddock, Mar. Poll. Bull. 44 (2002) 739.
[65] S.J. Macnaughton, R. Swannell, F. Daniel and L. Bristow, Spill Sci. Technol. Bull. 8
(2003) 179.
[66] B.E. Ornitz and M.A. Champ, Oil Spills First Principles. Elsevier, New York. 2002.
[67] T. Lunel, K. Lee, R. Swannell, P. Wood, J. Rusin, N. Bailey, C. Halliwell, L. Davies,
M. Sommerville, A. Dobie, D. Mitchell and M. McDonagh, Shoreline clean up during
the Sea Empress incident: the role of surf washing (clay oil flocculation), dispersants
and bioremediation. In Proceedings of the Nineteenth Arctic and Marine Oil Spill
Program Seminar. Environment Canada, 1996 pp. 1521-1540.
[68] J. Bohannon, X. Bosch and J. Withgott, Science 298 (2002) 1695.
[69] S.A. Nauman, Shoreline clean-up: equipment and operations. In Proceedings of the
1991 International Oil Spill Conference, American Petroleum Institute, Washington
DC, 1991 pp. 141-148.
[70] R.M. Atlas, Crit. Rev. Microbiol. 5 (1977) 371.
[71] E. Rosenberg, R. Legman, A. Kushmaro, R. Taube, E. Adler and E.Z. Ron,
Biodegradation 3 (1992) 337.
[72] R.P.J. Swannell, K. Lee and M, McDonagh, Microbiol. Rev. 60 (1996) 342.
[73] Q. Lin, LA. Mendelssohn, C.B. Henry, P.O. Roberts, M.M. Walsh, E.B. Overton and
R.J. Portier, Environ. Tech. 20 (1999) 825.
[74] M. Mathew, J.P. Obbard, Y.P. Ting, Y.H. Gin and H.M. Tan, Acta Biotechnol. 19
(1999)225.
[75] K. Lee and S. deMora, Environ. Tech. 20 (1999) 783.
[76] I. Head and R.P.J. Swannell, Curr. Opin. Biotechnol. 10 (1999) 234.
[77] S. Harayama, H. Kishira, Y. Kasai and K. Shutsubo, J Mol Microbiol Biotechnol. 1
(1999) 63.
[78] J.R. Bragg, R.C. Prince, E.J. Harner, and R.M. Atlas, Nature 368 (1994) 413.
[79] R.C. Prince, J.R. Clark, J.E Lindstrom, E.L. Butler, E.J. Brown, G. Winter, MJ.
Grossman, R.R. Parrish, R.E. Bare, J.F. Braddock, W.G. Steinhauer, G.S. Douglas, J.M.
Kennedy, P.J. Barter, J.R. Bragg, E.J. Harner and R.M. Atlas, Bioremediation of the
Exxon Valdez oil spill: monitoring safety and efficacy, hi Hydrocarbon Remediation(R.
E. Hinchee, B. C. Alleman, R. E. Hoeppel and R. N. Miller, eds.) Lewis Publishers,
Boca Raton, FL., 1994 pp. 107-124.
[80] R.C. Prince and J.R. Bragg, Bioremediation J. 1 (1997) 97.
[81] A. Ladousse and B. Tramier, Results of 12 years of research in spilled oil
bioremediation: Inipol EAP22. In: Proceedings of the 1991 International Oil Spill
Conference, American Petroleum Institute, Washington DC, 1991 pp. 577-582.
[82] R.P.J. Swannell, D. Mitchell, G. Lethbridge, D. Jones, D. Heath, M. Hagley, M. Jones,
S. Petch, R. Milne, R. Croxford and K. Lee, Environmental Technol. 20 (1999) 863.
[83] C.C. Guenette, G.A. Sergy, E.H. Owens, R.C. Prince and K. Lee, Spill Sci. Technol.
Bull. 8 (2003), 245.
[84] R.C. Prince, R.E. Bare, R.M. Garrett, M.J. Grossman, C.E. Haith, L.G. Keim, K Lee,
G.J. Holtom, P. Lambert, G.A. Sergy, E.H. Owens and C.C. Guenette, Spill Sci.
Technol. Bull. 8 (2003) 303.
[85] E. Rosenberg, R. Legman, A. Kushmaro, E. Adler, H. Abir and E.Z. Ron, J Biotechnol.
51 (1996)273.
512
[86] E. Rosenberg and E.Z. Ron, Non-polluting compositions to degrade hydrocarbons and
microorganisms for use thereof. US Patent 5780290 1998
[87] T. Hozumi, H. Tsutsumi, and M. Kono, Mar. Poll. Bull. 40 (2000) 308.
[88] H. Tsutsumi, M. Kono, K. Takai, T. Manabe, M. Haraguchi, I. Yamamoto, and C.
Oppenheimer, Mar. Poll. Bull. 40 (2000) 320.
[89] A.D. Venosa, J.R. Haines and D.M. Allen, J. Ind. Microbiol. 10 (1992) 1.
[90] J.M. Baker, Net environmental Benefit Analysis for oil spill response. In Proceedings of
the 1995 International Oil Spill Conference, American Petroleum Institute, Washington
DC, 1995 pp. 611-614.
[91] R.C. Prince, J.R. Clark and K. Lee, Bioremediation effectiveness; removing
hydrocarbons while minimizing environmental impact. In Proceedings of the Ninth
International Petroleum Environmental Conference, Albuquerque, NM. 2002,available
at: http://ipec.utulsa.edu/Ipec/Conf2002/prince dark lee 109.pdf
[92] K. Lee, G. Wohlgeschaffen, G.H. Tremblay, B.T. Johnson, G.A. Sergy, R.C. Prince,
C.C. Guenette and E.H. Owens, Spill Sci. Technol. Bull, 8 (2003) 273.
Chapter 19
Biotreatment of water pollutants from the petroleum

industry
E. Razo-Flores,a>b P. Olguin-Lora," S. Alcantara3 and M. Morales-Ibarria"
a
Instituto Mexicano del Petroleo, Programa de Biotecnologia. Eje Central
Lazaro Cardenas 152, C.P. 07730, Mexico D.F.
b
Instituto Potosino de Investigation Cientifica y Tecnologica,. Camino a la
Presa San Jose 2055,. C.P. 78216, San Luis Potosi, SLP, Mexico.
1. INTRODUCTION
Industrialization has resulted in the formation of waste products, which are

released into the environment in the form of wastewater, gaseous emissions and
solid residues leading to environmental pollution and deterioration. A good
example of this situation is the petroleum industry (oil and gas, chemical and
petrochemical). During decades, the production strategy aimed to maximize
product outputs with minimum production costs. Therefore, a large water usage,
soil contamination and energy wastage (oil by-products being lost into the
environment) were a normal practice.
Later, through the 1970s to 1980s, "end of the pipe" solutions were
developed to control pollution. This approach was effective but it can not be
affordable for much longer because the production and environmental protection
costs are added together, rising global costs and wasting a great amount of
energy and material resources (water, nutrients, metals, oil). A system approach
that integrates human activities with the protection and restoration of the
environment goes together with the sustainability concept in a world closely
linked through communications and markets.
More recently, during the last two decades, governmental regulatory
actions changed profoundly the wastewater treatment in the petroleum industry,
establishing effluent limitations for many specific organic and inorganic
compounds. Nowadays, water can be considered as one of the main raw
materials of the petroleum industry and its treatement and reuse with advanced
treatment technology is being developed.
514
1.1. Characterizing petroleum industry wastewater

Petroleum industry requires large water volumes for the oil and gas
refining and processing. PEMEX, the state owned Mexican Oil Company,
consumed 270.2 and 245.1 millions of cubic meters of water in the years 2001
and 2002, respectively, in its different processes. Besides, in 2002 the water
input per unit of throughput was 0.17, 0.86, 1.73 and 10.8 m3 ton"1 for
exploration and production, gas processing, refining, and petrochemical
operations, respectively. At national level, the petroleum and chemical industry
occupies the second place in industrial wastewater generation, both in volume
and organic load, after sugarcane industry. Table 1 presents a general list of the
main water pollutants in the petroleum industry. For a full review of the refinery
and petrochemical effluents and the common treatments used (physicochemical
and biological) [1-2].
1.2. Biological reactions applied to petroleum wastewater treatment

Biological processes are a cost-effective technology for the removal of
organic, sulfur and nitrogen compounds from wastewaters. Table 2 shows the
main transformations that can occur during biological petroleum wastewater
treatment.
Anaerobic processes are one of the most viable alternatives for the
treatment of complex effluents like those produced in the petroleum industry.
Up to date, methanogenic, sulfate-reduction and anoxic processes such as
heterotrophic denitrification, have been used for the biodegradation of organic
compounds. Most of these anaerobic processes were developed for the food
industry, but recently have been successfully applied in the chemical and
petrochemical industry wastewaters [3].
The most accepted high-rate process is that carried out in up-flow
anaerobic sludge blanket reactor (UASB), where the hydraulic and biomass
residence times are uncoupled, allowing a high biomass concentration inside it.
The granule formation and stability are essential for the right operation of the
UASB reactor.
The methanogenic treatment of organic compounds (e.g. phenols, organic
acids, etc.) is a complex microbial process involving many kinds of bacteria and
several intermediate steps. Generally, the first step is the hydrolysis of the
organic compounds producing simpler organics after which, they are fermented
to volatile fatty acids by the acidogens. Furthermore, the acetogenic bacteria
transform these compounds to acetate and hydrogen that are finally converted to
biogas (methane and CO2) by the methanogens [4].
515
Table 1
Main water-soluble contaminants generated by the petroleum
industry.
Family Compounds
Aromatic hydrocarbons Benzene
Toluene
Ethylbenzene
Xylene
Oxygenated compounds Phenols
Organic acids
Aldehydes
Metyl tert-butyl ether
Sulfur compounds Hydrogen sulfide
Mercaptans
Nitrogen compounds Ammonium
Amines
Urea
Sulfur bearing wastewaters can be treated by using the biological reactions

of the sulfur cycle. In the reductive side, both sulfate (SO42~) and elemental
sulfur (S°) act as electron acceptors in the metabolism of a wide range of
anaerobic bacteria, producing H2S. In the oxidative side of the cycle, sulfur
reduced compounds are biologically oxidized to sulfate or elemental sulfur
under either aerobic or anaerobic conditions by autotrophic bacteria.
Table 2
Main transformations carried out during biological petroleum wastewater treatment.
Biological Process (electron acceptor) Products

Methanogenesis (CO 2 , HCO3")
Organic compounds —> CH4 + CO2 + Biomass
Sulfate-Reduction (SO 4 , S°)
Organic compounds —> H2S + CO2 + Biomass
Metals -» MeS
Heterotrophic denitrification (NO3")
Organic compounds —> N2 + CO2 + Biomass
Autotrophic denitrification (NO3")
H2S N 2 + SO 4 (S°) + Biomass
Nitrification (O2)
NH 4 + -> NO3" + CO 2 + Biomass
Aerobiosis (O2)
H2S -> SO 4 (S°) + Biomass
Organic compounds -» H2O + CO2 + Biomass
516
Nitrogen, mainly as ammonium (NH4+), is one of the most abundant

contaminants in the petroleum industry wastewaters. Ammonium can be
biologically eliminated by means of a double process, nitrification and
denitrification, producing molecular nitrogen. Nitrification is a strict aerobic
process, litoautotrophic, where ammonium is the electron and nitrogen sources,
and it is oxidized to nitrate. Denitrification is a reductive process either
heterotrophic or litoautotrophic process, where nitrate is reduced to elemental
nitrogen.
In this chapter it will be presented some of the more recent developments
in biological wastewater treatment technology with application to the petroleum
industry. The topics that will be covered are:
a) Anaerobic biodegradation of aromatic compounds like phenol,
alkylphenols and terephthalate.
b) Biotransformation of S- and N-bearing inorganic compounds.
c) Methyl-tert-butyl ether (MTBE) biodegradation. MTBE is a high
recalcitrant compound and a potential water contaminant that only in few cases
can be treated with technology originally developed for biological wastewater
treatment. Conventional biological treatment, like activated sludge, is out of the
scope of this chapter.
2. ANAEROBIC BIODEGRADATION AND BIOTRANSFORMATION

OF AROMATIC COMPOUNDS
The implementation of anaerobic wastewater treatment in the petroleum

industry was initially limited due to the presumed toxicity and biodegradability
of aromatic compounds present in these waste streams. However, the treatment
of chemical and petrochemical wastewater has lately become a reality, due to a
better understanding of the microbial biodegradation process and the discovery
of the methanogenic granular sludge structure, which plays a key role in the
development of the so called high rate anaerobic processes.
The granular sludge is an aggregation of several metabolic groups of
bacteria living in synergism. The granules have a diameter between 0.5 to 3 mm
and a biomass concentration of approximately 100 g dry matter I'1.
2.1. Toxicity and biodegradability of phenolic compounds

Spent caustic is one of the refineries waste streams, rich in phenolic
compounds. This effluent is produced from nonregenerative desulfurization
processes that use caustic soda scrubbing in combination with air oxidation. This
process is used to remove H2S and CH3SH from gasoline and to remove H2S,
CO2 and HCN from sour condensate gas [2]. The effluent, although involves
very small volumes and its contain high concentration of phenols and sulfide.
517
The average phenol and alkyl phenols concentrations are 30.5 g 1" and 28.2 g 1' ,
respectively.
There are several reports about the toxic effect produced by phenolic
compounds on the acetoclastic methanogenic activity (AMA) of the granular
sludge. Table 3 shows the inhibitory concentrations that reduce in 50% (IC50)
the AMA. In general, the susceptibility of a granular sludge to the inhibitory
effect of phenolic compounds is affected by the impact of its "acclimation
history". The phenol-degrading acid-forming bacteria are more susceptible to
phenol inhibition than the methanogens [5]. Most granules have a layered
structure that protects bacteria, particularly methanogens. In the case of a
phenol-acclimated granular sludge, it is possible that a phenol-degraders layer
develops in the external zone of the granules, preventing the inward diffusion of
the toxic compounds. This outer layer can prevent the methanogens deactivation
either by reducing the exposure level or by a partial or complete
biotransformation into nontoxic intermediates such as volatile fatty acids [5].
The selection and multiplication of an acetoclastic flora more resistant to those
toxic compounds might be another protection mechanism.
The inhibitory mechanism of the phenolic compounds is governed by their
hydrophobicity that increases the ability of these compounds to solubilize into
the lipid bacterial membranes, altering the membrane functions, such as ion
transports causing cellular lysis. High linear correlations of methanogenic
toxicity data to the logarithm of octanol-water partition coefficients of phenolic
compounds (log P) have been proposed as shown in Fig. 1. This simple model
adequately estimates the IC50 values for anaerobic granular sludge in the
presence of phenolic compounds.
Table 3
Inhibitory concentrations that reduce in 50%
(IC50) the acetoclastic methanogenic activity
of granular sludge (phenol-acclimated and
non-acclimated) in batch assays [6-8].
IC50
Compound
(mg I"1)
Phenol 470 - 7802
o-cresol 433 - 844
m-cresol 443 - 919
p-cresol 389-1525
3,4-dimethylphenol 329-378
2-ethylphenol 195-207
4-methylphenol 657
4-ethylphenol 289
518
Fig. 1. Relationship between IC50 of phenolic compounds and the octanol/water partition
coefficient (Log P). Synthetic "spent-caustic phenols mixture" (X), data from reference [6]
(•), data from reference [7] (•) and data from reference [8]. (A). (1), phenol; (2), 4-methyl-
phenol; (3), 4-ethylphenol; (4), o-cresol; (5), m-cresol; (6), /)-cresol; (7), 3,4-dimethylphenol;
(8), 2-ethylphenol; (9), "synthetic phenols mixture".
Log (l/ICso) = 0.77 Log P - 2.28, r2 = 0.90
Phenol is a compound easily biodegradable under anaerobic conditions.

The biodegradation is initiated by phosphorylation of hydroxyl group followed
by carboxylation of the ring in the para position (benzoyl-CoA pathway). In the
case of the three cresol isomers (p-, m- and o-cresol) there are differences in
their anaerobic biodegradability pathways. Methanogenic consortia are capable
of /?-carboxylate the m-cresol and the o-cresol to their methylbenzoic acids.
After carboxylation, the main degradation mechanism is the oxidation of the
methyl group and in case of the /?-cresol it leads to the formation of a metabolic
intermediate, the p-hydroxybenzaldehyde. It has been reported that p-cresol
degradation also initiates by fumarate addition to the methyl group, forming
benzyl-succinate.
There are few reports about o-cresol biodegradation, since this compound
is considered hard to be degraded under anaerobic conditions. Biodegradation
rates of a mixture of phenol, p- and o-cresol obtained in batch experiments, with
an adapted granular sludge, were approximately two-orders of magnitude higher
than those observed with non-adapted sludge [9].
From evaluating the interaction of substrates, it was observed that p- and
o-cresol did not affect phenol biodegradation, however, both phenol and o-cresol
negatively affected /?-cresol biodegradation at the concentrations tested [9]. In
other study, degradation of />-cresol ceased when phenol was depleted. This
suggests that degradation of the most refractory p-cresol also requires phenol as
a co-substrate. However, after a period of acclimation to the phenol-free
environment, the biomass was able to degrade p-cresol without any co-substrate
[10]. So far, both xylenols and ethylphenol biodegradation has not been reported
519
under methanogenic conditions. A reversible reaction from 2-ethylphenol to 3-

hydroxy-4-ethylphenol seems to take place, but no further degradation has been
described.
2.1.2. Anaerobic treatment systems for the biodegradation of phenol

Anaerobic treatment of phenolic-bearing wastewaters produced from the
petroleum industry is a viable option. The bioreactor system most commonly
used for the anaerobic treatment of phenolics is the UASB, operating to a certain
organic loading rate (OLR), usually referred to chemical oxygen demand
(COD).
Lab scale UASB reactors have been applied to treat single phenolic
compounds at OLR as high as 6 and 7.2 g COD I"1 d"1 for phenol and />-cresol,
respectively, showing high compound removal efficiencies [11, 12]. However,
effluents from the petroleum industry are expected to contain mixtures of phenol
and cresols as the main COD bearing fractions. Thus, a successful treatment of
these effluents would require a simultaneous degradation of the major phenolic
substrates. Table 4 shows some results of anaerobic treatment of phenolic
compounds mixtures.
Table 4.
Continuous anaerobic treatment results of mixtures of phenolic
compounds treated in upflow anaerobic sludge bed reactors.
OLR
Mixture COD removal (%) Reference
(g COD I ' d 1 )
Phenol
7 94 [9]
p-Cresol
Phenol 7.1 91 [9]

/?-Cresol
Phenol
2.95 81.8 [9]
jt?-Cresol
o-Cresol
Phenol 8.12 85 [10]

p-Cresol
Phenol
0.66 85 [13]
/>-Cresol
o-Cresol
Phenol 4.3 - [15]

m-Cresol
520
The operational parameters of the UASB reactors have important

implications on the biodegradation efficiency of the phenolic compounds. In
general, the effect of OLR is more drastic in reactors with increased phenolic
concentrations, than in reactors with a constant phenolic concentration, and a
decreased hydraulic retention time [10]. In the same way, the phenol/cresols
ratio has to be controlled to avoid inhibitory or toxic effects to the living
biomass. Cresol concentrations higher than 600 mg 1" can cause severe
inhibition on the activity of the granular sludge [9, 10].
In a typical reactor wih phenol is used as sole source of carbon and
energy, granulation is reported to initiate after 3 months of the start-up operation
and develops for 6 months, to become fully mature. Granular sludge cultivated
has an average diameter of 1.8 mm and is highly settable with a settlement
volumetric index (SVI) of 14 ml g"1 [5].
The removal of phenolic mixtures can be improved in an UASB reactor
using bioaugmentation. This method not only improves the start-up time, but
also the COD removal. The bioaugmentation can be performed by simple
adsorption of the specific bacterial consortium onto the granules, to protect it
from being washed-out [13]. An increase of the enrichment from 2 to 5%
improved considerably the start-up of the reactors treating phenolic compounds
[14].
It was until 1981 that the two first full scale reactors treating chemical
wastes were built by Celanese Company in USA. A third reactor was built three
years later, and by 1989, 19 full-scale reactors were in operation treating
wastewater from the chemical and petrochemical industry. Since 1990, the rate
of digesters construction for that industrial sector increased from 2.1 to 4.6
reactors per year.
Although an UASB reactor has been in operation since 1986 to treat
phenol-bearing wastewater, no other reactor has been built to treat the same type
of effluents since then. This UASB reactor of 1280 m3 is treating a 30.5 g COD
I"1 with an OLR between 9 to 12 g COD l"'d"' and a COD removal of 95% has
been achieved [3].
2. 2. Toxicity and biodegradability of terephthalic acid

Phthalic acid isomers (benzene-dicarboxylic acid) are important
constituents of polyester fibers, films, polyethylene terephthalate (PET) bottles
and other plastics. During production of phthalic acids, an important volume of
wastewater is generated, approximately 3-10 m3 per ton of purified terephthalic
acid (PTA) containing 5-20 kg COD m"3 [16]. The main components in the
wastewater are terephthalic acid, acetic acid, benzoic acid and p-toluic acid in
decreasing order of concentration. After neutralization with NaOH, all acids are
present as sodium salts. Due to the characteristics of these wastes, anaerobic
521
pretreatment has been generally recognized as beneficial for wastewater

treatment.
2.2.1. Toxicity and biodegradation

Terephthalate concentration of 5 g COD I"1 does not produce any substrate
inhibition on its biodegradation and methanogenic activity [17].
Hydrogenotrophic methanogenesis inhibition by 4-carboxybenzaldehyde, p-
toluate and terephthalate generates IC50 values of 0.8 g I"1, 4.6 g I"1 and 16.6 g I"1,
respectively. Nonetheless, methane production can be inhibited by un-ionized
terephthalic acid, in near colloid state, using a settled terephthalic acid
wastewater (pH 4.5) or purified terephthalic acid (0.183 g g'VSS) adjusted to
pH6.15 [18].
The very low specific growth rate of terephthalate-biodegrading bacteria
(0.04 h"1) explains the long-lasting acclimation period and low loading rate
applied in UASB reactors [17]. The use of co-substrates like sucrose, benzoic
and acetic acids inhibits the terephthalate and p-toluate biodegradation. The
addition of benzoate delays the terephthalate biodegradation, which resembles a
diauxic inhibition [17].
The generally accepted metabolic pathway of terephthalate biodegradation
is the benzoyl CoA pathway after a probable decarboxylation leading to the
formation of benzoate. The decarboxylation step is thermodynamically
favorable under standard conditions, while the conversion of benzoate to acetate
is a highly endergonic process. The global conversion of terephthalate to acetate
and H2 becomes exergonic only when acetate and H2 are at very low
concentrations. The fermentation of co-substrate by methanogenic granular
sludge results usually in the production of H2 and acetate, generating an increase
in the AG 0 ' which, in turn, may limit the terephthalate biodegradation.
Analysis of specific activities of terephthalate and benzoate
biodegradation demonstrated that terephthalate biodegradation activity was
lower with a 33.6 mg COD g"'VSS d"1 value versus 117 mg COD g"1 VSS d'1
value for benzoate activity. Thus, the initial conversion of terephthalate to
benzoate seems to be the limiting step of the microorganisms involved in
terephthalate anaerobic biodegradation. Three bacterial populations were
involved: 1) a syntrophic organism similar to that described for Syntrophus
buswellii [19] able to convert terephthalate into acetate, CO2 and H2; 2) an
acetoclastic methanogen; and 3) a hydrogenotrophic methanogen.
2.2.2. Terephthalate anaerobic treatment at full scale

Crude terephthalic acid wastewaters must fulfill some conditions to be
successfully pre-treated with anaerobic process: Certain grade of effluent
neutralization, limiting concentration of other substrates than terephthalate or
benzoate or acetate, high biomass retention rate and low volumetric loading rate.
522
The best operation is obtained with a plug-flow process or staged reactor

system, because no substrate toxicity has been reported in normal operation with
neutralized effluent.
The company Amoco Petrochemicals Inc. operates a 15200 m3 full-scale
downflow fixed film reactor with an OLR of 4.0 kg COD m3 d"1, which
demonstrated the real feasibility of such pre-treatment [20]. The use of an
expanded granular sludge bed-type bioreactor allowed a terephthalate removal
higher than 80% and steady COD removal of 60% at an upflow velocity of 10 m
d"1; however, in such conditions, p-toluate appeared to be recalcitrant to
degradation [21].
Up to date, there are more than 10 full scale reactors treating terephthalic
acid, indicating that the anaerobic treatment has become a conventional
treatment for this kind of wastewater. The used bioreactor configurations are
UASB, expanded granular sludge bed and hybrid reactors [22].
3. BIOTRANSFORMATION OF S- AND N-BEARING INORGANIC

COMPOUNDS FROM SOUR STREAMS
The microbial treatment of sour wastewater resulting from either oil production
or refining and other fossil fuels has been subject of intensive worldwide
studies. The term "sour" was originated to describe those wastes contaminated
with sulfide [23]. In refineries, sour wastewaters are generated from sour steam
condensates that have been in contact with petroleum products, specifically from
thermal or hydrogen cracking operations, where a carrier steam is used for
injection or aeration [24]. Common total sulfur contents in sour water are around
1194 mg I"1. Because of the high sulfide, ammonium and phenols content, sour
wastewater must be treated before its release into the environment. Both, aerobic
and anaerobic processes have been reported to treat sour waste streams.
3.1. Aerobic processes

Aerobic Thiobacilli species, which oxidize reduced sulfur compounds to
obtain their growth energy, have been studied to promote the sulfur production
from partial sulfide oxidation as shown in Eq. (1) [25, 26, 27]. These bacteria
are gram-negative rods of about 0.3 um in diameter and 1 to 3 um long and
belong to the colorless sulfur bacteria. An important characteristic is their
capacity to excrete elemental sulfur, in contrast to filamentous colorless sulfur
bacteria, as Thiotrix sp., which accumulate it intracellularly.
Sulfur production from partial oxidation of sulfide instead of a complete
oxidation to sulfate has a significant relevance because elemental sulfur can be
recovered from the medium closing the sulfur cycle. Additionally, lower energy
consumption is required because the oxidation to sulfur requires 4-fold less
oxygen that the complete oxidation to sulfate, as shown in Eq. (2).
523
(1)
(2)
The reactor configuration, to promote both sulfur formation and

accumulation, was evaluated and reported by Janssen et al. [27] and Alcantara et
al. [28]. The configuration consisted mainly in the separation of aeration process
from the bioreactor. Thus the liquid saturated with oxygen from the aerator
vessel is sent to the reactor (reaction vessel) at a specific rate, which allows the
control of stoichiometric molar ratio between oxygen and sulfide (theoretical
molar ratio, Rmt, O2/S2~). When Rmt is close to 0.5, the sulfide oxidation is driven
to elemental sulfur formation, while a Rmt close to 2 promotes sulfate as the
main product. The performance of the system reported by Alcantara et al. [28]
was inoculated with a sulfoxidizing consortium and it is shown in Fig. 2.
Sulfide oxidation was studied under different dilution rates at steady state
conditions of 0.5, 1, 1.5, 2 and 3 d"1 (zones A, B, C and D, respectively),
maintaining a constant sulfide concentration in the feed solution at 4.0 g I"1.
Elemental sulfur was produced at dilution rates of 0.5, 1, 1.5 and 2. The
maximum sulfur formation occurred at Rmt of 0.5, where 85% of the total sulfur
added to the reactor as sulfide was transformed to elemental sulfur and 92% of it
was recovered from the bottom of the reactor.
Fig. 2. Performance of the recirculation reactor system under different culture conditions.
Capital letters corresponds to the following dilution rates (d"1): A, 0.5; B, 1; C, 2 and D, 3.
Subtitle letters show the Rmt evaluated: 2: b, c; 1.5: d; 1, e, k; 0.75: f, 1: m; 0.5: a, g; 0.35: h;
0.25: i; 0.15: j . Sulfide influent (—), sulfate (•), elemental sulfur (A), thiosulfate (o) and
sulfide effluent (A).
524
Elemental sulfur production was affected by the dilution rate applied to

the system. When the system operated at Rmt for sulfur production (0.5 and 0.75)
and dilution rates of 0.5, 1 and 2, the elemental sulfur produced was higher than
60%, while washout conditions were observed when the dilution rate was
increased from 2 to 3, at a Rmt of 0.75.
The Thiobacilli species are strict autotrophic bacteria, thus organic
compounds negatively affect their growth. However, sulfoxidizing consortia
have shown an adequate metabolism to oxidize reduced sulfur and organic
sulfur (CS2 for example) compounds [28], in presence of organic matter.
According to Sublette et al. [23] and Alcantara et al. [28, 29], the oxidation of
sulfur compounds is carried out by autotrophic bacteria while organic
compounds are used as energy and carbon source by heterotrophic
microorganisms. Phenol, o-, m- and />-cresol were degraded in a chemostat at
various organic loading rates by the consortium. Under all conditions sulfide
was completely oxidized to sulfate. Microcosm experiments showed that carbon
dioxide production increased under presence of phenols, suggesting that these
compounds were oxidized and they may be used as carbon and energy source by
heterotrophic microorganisms present in the consortium [28].
The expanded bed reactor reported by Janssen et al. [27] is actually built-
in to a family of processes called THIOPAQ, which are applied for the treatment
of wastewater containing sulfide. Also, this technology has been proposed for
the treatment of similar streams from petrochemical industries e.g. spent sulfidic
caustics and from liquefied petroleum gas (LPG) scrubbers [30].
3.2. Anaerobic processes

Thiobacillus denitrificans is a gram-negative, chemoautotroph and
facultative anaerobic bacteria, which oxidizes reduced sulfur compounds to
obtain its growth energy and it is able to use nitrate as electron acceptor.
According to Cadenhead and Sublette [31], this microorganism shows clear
advantages to oxidize sulfide over other Thiobacilli, such as Thiobacillus
thioparus, Thiobacillus versutus and Thiobacillus thiooxidans.
Sulfide is commonly oxidized to sulfate (Eq. 3) or elemental sulfur (Eq. 4)
under anoxic conditions, and where nitrate is used as a terminal electron
acceptor being reduced to elemental nitrogen.
1.25 S2" + 2 NO3" + 2 H+ -> 1.25 SO42" + N2 + H2O (3)
5 S2~ + 2 NO3" + 6 H2O - > 5 S ° + N 2 + 12 OH" (4)
Sour waste streams, including sour water, sour gases and refinery spent-
sulfidic caustics, have been successfully treated using Thiobacillus denitrificans.
For instance, the organic compounds such as benzene, toluene and phenol are
525
biodegraded by heterotrophic bacteria grown in co-culture with Thiobacillus

denitrificans [23, 32].
Sublette [23] identified some technical limitations to apply this
technology for the full-scale treatment of sour wastes. These include: substrate
inhibition (sulfide), product inhibition (sulfate), the need for septic operation,
biomass recycle and recovery, mixed waste issues, and the need for large-scale
cultivation of the organism for the process start up.
T. denitrificans strain F is sulfide tolerant [33] and it was used to treat oil-
field produced water containing sulfides under full-scale field conditions at
Amoco Production Co. in Salt Creek Field in Midwest, WY. More than 800 m3
d"1 of produced water containing 100 mg I'1 sulfide and total dissolved solids of
4800 mg I"1 were successfully biotreated in an earthen pit (3000 m3) over a six-
month period. Based on an average flow of 795 m3 d"1, sulfide influx to the pit
was about 80 kg d"'. Complete removal of sulfides and elimination of associated
odors were clearly observed.
More recently, there has been an increased interest about the oxidation of
reduced sulfur compounds in presence of organics under denitrifying conditions
[34, 35]. The novelty of this approach is the integration of biological processes
that frequently were studied and applied separately. The coupling of carbon,
nitrogen and sulfur cycles implicates the oxidation of reduced forms of sulfur,
organic compounds, as well as the reduction of nitrate [36, 37].
According to Betlach and Tiedje [38], the heterotrophic denitrification
process uses many organic compounds as carbon and energy source; thus
organic transformations were coupled to nitrate reduction and further to
molecular nitrogen. In the case of autotrophic denitrification, reduced sulfur
compounds are oxidized to non-toxic compounds and nitrate, which is used as
final electron acceptor, is reduced to molecular nitrogen.
Reyes-Avila et al. [36] reported that the critical parameters to steer the
nitrate reduction to molecular nitrogen are the C:N and N:S ratios for either
heterotrophic or autotrophic processes, respectively. The same authors reported
that biological denitrification was used to eliminate carbon, nitrogen and sulfur
in an anaerobic continuous stirred tank reactor. Acetate and nitrate at a C:N ratio
of 1.45 were fed at loading rates of 0.29 Kg C m"3 d"1 and 0.2 Kg N m"3 d"1,
respectively. Under steady state denitrifying conditions, the carbon and nitrogen
removal efficiencies were higher than 90%. Under these conditions, sulfide (S2")
was fed to the reactor at several sulfide loading rates (0.044 to 0.295 Kg S2" m"
3
d~'). The high nitrate removal efficiency of the denitrification process was
maintained along the whole process, whereas the carbon removal was 65%, even
at sulfide loading rates of 0.295 Kg S2" m^d"1. The sulfide removal increased up
to 99% via partial oxidation to insoluble elemental sulfur (S°) which
accumulated inside the reactor.
526
In the same way, a denitrifying fluidized bed reactor for effectively

remove sulfide, acetate and nitrate was proposed by Gommers et al., 1988 [39].
The authors reported that the rate-limiting step was the oxidation of sulfur to
sulfate, nevertheless, the biomass showed an overcapacity to oxidize sulfide to
sulfur and to degrade the acetate, under most tested loads.
However, in order to develop a denitrifying technology to treat
wastewaters from the petroleum industry, more studies are needed to elucidate
the effect of phenolic compounds on both sulfide oxidation and nitrate
reduction.
4. OXYGENATED FUEL ADDITIVES
Oxygenated gasoline additives have been used since mid-1970s to substitute

toxic lead compounds. The most common oxygenated used is methyl tert-hu\y\
ether (MTBE), that became the fourth chemical produced in USA [40] because
of their mixing properties, high octane level, low cost and good results in
reducing toxic emissions.
MTBE is manufactured from isobutene (isobutylene or 2-methylpropene),
a byproduct of petroleum refining, and methanol. Therefore MTBE can be
easily and inexpensively produced at refineries. The MTBE presence in refinery
effluents is due to discharges from facilities as a byproduct of the reprocessing
of contaminated or "out of spec" product from the refinery. The volume and
type of waste processed by refineries varies greatly over time, resulting in
order-of-magnitude variations in the MTBE discharges. Few studies have
evaluated the impact of this specific compound in complex wastewater in
refineries [41].
MTBE has been present as a pollutant in numerous water resources mainly
groundwater, The MTBE environmental impact is enhanced by the high
solubility in water, low retention on organic matter, low detection threshold (2.5
and 2.0 jig I"1, for odor and taste, respectively) and low biodegradability. In
1996, the first case of contaminated aquifers by MTBE was reported in Santa
Monica, CA. and 250,000 leaking underground fuel tank sites showed different
levels of MTBE contamination [40]. In Germany, traces of MTBE were
detected in rivers and influents and effluents of wastewater treatment plants
[42]. The MTBE half-life in groundwater systems is several years [43].
There are few reports in Mexico about MTBE occurrence in the
environment. Air concentrations of 11.5 ppb [44] and 4.4 ppb were monitored at
a service station [45] and emissions of on-road vehicles, respectively.
Additionally, concentrations between 100-1500 mg kg soil"1 were found in soils
at fuels distribution and storage stations [46]. Concentrations in the range of 4-
87 mg I"1 were found in groundwater at the surroundings of gas stations.
527
Fortunately, MTBE was detected in none of the nearby 33 monitored drinking

water wells [47].
Increasing reports of MTBE in groundwater produced great concern about
the toxicity and the carcinogenicity of this compound. Toxicological studies
classified the compound as a potential carcinogen for humans [48] and
regulations about the maximal concentration in groundwater were established.
An extreme case was adopted in California where MTBE phase out by 2003
was ordered. However, as long as the use of MTBE continues, the risk of its
presence in refinery effluents and water resources will be latent and treatments
will be required.
Due to its unique above-mentioned physicochemical properties, the clean
up using common techniques like air injection, activated carbon filtering, etc.
are inefficient for MTBE removal. Thus, biological techniques are of particular
interest. In this section a review of MTBE biodegradation and biotreatments is
done, in order to consider the experience adquired in this area for the eventual
treatment of wastewater polluted with MTBE.
4.1 MTBE biodegradation

MTBE has become a challenge for elucidation of its low biodegradability
and the scarcity of MTBE-degrading microorganisms using it as carbon and
energy source. The relatively recalcitrance of MTBE to microbial attack is
intrinsic to its structure containing a combination of an ether link and the
branched moiety. Alkyl ethers are stable molecules (AG° of the ether bond
formation is 360 kJ mol"1 [49]). The high-energy demand for MTBE
degradation is reflected by the low efficiency of biomass production on MTBE.
Fortin et al, [50] pointed out the low MTBE biomass yield obtained analyzing
different consortia. Salanitro [51] suggested that the slow growth on MTBE
might also be due to considerable feedback regulation metabolites on the
oxygenase responsible for the ether bond cleavage. The necessity of
regenerating cofactors, such as NADH, could also have an influence on the rate
of MTBE degradation, since reduced cofactors are required for several
oxidation steps.
Although initial works showed the high recalcitrance of this compound,
some authors have reported the biodegradation of MTBE as sole carbon source.
Moreover, cometabolism was shown to be an important mechanism for MTBE
biodegradation by microorganisms able to grow mainly on short-chain alkanes.
Anaerobic MTBE degradation has been recently observed under methanogenic
[52], nitrate [53] and Fe(III) reducing conditions [54] with longer adaptation
and degradation times. As far as we know, the highest value of MTBE
heterotrophic degradation rate of 454 mg g protein"' h"1 was reported for a strain
Hydrogenophaga flava ENV735 [55]. For cometabolism, the highest value was
528
obtained by the strain Mycobacterium vaccae JOB5 with a MTBE degradation

rate of 111 mg g protein"1 h"1 when hexane was used as growth source [56].
A metabolic pathway for MTBE degradation has been proposed (Fig. 3),
where the MTBE ether bond is enzymatically cleaved yielding tert-butyl alcohol
(TBA) and formaldehyde as the main metabolic intermediates. TBA has been
shown to further biodegrade to 2-methyl-2-hydroxy-l-propanol and 2-
hydroxyisobutyric acid [57]. Suspected further intermediates of the MTBE
degradation metabolic pathway include 2-propanol, acetone and
hydroxyacetone.
The complete understanding of poor MTBE biodegradability would
require the isolation of specialized microorganisms as will as the
characterization of genes and enzymes involved in the degradation and
regulation. Although microorganisms are able to grow using MTBE as a sole
carbon and energy source, we are still far from understanding all causes for its
low biodegradability. A number of excellent reviews are available on aerobic
biodegradation of MTBE [10, 43, 58, 59].
Fig. 3. Proposed metabolic pathway for aerobic MTBE biodegradation Adapted from Fayolle
et al. [49] and Steffan et al. [57].
529
4.2. MTBE removal biotreatments

Although MTBE can be removed from groundwater by physical
technologies such as activated carbon adsorption and air stripping, the cost-
effectiveness of these technologies in removal of MTBE is approximately 10
times higher than their application for removal of hydrocarbons, such as
benzene and toluene, in groundwater.
In December 2003, the USEPA established a database of 356 MTBE
polluted sites and treatment technologies [60] including 111 full-scale
completed cases. The main technologies used were: soil vapor extraction (18%),
pump and treat (17%), in situ bioremediation (21%), air sparging (14%) and
other technologies (30%). Bioremediation is a common technology and its cost
has been estimated [61].
There are two engineering challenges associated with the in situ aerobic
bioremediation of MTBE. First, groundwater polluted with MTBE has very low
dissolved oxygen, thus in all cases the addition of air/oxygen is a requirement
for the treatment; and the second, is the introduction of microorganisms able to
degrade it. Table 5 shows some of the reported cases for in situ treatments. Field
treatment includes the formation of a reactive zone named biobarrier by
introducing to the subsurface MTBE-degrading microorganisms, which is
placed to avoid the advance of the MTBE plume. Oxygen is supplied to the
subsurface either by pulse injecting oxygen gas, air or any oxygen release
compound. MTBE-contaminated water flowing through the biobarrier will
contact the microbes and be degraded to CO2 and water. Biobarriers that have
been applied successfully through biostimulation in some field studies, suggest
that native microorganisms can degrade MTBE through amendments of
nutrients and oxygen. However, the bioaugmentation, by adding
microorganisms already adapted to MTBE degradation, has probed to be a more
feasible option mainly when time-reduction in the treatment is required.
In Salanitro's work, a comparison between biostimulation and
bioaugmentation was performed [62]. The author found a notable MTBE
reduction in both cases, but there was a difference of approximately 150 days in
the lag phase between the treatments, achieving the total bioremediation of the
site in approximately 200 days. Other example of biostimulation versus
bioaugmentation was performed by Wilson et al. [63]. After six months a
noticeable decrease in MTBE was achieved in both inoculated (with PM1
strain) and non-inoculated zones. Polymerase chain reactions techniques
showed that in non-inoculated zone there was the presence of PM-1 like
bacteria [64].
Table 5
Technology Performance for MTBE biological removal
Treatment Scale Microorganism MTBE initial Treatment period or Reference

concentration removal rate
In situ treatment
Field MC-100 7 mg I"1 150-200 days [62]
Field Native microorganisms 1.5 mgT 1 4 days [63]
PM-1
Field ENV425 320 mg r 1 90 days [65]
Field Native microorganisms 19.6 mgl- 1 60 days [66]
Ex situ treatment
Membrane Laboratory Hydrogenophaga flava 1000 mgl"1 42 mg I"1 h"1 [67]
Laboratory Mixed culture 5 mg T1 2.5 mg I"1 rf1 [68]
Cytophaga-Flexibacter-Bacteroides
Fluidized Laboratory ENV735 10 mg T1 10 mg 1"' in 15 min [65]

bioreactor
Laboratory Mixed culture, cometabolism isopentane 10-50 mgT 1 29 mg r 1 h ' [69]
Field Mixed culture 9.6 mg r 1 9 mg r 1 If1 [69]
1
Laboratory Mixed bacterial culture 8.25 mg I"1 ls.smgr'h" [70]
Field N. S. 10 mgl"' and 4.5 g rf' MTBE [71]
15mgr'TBA and 6.2 g hf' TBA
Biotrickling filter Laboratory F-consortium 0.8 mg I"1 50 mg r1 h"1 [72]
Biofilter Laboratory PM-1 100 mg 1"' 58 mg r1 h"1 [73]
Biofilter Laboratory P. aeruginosa 1.1-12.3 mgl"1 l.Smgr'h' 1 [74]
N.S. not specified
531
On the other hand, the addition of a cosubstrate (propane) (US patent

5,814,514, Sept 29, 1998 and US patent 6,194,197 Feb 27,2001) to promote
the cometabolic biodegradation of MTBE was useful for groundwater in
situ bioremediation [65]. The authors inoculated a propane oxidizing strain
ENV425 to cleanup the polluted site by installing biosparging and propane
injection systems, and obtained a reduction of 90% in 90 days of treatment.
This treatment should be preferred when polluted sites present hydraulic
problems or for sites where groundwater extraction is required to stop the
migration of contaminant plumes toward neighboring receptors. Bioreactors
performance has been studied at lab scale and in some field applications
(see Table 5). Most of the investigated bioreactors use immobilized
microorganisms including membrane and fluidized reactors.
Membrane technology retains high biomass levels improving the
volumetric performance and reducing the area for treatment. However,
limitations of this technology are the economic cost associated with the
capital investment, low service-life and moderated operating costs
associated with the pressure-driven mechanism of separation. Membrane
fouling can also be a cost factor depending upon feed water conditions that
might require pretreatment. Table 5 shows some works using this
technology.
In fluidized bioreactors, the biomass is immobilized in a support
material (granular activated carbon, GAC, is commonly used) and this
particles are in continuos movement using an upward water flow.
Fluidization significantly increases the specific surface area available for
biomass and thus degradation of contaminants. Besides the use of GAC as
the fluidizing bed medium also increases specific surface area available for
microbial colonization. These reactors avoid the bed plugging problems
associated with a fixed bed bioreactor, but special care with operational
flows should be taken to avoid washout the bed. However, this type of
bioreactor requires a higher degree of operator maintenance and process
control than the other readily available treatment processes. Some of the
fluidized bioreactor studies are shown in Table 5, including two field
experiences.
MTBE treatment in vapor phase emissions is necessary when any
stripping technology (soil vapor extraction, air stripping, etc.) is used for
cleaning up groundwater containing MTBE (see chapter 17). Basically two
configurations have been proved: Biofilters and biotrickling filters (Table
5).
Biofilters use organic (diatomeaceous earth) or inert (vermiculate or
granular activated carbon) packing material to support the microorganisms
with non-addition or sporadic nutrient addition. Biotrickling filters are
532
similar to biofilters, but they have an aqueous phase trickling over the
packed bed. The liquid contains essential nutrients and it is usually
recycled. Biotrickling filters are more complex than biofilters but are
usually more effective, especially for the treatment of compounds that
generate acidic by-products (see chapter 17).
5. PERSPECTIVES
It is expected that more stringent environmental regulatory actions will be

taken by governments, worldwide. As water is the most important resource
for human, animal and plant life, holistic environmental wastewater
management will continue to gain in importance with time [75].
During the last decade significant efforts were devoted to the
development of technologies for process integration targeting energy
conservation and waste reduction. Great efforts have been done in industries
in order to increase the water conservation and reduce wastewater [76].
However, these integrated technologies will produce less and more
concentrated wastewater whose characteristic would lead to a complete
redesign of the biological wastewater treatment processes that are currently
applied on the process industry. Consequently, facility upgrading,
innovative and sustainable treatment technologies would reshape the
petroleum industry.
The anaerobic processes for the treatment of organic compounds in
industrial wastewater offer important advantages over conventional aerobic
processes. To date, less than 15% of the nearly 1600 full-scale anaerobic
wastewater treatment systems are used by the chemical and petrochemical
industry. However, as the range of compounds that are found to be
biodegraded under anaerobic conditions has increased enormously lately, a
large potential expansion seems possible in the future [22]. Thanks to a
combination of a simple construction and a high volumetric treatment
capacity, the UASB reactor is the dominant concept in the industrial
anaerobic wastewater treatment and it probably will keep reigning in the
future. Nonetheless, higher loaded expanded granular sludge bed reactors
will gradually replace at least part of the UASB applications.
In the case of wastewater streams rich in reduced sulfur compounds,
the new sulfur biotechnology has allowed the development of reactor
systems to remove sulfide producing elemental sulfur. This technology has
been adapted for the sweetening of natural gas [30] and more recently for
liquefied petroleum gas (LPG), which contains predominantly sulfide and
lower alkylthiols [77]. The latter process involves three steps: 1) extraction
of the sulfur compounds from the liquefied hydrocarbon phase to a mild
533
carbonate solution in an absorption column; 2) anaerobic conversion of

alkylthiols to sulfide and methane in an UASB reactor; and 3) partial
oxidation of sulfide into elemental sulfur.
Noteworthy, biological processes developed specifically for wastewater
treatment will play a key role in the treatment of gas streams from the
petroleum industry. Additionally, it is expected that the combination of the
biological carbon, nitrogen and sulfur cycles under anaerobic conditions
would be a potential technology for the removal of such contaminants in a
single step.
In conclusion, the application of biological wastewater treatment in the
frame of a process integration treatment technology will hopefully close the
water cycle allowing the "zero discharge" in the petroleum industry as
shown in Fig. 4.
Fig. 4. Schematic representation of the close water cycle in the petroleum industry.
534
REFERENCES
[I] M.R. Beychock, Aqueous Wastes from Petroleum and Petrochemical Plants, John
Wiley & Sons, London (1967).
[2] F. Berne and J. Cordonnier, Industrial Water Treatment: Refining, Petrochemicals
and Gas processing Techniques, Gulf Publishing, Houston (1995).
[3] H. Macarie, Water Sci. Technol., 42:5-6 (2000) 201.
[4] R.E. Speece, Anaerobic biotechnology for Industrial Wastewaters, Archaea Press,
Nashville (1996) 25-68.
[5] J.H Tay, Y.X He and Y.G Yan, Water Environ. Res. 72 (2000) 189.
[6] P. Olguin-Lora, L. Puig-Grajales and E. Razo-Flores, Environ. Technol., 24 (2003)
999.
[7] R. Sierra-Alvarez and G. Lettinga, Appl. Microbiol. Biotech., 34 (1991) 544.
[8] B. Donlon, E. Razo-Flores, J. Field and G, Appl. Environ. Microbiol., 61 (1995)
3889.
[9] E. Razo-Flores, M. Iniestra-Gonzalez, J.A. Field, P. Olguin-Lora and L. Puig-
Grajales, J. Environ. Eng., 129 (2003) 999.
[10] H. Fang and G. Zhou, Water Sci. Technol., 42:5 (2000) 237.
II1] P. Hwang and S. Cheng, Water Sci. Technol., 24:5 (1991) 133.
[12] H. Fang, T. Chen, Y. Li and H. Chui, Water Res., 30 (1996) 1353.
[13] K. Tawfiki, K. Lepine, J. Bisaillon, R. Beaudet, J. Hawari and S. R Guiot,
Biotechnol. Bioeng., 67 (2000) 419.
[14] S.R. Guiot, K. Tawfiki-Hajji and F. Lepine, Water Sci. Technol., 42:5-6 (2000)
245.
[15] G. Zhou and H. Fang, Bioresource Technol., 61 (1997) 47.
[16] R. Kleerebezem, J. Mortier, L.W. Hulshoff-Pol and G. Lettinga, Water Sci.
Technol., 36:2-3 (1997) 237.
[17] R. Kleerebezem and G. Lettinga, Water Sci. Technol., 42:5-6 (2000) 259.
[18] H. Macarie, Ph. D. Thesis, Universite de Provence Marseille, France (in Freeh)
1992.
[19] D.O. Mountfort, W.J. Krumholz and M.P. Bryant, Int. J. System. Bacteriol, 134
(1984)216.
[20] S. Shelley, Chem. Eng., 98 (1991) 90.
[21] S.S. Cheng, C.Y. Ho and J.H. Wu. 8th Int. Conf. On Anaerobic Digestion. Sendai,
Japan (1997)
[22] R. Kleerebezem and H. Macarie, Chem. Eng., April, (2003) 56.
[23] K. Sublette, R.M. Kolhatkar and K. Raterman, Biodegradation, 9 (1998) 259.
[24] C. Buisman, R. Post, P. Ijspeert, G. Geraats and G. Lettinga, Acta Biotechnol., 9
(1989)255.
[25] A. Janssen, R. Sleylter, C. van der Kaa, J. Jochemsen, J. Bontsema, S. Ma and G.
Lettinga, Biotechnol. Bioeng., 47 (1995) 327.
[26] J. Visser, L. Robertson, H. Verseveld and J. Kuenen, Appl. Environ. Microbiol., 63
(1997)2300.
[27] A. Janssen, S. Ma, P. Lens and G. Lettinga, Biotechnol. Bioeng., 53 (1997) 32.
[28] S. Alcantara, A. Velasco, A. Munoz, J. Cid, S. Revah and E. Razo-Flores, Environ.
Sci. Technol., 38 (2004) 918.
[29] S. Alcantara, I. Estrada, M. Vasquez and S. Revah, Biotechnol. Lett., 21 (1999) 81.
535
[30] B.J. Arena, H.N. Robson, A.L. de Vegt and C.J. Buisman, National Petroleum
Refiners Asociation, Annual Meeting (1988).
[31] P. Cadenhead and K. Sublette, Biotechnol. Bioeng., 35 (1990) 1150.
[32] B. Rajganesh, M. Selvaraj, K. Sublette and C. Camp, Appl. Biochem. Biotechnol.,
51/52(1995)735.
[33] K. Sublette and M. Woolsey, Biotechnol. Bioeng., 34 (1989) 565.
[34] E. W. Kim and J. H. Bae, Water Sci. Technol, 42:3-4 (2000) 233.
[35] B. Krishnakumar and V. B. Manilal, Biotechnol. Lett., 21 (1999) 437.
[36] J. Reyes-Avila, E. Razo-Flores and J. Gomez, Water Res., (2004) submitted.
[37] F. Fdez-Polanco, M. Fdez-Polanco, N. Fernandez, M. Uruena, P. Garcia and S.
Villaverde, Water Sci. Technol., 44: 4 (2001) 15.
[38] M.R. Betlach and J.M. Tiedje, Appl. Environ. Microbiol., 42 (1981) 1074.
[39] P.J. Gommers, W. Buleveld, F.J. Zuiderwijk and J.G. Kuenen, Water Res. 22
(1988) 1075.
[40] R. Johnson, J. Pankow, D. Bender D., C. Price and J. Zogorsky, Environ. Sci.
Technol., 32 (2000) 210.
[41] J. Brown, S. Bay, D. Greenstein and W. Ray, Report Southern California Coastal
Water Research Project. www.sccwrp.org/pubs/annrpt/99_00/abstl I_ar34.htm
[42] C.Achten, A. Kolb and W. Puttmann, Environ. Sci.Technol, 36 (2002) 3652.
[43] S. Fiorenza and H. Rifai, Bioremediation Journal, 7 (2003) 1.
[44] G. Reyna, E. Vega, E. Reyes, V. Mugica, V. Chow, J. Watson and J. Arriaga. 94th
Annual Conference and Exhibition of the Air and Waste Management Association.
Orlando, FL (2001).
[45] L. Manzanares, L. Mufioz, C. Romero, V. Nevarez, E. Ramirez, M. Delgado and A.
Keer. 94th Annual Conference and Exhibition of the Air and Waste Management
Association. Orlando, FL (2001).
[46] R. Iturbe, R. Flores and L.Torres, Water Air Soil Poll., 14 (2003) 261.
[47] C. Buenrostro and A. Dovali. DGCH. (2001) Study for determining and evaluating
methyl tert-buryl ether (MTBE). Oficio GDF-SOS/01-523. In Spanish.
[48] USEPA 1997, EPA822-F-97-009, Office of Water, 34.
[49] F. Fayolle, J. Vandecasteele and F. Monot, Appl. Microbiol. Biotechnol., 56 (2001)
339.
[50] N. Fortin, M. Morales, Y. Nakagawa, D. Focht and M. Deshusses, Environ.
Microbiol., 3(2001)407.
[51] J. Salanitro, Curr. Op. Biotechnol., 6 (1995) 337.
[52] J. Wilson, J. Cho, B. Wilson and J. Vardy. U.S. Environmental Protection Agency.
2000. Natural attenuation of MTBE in the subsurface under methanogenic
conditions. EPA/600/R-00/006. U.S. EPA Office of Research and Development:
Washington, D.C.
[53] P. Bradley, F. Chapelle and J. Landmeyer, Appl. Environ. Microbiol., 67 (2001)
1975.
[54] K. Finneran and D. Lovley, Environ. Sci. Technol., 35 (2001) 1785.
[55] R. Steffan, S. Vainberg, C. Condee, K. McClay and P. Hatzinger. In G.
Wrickramayake, A. Gavaskar, B. Alleman, V. Magar (eds.) Bioremediation and
phytoremediation of chlorinated and recalcitrant compounds. Batelle, Columbus,
OH (2000) 165.
536
[56] M. Hyman and K. O'Reilly. In: Alleman B, Lesson A (eds.) In situ bioremediation
of petroleum hydrocarbon and other organic compounds. Battelle, Columbus, OH.
(1999)7.
[57] R. Steffan, K. Me Clay, S. Vainberg, C. Condee and D. Zhang, Appl. Environ.
MicrobioL, 63(1997)4216.
[58] R. Deeb, K. Scow and L. Alvarez, Biodegradation, 11 (2000) 171.
[59] R. Prince, Crit. Rev. MicrobioL, 26 (2002) 163.
[60] USEPA 2003. MTBE treatment case study website, http://clu-in.org/products/mtbe/
December 2003.
[61] B. Wilson and J. Wilson, Contaminated Soil Sediment and Water (2002) 47.
[62] J. Salanitro, P. Johnson, G. Spinnler, P. Maner, H. Wisniewski and C. Bruce,
Environ. Sci. TechnoL, 34 (2000) 4152.
[63] R. Wilson R., K. Scow and D. Mackay, Environ. Sci. TechnoL, 36 (2001) 190.
[64] K. Hrystova, B. Gebreyesus, D. Mackay and K. Scow, Appl.Environ. MicrobioL,
69(2003)2616.
[65] R. Steffan, P. Hatzinger, Y. Farhan and S. Drew. NGWA/API Conference on
Petroleum Hydrocarbons and Organic Chemicals in groundwater: Prevention,
detection and Remediation. Westville, OH (2001).
[66] J. Landmeyer, F. Chapelle, H. Herlong and P. Bradley, Environ. Sci. TechnoL, 35
(2001) 1118.
[67] R. Steffan, J. Johnson and S. Drew. In Sublette K (eds.) Proceedings of the 7th
International Petroleum Environmental Conference. IPEC, Albuquerque, NM
(2000) 1-12.
[68] J. Morrison, M. Suidan and A. Venosa, J. Environ. Eng., 128 (2002) 836.
[69] W. Stringfellow and K. Oh, J. Environ. Eng., 128 (2002) 852.
[70] S. Vainberg, A. Togna, P. Sutton and R. Steffan, J. Environ. Eng., 128 (2002) 852.
[71] J. O'Connell and S. Zigan. In, E. Moyer, P. Kostecki (eds.) MTBE remediation
Handbook. Amherst Scientific Publishers. Amherst Massachusets. (2003)
[72] N. Fortin and M. Deshusses, Environ. Sci. TechnoL, 33 (1999) 2980.
[73] J. Eweis, J. Scarano, B. Converse, D. Chang and E. Schroeder. Report American
Petroleum Institute Contract # 97000-2577(1999).
[74] D. Dupasquier, S. Revah and R. Auria, Environ.Sci.TechnoL, 36 (2002) 247.
[75] W.W. Eckenfelder Jr. and A.J. Englande Jr., Water Sci. TechnoL, 34:10 (1996) 1.
[76] R.F. Dunn and M.M. El-Halwagi, J. Chem. TechnoL BiotechnoL, 78 (2003) 1011.
[77] J. Sipma, A.H. Janssen, L.W. Hulshoff Pol and G. Lettinga, BiotechnoL Bioeng.,
82(2003)1.
537
STUDIES IN SURFACE SCIENCE AND CATALYSIS
Advisory Editors:
B. Delmon, Universite Catholique de Louvain, Louvain-la-Neuve, Belgium
J.T. Yates, University of Pittsburgh, Pittsburgh, PA, U.S.A.
Volume 1 Preparation of Catalysts I. Scientific Bases for the Preparation of Heterogeneous

Catalysts. Proceedings of the First International Symposium, Brussels,
October 14-17,1975
edited by B. Delmon, P.A. Jacobs and G. Poncelet
Volume 2 The Control of the Reactivity of Solids. A Critical Survey of the Factors that
Influence the Reactivity of Solids, with Special Emphasis on the Control of the
Chemical Processes in Relation to Practical Applications
by V.V. Boldyrev, M. Bulens and B. Delmon
Volume 3 Preparation of Catalysts ll.Scientific Bases for the Preparation of Heterogeneous
Catalysts. Proceedings of the Second International Symposium, Louvain-la-
Neuve,
September 4 - 7 , 1978
edited by B. Delmon, P. Grange, P. Jacobs and G. Poncelet
Volume 4 Growth and Properties of Metal Clusters. Applications to Catalysis and the
Photographic Process. Proceedings of the 32 nd International Meeting of the
Societe de Chimie Physique, Villeurbanne, September 2 4 - 2 8 , 1979
edited by J. Bourdon
Volume 5 Catalysis by Zeolites.Proceedings of an International Symposium, Ecully (Lyon),
September 9 - 1 1 , 1980
edited by B. Imelik, C. Naccache,Y. BenTaarit, J.C.Vedrine, G. Coudurier and
H. Praliaud
Volume 6 Catalyst Deactivation. Proceedings of an International Symposium, Antwerp,
October 1 3 - 15,1980
edited by B. Delmon and G.F. Froment
Volume 7 New Horizons in Catalysis. Proceedings of the 7 th International Congress on
Catalysis, Tokyo, June 30-July 4, 1980. Parts A and B
edited by T. Seiyama and K.Tanabe
Volume 8 Catalysis by Supported Complexes
by Yu.l.Yermakov, B.N. Kuznetsov and V.A. Zakharov
Volume 9 Physics of Solid Surfaces. Proceedings of a Symposium, Bechyne,
September 29-October 3,1980
edited by M. Laznicka
Volume 1 0 Adsorption at the Gas-Solid and Liquid-Solid Interface. Proceedings of an
International Symposium, Aix-en-Provence, September 2 1 - 2 3 , 1981
edited by J. Rouquerol and K.S.W. Sing
Volume 11 Metal-Support and Metal-Additive Effects in Catalysis. Proceedings of an
International Symposium, Ecully (Lyon), September 1 4 - 1 6 , 1982
edited by B. Imelik, C. Naccache, G. Coudurier, H. Praliaud, P. Meriaudeau,
P. Gallezot, G.A.Martin and J.C.Vedrine
Volume 12 Metal Microstructures in Zeolites. Preparation - Properties - Applications.
Proceedings of a Workshop, Bremen, September 2 2 - 2 4 , 1982
edited by P.A. Jacobs, N.I. Jaeger, P. Jirii and G. Schulz-Ekloff
Volume 13 Adsorption on Metal Surfaces. An Integrated Approach
edited by J. Benard
Volume 14 Vibrations at Surfaces. Proceedings of the Third International Conference,
Asilomar, CA, September 1-4, 1982
edited by C.R. Brundle and H.Morawitz
Volume 15 Heterogeneous Catalytic Reactions Involving Molecular Oxygen
by G.I. Golodets
538
Volume 1 6 Preparation of Catalysts III. Scientific Bases for the Preparation of Heterogeneous
Catalysts. Proceedings of the Third International Symposium, Louvain-la-Neuve,
edited by G. Poncelet, P. Grange and P.A. Jacobs
Volume 17 Spillover of Adsorbed Species. Proceedings of an International Symposium,
Lyon-Villeurbanne, September 12-16, 1983
edited by G.M. Pajonk,S.J.Teichner and J.E. Germain
Volume 1 8 Structure and Reactivity of Modified Zeolites. Proceedings of an International
Conference, Prague, July 9 - 1 3 , 1984
edited by P.A. Jacobs, N.I. Jaeger, P. Jiru, V.B. Kazansky and G. Schulz-Ekloff
Volume 19 Catalysis on the Energy Scene. Proceedings of the 9 th Canadian Symposium
on Catalysis, Quebec, P.Q., September 30-October 3, 1984
edited by S. Kaliaguine and A.Mahay
Volume 20 Catalysis by Acids and Bases. Proceedings of an International Symposium,
Villeurbanne (Lyonl, September 2 5 - 2 7 , 1984
edited by B. Imelik, C. Naccache, G. Coudurier.Y. Ben Taarit and J.C.Vedrine
Volume 21 Adsorption and Catalysis on Oxide Surfaces. Proceedings of a Symposium,
Uxbridge, June 2 8 - 2 9 , 1984
edited by M. Che and G.C.Bond
Volume 22 Unsteady Processes in Catalytic Reactors
by Yu.Sh. Matros
Volume 23 Physics of Solid Surfaces I984
edited by J.Koukal
Volume 24 Zeolites:Synthesis, Structurejechnology and Application. Proceedings of an
International Symposium, Portoroz-Portorose, September 3 - 8 , 1984
edited by B.Drzaj.S. Hocevar and S. Pejovnik
Volume 25 Catalytic Polymerization of Olefins. Proceedings of the International Symposium
on Future Aspects of Olefin Polymerization, Tokyo, July 4 - 6 , 1985
edited by T.Keii and K.Soga
Volume 26 Vibrations at Surfaces 1985. Proceedings of the Fourth International Conference,
Bowness-on-Windermere, September 15-19, 1985
edited by D.A. King, N.V. Richardson and S. Holloway
Volume 27 Catalytic Hydrogenation
edited by L. Cerveny
Volume 28 New Developments in Zeolite Science and Technology. Proceedings of t h e
7th International Zeolite Conference, Tokyo, August 17-22, 1986
edited by Y.Murakami,A. lijima and J.W.Ward
Volume 29 Metal Clusters in Catalysis
edited by B.C. Gates, L. Guczi and H. Knozinger
Volume 30 Catalysis and Automotive Pollution Control. Proceedings of t h e First
International Symposium, Brussels, September 8 - 1 1 , 1986
edited by A. Crucq and A. Frennet
Volume 31 Preparation of Catalysts IV. Scientific Bases for the Preparation of Heterogeneous
Catalysts. Proceedings of the Fourth International Symposium, Louvain-la-
Neuve,
September 1-4, 1986
edited by B. Delmon, P. Grange, P.A. Jacobs and G. Poncelet
Volume 32 Thin Metal Films and Gas Chemisorption
edited by P.Wissmann
Volume 33 Synthesis of High-silica Aluminosilicate Zeolites
edited by P.A. Jacobs and J.A.Martens
Volume 34 Catalyst Deactivation 1987. Proceedings of the 4 t h International Symposium,
Antwerp, September 29-October 1, 1987
Volume 35 Keynotes in Energy-Related Catalysis
edited by S. Kaliaguine
539
Volume 36 Methane Conversion. Proceedings of a Symposium on the Production of Fuels and

Chemicals from Natural Gas, Auckland, April 2 7 - 3 0 , 1987
edited by D.M. Bibby, C.D.Chang, R.F.Howe and S.Yurchak
Volume 37 Innovation in Zeolite Materials Science. Proceedings of an International
Symposium, Nieuwpoort, September 1 3 - 1 7 , 1987
edited by P.J. Grobet, W.J.Mortier, E.F.Vansant and G. Schulz-Ekloff
Volume 38 Catalysis l987.Proceedings of the 10 th North American Meeting of the Catalysis
Society, San Diego, CA, May 17-22, 1987
edited by J.W.Ward
Volume 39 Characterization of Porous Solids. Proceedings of the IUPAC Symposium
(COPS I), Bad Soden a. Ts., April 26-29,1987
edited by K.K.Unger, j . Rouquerol, K.S.W.Sing and H. Krai
Volume 40 Physics of Solid Surfaces I987. Proceedings of the Fourth Symposium on
Surface Physics, Bechyne Castle, September 7 - 1 1 , 1987
edited by J.Koukal
Volume 41 Heterogeneous Catalysis and Fine Chemicals. Proceedings of an International
Symposium, Poitiers, March 1 5 - 1 7 , 1988
edited by M. Guisnet, |. Barrault, C. Bouchoule.D. Duprez, C. Montassier and
G. Perot
Volume 42 Laboratory Studies of Heterogeneous Catalytic Processes
by E.G. Christoffel, revised and edited by Z. Paal
Volume 43 Catalytic Processes under Unsteady-State Conditions
by Yu. Sh. Matros
Volume 44 Successful Design of Catalysts. Future Requirements and Development.
Proceedings of the Worldwide Catalysis Seminars, July, 1 988, on the Occasion
of the 30 th Anniversary of the Catalysis Society of Japan
edited by T. Inui
Volume 45 Transition Metal Oxides. Surface Chemistry and Catalysis
by H.H.Kung
Volume 46 Zeolites as Catalysts, Sorbents and Detergent Builders. Applications and
Innovations. Proceedings of an International Symposium, Wurzburg,
September 4 - 8 , 1 9 8 8
edited by H.G. Karge and j.Weitkamp
Volume 47 Photochemistry on Solid Surfaces
edited by M.Anpo and T. Matsuura
Volume 48 Structure and Reactivity of Surfaces. Proceedings of a European Conference,
Trieste, September 13-16, 1988
edited by C.Morterra.A. Zecchina and G. Costa
Volume 49 Zeolites: Facts, Figures, Future. Proceedings of the 8th International Zeolite
Conference, Amsterdam, July 10-14, 1989. Parts A and B
edited by P.A. Jacobs and R.A. van Santen
Volume 50 Hydrotreating Catalysts. Preparation, Characterization and Performance.
Proceedings of the Annual International AlChE Meeting, Washington, DC,
November 27-December 2, 1988
edited by M.L. Occelli and R.G.Anthony
Volume 51 New Solid Acids and Bases.Their Catalytic Properties
by KJanabe, M. Misono.Y. Ono and H. Hattori
Volume 52 Recent Advances in Zeolite Science. Proceedings of the 1989 Meeting of the
British Zeolite Association, Cambridge, April 1 7 - 1 9 , 1989
edited by J. Klinowsky and P.j. Barrie
Volume 53 Catalyst in Petroleum Refining 1989. Proceedings of the First International
Conference on Catalysts in Petroleum Refining, Kuwait, March 5 - 8 , 1989
edited by D.LTrimm.S.Akashah, M.Absi-Halabi and A. Bishara
Volume 54 Future Opportunities in Catalytic and Separation Technology
edited by M. Misono,Y.Moro-oka and S. Kimura
540
Volume 55 New Developments in Selective Oxidation. Proceedings of an International

Symposium, Rimini, Italy, September 18-22, 1989
edited by G. Centi and F.Trifiro
Volume 56 Olefin Polymerization Catalysts. Proceedings of the International Symposium
on Recent Developments in Olefin Polymerization Catalysts, Tokyo,
October 2 3 - 2 5 , 1989
edited by T.Keii and K.Soga
Volume 5 7A Spectroscopic Analysis of Heterogeneous Catalysts. Part A: Methods of
Surface Analysis
edited by J.L.G. Fierro
Volume 57B Spectroscopic Analysis of Heterogeneous Catalysts. Part B: Chemisorption of
Probe Molecules
edited by J.L.G. Fierro
Volume 58 Introduction to Zeolite Science and Practice
edited by H. van Bekkum, E.M. Flanigen and ].C. Jansen
Volume 59 Heterogeneous Catalysis and Fine Chemicals II. Proceedings of t h e 2 n d
International Symposium, Poitiers, October 2 - 6 , 1 990
edited by M. Guisnet, J. Barrault, C. Bouchoule.D. Duprez, G. Perot, R.Maurel
and C. Montassier
Volume 60 Chemistry of Microporous Crystals. Proceedings of the International Symposium
on Chemistry of Microporous Crystals, Tokyo, June 2 6 - 2 9 , 1990
edited by T. Inui.S. Namba and T.Tatsumi
Volume 61 Natural Gas Conversion. Proceedings of the Symposium on Natural Gas
Conversion, Oslo, August 12-17, 1990
edited by A. Holmen, K.-j. Jens and S.Kolboe
Volume 62 Characterization of Porous Solids II. Proceedings of t h e IUPAC Symposium
(COPS II), Alicante, May 6 - 9 , 1990
edited by F. Rodriguez-Reinoso, j . Rouquerol, K.S.W. Sing and K.K.Unger
Volume 63 Preparation of Catalysts V. Scientific Bases for the Preparation of Heterogeneous
Catalysts. Proceedings of the Fifth International Symposium, Louvain-la-Neuve,
edited by G. Poncelet, P.A. Jacobs, P. Grange and B. Delmon
Volume 64 New Trends in CO Activation
edited by L. Guczi
Volume 65 Catalysis and Adsorption by Zeolites. Proceedings of ZEOCAT 9 0 , Leipzig,
August 20-23, 1990
edited by G. Ohlmann, H. Pfeifer and R. Fricke
Volume 66 Dioxygen Activation and Homogeneous Catalytic Oxidation. Proceedings of t h e
Fourth International Symposium on Dioxygen Activation and Homogeneous
Catalytic Oxidation, Balatonf tired, September 10-14, 1990
edited by L.I. Simandi
Volume 67 Structure-Activity and Selectivity Relationships in Heterogeneous Catalysis.
Proceedings of the ACS Symposium on Structure-Activity Relationships in
Heterogeneous Catalysis, Boston, MA, April 22-27', 1990
edited by R.K. Grasselli and A.W.SIeight
Volume 68 Catalyst Deactivation 1991. Proceedings of the Fifth International Symposium,
Evanston, IL, June 2 4 - 2 6 , 1991
edited by C.H. Bartholomew and J.B. Butt
Volume 69 Zeolite Chemistry and Catalysis. Proceedings of an International Symposium,
Prague, Czechoslovakia, September 8 - 1 3 , 1991
edited by P.A. Jacobs, N.I. Jaeger, L.Kubelkova and B.Wichterlova
Volume 70 Poisoning and Promotion in Catalysis based on Surface Science Concepts and
Experiments
by M. Kiskinova
541
Volume 71 Catalysis and Automotive Pollution Control II. Proceedings of t h e 2 n d

International Symposium (CAPoC 2), Brussels, Belgium, September 10-13,
1990
edited by A. Crucq
Volume 72 New Developments in Selective Oxidation by Heterogeneous Catalysis.
Proceedings of the 3rd European Workshop Meeting on New Developments in
Selective Oxidation by Heterogeneous Catalysis, Louvain-la-Neuve, Belgium,
April 8 - 1 0 , 1991
edited by P. Ruiz and B. Delmon
Volume 73 Progress in Catalysis. Proceedings of the 12th Canadian Symposium on Catalysis,
Banff, Alberta, Canada, May 2 5 - 2 8 , 1992
edited by K.J. Smith and EX. Sanford
Volume 74 Angle-Resolved Photoemission.Theory and Current Applications
edited by S.D. Kevan
Volume 75 New Frontiers in Catalysis, Parts A-C. Proceedings of the 10 th International
Congress on Catalysis, Budapest, Hungary, 19-24 July, 1992
edited by L. Guczi, F. Solymosi and P.Tetenyi
Volume 76 Fluid Catalytic Cracking: Science and Technology
edited by J.S.Magee and M.M. Mitchell, Jr.
Volume 77 New Aspects of Spillover Effect in Catalysis. For Development of Highly Active
Catalysts. Proceedings of the Third International Conference on Spillover, Kyoto,
Japan, August 17-20, 1993
edited by T. Inui, K. Fujimoto.T.Uchijima and M. Masai
Volume 78 Heterogeneous Catalysis and Fine Chemicals III.
Proceedings of the 3rd International Symposium, Poitiers, April 5 - 8, 1993
edited by M. Guisnet, J. Barbier, J. Barrault, C. Bouchoule.D. Duprez,
G. Perot and C. Montassier
Volume 79 Catalysis: An Integrated Approach to Homogeneous, Heterogeneous and
Industrial Catalysis
edited by J.A. Moulijn, P.W.N.M. van Leeuwen and R.A. van Santen
Volume 80 Fundamentals of Adsorption. Proceedings of the Fourth International Conference
on Fundamentals of Adsorption, Kyoto, Japan, May 17-22, 1992
edited by M. Suzuki
Volume 81 Natural Gas Conversion II. Proceedings of the Third Natural Gas Conversion
Symposium, Sydney, July 4 - 9 , 1993
edited by H.E. Curry-Hyde and R.F.Howe
Volume 82 New Developments in Selective Oxidation II. Proceedings of t h e Second World
Congress and Fourth European Workshop Meeting, Benalmadena, Spain,
September 2 0 - 2 4 , 1993
edited by V. Cortes Corberan and S.Vic Bellon
Volume 83 Zeolites and Microporous Crystals. Proceedings of the International Symposium
on Zeolites and Microporous Crystals, Nagoya, Japan, August 2 2 - 2 5 , 1993
edited byT. Hattori and T.Yashima
Volume 84 Zeolites and Related Microporous Materials: State of the Art I994.
Proceedings of the 10th International Zeolite Conference,
Garmisch-Partenkirchen, Germany, July 17-22, 1994
edited by J.Weitkamp, H.G. Karge.H. Pfeifer and W. Holderich
Volume 85 Advanced Zeolite Science and Applications
edited by J.C. jansen, M. Stocker, H.G. Karge and J.Weitkamp
Volume 86 Oscillating Heterogeneous Catalytic Systems
by M.M. Slinko and N.I. Jaeger
Volume 87 Characterization of Porous Solids III. Proceedings of t h e IUPAC Symposium
(COPS III), Marseille, France, May 9 - 1 2 , 1993
edited by j.Rouquerol, F. Rodriguez-Reinoso, K.S.W. Sing and K.K.Unger
542
Volume 88 Catalyst Deactivation 1994. Proceedings of the 6 th International Symposium,

Ostend, Belgium, October 3 - 5 , 1994
Volume 89 Catalyst Design for Tailor-made Polyolefins. Proceedings of t h e International
Symposium on Catalyst Design for Tailor-made Polyolefins, Kanazawa,
Japan, March 10-12, 1994
edited by K.Soga and M.Terano
Volume 90 Acid-Base Catalysis II. Proceedings of the International Symposium on
Acid-Base Catalysis II, Sapporo, Japan, December 2 - 4 , 1993
edited by H. Hattori.M. Misono and Y.Ono
Volume 91 Preparation of Catalysts VI. Scientific Bases for the Preparation of
Heterogeneous Catalysts. Proceedings of the Sixth International Symposium,
Louvain-La-Neuve, September 5-8, 1994
edited by G. Poncelet, J.Martens.B. Delmon, P.A. Jacobs and P. Grange
Volume 92 Science and Technology in Catalysis I994. Proceedings of t h e Second Tokyo
Conference on Advanced Catalytic Science and Technology, Tokyo,
August 2 1 - 2 6 , 1994
edited by Y. Izumi, H.Arai and M. Iwamoto
Volume 93 Characterization and Chemical Modification of the Silica Surface
by E.F.Vansant, P.Van Der Voort and K.C.Vrancken
Volume 94 Catalysis by Microporous Materials. Proceedings of ZEOCAT'95, Szombathely,
Hungary, July 9-13, 1995
edited by H.K. Beyer, H.G.Karge, i. Kiricsi and J.B.Nagy
Volume 95 Catalysis by Metals and Alloys
by V. Ponec and G.C.Bond
Volume 96 Catalysis and Automotive Pollution Control III. Proceedings of
the Third International Symposium (CAPoC3), Brussels, Belgium,
April 2 0 - 2 2 , 1994
edited by A. Frennet and J.-M. Bastin
Volume 97 Zeolites:A Refined Tool for Designing Catalytic Sites. Proceedings of
the International Symposium, Quebec, Canada, October 15-20, 1995
edited by L. Bonneviot and S. Kaliaguine
Volume 98 Zeolite Science 1994: Recent Progress and Discussions. Supplementary Materials t o t h e
10th International Zeolite Conference, Garmisch-Partenkirchen, Germany, July
17-22, 1994
edited by H.G. Karge and J.Weitkamp
Volume 99 Adsorption on New and Modified Inorganic Sorbents
edited by A.Dabrowski and V.A.Tertykh
Volume 100 Catalysts in Petroleum Refining and Petrochemical Industries 1995.
Proceedings of the 2nd International Conference on Catalysts in Petroleum
Refining and Petrochemical Industries, Kuwait, April 22-26, 1995
edited by M.Absi-Halabi, J. Beshara, H. Qabazard and A. Stanislaus
Volume 101 II th International Congress on Catalysis -40 t h Anniversary.
Proceedings of the 11 t h ICC, Baltimore, MD, USA, June 30-July 5, 1996
edited by J.W. High tower, W.N. Delgass, E. Iglesia and A.T. Bell
Volume 102 Recent Advances and New Horizons in Zeolite Science and Technology
edited by H. Chon.S.I.Woo and S. -E. Park
Volume 1 03 Semiconductor Nanoclusters - Physical, Chemical, and Catalytic Aspects
edited by P.V. Kamat and D. Meisel
Volume 104 Equilibria and Dynamics of Gas Adsorption on Heterogeneous Solid Surfaces
edited by W. Rudzihski.W.A. Steele and G. Zgrablich
Volume 105 Progress in Zeolite and Microporous Materials
Proceedings of the 11 t h International Zeolite Conference, Seoul, Korea,
August 12-17, 1996
edited by H. Chon.S.-K. Ihm and Y.S.Uh
543
Volume 1 06 Hydrotreatment and Hydrocracking of Oil Fractions

Proceedings of the 1 s t International Symposium / 6 th European Workshop,
Oostende, Belgium, February 17-19, 1997
e d i t e d b y G.F. Froment.B. Delmon and P. Grange
Volume 1 07 Natural Gas Conversion IV
Proceedings of the 4 th International Natural Gas Conversion Symposium,
Kruger Park, South Africa, November 19-23, 1995
e d i t e d b y M. de Pontes, R.L. Espinoza, C.P. Nicolaides, J.H. Scholtz and
M.S. Scurrell
Volume 108 Heterogeneous Catalysis and Fine Chemicals IV
Proceedings of the 4 th International Symposium on Heterogeneous Catalysis and
Fine Chemicals, Basel, Switzerland, September 8-12, 1996
e d i t e d by H.U. Blaser, A. Baiker and R. Prins
Volume 109 Dynamics of Surfaces and Reaction Kinetics in Heterogeneous Catalysis.
Proceedings of the International Symposium, Antwerp, Belgium,
September 15-17, 1997
edited by G.F. Froment and K.C.Waugh
Volume 110 Third World Congress on Oxidation Catalysis.
Proceedings of the Third World Congress on Oxidation Catalysis, San Diego, CA,
U.S.A., 21-26 September 1997
e d i t e d b y R.K. Grasselli,S.T.Oyama, A.M. Gaffney and J.E. Lyons
Volume 111 Catalyst Deactivation I997.
Proceedings of the 7th International Symposium, Cancun, Mexico, October 5-8,
1997
e d i t e d b y C.H. Bartholomew and G.A. Fuentes
Volume 11 2 Spillover and Migration of Surface Species on Catalysts.
Proceedings of the 4 th International Conference on Spillover, Dalian, China,
September 15-18, 1997
e d i t e d b y Can Li and Qin Xin
Volume 113 Recent Advances in Basic and Applied Aspects of Industrial Catalysis.
Proceedings of the 13th National Symposium and Silver Jubilee Symposium of
Catalysis of India, Dehradun, India, April 2-4, 1997
e d i t e d b y T.S.R. Prasada Rao and G.Murali Dhar
Volume 114 Advances in Chemical Conversions for Mitigating Carbon Dioxide.
Proceedings of the 4 th International Conference on Carbon Dioxide Utilization,
Kyoto, Japan, September 7-11, 1997
e d i t e d b y T. Inui, M.Anpo.K. liui.S.Yanagida and T.Yamaguchi
Volume 11 5 Methods for Monitoring and Diagnosing the Efficiency of Catalytic Converters.
A patent-oriented survey
by M. Sideris
Volume 116 Catalysis and Automotive Pollution Control IV.
Proceedings of the 4 th International Symposium (CAPoC4), Brussels, Belgium,
April 9-11, 1997
e d i t e d b y N. Kruse, A. Frennet and J.-M. Bastin
Volume 11 7 Mesoporous Molecular Sieves 1998
Proceedings of the 1 s t International Symposium, Baltimore, MD, U.S.A.,
July 10-12, 1998
e d i t e d b y L.Bonneviot, F. Beland, C.Danumah, S. Giasson and S. Kaliaguine
Volume 118 Preparation of Catalysts VII
Proceedings of the 7 th International Symposium on Scientific Bases for the
Preparation of Heterogeneous Catalysts, Louvain-la-Neuve, Belgium,
September 1-4, 1998
e d i t e d b y B. Delmon, P.A. Jacobs, R. Maggi, J.A.Martens, P. Grange and G. Poncelet
Volume 11 9 Natural Gas Conversion V
Proceedings of the 5th International Gas Conversion Symposium, Giardini-Naxos,
Taormina, Italy, September 20-25, 1998
e d i t e d b y A. Parmaliana, D. Sanfilippo, F. Frusteri, A.Vaccari and F.Arena
Volume 12 0 A Adsorption and its Applications in Industry and Environmental Protection.
Vol I: Applications in Industry
e d i t e d b y A. Dijbrowski
544
Volume 1 2 0 B Adsorption and its Applications in Industry and Environmental Protection.

Vol II: Applications in Environmental Protection
edited b y A. Dqbrowski
Volume 121 Science and Technology in Catalysis 1998
Proceedings of the Third Tokyo Conference in Advanced Catalytic Science and
Technology, Tokyo, July 19-24, 1998
edited b y H. Hattori and K. Otsuka
Volume 122 Reaction Kinetics and the Development of Catalytic Processes
Proceedings of the International Symposium, Brugge, Belgium, April 19-21,
1999
edited b y G.F. Froment and K.C.Waugh
Volume 1 23 Catalysis:An Integrated Approach
Second, Revised and Enlarged Edition
edited b y R.A. van Santen, P.W.N.M. van Leeuwen, J.A. Moulijn and BAAveriil
Volume 1 24 Experiments in Catalytic Reaction Engineering
by J.M. Berty
Volume 125 Porous Materials in Environmentally Friendly Processes
Proceedings of the 1 s t International FEZA Conference, Eger, Hungary,
September 1-4, 1999
edited b y I. Kiricsi, G. Pal-Borbely, J.B.Nagy and H.G. Karge
Volume 1 26 Catalyst Deactivation 1999
Proceedings of the 8th International Symposium, Brugge, Belgium,
October 10-13, 1999
edited b y B. Delmon and G.F. Froment
Volume 127 Hydrotreatment and Hydrocracking of Oil Fractions
Proceedings of the 2nd International Symposium/7th European Workshop,
Antwerpen, Belgium, November 14-17, 1999
edited b y B. Delmon, G.F. Froment and P. Grange
Volume 1 28 Characterisation of Porous Solids V
Proceedings of the 5th International Symposium on the Characterisation of
Porous Solids (COPS-V), Heidelberg, Germany, May 30- June 2, 1999
edited b y K.K.Unger,G.Kreysa and J.P. Baselt
Volume 129 Nanoporous Materials II
Proceedings of the 2nd Conference on Access in Nanoporous Materials,
Banff, Alberta, Canada, May 25-30, 2000
edited byA. Sayari.M. jaroniec and T.J. Pinnavaia
Volume 1 30 12 th International Congress on Catalysis
Proceedings of the 12 t h ICC, Granada, Spain, July 9-14, 2000
edited byA. Corma, F.V. Melo.S. Mendioroz and J.L.G. Fierro
Volume 131 Catalytic Polymerization of Cycloolefins
Ionic, Ziegler-Natta and Ring-Opening Metathesis Polymerization
By V. Dragutan and R. Streck
Volume 1 32 Proceedings of the International Conference on Colloid and Surface Science,
Tokyo, Japan, November 5-8,2000
25 th Anniversary of the Division of Colloid and Surface Chemistry,
The Chemical Society of Japan
edited b y Y. Iwasawa, N.Oyama and H.Kunieda
Volume 1 33 Reaction Kinetics and the Development and Operation of Catalytic Processes
Proceedings of the 3rd International Symposium, Oostende, Belgium, April 22-
25, 2001
edited b y G.F. Froment and K.C.Waugh
Volume 134 Fluid Catalytic Cracking V
Materials and Technological Innovations
edited b y M.L. Occelli and P. O'Connor
545
Volume 135 Zeolites and Mesoporous Materials at the Dawn of the 21 st Century.
Proceedings of the 13lh International Zeolite Conference, Montpellier, France,
8-13 July 2001
edited by A. Galameau, F. di Renso, F. Fajula ans J. Vedrine
Volume 136 Natural Gas Conversion VI
Proceedings of the 6th Natural Gas Conversion Symposium, June 17-22, 2001,
Alaska, USA.
edited by J J . Spivey, E. Iglesia and T.H. Fleisch
Volume 137 Introduction to Zeolite Science and Practice.
2nd completely revised and expanded edition
edited by H. van Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jansen
Volume 138 Spillover and Mobility of Species on Solid Surfaces
edited by A. Guerrero-Ruiz and I. Rodriquez-Ramos
Volume 139 Catalyst Deactivation 2001
Proceedings of the 9th International Symposium, Lexington, KY, USA, October 2001
edited by J.J. Spivey, G.W. Roberts and B.H. Davis
Volume 140 Oxide-based Systems at the Crossroads of Chemistry.
Second International Workshop, October 8-11, 2000, Como, Italy.
Edited by A. Gamba, C. Colella and S. Coluccia
Volume 141 Nanoporous Materials III
Proceedings of the 3rd International Symposium on Nanoporous Materials,
Ottawa, Ontario, Canada, June 12-15, 2002
edited by A. Sayari and M. Jaroniec
Volume 142 Impact of Zeolites and Other Porous Materials on the New Technologies
at the Beginning of the New Millennium
Proceedings of the 2nd International FEZA (Federation of the European Zeolite
Associations) Conference, Taormina, Italy, September 1-5, 2002
edited by R. Aiello, G. Giordano and F.Testa
Volume 143 Scientific Bases for the Preparation of Heterogeneous Catalysts
Proceedings of the 8th International Symposium, Louvain-la-Neuve, Leuven,
Belgium, September 9-12, 2002
edited by E. Gaigneaux, D.E. De Vos, P. Grange, P.A. Jacobs, J.A. Martens,
P. Ruiz and G. Poncelet
Volume 144 Characterization of Porous Solids VI
Proceedings of the 6lh International Symposium on the Characterization of Porous
Solids (COPS-VI), Alicante, Spain, May 8-11, 2002
edited by F. Rodriguez-Reinoso, B. McEnaney, J. Rouquerol and K. Unger
Volume 145 Science and Technology in Catalysis 2002
Proceedings of the Fourth Tokyo Conference on Advanced Catalytic Science and
Technology, Tokyo, July 14-19, 2002
edited by M. Anpo, M. Onaka and H. Yamashita
Volume 146 Nanotechnology in Mesostructured Materials
Proceedings of the 3rd International Mesostructured Materials Symposium,
Jeju, Korea, July 8-11, 2002
edited by Sang-Eon Park, Ryong Ryoo, Wha-Seung Ahn, Chul Wee Lee and
Jong-San Chang
Volume 147 Natural Gas Conversion VII
Proceedings of the 7lh Natural Gas Conversion Symposium, Dalian, China, June
6-10, 2004
edited by X. Bao and Y. Xu
Volume 148 Mesoporous Crystals and Related Nano-Structured Materials
Proceedings of the Meeting on Mesoporous Crystals and Related Nano-Structured
Materials, Stockholm, Sweden, 1-5 June, 2004
edited by O. Terasaki
Volume 149 Fluid Catalytic Cracking VI: Preparation and Characterization of Catalysts
Proceedings of the 6th International Symposium on Advances in Fluid Cracking
Catalysts (FCCs), New York, September 7 - 1 1 , 2003
Edited by M. Occelli
Volume 150 Coal and Coal-Related Compounds
Structures, Reactivity and Catalytic Reactions
edited by T. Kabe, A. Ishihara, W. Qian,
I.P. Sturisna and Y. Kabe

Petroleum Biotechnology - Duhalt - (NAFTI - IR) PDF

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Petroleum Biotechnology - Duhalt - (NAFTI - IR) PDF

Încărcat de

Drepturi de autor:

Formate disponibile

Studies in Surface Science and Catalysis 151

First edition 2004

ISBN: 0 444 51699 9

Innovative new processes could be explored, such as methanol production from

K.M. Kirk wood

Use of Petroleum Biotechnology throughout the value

Sintef Materials and Chemistry, bDept. Marine Environmental Technology,

1. INTRODUCTION TO AN INTEGRATED APPROACH

looking into the possibility of using petroleum biotechnology as an integrated

Combined approaches of microbiology, biochemistry and DNA technology

functions. Such organisms can be applied in reservoirs for the production of

Environmental aspects/public awareness: Apart from providing technical

Value generation: This program will contribute to increasing and maturing

Fig. 1. Biotechnology throughout the value chain.

The main challenges are related to:

Acquiring new knowledge: In order to balance the beneficial and detrimental

2. MICROBIAL DNA FINGERPRINT TECHNIQUES IN EXPLORATION AND

Several studies have documented microbial communities in hot oil reservoirs

Fig. 2. System for characterization of microbes in exploration cores by culture-dependent and

2.1. Microbial diversity in oil reservoirs

2.1.1. Culture-independent methods

2.1.2. Culture-based methods

Thermophilic species of Thermotogales, Archaeoglobus, Thermoanaero-

2.1.3. Detection of specific microbes

2.1.4. Characterization ofmicrobial dynamics by microarrays

3. BIOREACTOR: POTENTIAL USE OF BIOCATALYSTS IN CRUDE

Until recently, research within oil biotechnology mainly focused on bio-

Reduction of the viscosity of heavy crudes through partial degradation of

3.1.1. Increased transportability by biocatalytic cleavage of heavy compounds

Biodegradation of asphaltenes seems to represent a more challenging

3.1.2. Demineralization - Biosorption of heavy metals

3.2. Biocatalytic refining, distillate quality improvements

3.2.1. Selective ring opening

Bioconversion of aromatic compounds in a real feedstock from crude oil in a

The P. fluorescence LP6a 21-41 was obtained by transposon mutagenesis

In order to apply the concept to a real industrial process, higher degrees of

Fig. 7. Efficient bioconversion by a mixed biocatalyst.

The PAH degradation pathway of Sphingomonas yanoikuyae DSM 6900

compounds and approximately 30 % of the sulfur containing substrates were

In future refinery processes this might replace the energy-expensive

Bioconversion for more

Fig. 10. Schematic of a bioreactor for continuous feed of LGO.

Continuous processes are well suited for multiphase processes. In the

Realistic cost of developing new technology. New technologies are often

4. WELL TREATMENTS TO SECURE CONTINUOUS PRODUCTION

Preventive medication could be defined as intelligent treatment concepts

Syncrude Canada OPEX

4.1. Preventive treatment: Increased productivity by self-generating- or

In situ production of treatment chemicals

Fig. 12. Schematic view of in situ production of treatment chemicals.

4.2 Green treatment products

Down hole: Work is in progress, introducing the corresponding synthetic

5. NEW APPLICATION OF EXTREMOPHILES IN OIL RELATED

5.1. Bioprospecting of the gene pools

5.2. Thermophilic/extremophilic enzymes

Such enzymes are tailor-made as catalysts in industrial processes performed at

5.3. Future prospective

[11] C. Tardy-Jackuenod, P. Caumette, R. Matheron, C. Lanau, O. Arnauld and M. Magot,

Petroleum biorefining: the selective removal of sulfur,

The quality of petroleum is progressively deteriorating as the highest quality

The geochemical conversion of organic matter into petroleum is a slow

1.1 Composition of crude oils

of dibenzothiophene are the most common organosulfur compounds typically