LNG Industry May 2013

L
N
G
I
N
D
U
S
T
R
Y

|

M
a
y
/
J
u
n
e

2
0
1
3

w
w
w
.
e
n
e
r
g
y
g
l
o
b
a
l
.
c
o
m
May / June 2013
LNG_MayJune_2013_OFC.indd 1 21/05/2013 12:50
www.fmctechnologies.com
Copyright FMC Technologies, Inc. All Rights Reserved.
FMC Technologies invented the rst ofshore LNG loading technology. Today we are
developing new solutions built on proven components. Our Articulated Tandem Ofshore
Loader (ATOL) safely performs high-velocity LNG transfers in severe conditions with
waves up to 18 feet (5.5 meters). Our Ofshore Loading Arm Footless (OLAF) side by
side transfer solution accommodates massive new FLNG freeboards in the range of
82 feet (25 meters). And for tomorrow? Were practically there already.
Meeting
challenges
you havent even thought of yet.
LNG_MayJune_2013_IFC.indd 1 21/05/2013 12:58
LNG Industry is audited by the Audit Bureau of Circulations (ABC).
An audit certificate is available on request from our sales department.
Copyright Palladian Publications Ltd 2013. All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying,
recording or otherwise, without the prior permission of the copyright owner. All views expressed in this journal are
those of the respective contributors and are not necessarily the opinions of the publisher, neither do the publishers
endorse any of the claims made in the articles or the advertisements. Printed in the UK.
CONTENTS
ISSN 1747-1826
O
N

T
H
I
S

M
O
N
T
H
S

C
O
V
E
R
L
N
G
IN
D
U
S
T
R
Y
|
M
/J
2
0
1
3
ener
lo
balco
m
May / June 2013
LNG MayJune 2013 OFC indd 1 21/05/2013 12:50
LNGs LNG s
European core
David Stokes and Oliver Spinks,
Timera Energy, UK, examine
European LNG dynamics in the
context of the global gas market.
MAY/JUN 2013 LNGINDUSTRY 11
T
he global LNGmarket is entering a period of rapid
evolution. Market growth is being driven by a structural
shift towards natural gas as a cleaner fuel, as well as an
increase in gas import dependence in Asia and Europe. In order to
meet this increase in demand, global liquefaction capacity may
need to double by 2030, requiring an estimated US$ 1 trillion of
newinvestment. At the same time, pronounced inter-regional
price differentials and improvements in gas hub liquidity have
been a catalyst for LNGcontract re-negotiation and growth in the
shorter termtrading of cargoes.
Europe will play a central role in shaping this growth phase in
the LNGmarket. In turn, LNGpricing and flows are becoming
increasingly important drivers of the European gas market.
European gas suppliers are diversifying portfolios of pipeline gas
supply to include contracted LNGand access to regasification
capacity. As a result, LNGhas become a key source of
incremental supply into European gas portfolios. LNGimport
volumes are still lowrelative to pipeline imports, but they have a
disproportionate influence on marginal pricing at European gas
hubs.
However, Europe cannot be considered in isolation. In order
to understand European LNGdynamics it is important to take a
step back and consider Europe in a global gas market context.
European interaction with the
global gas market
The global LNGmarket is still in a stage of relative infancy. Only
about 10%of global gas consumption is currently satisfied by
LNG, and only a relatively small volume of LNGsupply has the
contractual flexibility to respond optimally to price dynamics.
The constraints around supply flexibility are reflected in the
current regional price divergence across Asia (tight market
post Fukushima), Europe (broadly tracking oil-indexed contract
MAY/ JUN
2013
10
16 Sea change
Nick Prowse, Scott McCabe and Rod Chooramun,
Norton Rose, look at trends in the LNG shipping industry.
19 Unlocking markets
Cyril Widdershoven and Bas van den Beemt, TNO, the
Netherlands, explain why technical-economic cooperation
is needed between European organisations and Arab gas
producers to kick-start small scale LNG markets in the
Middle East.
22 Jumping the hurdles
Jrgen Harperscheidt, TGE Marine Gas Engineering,
Germany, looks at the technical issues facing LNG bunkering.
29 Membrane market
Julie Sakhrai, Lorenz Claes, Stphane Maillard and
Thomas Crmire, GTT (Gaztransport & Technigaz), France,
describe the benefits of using membrane systems along the
LNG chain.
33 From flare to fuel
David A. Franklin, WorleyParsons, USA, establishes the case
for mini-LNG.
39 Flexible processing
Rebecca Liebert, UOP, a Honeywell Company, provides a
guide to selecting appropriate gas processing technologies.
45 Packing for CO
2
removal
Ralph H. Weiland and Nathan A. Hatcher, Optimized Gas
Treating, Inc. and Jenny Seagraves, INEOS Oxide, USA,
discuss the benefits of choosing the right packing for
absorbers and regenerators in a gas processing facility.
53 Mercury rising
John Markovs, Adsorption Solutions LLC, USA, and
Robert W. Soffel, Selective Adsorption Associates Inc., USA,
examine solutions to mercury problems in LNG plants.
57 The cryogenic challenge
Barry Wilder, Severn Glocon, UK, looks at how to maximise
cryogenic control valve performance.
61 Take the weight off
Femke Schaefer, Bronswerk Heat Transfer, the Netherlands,
looks at size, weight and power reduction of process
equipment in LNG liquefaction and regasification.
64 Good vibrations
Israel Ferreres, Cryoperl Australia Pty Ltd, Australia, outlines
the benefits of vibration prior to operation.
69 Detuning plants
Attilio Brighenti, Systems and Advanced Technologies
Engineering S.r.l. (S.A.T.E.), Italy, and Luigino Vitali, SAIPEM,
Italy, discuss how riser singing generated by inner wall
turbulence of flexible risers can be a severe problem for LNG
offloading and regasification plants.
74 Pulling out the stops
Alfred Hbner, Ultratug, Chile, explains the beginning of LNG
terminal services in Chile and how this has impacted the local
market.
79 Maintaining tension
Blaine Dempke, Markey Machinery Company Inc., USA,
discusses new winch technology that helps to improve tanker
transit safety.
83 Predicting the weather
Olivier Gorieu and Max Nussbaum, GDFSUEZLNG, France,
and Dr Christophe Messager, Extreme Weather Expertises,
France, look at extreme weather conditions generated by
Mesoscale Convective Systems over the Cameroon LNG
production site.
03 Comment
05 LNG news
10 LNGs European core
David Stokes and Oliver Spinks, Timera Energy, UK, examine
European LNG dynamics in the context of the global gas
market.
Qatargas, established
in 1984, pioneered
the LNG industry in
Qatar. Today, Qatargas
is the largest LNG
producing company in
the world, with an LNG
production capacity
of 42 million tpy. It is
through our operating
excellence and our
innovation in global
energy that we are
safely and reliably
delivering LNG to our
customers around the
globe.
LNG_MayJune_2013_01-02.indd 1 23/05/2013 15:02
1
2
1
9
_
e
Whatever the impurity, whatever the composition, Air Liquide Global E&C Solutions
has the right treatment.
, ' _ ' ,_
'
The composition of natural gas varies
tremendously: almost every source
contains a diferent blend of impurities.
The options for treatment are almost
as diverse. That's why ofering a solution
specifcally designed for your gas feld
is crucial. We as your partner of choice
provide solutions for all types of natural gas,
including associated and unconventional
gas, from a single source. Customised
and em cient.
Come and visit us at:
MIOGE 2013, 25 - 28 June
in Moscow, Russia
Pavilion 1 | Booth A 524
www.engineering-solutions.airliquide.com
LNG_MayJune_2013_01-02.indd 2 23/05/2013 12:06
COMMENT
CALLUM OREILLY EDITOR
C
O
N
T
A
C
T

I
N
F
O
R
M
A
T
I
O
N
Managing Editor
James Little
james.little@lngindustry.com
Editor
Callum OReilly
callum.oreilly@lngindustry.com
Editorial Assistant
Peter Farrell
peter.farrell@lngindustry.com
Advertisement Director
Rod Hardy
rod.hardy@lngindustry.com
Advertisement Manager, USA/Canada
Chris Atkin
chris.atkin@lngindustry.com
Advertisement Manager, EMEA
John Baughen
john.baughen@lngindustry.com
Editorial/Advertisement Offices, Palladian Publications Ltd, 15 South Street, Farnham, Surrey, GU9 7QU, ENGLAND,
Tel: +44 (0) 1252 718 999 Fax: +44 (0) 1252 718 992 Website: www.energyglobal.com
Production
Stephen North
stephen.north@lngindustry.com
Website Editor
Callum OReilly
callum.oreilly@energyglobal.com
Circulation Manager
Vicki McConnell
victoria.mcconnell@lngindustry.com
Subscriptions
Laura Cowell
laura.cowell@lngindustry.com
Reprint / Marketing Assistant
Catherine Gower
catherine.gower@lngindustry.com
Publisher
Nigel Hardy
LNG Industry Subscription rates:
Annual subscription: 50 UK including postage
60/d85 overseas (postage airmail)
US$ 85 USA/Canada (postage airmail).
Two year discounted rate: 80 UK including postage
96/d136 overseas (postage airmail)
US$ 136 USA/Canada (postage airmail).
Subscription claims:
Claims for non receipt of issues must be made within
3 months of publication of the issue
or they will not be honoured without charge.
Applicable only to USA & Canada.
LNG INDUSTRY (ISSN No: 1747-1826, USPS No: 006-760) is published six
times per year: February, April, June, August, October and December, by
Palladian Publications and is distributed in the USA by by SPP, 17B S Middlesex
Ave, Monroe NJ 08831. Periodicals postage paid at New Brunswick, NJ.
POSTMASTER: send address changes to LNG INDUSTRY, 17B S Middlesex
Ave, Monroe NJ 08831.
Uncaptioned Images courtesy of www.bigstockphoto.com
S
mall scale LNG was one of the big talking points at this
years LNG 17 Conference in Houston, Texas. In addition
to hosting its own conference session, the topic of LNG
as a transportation fuel was also creating quite a stir in the
exhibition hall. The LNG for transport exhibit pavilion was
bursting to the brim for each of the 20+ presentations that
took place over the three day seminar programme, and the
exhibition floor was full of companies showcasing their latest
technology to meet this rapidly developing sector of the
industry.
Previously, I have written comments alluding to LNG fuel
as the industrys next big thing, and recent developments
would appear to support this line of thought. In the last few
months, several major proponents of this market have started
to aggressively pursue infrastructure development.
In April, Royal Dutch Shell finalised an agreement with
TravelCenters of America (TA) to develop a network of LNG
fuelling stations across the US for heavy-duty road transport
customers. The proposed plans will see the construction of at
least two LNG fuelling lanes and a storage facility at up to 100
existing TA and Petro stopping centres. Shell had previously
announced that it is to invest in two small scale LNG
production units that will form the basis of two LNG transport
corridors in the Great Lakes and Gulf Coast regions. In addition
to this, the company is also developing LNG stations at
Flying J truck stops in Alberta, Canada, and it will charter the
worlds first inland barges that run solely on LNG, set to sail on
the Rhine later this year.
Chinas ENN Group has also put in place plans to establish
a network of natural gas fuelling stations for trucks across the
US. The group has partnered with Utah-based CH4 Energy to
form Transfuels LLC, which operates as Blu LNG. It is reported
that the company will build around 50 natural gas filling
stations this year, roughly equal to the number of stations that
Clean Energy Fuels Corp. is expected to open.
On the tracks, BNSF Railway recently announced that it is
to begin testing a number of locomotives using LNG later this
year, with its Chief Executive, Matthew Rose, describing the
use of LNG as a fuel as a potential transformational change
for [the] industry.
The two main drivers for this recent flurry of activity are
cost and the environment. The abundance of natural gas in
North America has driven prices to ten-year lows, and LNG has
the potential to significantly reduce greenhouse gas emissions.
This latter point is particularly important for the marine
industry, with ship owners in Europe and North America under
pressure to comply with strict emission regulations due to
come into force in 2015. The European Commission has also
put aside 2.1 billion to equip 139 seaports and inland ports
with LNG bunker stations by 2025, as part of the EUs new
clean fuels strategy.
There are, of course, some daunting hurdles to overcome
before this market can really take-off, not least the cost of
building the required infrastructure. However, it seems that the
potential benefits are encouraging companies to push forward
with their strategies ahead of the market, as they attempt to
knock down these hurdles.
This issue of LNG Industry takes a closer look at the topic
of LNG as a transportation fuel, with particular focus on the
marine industry. You can also keep up to date with the latest
developments in this burgeoning market via our website:
www.energyglobal.com.
LNG_MayJune_2013_03-04.indd 3 23/05/2013 15:07
versalis chemistry to evolve
versalis the new face of chemicals
versalis teams up with you over the ever-expanding challenges around the industry.
we stand by you in driving innovative solutions with marketable end alongside
a timely advantage. at versalis we believe that excellence goes beyond delivery;
it reaches out to after-sale assistance.
O 1306 210 297 LNG I d t i i dd 1 22/04/13 11 49
LNG_MayJune_2013_03-04.indd 4 23/05/2013 12:09
LNGNEWS
USA
Freeport LNG receives DOE export approval
T
he Energy Department has conditionally authorised
Freeport LNG Expansion, L.P. and FLNG Liquefaction,
LLC (Freeport) to export domestically produced LNG
to countries that do not have a Free Trade Agreement
(FTA) with the US from the Freeport LNG terminal on
Quintana Island, Texas. Freeport previously received
approval to export LNG from this facility to FTA countries
in February 2011.
Subject to environmental review and final regulatory
approval, the facility is conditionally authorised to
export at a rate of up to 1.4 billion ft
3
/d of natural gas
for a period of 20 years. The Department granted the
first authorisation to export LNG to non-FTA countries in
May 2011 for the Sabine Pass LNG terminal in Cameron
Parish, Louisiana at a rate of up to 2.2 billion ft
3
/d.
Michael S. Smith, Chief Executive Officer of Freeport
LNG, welcomed the news: We applaud the DOE on its
thorough and balanced analysis of the public benefits
of LNG exports to the United States. The overwhelming
evidence in favour of LNG exports generated during the
DOEs two-and-a-half year review, definitively confirms
that the DOE should act swiftly to approve additional
pending LNG export applications.
UK
South Hook Gas welcomes the arrival of its 300
th
cargo
S
outh Hook Gas has cemented its position as the UKs
largest LNG terminal shipper with the arrival and safe
off-loading of its 300
th
cargo. The Umm Slal recently sailed
from the South Hook LNG terminal in Milford Haven.
Sheikh Saoud Al-Thani, South Hook Gas Company
Chairman, said: The arrival of LNG has transformed gas
supply to the UK as it can now access supplies from all
over the world and this has improved energy security.
The arrival of our 300
th
cargo demonstrates our long-term
commitment to the UK.
As a Q-Max vessel, the Umm Slal is one of the biggest
LNG vessels in the world, and can provide enough gas to
meet Londons needs for a week.
South Hook Gas has supplied gas every day since
commencing operations and can supply around 20% of the
UKs gas needs.
Uruguay
GDF SUEZ wins regas terminal contract
G
DF SUEZ has been awarded the contract to build an
LNG regasification terminal in the port of Montevideo
by the government of Uruguay, at an estimated cost of
US$ 1.125 billion. The terminal will have a processing capacity
of up to 10 million m
3
/d of LNG.
The LNG project will consist of a floating LNG regasification
vessel and will also include construction of a breakwater and
works to connect the vessel to Uruguays domestic network of
gas pipelines.
Meanwhile, GDF SUEZ has also entered into a joint venture
agreement with Sempra Energy, Mitsubishi and Mitsui to develop,
finance and build the Cameron LNG liquefaction plant in Louisiana,
Texas, USA. Mitsubishi, Mitsui and GDFSUEZ will all have a 16.6%
equity stake, while Sempra Energy will retain a 50.2% stake.
GDF SUEZ already signed a 20-year liquefaction supply
agreement for 4 million tpy of LNG with Sempra last year and
the three minority stakeholders have collectively subscribed
the full capacity of the three-train facility.
The joint venture agreements are subject to a final
investment decision by each party, to final permit
authorisations, and to securing financing commitments, all
of which are expected by early 2014. The new LNG plant is
expected to start operations in late 2017.
LNG_MayJune_2013_05-09.indd 5 24/05/2013 09:12
LNGNEWS
6 LNGINDUSTRY MAY/JUN 2013
N
E
W
S

H
I
G
H
L
I
G
H
T
S
Scan for the
Energy Global
iPhone/ iPad App
To read more
about these
stories go to:
Get the free mobile app at
http:/ / gettag.mobi
USA
Golden Pass affiliates sign Commercial
Framework Agreement for exports
G
olden Pass Products LLC (GPP) shareholders affiliates
Qatar Petroleum International (QPI) and ExxonMobil
have signed a Commercial Framework Agreement for the
proposed liquefaction project at the Golden Pass LNG
facility.
This agreement establishes the commercial framework
to sell the 16.5 million tpy output of LNG from the facility.
This agreement sets out a highly competitive
commercial blueprint for Golden Pass Products, with a
commitment that builds on the unique combined strengths
of QPI and ExxonMobil throughout the global downstream
LNG value chain, said Bill Collins, president of Golden
Pass Products.
Golden Pass products plans to invest approximately
US$10billion to build the liquefaction facility. The
company has already received US DOE authorisation for
exports to Free Trade Agreement (FTA) countries, and is
awaiting DOE approval to export to non-FTA countries.
Clean Air Power welcomes Gasrecs Bio-LNG station
FMC chooses RINA technology qualification
Approve remaining LNG export permits, urges API
Australia
Severn Glocon wins Ichthys contract
C
ontrol valve specialist Severn Glocon has been awarded
the contract to supply 536 engineered control valves
for the processing plant on the Ichthys project onshore LNG
facilities. Included in the contract are cryogenic valves of up
to 42 in. bore, weighing around 20 t, as well as valves with
pressure ratings of up to 2500 ANSI.
This is the largest and one of the most technically
challenging contracts in the firms 50 year history. The valves
will deal with volatile fluids at temperatures as low as -160 C,
and they need to handle extreme requirements surrounding
cryogenics, thermal dynamics and velocity control.
EPC activities for the project are being undertaken by
the JKC Australia LNG Pty Ltd (JKC). The project is now in the
construction phase and the first valves are being shipped.
Manufacture and supply of valves will continue into 2014.
The scope, scale and harsh demands of this project
command highly-engineered, technically advanced
hardware, says Roger Spiers, Commercial Manager at Severn
Glocon.
LNG_MayJune_2013_05-09.indd 6 23/05/2013 16:51
LNG_MayJune_2013_05-09.indd 7 23/05/2013 16:51
LNGNEWS
D
I
A
R
Y

D
A
T
E
S
UK
Gasrec to open UKs first Bio-LNG filling station
G
asrec is opening the UKs first Bio-LNG filling station in
Daventry, which will store enough fuel to fill 700 HGVs a
day. Bio-LNG is Gasrecs proprietary blend of LNG and liquid
biomethane (LBM), which is produced from organic matter
such as household food waste. Gasrec is Europes largest
producer of this fuel.
Daventry is the first site identified by Gasrec as part of its
national Bio-LNG refuelling network planned to be up and
running by the end of 2015.
Transport Minister Norman Baker said: This
announcement represents a significant step towards
achieving carbon and air quality benefits and I am very
pleased that demonstration fleets from the Governments
Low Carbon Truck Demonstration Trial will be refuelling from
this site.
Compared with pure diesel equivalents, Bio-LNG can cut
fuel costs by 20 30% and CO
2
by a minimum of 20%, while
delivering a 90% reduction in NO
x
and particulate matter
emissions.
17 - 19 June 2013
Small-Mid Scale LNG Summit
Amsterdam, the Netherlands
www.smallmidlng.com
3 - 6 September 2013
SPE Offshore Europe 2013
Aberdeen, UK
www.offshore-europe.co.uk
24 - 26 September 2013
LNG Global Congress
London, UK
www.lnggc.com
30 September - 3 October
2013
Pump/Turbomachinery Symposia
Houston, Texas, USA
turbolab.tamu.edu/articles/
pump_turbomachinery_symposia
14 - 16 October 2013
LNG Tech Global Summit
Barcelona, Spain
www.lngsummit.com
18 - 21 November 2013
World LNG Summit
Paris, France
world.cwclng.com
USA
BPC reports on impact of US LNG exports
A
report by the Bipartisan Policy Center (BPC) has
concluded that LNG exports are unlikely to have a great
impact on domestic gas prices in the US.
Levels of LNG exports will be determined by the domestic
price of natural gas, costs for transportation and liquefaction,
and the price and level of demand for natural gas in target
foreign markets. One of the fundamental drivers in this
equation is the US price of natural gas, which will influence
LNG export levels far more than LNG exports will influence
domestic prices.
The BPC estimates LNG exports will range from
26.4billion ft
3
/d by 2030. These estimates take into
account the barriers facing LNG projects, including the high
cost of building facilities, liquefying and transporting natural
gas overseas, and finding US producers willing to enter into
long-term low price contracts. Overall, the US is projected to
become a net exporter of natural gas between 2017 and 2021.
LNG_MayJune_2013_05-09.indd 8 23/05/2013 16:51
LNG_MayJune_2013_05-09.indd 9 23/05/2013 16:51
LNGs LNG s
European core
David Stokes and Oliver Spinks,
Timera Energy, UK, examine
European LNG dynamics in the
context of the global gas market.
LNG_MayJune_2013_10-15.indd 10 20/05/2013 11:00
T
he global LNG market is entering a period of rapid
evolution. Market growth is being driven by a structural
shift towards natural gas as a cleaner fuel, as well as an
increase in gas import dependence in Asia and Europe. In order to
meet this increase in demand, global liquefaction capacity may
need to double by 2030, requiring an estimated US$ 1 trillion of
new investment. At the same time, pronounced inter-regional
price differentials and improvements in gas hub liquidity have
been a catalyst for LNG contract re-negotiation and growth in the
shorter term trading of cargoes.
Europe will play a central role in shaping this growth phase in
the LNG market. In turn, LNG pricing and flows are becoming
increasingly important drivers of the European gas market.
European gas suppliers are diversifying portfolios of pipeline gas
supply to include contracted LNG and access to regasification
capacity. As a result, LNG has become a key source of
incremental supply into European gas portfolios. LNG import
volumes are still low relative to pipeline imports, but they have a
disproportionate influence on marginal pricing at European gas
hubs.
However, Europe cannot be considered in isolation. In order
to understand European LNG dynamics it is important to take a
step back and consider Europe in a global gas market context.
European interaction with the
global gas market
The global LNG market is still in a stage of relative infancy. Only
about 10% of global gas consumption is currently satisfied by
LNG, and only a relatively small volume of LNG supply has the
contractual flexibility to respond optimally to price dynamics.
The constraints around supply flexibility are reflected in the
current regional price divergence across Asia (tight market
post Fukushima), Europe (broadly tracking oil-indexed contract
LNG_MayJune_2013_10-15.indd 11 20/05/2013 11:00
supply) and the US (awash with unconventional domestic gas
production).
Only a decade ago the European gas market existed in
relative isolation. But growth in LNG flows and price arbitrage
have connected Europe with the gas markets of the US and Asia.
At the same time, the divergence of European gas hub prices
below oil-indexed contract prices has caused the rapid
development of hub liquidity and market price signals. These
factors have resulted in gas hub pricing becoming the key driver
of commercial decision making across European gas portfolios.
This is despite the fact that the majority of gas is still sourced via
long-term oil-indexed supply contracts.
Europe currently accounts for about 20% of global LNG
demand compared to 70% for Asia. But Europe punches above
its weight when it comes to its impact on global LNG pricing. The
interaction between the European and global gas markets is well
illustrated by looking at the evolution of global price benchmarks
over the last half decade, shown in Figure 1.
Price evolution over this period can be broadly split into three
phases:
Commodity super cycle (2007 2008): global gas prices
tracked oil prices through the boom/bust of the commodity
super cycle. At the same time, there was broad gas price
convergence across Europe, Asia and the US in line with
contract oil-indexation.
Global supply glut (2009 2010): the financial crisis, new
global liquefaction capacity and the US shale gas revolution
rapidly combined to create a global gas supply glut.
European gas hub prices fell below oil-indexed contract
prices as surplus LNG flowed into Europe, driving hub prices
down towards Henry Hub (HH) levels. Spot Asian LNG prices
were strongly influenced by the transport cost differential to
UK NBP.
Fukushima led divergence (2011 2013): a recovery in oil
prices, followed by aggressive Japanese buying of LNG
after the Fukushima disaster, opened up a substantial Asian
LNG price premium. European hub prices reconnected with
oil-indexed contract prices. This was helped by a reduction in
LNG flow into Europe as cargoes were diverted to Asia.
Europe lies in close proximity to the key producers in the
Middle East and Africa, so it is an attractive destination for LNG
sales, given relatively low transport costs. The supply of LNG into
Europe is also relatively flexible compared to Asian supply.
Location and supply flexibility mean that Europe plays an
important role in influencing the marginal pricing of LNG. Liquidity
at traded hubs, such as the UK NBP and Dutch TTF, provide key
price signals to the global gas market.
For the time being, US gas prices are in a world of their own
as robust domestic shale gas production flows into the HH spot
market. However, US and European gas prices are re-converging
on a forward basis, in anticipation of the ramp up of US LNG
exports from the middle of this decade and the potential for US
production costs to rise. The volume of US exports remains
subject to regulatory review, but liquefaction terminal developers
in the Gulf of Mexico are aggressively marketing HH indexed
supply contracts to European buyers. This should increase the
relationship between HH and European gas hub prices, but also
presents some key challenges for European LNG buyers as set
out in the following case study.
Case study: buying LNG supply
from a US exporter
A US export contract provides access to HH indexed gas. The
contract has the advantage of being a supply source that can be
managed (both physically and financially) against a liquid hub
and forward market. But contracting gas from a US exporter will
typically involve paying a premium to hub prices (e.g.115%HH)
as well as a capacity fee. So there is an important value
consideration in signing the contract. There is also an important
risk consideration since the contract involves buying gas on a
different price basis (HH) to European end user demand
(NBP/TTF), e.g. for power stations and retail load.
Contract value is driven by regional price spreads between
the US (HH), Europe (NBP/TTF) and Asia. This is in part a question
of the evolution of global gas market fundamentals. For example,
if the flow of gas out of the US
remains constrained by a lack of US
liquefaction capacity then the
transatlantic price spread (NBP minus
HH) may exceed the transport
differential for sustained periods (i.e.
access to HH gas has intrinsic value).
On the other hand, as US export
capacity becomes less constrained
the correlation between NBP and HH
should increase. But as well as
fundamental considerations there are
also specific portfolio value issues.
From the perspective of a
European supplier, HH indexation
offers a liquid/hedgeable alternative
to oil indexed supply, which can be
attractive if the supplier is constrained
in their ability to buy gas indexed to
NBP/TTF (i.e. the price basis for sale to
end users). A US export contract also
adds hedgeable hub based supply
flexibility to an LNG portfolio that can
unlock portfolio constraints; e.g.the
Figure 1. Key global gas price benchmarks (source: Timera Energy).
LNG_MayJune_2013_10-15.indd 12 23/05/2013 10:19
OIL-FREE . MAX MTBO
LABY
UNIQUE SOLUTION FOR YOUR

BOIL-OFF GAS APPLICATION
Best for LNG, LPG, LEG and other
hydrocarbon gases
The ultimate contactless labyrinth
sealing technology no piston rings
no wear
Flexible compression of boil-off
gas such as LNG at suction
temperatures down to -170 C (-250F)
Gas-tight design for zero emissions
For immediate start-ups without
pre-cooling of the compressor

Highly reliable design and com-
pressor components for highest
availability of your compressor
Count on our experience:
Thousands of compressors in
operation, hundreds of long-term
customers
www.recip.com/laby
YOUR BENEFIT:
LOWEST LIFE CYCLE COSTS
LNG_MayJune_2013_10-15.indd 13 23/05/2013 10:54
HH indexed gas could be swapped within a portfolio for a
destination clause constrained Middle Eastern supply source and
then sold on to an Asian buyer at a premium.
Europes role in LNG spot and
contract markets
LNG delivered to Europe is predominantly sourced under
long-term oil-indexed contracts. Contract terms vary by delivery
location and seller, but typically feature indexation to either a
basket of oil products or crude. Hub indexed contracts have
become increasingly common in North West Europe as hub
prices have fallen below oil-indexed contract prices and hub
liquidity has developed. To date, hub contract indexation has
been focused on the UK NBP, but US HH is becoming more
important, given that it is the basis for US export contract pricing
(as set out in the previous case study).
While long-term oil-indexed contracts still dominate the LNG
market, there has been a sharp increase in LNG spot market
trading volumes over the last five years. This is in part the result of
an increase in new uncontracted LNG supply, particularly from
Qatar. But global spot price signals have also become more
pronounced and volatile, increasing the commercial incentives for
long-term contract buyers to re-negotiate increases in contract
flexibility. As a result, there is a growing volume of contracted
LNG with flexibility that can be exercised against spot prices (e.g.
via cargo diversions). This is improving the response of LNG flows
to market price signals.
The spot market for LNG may still be small relative to the
contract market, but it is important because prices at the margin
drive the behaviour of flexible LNG flows. The interaction
between the LNG spot market and the European gas market is
well illustrated by the price dynamics over the period from
summer 2012 to winter 2013, as described in the LNG spot
market gyrations (2012 2013) sidebar.
Price behaviour over this period illustrates the inelasticity of
spot LNG supply. The resulting spot price volatility is caused by a
limited volume of flexible cargoes available to respond to spot
price signals. Volatility is exacerbated by significant shorter term
swings in demand from large buyers with lower levels of contract
cover such as China, India, Brazil and Argentina. These conditions
are set to remain for the next three to five years, with LNG spot
price volatility likely to continue. This has direct implications for
the European gas market.
The relationship of LNG pricing in relation to other European
gas imports is summarised in Figure 2.
The interaction between European gas hub pricing and the
spot LNG market can be seen over summer 2012 as European
hubs provided a soft floor for surplus LNG cargoes in the LNG
spot market. At the point that a flexible exporter, such as Qatar,
can get a better netback price at European hubs than in Asia, spot
LNG flows back into Europe, alleviating downward price pressure.
In turn, Asian or South American buyers need to pay a price
premium above NBP, plus the relevant transport differential, to
attract cargoes away from Europe.
The prospect of having to compete for LNG with Asia and
South American buyers is a key concern for European gas
suppliers. A gas price squeeze, e.g. from a cold winter or supply
disruptions, may drive gas hub prices and volatility sharply higher.
This happened in March 2013, where UK NBP prices spiked as
LNG became the marginal source of supply. The risk of Europe
having to compete for LNG in the global market also influences
forward market price behaviour and this risk will rise as Europe
becomes increasingly reliant on LNG as an incremental supply
source. Europe is facing up to the reality that pricing in its gas
market will at times be directly subject to the volatility of the
global spot market.
The business impact of LNG
market evolution
Europe is set to play a core role in the evolution of the LNG
market as an important incremental source of demand. European
hub price signals will also be increasingly important in driving the
pricing dynamics and flow of LNG. This environment presents
two clear commercial and analytical challenges for businesses
with a European LNG market exposure.
The first challenge is to understand the evolution of the
dynamics of LNG pricing. This is clearly a problem that is relevant
to companies with a direct exposure to the LNG supply chain. But
LNG spot market gyrations (2012 2013)
The LNG flow and pricing dynamics described below can be observed in Figure 1.
Spring peak
The spring of 2012 was a sellers market. Asian spot prices exceeded US$ 18/million Btu in May 2012 as robust demand
from the large Asian spot buyers, China and India, coincided with aggressive post Fukushima Japanese buying to hedge
gas-fired power generation. The flow of LNG into Europe had virtually dried up as contracted cargoes were re-routed (or in
some cases re-loaded) for diversion to the higher priced Asian market.
Summer slump
Over several weeks between May and July 2012, spot prices fell precipitously by US$ 6/million Btu, with the debt
crisis intensifying and Asian buyers well hedged in preparation for the summer. Spot LNG flows into Europe suddenly
re-emerged, as the incentive to divert contracted supply evaporated and surplus uncontracted LNG cargoes were sold into
European hubs. European gas hubs played a key role in supporting LNG price levels around the US$ 12/million Btu mark.
Winter rebound
With the onset of winter, spot prices rebounded back towards US$ 20/million Btu as quickly as they had fallen into the
summer. But this time the aggressive buying was from South America, specifically Brazil, which was struggling to replace
low hydro production with gas-fired power. In turn, this forced Chinese and Indian buyers to pay a premium to attract spot
cargoes away from the Atlantic Basin. European contract diversion, particularly Spanish LNG to Brazil and Argentina, has
returned as a dominant driver.
LNG_MayJune_2013_10-15.indd 14 20/05/2013 11:00
it is also important for any company with exposure
to the European gas market, given the increasing
influence of the LNG market on the level and
volatility of hub prices. Analysis of LNG price
dynamics is not a task that is well suited to
traditional gas supply and demand modelling, given
the global nature of market drivers. It is more
powerful (and less complex) to develop a top down
analytical approach that focuses on the impact of
incremental supply and demand on the marginal
pricing of LNG.
The second challenge is understanding how LNG
asset value, e.g. a supply contract or re-gas capacity,
interacts with the value of other assets in a gas
portfolio. The interdependent nature of LNG portfolio
flexibility means that an investment decision in a
single asset or contract usually cannot be made
without analysing its impact on portfolio value. When
it comes to LNG portfolio value, the whole does not
equal the sum of the parts. So developing an
effective portfolio valuation capability provides a key
foundation for value creation from investing in and
originating LNG assets and contracts.
These challenges are exacerbated by the scale
and pace of change in the LNG market. This has been
well illustrated by the events of the last five years, with a
progression from commodity supercycle, to gas supply glut, to a
post Fukushima world. With little new liquefaction capacity
coming online before 2015, demand dynamics are likely to shape
the next phase of LNG market evolution. These conditions do not
lend themselves to making commercial decisions on the back of
LNG market forecasts. The path of LNG market evolution may be
uncertain, but it is a little easier to sleep at night with a confident
understanding of the changing impact of LNG market dynamics
on portfolio value.
Figure 2. LNG and pipeline gas import pricing into Europe (source: Timera
Energy).
LNG_MayJune_2013_10-15.indd 15 24/05/2013 09:50
SEA CHANGE
Nick Prowse, Scott McCabe and
Rod Chooramun, Norton Rose,
look at trends in the LNG
shipping industry.
LNG_MayJune_2013_16-18.indd 16 20/05/2013 11:03
D
uring the past decade, LNG has
increasingly been at the forefront
of meeting global energy demand,
with transportation by LNG carriers being an
essential component in the LNG supply chain.
In this article, Norton Rose considers certain key
trends in LNG shipping.
Major expansion of the
worlds LNG carrier
fleet
Global production of and demand for LNG is
rising, driving expansion of the LNG shipping
industry. The number of LNG exporters
continues to increase with many countries
including the US and Canada currently planning
to monetise their gas reserves, and access
global gas markets, using LNG. According to
the International Energy Agency, LNG plants
with capacity to produce a total of 108 billionm
3

of LNG are under construction and another
64billion m
3
of capacity is planned. Chevron
sees LNG demand growing to approximately
450million tpy by 2025 and Shell estimates
that global demand for gas in 2050 (mainly
served by LNG rather than pipeline gas) will
be double that of 2010. This growth in LNG
production will require further significant
expansion of the worlds LNG carrier fleet.
Reports suggest that in addition to the 97 LNG
carriers currently on order, a further 120 carriers
may need to be constructed by the end of
the decade to cater for the new LNG volumes
predicted to come on stream in Australia,
Indonesia, East Africa, Eastern Mediterranean,
Russia, US, Canada and other parts of the
world. Exact requirements are difficult to
predict. Some commentators have suggested
that with approximately 39million tpy of LNG
exports from the US (including under the
reported offtake agreements at Sabine Pass and
Cameron LNG, tolling agreements at Freeport
and preliminary agreements at Cove Point), at
least 65 LNG carriers will be required for US
LNG exports alone. Many more may be required
depending on the final number of US LNG
export projects, with some believing there will
be 80 million tpy of LNG exports from the US.
Flexible LNG shipping
facilitates major change
in global gas markets
Asia is still the largest market for LNG with
countries such as Japan, South Korea and
Taiwan continuing to import significant
quantities. In particular, following the Tohoku
earthquake in March 2011 and the subsequent
issues arising from the Fukushima Dai-ichi
nuclear disaster, Japan has moved away from
nuclear power generation, with only two of its
nuclear power plants currently operational, and
has increased its gas-fired power generation.
As a result, nearly 40% of current global LNG
production is destined for Japan, the worlds
largest LNG importer. This situation looks
set to continue. BG Group, for example, has
LNG_MayJune_2013_16-18.indd 17 20/05/2013 11:03
suggested that Japanese nuclear restarts will be very slow,
possibly with only two to three facilities starting up in 2013.
Tokyo Gas has indicated that Japanese LNG demand may be
as high as 94million tpy by 2020, even in a scenario where
the nuclear plants are restarted. This surge in Asian demand
has been accompanied by a significant decline in demand for
LNG in the US, with the domestic market over supplied by
gas produced from shale, and a substantial slump in demand
for gas and LNG in Europe. This fundamental shift of LNG
volumes to Asia required a dramatic change in deployment of
the worlds LNG carrier fleet. This was the first time the LNG
industry demonstrated its flexibility and capability to respond to
fundamental shifts in global gas markets, something which was
not considered possible to such an extent in the past.
Tight market continues to
hamper trading
The continued development of the LNG spot and short-term
physical trading market also requires flexibility in the
deployment of LNG carriers. However, following the slow-down
in supply growth and significant increase in Asian demand,
there is currently a shortage of LNG available for spot and
short-term physical trading. Production in 2012 (reported
at approximately 239 million t) was slightly down on 2011
(reported at approximately 242 million t). Last year, in addition
to scheduled maintenance in Australia, Trinidad and Qatar,
there were unplanned outages due to flooding at Nigeria
LNG, a fire at train2 of Tangguh LNG and sabotage of the gas
pipeline at Yemen LNG. There are also delays in start-up of
a number of major LNG projects. For example, Angola LNG,
originally scheduled to start-up in early 2012, is now expected
to start-up in the second quarter of 2013. This shortage of LNG
availability is restricting the development of the LNG trading
market, despite good arbitrage opportunities and availability
of LNG shipping capacity. According to Arctic Securities and
Fearnleys, this lack of volume in the market is currently causing
freight rates to decrease with spot modern steam vessels now
at US$110000/d, with potential to reduce further towards the
summer. Experts including BG are predicting that LNG supply
will remain tight in 2013.
Financing requirements and
impact of shorter charters
Historically, LNG was sold under long-term (20 year plus)
LNG sale and purchase agreements (LNG SPAs), typically on
a tram-line basis with the LNG being shipped to a specific
destination during the term of the LNG SPA. Consequently,
LNG carriers were chartered on a long-term basis (often closely
associated with the term of the LNG SPA) and deployed on
the same tram-line basis to ship LNG from export terminal to
destination market. Owners of LNG carriers relied on long-term
charter commitments to support their investment decision and
obtain financing for the construction of the LNG carriers. Lenders
regarded such long-term charters as attractive investments
given the long-term nature and stability of cash-flows. Todays
LNG industry has moved a long way from the traditional model.
For example, the term of many recent charters has been reduced
to five to eight years. Shorter charters means shorter loan tenors.
In addition, todays lenders may also require a more robust
security package, more extensive parent company support and
possibly increased equity contributions.
Reduced boil-off rates for new
LNG carriers
Boil-off is the process whereby the LNG cargo regasifies during
a voyage and then escapes into the atmosphere through valves
designed to prevent excessive pressure building up in LNG
cargo tanks. Boil-off rates of 0.15% of the LNG cargo volume
per day are common. However, new LNG carriers utilising the
latest cargo containment technology can reduce the boil-off
rate to 0.1%. This is a significant commercial advantage. For an
LNG carrier with a cargo capacity of 170 000 m
3
, a reduction
in the boil-off rate from 0.15% to 0.1% is equivalent to saving
approximately 85 m
3
or 36 tpd of LNG, making such vessels
highly attractive.
Speculative orders
Given the opportunities outlined above, a number of shipowners
have placed orders for newbuild LNG carriers on a speculative
basis, aiming to benefit from the expected demand for new
vessels.
Continued expansion of the
worlds FLNG vessel and FSRU
fleet
Shells first floating LNG (FLNG) vessel is currently under
construction. It will be deployed at the Prelude field offshore
Australia and will be the largest offshore floating facility ever
built (488 m long x 74 m wide, with a production capacity
of at least 5.3 million tpy of liquids, 3.6 million tpy of LNG,
0.4million tpy of LPG and 1.3 million tpy of condensate). Other
FLNG projects are also underway or in development. Petronas
first FLNG vessel is smaller (1.2 million tpy) and is expected to
be deployed in 2015 at the Kanowit gas field, 180 km offshore
Bintulu, Malaysia. Some players such as Golar LNG are adopting
a different approach, intending to convert LNG carriers to FLNG
vessels. Other IOCs, NOCs and participants in the LNG industry
are looking to follow these leads and develop offshore gas fields
using FLNG vessels.
Infrastructure in destination markets is also moving offshore,
with a vast increase in the number of LNG floating storage and
regasification units (FSRUs) now in operation or under
development in many import markets. This increased trend
towards the use of FSRUs is attributed to their flexibility of
deployment, relatively low cost and potentially quick timeframe
in which FSRUs can be in service. These are attractive attributes
for companies seeking to import LNG. Shipowners operating in
this relatively new FSRU market include Golar LNG, Hegh LNG,
Excelerate and BW. With new build FSRUs having greater
storage capacity and using the latest technology, shipowners
with new build FSRUs may have a competitive advantage when
tendering for LNG projects compared to those planning on using
older, converted FSRUs.
Conclusion
The LNG industry is set to continue to grow significantly in the
future. With more LNG carriers required to cater for increased
global LNG production and further growth in the LNG spot and
short-term trading market, as well as increasing demand for
FLNG vessels and FSRUs, there are exciting opportunities ahead
for shipowners and others in the LNG shipping industry as the
LNG sector continues to grow, evolve and respond to the rapidly
changing dynamics of global gas markets.
LNG_MayJune_2013_16-18.indd 18 20/05/2013 11:03
T
he Middle East transportation
sector, especially its vast
shipping fleet, is being
confronted by increasing global
challenges. In addition to growing challenges in key LNG markets,
such as in Asia (Japan) and Europe (where strict emissions
legislation is being put in place), LNG exports are also confronted
by increased global competition. The shale gas revolution in
the US, in combination with the emergence of Australia as the
upcoming leading LNG exporter, is forcing Arab (and Iranian) gas
exporters to reconsider their options.
LNG now has an opportunity to become a fuel of choice. Its
ultra-low sulfur content and clean burning properties, combined
with the industrys experience of using boil-off gas as a fuel for
LNG carriers, makes LNG a promising alternative to low sulfur
marine fuels that will become mandatory by International
Maritime Organization (IMO) regulations in a decade. This new
market potential could present exporting countries, such as Qatar,
the UAE or even Iran, with new opportunities for growth. Instead
of fighting an uphill battle in the SECAs (Sulfur Emission
Controlled Areas) and inland shipping areas such as ARA
(Antwerp-Rotterdam-Amsterdam), with governments putting in
place strict emissions targets, LNG shippers and exporters could
now even lead the pack. A growing number of governments,
shippers, ports and environmental groups are keen to find out
more.
The potential market is huge, over 90% of the worlds tanker
and container fleet still use the most polluting form of heavy fuel
oil, the residual oil sludge from refining (Mazuth). It is cheap, but
not environmentally friendly. By substituting this fuel with LNG,
the sector could become greener. The current price setting of LNG
is also a benefit. At the same time, the increased recovery
potential of natural gas due to successful production from
unconventional and previously stranded reserves is bringing it into
direct competition with oil based fuels. The IMO regulations, in
effect from January 2012, aim to cut emission levels by 30% by
2030. If parties do not comply, seaborne transportations share of
global carbon emissions is expected to increase to 18% in 2050,
based on current growth rates. Several European ports, such as
Rotterdam or Antwerp, as well as Singapore, are already setting
the necessary framework to enable bunkering of LNG for fuel.
Although interest in LNG fuelled vessels is growing, several
(technical) hurdles still need to be overcome. In a report entitled:
LNG fuelled deep-sea shipping Outlook for LNG bunker and
fuelled new building demand up to 2025, classification company
Lloyds Register reported that the technology issue is one of the
main constraints at present. Other hurdles include LNG prices and
acceptance by consumers and producers. Still, analysts expect
that the market for LNG fuelled vessels could grow to just under
2000 vessels by 2025.
Business drivers
For the MENA (Middle East and North Africa) region, other
business drivers are becoming important. Due to a substantial
growth in domestic demand for natural gas, Arab producers
Unlocking
Cyril Widdershoven and Bas van den Beemt, TNO, the Netherlands,
explain why technical-economic cooperation is needed between
European organisations and Arab gas producers to kick-start
small scale LNG markets in the Middle East.
markets
LNG_MayJune_2013_19-21.indd 19 20/05/2013 11:19
and consumers are looking for new supply sources. Increased
investments are expected to develop indigenous gas production,
especially in Abu Dhabi, Iraq and Saudi Arabia. Transportation
of these volumes will partly be carried out via conventional
infrastructure options, pipelines and large-scale LNG. However,
experts are proposing that the regional Gulf based economies
also need to assess the options of small scale LNG. The Gulf
market is large enough to introduce small scale options, such as
shipping and storage.
The current market, with LNG regasification projects proposed
in Abu Dhabi, Dubai and Bahrain, shows a growing need for LNG
supplies in smaller quantities (breaking bulk). Taking on smaller
quantities of LNG, from ships of around 70000 t, is not only
commercially attractive, but will also increase the possibility of
bringing onshore LNG into almost all of the existing ports in the
Gulf region. Another of the underlying drivers for small scale LNG
is a need to develop LNG as a fuel for transport. Arab producers
are also looking at increasing their market share in emerging
decentralised markets (e.g. the Pacific and Europe) with a clean
fuel as an alternative to sulfur containing fuel oils.
In order to establish these new diversified LNG markets,
access to efficiently produced LNG supply chains is essential. That
is why traditional LNG players, such as Qatar, are at the front of
the queue to monetise these new LNG markets. Based on the
current lack of international knowledge and experience, these
interesting development paths have not yet been taken in the
GCC region.
Domestic gas challenges in the
GCC
The Gulf regions gas sector is currently facing domestic
challenges. Increased domestic demand in the Arab countries,
especially Saudi Arabia and the UAE, are a potential threat to
export plans. Apart from Qatar, which has vast gas reserves
in the North Field, all the other Arabian Gulf states are short of
gas. In addition to pre-existing regional gas export pipelines,
new LNG regasification projects are being put in place in several
Arab countries. Abu Dhabi is planning LNG imports via Fujairah,
while Dubai and Kuwait already use expensive liquefied fuel.
Development of new domestic gas fields in Abu Dhabi and
Kuwait has proved costly because of high levels of toxic hydrogen
sulfide, and has therefore lagged. More than half of Saudi Arabias
electricity comes from burning crude oil products, as there is
a lack of sufficient gas supplies. The decision to burn oil is also
hampering Arab countries export potential.
At the same time, gas is also being used to support their
respective petrochemical industries or energy intensive sectors,
such as steel and aluminium production. The desalination projects
are also expected to increase demand in the future. The driver
behind most of the domestic demand increase is the fact that gas
is underpriced. This price strategy is now reaching its limits. Not
only is demand increasing faster than supply, but the overall
competitiveness of GCC based industries is also being
undermined by cheap shale gas production in the US. US
companies are already returning their operations from the Arab
world to the US.
Changes in the global markets are dramatic. Shale gas is a
major game changer, but it could also prove beneficial for some
Arab countries. Increased competition, a possible oversupply of
pipeline gas and LNG could be of benefit to the struggling
domestic gas markets in the GCC. New gas consuming markets
need to be found by exporters such as Qatar. A revamp of their
domestic and regional gas strategy may not be a choice but a
necessity.
Research and development
In stark contrast to most European and US market players,
the GCC countries face technical hurdles. A lack of experience,
technical knowledge or an applicable infrastructure is blocking
possible endeavors into new markets by Qatar, Egypt, Oman and
in the long-term Iraq, Iran and Algeria. A technologically driven
LNG strategy needs to be established in MENA. A possible option
would be to target the emerging opportunities in small scale
LNG applications. R&D is essential for this, based on cooperation
between Western applied technology organisations, such as
TNO, and Arab national gas producers, such as Qatar Petroleum.
The market has not been standing still. Although enabling
technologies for small scale LNG distribution and the use of LNG
as a fuel have been available for decades, they still do not meet
the new market requirements on costs, safety and emissions. This
is because small scale LNG distribution has been developed
mainly in order to extend the coverage area of existing pipe grids
and gas storage capacity by natural gas importers. For this
purpose, break bulk is organised at regasification terminals,
leaving other more cost effective break bulk supply routes, such as
ship-to-ship transfer supply, untouched. In addition, with the
preference for utilising the existing pipe grid, gas importers have
mainly developed compressed natural gas (CNG) as a commercial
fuel for light duty vehicles without monetising LNG directly as an
alternative for gas oil and marine diesel. This has largely resulted
in a situation where engine technology has been predominantly
developed for CNG with limited power capacity, which is not
attractive for heavy duty vehicles such as ships and trucks. A large
potential market has, until now, been unutilised. New
requirements, such as zero emissions from venting, cost effective
small scale LNG supply and dedicated LNG engine technology,
have never been a subject for innovation. R&D dedicated to these
new market trends will be essential in order to make the small
scale supply chain cost effective, reliable and safe enough for the
retail of LNG as a commercial fuel.
Conclusion
Breaking bulk to unlock new emerging small scale LNG
markets, especially with regards to LNG as a fuel, is needed.
The technology exists in part, but it is still facing hurdles that are
limiting the potential of this new but exciting sector. Technical-
economic cooperation between applied high-technology
organisations in Europe and Arab gas producers in the Gulf region
is required to kick-start a new market approach. In the coming
years, more research is needed to address these issues, not only
to open up new markets in the Gulf region but also to counter
growing gas supply difficulties. At the same time, Arab Gulf
countries are in dire need of their own technological research
capabilities, underpinning their own needs, while presenting the
next generation with potential new, challenging and exciting
opportunities in a more diversified hydrocarbon-based sector.
The countries must set up gas research centres, targeting the
development of new technologies to monetise their own
resources in the long-term. Western organisations should look to
establish research programmes in the Gulf region, addressing not
only the technical issues but also willing to transfer knowledge to
younger generations.
Part of this could be the set-up of a (small scale) LNG
technology and test centre, in which gas producers would join up
LNG_MayJune_2013_19-21.indd 20 20/05/2013 11:19
with applied technology companies to establish a high-class,
techno-economically viable research institute focusing on the
future challenges of the Gulfs gas sectors. For Arab governments
and their national oil and gas companies, such a step would not
only assist their capabilities but also support the development of
their own population. The development of (small scale) LNG
technology should be driven by the forces behind the national gas
producers, not by IOCs. New technology development related to
emerging LNG markets will increase the adaptability of the
traditional LNG industry for the changing LNG sectors. TNO is
currently assessing the options of bringing part of its own research
capabilities and experience to the centre of LNG production in the
Gulf region.
The company, in cooperation with the Dutch metrology
institute VSL and the federation of three leading Universities of
Technology in the Netherlands 3TU, set up the LNG TR&D
research foundation in 2010. The LNG TR&D aims to facilitate
international collaborative technology innovation in the LNG value
chain. Its main focus will be on providing consolidated information
on ongoing and new collaborative research initiatives regarding
innovative areas of the LNG value chain (small scale LNG, mid-size
LNG production, application of LNG as a transportation fuel, etc.),
to develop and sustain an internationally recognised technology
roadmap, in which technology development challenges are
aligned with international business objectives and requirements
identified by leading industrial and governmental stakeholder and
research organisations. For the GCC region, the LNG TR&D also
focuses on sustaining new and existing knowledge and research
capacity (including test and calibration facilities) through
stimulation of cooperation between research organisations. Part
of this research will be carried out at the premises of the
international parties taking part.
A strategy is being set up to advise Gulf governments and
operators to qualify and optimise small scale LNG technologies,
taking into account aspects such as reliability, safety, security,
economics and environment. Based on TNOs experience and
work with players such as the Port of Rotterdam on environmental
chain analyses for LNG as a fuel for shipping, including legal and
safety assessments, new proposals are currently being tweaked
for the Gulf markets. Research will be focused on the following:
LNG transfer and bunkering systems and operations.
LNG fuelled vessel storage and fuel systems.
Gas and dual fuel engine technology.
Small scale LNG distribution, storage systems and equipment.
LNG quality specifications and measurement
technologies.
The coming years will reveal whether or not Arab countries are
willing to put money and research in place to address these issues.
If countries such as Qatar, Abu Dhabi or Iraq are not willing to
invest in new developments, the current situation will prevail.
Knowledge and technology developments will be in the hands of
the developed world countries, leaving emerging giants such as
Qatar and the UAE with the spoils of their current economic
systems. Monetising resources does not only mean selling
volumes but also setting up a more integrated economic-technical
chain of operations. Developing new technology, and new
capabilities, will only increase Arab countries options to find new
markets.
ENABLING CLEAN
TRANSPORT WITH LNG
TNO is the independent partner to
accelerate innovation. We are keen
to help you in qualification and
optimisation of your safe and reliable
offshore LNG or in the operations and
application of LNG as fuel.
SERVICE AREAS
LNG transfer systems
processing for floating LNG
LNG as fuel for transport
Collaborative LNG R&D
Safe offshore LNG operations
For more info go to TNO.NL/LNG
TNO.NL
LNG_MayJune_2013_19-21.indd 21 23/05/2013 11:36
LNG_MayJune_2013_22-28.indd 22 20/05/2013 11:24
E
ven though shipping is accepted to be the most eco-friendly means of
transportation for the majority of goods transported, there is no doubt
today that improvement of ships emissions is urgently required. The
shipping industry now faces local ECAs/SECAs (Sulfur Emission Control Areas)
in order to reduce SO
X
, NO
X
and particulate matter emissions in areas with high
shipping activities, as well as global challenges from greenhouse gas emissions.
There are different ways to cut NO
X
, SO
X
, particulate matters and CO
2
emissions, but
only one solution provides an all in one reduction. The use of LNG as a ships fuel will
reduce NO
X
to below Tier III levels (with Otto principle engines), SO
X
to zero, particulate
matters to approximately zero and CO
2
by about 20% without any after treatment of
combustion gases.
But even if the technology is there for storage and processing LNG on board consumer
vessels, there are still some hurdles for the use of LNG in a wider range:
Development of LNG bunker infrastructure is embryonic, so availability is an issue to be
solved. Owners/operators will not switch fuel unless they know where to bunker.
Pricing of LNG as a bunker fuel is expected to be below marine diesel oil/marine gas oil,
and maybe even below heavy fuel oil in some regions. But the market is not established,
and so price estimations are not sufficiently reliable to allow for significant investments in
LNG fuelled ships.
Technical solutions for storage and processing of LNG on board as well as engines are
available, but additional cost and potential loss of cargo space create complex business
models.
Jrgen
Harperscheidt,
TGE Marine Gas
Engineering,
Germany, looks
at the technical
issues facing LNG
bunkering.
LNG_MayJune_2013_22-28.indd 23 23/05/2013 10:21
Burden sharing between the owner (investment cost)
and charterer (potential savings in fuel cost) is subject to
discussions.
This article will look at the technical aspects of these
hurdles, with a particular focus on those connected to LNG
bunkering.
LNG infrastructure
Looking at todays LNG infrastructure, there are large
liquefaction terminals in the supplier countries, large LNG
carriers of approximately 140 000 to 265 000 m
3
tank capacity
to transport it, and regasification terminals operated by major
energy companies to receive the LNG in order to supply it to
the local gas grids after regasification.
Small scale LNG distribution is already in place in some
dedicated areas, such as Norway. Trucks and small LNG carriers
such as Pioneer Knutsen, Coral Methane and CoralEnergy are
trading to supply shore terminals with a few vacuum insulated
tanks or flat bottom tanks of limited size. These terminals
provide LNG to small local grids, industrial plants or ferries and
offshore supply ships. These LNG-fuelled ships have dedicated
routes and always use the same location for bunkering.
Bunkering procedures and equipment are tailor-made, no
parallel operation is taking place and time is not a critical factor.
Some ferries are taking bunkers directly from trucks. The
majority of the LNG for the mentioned supply chain is produced
by liquefaction from pipeline gas, but operations of Coral
Methane have proven the interconnection between the large
and small chain (Figure 1).
One solution that could be available today for some
sectors of shipping could be the use of mobile LNG tanks
LNG transport containers or trailers (Figure 2). The
loading/unloading of such mobile tanks would fit perfectly
with the normal cargo operations, especially for RoRo/RoPax
ferries no dedicated bunkering operations would be required.
In order to achieve an equivalent level of safety for such
concepts compared to fixed installed tanks, a number of
safeguards need to be followed. These include the following:
Positive fixation that cannot be unlocked unintentionally.
Connection/disconnection operation only in port.
Fuel area separated from cargo area.
Gas/fire detection sensors.
Secondary barrier for tank connections to ship systems.
Protection against damage by cargo operations.
These safeguards, and the accompanying cost and space
restrictions, limit the number of such tanks to be installed on a
ship. As such, this solution will only be an option for short sea
shipping to overcome the LNG bunker process.
In order to build up a reliable infrastructure for LNG
bunkering that can be accepted by the shipping industry, the
major ports need to provide a sufficient amount of bunker LNG
to the vessels calling there. But how will this work? Feeder
vessels may pick up the LNG either at large terminals or at
liquefaction plants and feed it either directly to bunker vessels
(ship to ship) or to local bunker terminals (floating or onshore).
The local terminals could either fill trucks for bunkering small
Figure 1. Coral Methane loading in Zeebrugge in May 2010 (courtesy of Anthony Veder).
LNG_MayJune_2013_22-28.indd 24 23/05/2013 10:22
LNG_MayJune_2013_22-28.indd 25 23/05/2013 11:00
capacity vessels, directly bunker consumer vessels calling
there, or feed LNG to bunker vessels. Finally, bunker vessels
would go alongside the consumer vessels, in a similar way to
the majority of todays oil bunkering procedures.
The amount of LNG consumed during the upcoming years
will be rather limited, which makes investment in the
infrastructure very risky, as the cost of the assets is
significantly higher than for oil bunker assets due to cryogenic
technology. So the world is waiting for the first movers to
believe strongly enough in the LNG fuel market to invest.
Decisions must be taken urgently to ensure supply is in place
to cover the restrictions of the 2015 sulfur cap.
LNG bunkering
LNG bunkering will have to be as close as possible to
traditional bunkering if it is to be accepted by the majority of
the shipping industry. However, there are few details in place
about how to bunker large amounts of LNG to a ship while it is
at the terminal for cargo operations. The main challenges are
the procedures and the global and local regulations that still
need to be developed.
Some of the current challenges are described in the
following section, and the importance of solutions to these
issues for the development of LNG as a fuel cannot be
overestimated. Ship owners and operators will not commit to
LNG as a fuel while major questions remain on how to get it on
board.
First of all there is the connection between the ships
piping interface, data interface, as well as practical
considerations such as the location of the bunker station with
regard to berthing of the bunker ship, water curtain falling
down on the deck of the bunker vessel and height difference.
Standardisation of the piping and data interface is under
progress with ISO TC 67 working group 10. The
standardisation for bunkering LNG is much more important
than for oil bunkering, as each connection or spool piece
presents a potential leakage. Therefore, the ideal scenario is a
quick connect coupling that directly fits with the coupling on
the receiver vessel without any intermediate piece. The data
interface will exchange level and pressure information
between the systems to carefully control the bunker process
and avoid overpressure or overfilling. An emergency
shut-down (ESD) signal also needs to be exchanged to ensure
that both vessels can safely stop the bunker process in case of
emergency.
It seems to be agreed that quick connect/disconnect
coupling would be ideal to establish the connection in a safe
but timely manner. Those couplings available ensure that a
tight connection between both sides is established, before
either side is opened. This will significantly reduce the risk of
spills created by couplings or flanges that are not connected
properly. Furthermore, a dry breakaway coupling for
emergency release will be part of the connection, be it
manually activated or as a PERC (powered emergency release
coupling) with hydraulic or pneumatic release.
Taking into account this equipment and the required
bunker rates, it is quite clear that such connections will be too
heavy to be moved manually. A hydraulic or mechanic support
of the loading arm or hose handling crane is essential. It is
unclear whether the composite hoses used today in LNG ship
to ship operation are suitable for bunkering, or if double
walled hoses or fixed pipes with swivels are necessary to
achieve the required safety levels.
Quality and quantity measurement is also quite a complex
but important topic due to the fact that LNG composition
varies and energy content, as well as methane number,
change with the composition. Volume or mass flow
measurement need to go along with composition certification
or measurement. Energy content calculated from this will
govern the price, but the methane number will tell the owner
whether he will be able to make full use of the engine capacity
with the bunkered LNG. Dual or single fuel gas engines
following the Otto principle are sensitive to knocking,
therefore a low methane number will cause derating of the
engines and prevent full capacity.
The keyword SIMOPS also creates lively discussions,
touching the question whether LNG bunkering can be done
SIMultaneously with cargo OPerationS (Figure 3). Key
questions include how to ensure that workers are fully focused
on their duty during bunkering and how to ensure that cargo
operations do not create dangerous situations? No final
decision has been made, but for some segments of the
shipping industry it seems clear that LNG bunkering can only
be accepted if done in parallel.
Training and education of workers on both bunker vessels
and LNG fuelled vessels need to be defined. Best
management practices combined with repeated trainings and
exercises are mandatory. However, the extent of training in
order to maintain high safety levels is under discussion.
Today, most LNG bunkering is carried out with moderate
speed and by utilising spray lines in the tank top to condense
the displaced vapour inside the tank. Higher bunkering rates
go along with the requirement to send the displaced vapour
back to the bunker source. This vapour return is under
discussion as it creates another connection that needs to be
Figure 2. Design example of mobile tanks (TGE, MAN &
Flensburger).
Figure 3. How LNG bunkering may look (TGE).
LNG_MayJune_2013_22-28.indd 26 20/05/2013 11:24
2013 Air Products and Chemicals, Inc.
Air Products has contributed to the success of more LNG operations than
any other company. And we bring our full capabilities to LNG projects of
any scale, from peak-shaving plants producing less than 0.1 MMTPA to
the largest base-load facilities, on land or off-shore. Our LNG team can
help you get a plant up and running at the highest efficiencyon time,
on budget, and in any climate. To learn more, call 1-800-654-4567 (US),
1-610-481-4861 (worldwide) or visit us online.
tell me more
airproducts.com/lngplant
Big LNG expertise.
Also available in small LNG plants.
DOWNLOAD OUR FREE
GAS WEIGHT & VOLUME
CONVERSION APP.
LNG_MayJune_2013_22-28.indd 27 23/05/2013 11:01
handled and supervised. It also adds complexity to the
custody transfer issue, as a measurable amount of valuable
gas goes back to the supplier.
The preparation of the connection hose or pipe for transfer
before bunkering and for disconnection after bunkering
requires inerting, typically by nitrogen purge. The purge gas at
startup is typically nitrogen in air, which can easily be vented.
However, the purge gas after bunkering is a mix of nitrogen
and natural gas. It could be vented or kept inside the system,
where nitrogen is likely to accumulate and create pressure
build-up problems. Removing the purge gas is one of the
issues to get the no venting strategy in place over the entire
supply chain.
It is clear that local and port regulations are going to play a
significant role in the development of bunker procedures. It is
difficult to predict what regimes will be in place in ten years
time, but ports will hopefully find a common way to deal with
LNG bunkering.
There are a number of further operations in LNG as a fuel
that will be required from time to time. To prepare a ship for
dry dock it would typically need to be gas-free. So an
additional service that bunker ships could provide would be to
receive the heel from the fuel tanks when stripping, and to
provide hot gas to warm up the tanks. Further, most vessels
will have limited nitrogen capacity for purging lines only, so
the nitrogen plant on a bunker vessel could be designed to
inert the entire fuel gas system of the consumer vessel,
including tanks. Not all bunker vessels would be equipped like
that, but some would make sense.
LNG containment systems
There is no need in this article to discuss the variety of LNG
containment systems. However, the bunker chain and fuel gas
systems will mainly focus on IMO type C pressurised tanks,
which are not yet widely used in the LNG sector. These tanks
are safe and reliable, and their high design pressures allow for
high loading rates and pressure increase due to boil-off. Partial
filling is not an issue for these tanks, they are easy to fabricate
and install and do not require a secondary barrier. As such,
they are attractive for the small scale chain and for fuel gas
systems.
An established solution for small fuel tanks is to use
vacuum insulation to keep heat ingress and boil-off gas (BOG)
generation low. For larger fuel tanks and cargo tanks on feeder
and bunker ships, single shell tanks with foam insulation
would provide a better solution, comprising less weight and
cost. BOG handling for such tanks can be done by pressure
increase or by burning the gas in Otto cycle engines.
Conclusion
Technical solutions to establish the supply chain for the use
of LNG as a fuel are available. Emission control and stable
LNG prices should be the main drivers to develop LNG as a
bunker fuel sooner rather than later, especially in Northern
Europe and the US. The lack of LNG bunker infrastructure
and valid procedures to perform LNG bunkering is the major
challenge, which must be addressed urgently to enable
a significant number of ships to sail on LNG from 2015
onwards.
NEED A
REPRINT?
WE CAN TAILOR
TO YOUR
REQUIREMENTS,
PRODUCE 1-12
PAGE FORMATS,
PRINT COLOUR OR
MONO AND MORE!
Contact Catherine Gower
+44 (0j1252 718999 or email:
catherine.gower@palladian-publications.com
LNG_MayJune_2013_22-28.indd 28 23/05/2013 11:02
T
he LNG market is changing rapidly. The increase in gas
resources and the new international regulations on
pollution standards will change the game for all players.
New market forces require adaption and innovation so that
professionals and consumers can take advantage of all that
LNG has tooffer.
Use of LNG as a fuel for
sea-going vessels
One of the key drivers in Europe and in the US is the
introduction of Sulfur Emission Control Areas (SECAs) from
2015, coupled with the International Maritime Organizations
requirement for 0.1% sulfur content in fuels used in these
SECAs. More ECAs are likely to be implemented soon. A
US-Carribbean ECA has been approved and Japan is reported
to be preparing an ECA application. Ultimately, there are only
a few viable options enabling vessels to comply with SO
x

emission regulations: use distillates and low sulfur products, fit
exhaust gas cleaning systems, or pioneer clean fuels such as
LNG. As such, many shipowners are considering switching their
ships fuel to LNG.
These new market requirements can be adressed with
membrane technology solutions. Membrane systems provide
key advantages, including the following:
Compact: minimising loss of commercial space due to the
tanks adaptability to vessel shape.
MEMBRANE MARKET
Julie Sakhrai, Lorenz Claes,
Stphane Maillard and Thomas
Crmire, GTT (Gaztransport
& Technigaz), France, describe
the benefits of using membrane
systems along the LNG chain.
Figure 1. The
NO 96 technology
inside a GTT
membrane tank.
LNG_MayJune_2013_29-32.indd 29 23/05/2013 15:08
The cheapest solution per m
3
for large fuel tanks
(> 1000 m
3
).
Impressive track record for LNG transportation at sea.
The lightest existing system.
LNG-fuelled ships are a viable option in the long-term,
particularly for sea-going vessels.
One of the challenges faced by many ship owners is the
loss of cargo space on LNG-fuelled ships compared to the
equivalent conventional ship. Generally, the cargo space
reduction occurs due to the increased volume of liquid required
(LNG vs heavy fuel oil and distillates), as well as space for
insulation, inspection, and the fuel gas handling and supply
system (FGHSS).
Numerous design calculations demonstrate that membrane
technology allows ship owners to minimise the loss of cargo
space.
Moreover, the LNG-fuelled vessels equipped with
membrane technology provide an economical advantage based
on the key points of basic design, driving the construction cost
down significantly.
In addition, the following subjects are currently being
evaluated for membrane LNG fuel tanks :
Low boil-off rate (BOR).
Standardisation approach to sloshing load verification for
partial tank filling.
Methods for pre-fabrication of complete tanks or parts of
tanks.
A larger range or complex tank shapes.
Gravity drainage of tank (no pump tower, no liquid dome).
Self-standing membrane tanks, including micro membrane
tanks (200 m
3
).
The entire LNG chain
The membrane technologies developed by GTT are known for
large capacities (the latest market requirement) and are able to
respond to the different storage requirements along the LNG
bunkering chain.
Another important aspect when considering LNG as a fuel is
the entire infrastructure and supply chain. This is considered as a
significant barrier to the widespread adoption of LNG as a fuel.
GTT is able to address the main components of the LNG chain in
the following way:
Small land storage solutions in port areas or by the sea and
connected to an import/export jetty.
Offshore storage and bunker station the concept of
gravity based bunkering at an offshore LNG refilling station
in sheltered water outside of the port area.
LNG bunkering ships and coastal feeders (eg. 4000m
3
LNG
plus 1000 m
3
marine diesel oil).
LNG fuel tanks on board ships, minimising the loss of cargo
space.
Membrane solutions are easily implemented via either
retrofit or newbuild and can be installed afloat in parallel to
other ongoing construction activities, thereby optimising the
building schedule. Moreover, the pressure in the tank of up to
0.7 barg offers the flexibility of boil-off gas (BOG) management.
Furthermore, GTT has also considered other challenges such as
ship-to-ship transfer, which presents some complexity,
especially in adverse weather conditions.
Solutions for small and
mid-scale LNG
LNG storage and transportation by sea was initially developed
on a small scale. In the early days, membrane technologies
were considered by many to be suitable only for small scale
applications. Sizes increased during the market evolution of the
1970s and 1980s. With the major orders from Qatar in 2004
onwards, membrane technology was adopted for the largest
LNG carriers ever built and the full capacity range has been
successfully covered (15 000 266 000 m
3
).
The small scale membrane type LNG carriers have
performed reliably. Some of the early generation ships are still
working, while others such as the Tellier have recently been
retired from service after 40 successful years of operation.
More small size membrane ships were built in the 1990s,
covering the 19000 22 500 m
3
range, while others were
added in the midscale range of 75 000 m
3
(Med-Max).
Today, GTT is actively working on a new range of small
scale LNG carrier designs, proposing bunker and LNG feeder
vessels over the full range. Examples include a 4000 m
3
LNG
bunker ship (two 2000 m
3
LNG fuel tanks), a 15 000 m
3
LNG
carrier for small scale distribution, and a 16 500 m
3
shallow draft
LNG carrier, intended for coastal and river trade. GTT is also
designing a 32 000 m
3
carrier, which is projected for regional
distribution. Any other size may be proposed and developed
with the shipyards and their clients. Specific requirements can
be addressed, giving operational advantages based on proven
solutions.
Completing the supply and distribution chain, LNG can be
stored using floating barge solutions or land based membrane
tanks. Membrane technologies offer a wide range of
containment capacity and optimise the footprint/volume of
containment.
Offshore platforms
Challenges for offshore platforms cover several fields
unexplored until now. These units require a continuous
production, entailing both onsite maintenance and operations Figure 2. Typical arrangement of gas system on a cargo ship.
LNG_MayJune_2013_29-32.indd 30 20/05/2013 11:38
LNG_MayJune_2013_29-32.indd 31 20/05/2013 11:38
with partially filled tanks. The cargo handling system must also
be adapted to meet production and maintenance requirements.
Membrane units provide a flat deck space for easy topside
arrangement and a full occupation of the cargo hold, providing
compactness and optimisation of the hull design. The tank
arrangement is selected either as a one-row arrangement or a
two-row arrangement. Membrane solutions have already been
validated for several barge designs and site conditions all over
the world. The advantages of two rows of tanks include a
central cofferdam structure to withstand high loads of topside
modules, a large range of filling levels free of sloshing impacts,
and a reduced set of sea states inducing sloshing.
The pump tower has been adapted to implement
retractable pumps for pump maintenance without tank
decommissioning. One or several tanks can be temporarily
decommissioned for maintenance purposes while other tanks
can be kept in normal operation.
Formal safety assessment and maintenance analysis
(FSAMA), and failure mode identification and risk ranking (FMI &
RR) have been performed to demonstrate the suitability of
membrane systems for offshore applications. These analyses
have received a Statement of Endorsement from a classification
society. The control of both membranes is performed while the
tanks are in service to avoid any downtime. Standard and easy
procedures are applicable offshore, in case repairs are necessary.
Access can be provided to any part of the tank if required.
Owners and operators receive the support of the GTT operations
department during operations and maintenance phases.
The company has also developed a system for fibre optic
monitoring of the containment, as well as a hydraulic arm that
can be introduced into the tank through the gas dome to
facilitate access to all areas of the tank for inspection and
eventual repair, thereby minimising the need for scaffolding.
The reliability of membrane systems is proven based on
their long track record and diversity in terms of application and
size of vessels. Systems and procedures have been refined and
enhanced throughout the years, without compromising the
safety of the vessels. The membrane system design guarantees
that the tank construction schedule is not on the critical path.
The large network of experienced yards and approved suppliers
are all limiting the industrial risk, while bringing cost and
schedule advantages to offshore units fitted with membrane
systems.
Since 2000, membrane technologies have gathered
experience in LNG offshore operations. The early regasification
vessel projects used both NO 96 and Mark III. Several projects
to convert LNG carriers into FSRUs or FSUs have followed, and
membrane solutions have now become the logical choice for
newbuild FSRUs. GTT has completed many feasibility studies as
well as Front-End Engineering and Design studies for LNG
FPSOs. These studies were sanctioned by approvals from
classification societies. The first two orders of LNG-FPSOs
employ GTT solutions.
Conclusion
In conclusion, membrane systems offer many advantages
to the LNG industry. In particular, with the arrival of the ECA
navigation zone in northern Europe and North America, LNG is
being seen as the solution for an alternative fuel. Membrane
systems can be used for LNG fuel tanks onboard commercial
vessels and as a containment solution all along the LNG
chain.
Case study: Arctic revolution
In November 2012, the LNG carrier Ob River became the
first ship of its kind to sail across the Arctic during winter.
The Ob Rivers odyssey lasted 27 days, over a distance
of more than 6000 miles through Russias Northern Sea
Route (NSR) from Norway to Japan. The increased use of
the NSR will open up new transport and trade opportunities
that may revolutionise shipping. Using the Arctic route can
cut journey times by 40% compared with the Suez Canal.
Built in 2007 by Hyundai Heavy Industries, Ob River is
one of a few ice-class LNG carriers. It is equipped with four
Mark III membrane tanks, and chartered from Dynagas by
Gazprom. Ob River is, in part, recognition of the work GTT
has carried out to demonstrate that its technologies are fit
for use in arctic conditions.
Many studies have been carried out by GTT over recent
years in order to demonstrate that its membrane
technologies are suitable for use in LNG carriers for arctic
navigation. Most of these studies were performed in
cooperation with ice specialists and institutes with
experience of navigation in arctic waters. GTT has worked
with the Krylov Shipbuilding and Research Institute,
Russian Maritime Register of Shipping and Servernoye
Design Bureau. These studies have highlighted the high
safety margins for GTT containment systems under the
specific solicitations encountered during arctic navigation
compared to open water routes.
The fatigue effect of vibrations induced by the progress
of the ship in ice was investigated, including theoretical
and experimental work, which resulted in positive
conclusions for the membrane containment systems, even
under severe ice conditions (ARC 7 vessels). Iceberg
collisions have also been treated considering various
scenarios and hypotheses through calculations, test
campaigns and risk analyses. It was concluded that the
companys containment systems can sustain these
solicitations whatever the collision scenario considered.
Finally, cases of sloshing induced by sudden stops in ice
were also studied, leading to pressure levels far below the
membrane capacity.
Membrane systems can be considered as
fit-for-purpose for navigation under arctic conditions.
Figure 3. Ob River plying the Northern Sea Route (courtesy
of Dynacom).
LNG_MayJune_2013_29-32.indd 32 23/05/2013 10:27
J
ust ten years ago the clarion call of US energy
prognosticators and regulators was that the
US was running out of natural gas. The industry
began scouring the coasts of the US for sites to build
LNG import terminals. The Federal Energy Regulatory
Commission (FERC) and the US Coast Guard (USCG)
were overwhelmed with proposals and permit
applications to build onshore and offshore regasification
terminals. Prices for natural gas and natural gas liquids
(NGLs) had reached uneconomic levels for many
industries (fertiliser, ethylene, plastics, chemicals, etc.)
and many manufacturers invested in foreign facilities
to minimise costs. Utilities began to look favourably at
coal and nuclear energy for base load power production
as natural gas prices soared from US$2/million Btu to
peaks exceeding US$ 15/million Btu.
The presence of shale and tight gas seams had been
well established for many years; shallow small scale
horizontal drilling and fracturing had been demonstrated
in the 1930s and 1940s. However, oil and gas drillers
lacked the large scale technology to economically
recover the vast potential of the deep, narrow tight gas
seams. In the mid-1970s, the US Department of Energy
(DOE) commissioned the Eastern Gas Shales Project.
The findings of this report ignited the imagination of
David A. Franklin,
WorleyParsons,
USA, establishes the
case for mini-LNG.
From flare
to fuel
LNG_MayJune_2013_33-38.indd 33 20/05/2013 11:44
wildcat driller, George P. Mitchell, founder of Mitchell
Energy and Development Corp. Subsidised by tax credits
under the 1980 Energy Act and guided by Sandia Labs
superior seismic mapping technology of the day, Mitchell
and his team of drillers developed directional drilling
methods to guide the drill string into the deep tight
horizontal seams of the Barnett Shale. Mitchell worked
diligently with cutting edge well fracturing contractor,
Union Pacific Resources, developing multi-fracturing
techniques to stimulate production. Mitchells persistence
paid off handsomely. The methods developed were so
different from the conventional techniques that the
produced gas was classified as unconventional gas.
In the past 10 15 years, the techniques and
technologies have been refined and improved. In 2000,
shale gas contributed less than 1% of the US gas supply.
Today, the US is awash with natural gas and shale gas
currently provides 25% of the production; some are
projecting as much as 50% by 2035. US Energy Information
Administration reports indicate that proven reserves of oil
and natural gas are climbing steadily (Figure 1). Excess
supply of natural gas has depressed the price to the point of
being uneconomic to drill or import. The horizontal drilling
revolution has been disruptive to many industries, including
the recently built LNG import terminals, which now sit idle.
Several of these facilities have applied for permits to liquefy
and export LNG to try to recover their investments.
The shale oil revolution
As natural gas prices have fallen, exploration and production
companies have employed and optimised the horizontal
drilling and production techniques to produce wet gas
and oil from shales and tight sands. The Bakken, Eagle
Ford, Niobrara and Marcellus basins have sparked one
of the largest oil and gas booms in US history. Recent
developments in the vast California Monterey Shales
indicate reserves four times greater than those contained
in the Bakken. The US, once dependent on foreign oil, may
soon be the largest oil producer in the world.
Natural gas is also co-produced with shale oil. This
associated gas is often dripping wet with condensate,
butane, propane and ethane. Condensate is easily extracted
on site and is often recovered and put into the crude stream
that is trucked to market. In regions with ample pipelines,
such as the Eagle Ford, the rich gas is transported to gas
processing facilities to recover the ethane, propane and
butane liquids, while the methane is diverted to the gas
network. However, some of the most productive shale oil
regions are so remote, and the local wellhead value of the
natural gas is so low, that it is currently uneconomic to
Figure 1. US oil and natural gas proven reserves.
Table 1. Liquid fuel price comparison (date: 4 February 2013)
Commodity
price (US$)
Unit Btu
HHV/gallon
US$/million
Btu
HHV
Discount ratio
NG:fuel
Price ratio
LNG:fuel
WTI crude oil 97.77 bbl = 42 gal. 140 500 16.57 0.20 0.72
ULS diesel 3.23 gal. 138 490 23.32 0.14 0.51
RBOB gasoline 3.05 gal. 121 848 25.03 0.13 0.48
Condensate: est (crude - US$ 17) 80.77 bbl = 42 gal. 112 000 17.17 0.19 0.70
Butane: (77 F, SG = 0.542) 1.70 gal. 95 553 17.79 0.19 0.67
Propane: (77 F, SG = 0.493) 0.87 gal. 88 370 9.84 0.34 1.22
Ethane: (-50 F, SG = 0.50) 0.27 gal. 92 511 2.92 1.14 4.11
Natural gas (mostly methane) 3.34 Million Btu
HHV
NA 3.34 1.00 3.59
LNG: (NG price +/~production cost) 12.00 Million Btu
HHV
84 820 12.00 0.28 1.00
Note: Federal and state excise taxes, marketing and transportation costs are not included
Sources: Wall Street Journal cash price; CME (OPIS): RBN Energy
LNG_MayJune_2013_33-38.indd 34 20/05/2013 11:44
Investigate Lydalls specialty cryogenic insulation technologies.
www.lydallthermal.com

Be Lighter

Keep Colder

Hold Longer

Deliver More
LNG_MayJune_2013_33-38.indd 35 20/05/2013 11:44
recover the associated gas and NGLs. Over 30%, about
250million standard cubic feet per day (MMSCFD), of the
associated gas produced from the Bakken shales is currently
being flared in the process of recovering the oil. Similar
flaring in other shale basins is common (i.e. Eagle Ford,
Niobrara, Marcellus, Utica, etc.).
Environmental regulators and energy producers are
scrambling to find solutions to avoid this flaring. Finding or
creating beneficial markets and developing economic
processes to recover these hydrocarbons will reduce this
wasteful flaring practice. One potential solution to this issue
is to totally condense all of the hydrocarbons locally and
truck these hydrocarbons to market in the same manner that
the crude oil is currently hauled.
The mini-LNG solution
LNG liquefaction is often viewed as unprofitable unless it is
done on a grand scale of 4.5 6 million tpy. Large and
mid-scale based load plants are optimised for ship borne
cargoes, to reach distant utility markets. Small scale
liquefaction is usually provided in utility peak shaving
applications with pipeline quality gas as a feedstock. A
mini-LNG plant can provide an attractive solution to remote
gas flaring, with LNG targeted at the transportation fuel
market, where it can compete favourably with traditional
petroleum based fuels. However, producing an LNG product
from the rich gas associated with shale oil production
requires considerable care and attention to composition and
products. The flared gas quantities at each well site are fairly
small, averaging less than 300 000 standard cubic feet per
day (SCFD). A small gathering system must be developed to
accumulate sufficient quantities of gas to justify production.
To maximise participation, the facilities must be small
enough to be mobile and only require a small gathering pipe
network. The minimum economic size of the facility should
accommodate about 2 5 MMSCFD of the flared gas.
All of the hydrocarbons and contaminants must be
accounted for. For simplicity and economics, it is assumed
that the mini-LNG facility will produce three products: LNG,
liquid ethane and C3+ NGLs. The value of ethane is currently
less than the value of pipeline natural gas and may be used
as feedstock to chemical plants or as fuel in thermal power
plants. Removal of most of the ethane is required to
produce marketable quality LNG, optimise the value for the
remaining C3+ hydrocarbons and to provide economic
storage and transportation options.
LNG as a transportation fuel
There is growing demand for LNG as a transportation fuel
for trucks, drill rigs, mine haul and mining equipment, rail
locomotives, ship fuel and remote power generation. LNG
currently offers significant economic and environmental
benefits to the users. These benefits have been proven
by the recent California Clean Ports initiative at the Ports
of Los Angeles and Long Beach, where LNG and CNG
fuelled vehicles were given advantageous access to the
ports. LNG currently sells at US$ 1 1.50 discount to
diesel on an equivalent diesel basis (dge). LNG also offers
companies an economic way to achieve the Environmental
Protection Agency Tier III and Tier IV emission guidelines for
off-road vehicles and stationary equipment, such as power
generators. LNG is transportable and is cost effective at
current pricing.
Cleaning up the gas
Natural gas occurs with a myriad of compositions. Every
gas formation has a characteristic composition and that
composition can also vary over time as the field is produced.
Natural gas is always comprised of mainly methane (C1),
but natural gas liquids (NGLs), ethane (C2), propane
(C3), butanes (iC4, nC4), pentanes (C5s), hexanes (C6s)
and other hydrocarbons are also common components.
Water and inert gases, such as CO
2
, nitrogen and helium,
sulfur compounds (including H
2
S, COS, CS
2
) and organic
sulfurs (such as various mercaptans) may also be present
in natural gas. The producer of natural gas or LNG must
understand the markets that are available for the gas. If
the gas will be sold into the pipeline grid for utility use,
then it must meet the tariff composition specifications of
the pipeline. Every pipeline in the US has slightly different
Table 2. Case study: Bakken associated gas stream gas composition
Typical Bakken composition (ND EERC End-use technology study 5 November 2012)
Average flare flow rate: 300 000 SCFD Assume gathering system: 8 flares
Design flowrate: 2.50 MMSCFD
Composition Mol% MW Cleaned Normalised Recovered
(gal./d)
Wellhead
(US$/gal.)
Revenue (US$)
CO
2
0.57 44 0.00% 0.00% 0 0.00 0
N
2
5.21 28 2.00% 2.08% 0 0.00 0
Total S 0.00 NA 0.00% 0.00% 0 0.00 0
H
2
O 0.02 18 0.00% 0.00% 0 0.00 0
C1 57.67 16.7 57.67% 59.95% 19 081 0.75 14 311 (5% C2)
C2 19.94 30 19.94% 20.73% 7751 0.20 1550
C3+ 16.59 49.7 16.59% 17.25% 13 334 0.60 8000
Total 100 96.20% 100.00% 40 167 Revenue
(US$/d)
23 862
Monthly 715 870
340 day annual 8 113 191
LNG_MayJune_2013_33-38.indd 36 23/05/2013 10:28
tariff requirements, but typically the water must
be removed to less than 7 lbs/MMSCF, nitrogen
and CO
2
must not exceed 2 3% and the higher
heating value (HHV) must be in the range of
1000 1100 Btu/SCF. Gas processing facilities
are provided to remove the sulfur, CO
2
, water,
and NGLs to meet the pipeline specifications. The
NGLs are recovered and further fractionated and
purified to be sold into other fuel and chemical
feedstock markets.
LNG shelf life and
market composition
considerations
Producing LNG requires cryogenic refrigeration to
cool and condense methane at about -260 F. This
cryogenic refrigeration can be provided by one
of many various open art or proprietary licensed
liquefaction processes. The cryogenic refrigeration system
may also be used beneficially to separate the methane,
ethane and C3+ product streams. Some components of
natural gas can freeze and foul the cryogenic heat exchanger
and must be removed. These include water (<1.0 ppm),
CO
2
(<50 ppm), C5 (<0.1 mol%) and C6+(<20 ppm). To
meet most LNG specifications, H
2
S must also be removed
(<4ppm) and total sulfur must be minimised (<20 ppm). If
the LNG end user is a pipeline then the tariff composition
limits will apply. Dual fuel diesel engines, which utilise a
small amount of diesel for ignition and LNG for the balance
of fuel, are typically tolerant to a wide range of LNG
compositions. A gas turbine driven power generation facility
will require a gas composition that meets the Wobbe Index
(or Modified Wobbe Index) specified by the gas turbine
supplier. If end users include spark ignited reciprocating
engines then the Methane Number (MN) of the fuel must
typically be greater than 85 to avoid detonation.
Further complicating the LNG sellers life is the fragile
nature of LNG. LNG only stays in a liquid state at very cold
Table 3. Case study: Bakken associated gas stream cases with CAPEX
Option 1. Machinery based process: CO
2
and mol sieves, machinery based liquefier
CAPEX US$ 20 million
Revenue US$ 715 870
OPEX US$ 150 000 monthly: electric power, operators,
maintenance
Gross income US$ 565 870
Option 2. Stationery equipment: CO
2
and Dehy mol sieves, LIN assist gas plant/
liquefier
CAPEX US$ 10 million
Revenue US$ 715 870
OPEX US$ 280 000 monthly: electric power, operators, LIN,
maintenance
Gross income US$ 435 870
The Marshall Islanos Registry
servIce and qualIty are nIthIn your reach
International Registries ,U.K., Limiteo
io offliotio oit/ t/ Mor/oll Ilooc Moritio C Crrot Jcoioitrotr
tel: -!! 20 7o38 !7!8
|
lonoonregister-iri.com www.register-iri.com
LNG_MayJune_2013_33-38.indd 37 24/05/2013 09:52
temperatures. It is stored at its bubble point (boiling) in
highly insulated storage tanks designed to minimise heat
leaking into the fuel from the environment. The LNG is
constantly boiling off as heat leaks into the tank. If stored in
a pressure vessel with a relief valve, the pressure in the tank
will increase until it is vented. If the pressure is fairly
constant in a large storage tank, then the vapour boiling off
(boil-off gas or BOG) must be removed by compressors and
either discharged to a pipeline, used as fuel or recondensed
in a liquefier. Nitrogen and methane have the lowest boiling
points in LNG and will preferentially boil off first, leaving
behind the heavier fractions of ethane, propane and butanes.
Unless the LNG is constantly sold and replenished, the
composition of the stored gas will change over time,
gradually increasing in heating value as the methane and
nitrogen boil off. At a production facility this is not an issue;
however, if the end user is a fleet of vehicles and some of
the vehicles sit idle for a considerable length of time, the
heating value/MN of the contents inside the LNG tank could
trend outside of the engine manufacturers guidelines for
detonation, and could cause serious damage to the engines.
The LNG transportation fuels industry has established
some guidelines to ensure that fuel will not likely fall out of
specification. The recommended composition for
transportation fuel that includes spark ignited engines is to
provide an LNG that is >97% methane, <1% ethane and <2%
inerts. LNG of this quality is difficult and expensive to make
and many LNG fuel marketers have relaxed the guidelines to
LNG having a MN>90 to assure adequate supply for their
customers.
Dollars and cents
What are the economics for mini-LNG applied to shale oil
associated gas? The current market prices for various fuels
are shown in Table 1. Fuel prices are compared on a HHV
basis with natural gas and LNG.
The market prices clearly show an opportunity for
exploiting LNG as a replacement fuel for diesel. LNG
currently sells at about 50% of the diesel value before taxes,
marketing and transportation costs are applied. Condensate,
propane and butane also have significant value. Ethane is
the lowest value hydrocarbon. It is a refrigerant, a fuel, a
feedstock for ethylene crackers and can also be converted to
synthetic liquid fuels in a gas-to-liquids (GTL) plant. The
abundant supply of ethane from the shale oil revolution has
flooded the market and it is currently selling at record lows.
Many ethylene plants are being built to exploit this low cost
feedstock.
If the mini-LNG facility is simplified to only produce the
three proposed products, what prices can the producer
expect to receive? Some general assumptions and
simplifications have been made as follows:
A composition for Bakken associated gas has been used
(published by the North Dakota EERC in its recent flare
study published in November 2012).
LNG is selling for approximately US$ 12/million Btu HHV
(about US$1/gal.). Assume that the facility can capture
75% of this price.
Ethane is currently selling for approximately
US$ 0.20/gal. at the wellhead in the Eagle Ford.
C3+ (propane + heavier hydrocarbons) will sell at a
discount to purity propane to cover transportation and
fractionation costs. Assume the facility can capture
US$ 0.60/gal. for the C3+ mix.
After considering the current market prices for fuels, the
economics for a mini-LNG facility can be estimated. A
2.5MMSCFD mini-LNG facility is shown in Tables 2 4. Two
different liquefaction technologies are evaluated. The first is
a conventional molecular sieve based CO
2
and dehydration
system with a rotating machinery based cryogenic
refrigeration system. The second option is a simple
molecular sieve pretreatment and a refrigeration system
utilising liquid nitrogen as a disposable refrigerant.
The analysis demonstrates that flared gas can be
economically recovered and sold into the various markets. It
also shows the fuel savings after taxes, transportation and
fuel service costs are applied for a truck fleet converting to
LNG fuel.
Conclusion
Mini-LNG facilities can offer a flare-to-fuels solution for
associated gas that is currently being flared and
wasted in the remote shale oil regions. With
environmental regulatory pressure mounting
to reduce flaring, an adequate supply of
inexpensive flared gas in the wet and oil shales
is not in question. The hydrocarbons can be
economically captured and sold into the fuels and
petrochemical markets. There is sufficient margin
for profitability to fully produce the flared gas by
condensing it into liquid products, even on a small
scale of only 2.5 MMSCFD. There is sufficient
demand for the LNG fuels and adequate pricing
for NGLs to justify many projects, providing the
US with the opportunity to quench the flares
and reap the additional benefits of the shale oil
boom, by recovering the hydrocarbon fuels in the
associated gas.
Table 4. Case study: Bakken associated gas stream cost savings for the
end user
Case: fleet of 190 over-the-road trucks consuming an average of 100 gal./d of LNG
LNG cost US$ 0.75/g of LNG = US$ 1.28 dge
Federal tax US$ 0.24/g of LNG = US$ 0.41 dge
State tax US$ 0.12/g of LNG = US$ 0.20 dge
Transportation US$ 0.10/g of LNG = US$ 0.17 dge
Fuel service US$ 0.25/g of LNG = US$ 0.43 dge
Total US$ 1.46/g of LNG = US$ 2.49 dge
Average retail diesel cost US$ 4.00/gal.
Saving US$ 1.51/gal.
Annual savings for fleet US$ 5 918 109
LNG_MayJune_2013_33-38.indd 38 20/05/2013 11:44
M
odularised gas processing equipment is a proven solution to
deliver fast, effective acid gas and contaminant removal, as well
as natural gas liquids (NGL) recovery for North American shale
gas resources. Through a packaged plant delivery model, modularised
amines, adsorption, and cryogenic turboexpander equipment skids can be
prefabricated and delivered to customer locations, reducing costs and
increasing profits due to speed to gas. Modularised solutions can be
particularly advantageous for gas flow rates below 500 million ft
3
/d
and scalable to higher flows by incorporating multiple equipment
trains. Packaged equipment simplifies gas processing projects
by offering faster construction times, greater reliability as a
result of proven shop-fabricated quality, and easier installation
Flexible processing
Rebecca Liebert, UOP, a Honeywell
Company, provides a guide to
selecting appropriate gas
processing technologies.
LNG_MayJune_2013_39-44.indd 39 21/05/2013 11:33
via skid-mounted delivery, compared to field fabricated
solutions.
This article incorporates these themes to provide a
decision framework to guide appropriate selection of gas
processing technologies. Within the presented framework, it
is not only important to select the proper processing
technology but also to consider the interactions among each
gas process unit and to adjust sequencing of the process
steps to optimise operations, thereby increasing operating
flexibility within the overall gas processing system.
Experience within the gas processing space has proven that
flexibility should serve as a key attribute when selecting the
proper solution for gas processing, including LNG
applications. Flexible operating and investment capabilities
are becoming more important as the industry develops
more sub quality and distributed gas reserves, as it is likely
the variability of the feed gas composition and feed flow
rates will increase.
Gas processing technology
decision guide
Numerous factors impacting the upstream resource,
midstream project specifics, and downstream commitments
affect the selection of gas processing technologies. The key
decision blocks and variables underlying a design decision
making matrix are displayed in Figure 1.
The upstream factors impacting gas processing
technology selection include:
Hydrocarbon (HC) distribution composition of NGLs
and/or crude oil within the feed stream.
Production profile of the hydrocarbon reserve.
Table 1. AGR (acid gas removal) functional process block detail
Functional requirements Process options Key selection criteria Interactions/comments
Reduce CO
2
, H
2
S,
mercaptan and COS
Solvents In Out Interference of heavy
hydrocarbons
Concentrate for further
processing (tail gas treating)
Membranes Gas flow rate Gas quality targets Sequencing with
dehydration
Minimise hydrocarbon
losses
Adsorbents Concentration AG disposal/product
options
Relative H
2
S/CO
2
Chemical (chem-sorb,
scavengers
Pressure Value/penalty for CO
2
use/
discharge?
Heavy hydrocarbon content
Figure 1. Flow schematic representing key decision criteria for gas processing projects.
LNG_MayJune_2013_39-44.indd 40 21/05/2013 11:33
Feed gas contaminants including acid gas (CO
2
, H
2
S,
mercaptans), mercury, arsenic and nitrogen.
Feed gas pressure and flow rate.
Ambient conditions such as temperature.
Onshore/offshore situation of resource and processing
equipment.
Regional fiscal regime impacting project funding.
Environmental constraints, either local or global.
When considering plant configurations and design
decisions, operating flexibility should serve as a key design

objective for the processing system. Flexibility is important
due to the complex interrelationship and relative
uncertainties among input upstream factors, interactions
among the processing plant functional blocks, and overall
delivered project economics and/or downstream
commitments.
Within the gas plant configuration, the dynamic
interactions among functional blocks within Figure 1 will
also drive processing systems towards more flexible
operations. As an example, the AGR process block within
Figure 1 is detailed in Table 1 to show the various decision
elements underlying appropriate technology selection.
Figure 2. AGR, HCM and OCM process block decision tree for sample scenario.
Figure 3. Integrated natural gas processing flow scheme for case study.
LNG_MayJune_2013_39-44.indd 41 23/05/2013 10:31
An example of an integrated AGR, HCM and OCM decision
tree for a sub quality gas feed and specified products for export
by pipeline is included in Figure 2. Of course, the decision tree
would be similar for LNG applications, with the added
complexity of adding the downstream liquefaction step.
When determining the proper solution set for a gas
processing project, it is important to conduct the following
three exercises in unison:
Select the proper technology solution within each
processing block.
Account for interactions across different processing blocks.
Adjust the sequence of processing blocks for optimisation.
For this particular feed, which is high in CO
2
and H
2
S, an
amine unit integrated with a licensed turboexpander and
mercury guard bed, highlighted in red in Figure 2, is identified
as the appropriate technology selection to meet the product
specifications.
Integrated packaged
solutions
Delivering gas treating solutions via
packaged modular process units can
provide clear economic and schedule
advantages for small and mid-sized
LNG projects, particularly in situations
of distributed (or remote) gas reserves
or when the lead-time for start-up of
gas processing units represents a critical
path element. Some industry participants
are also considering packaged modular
process units for larger-scale projects by
incorporating multiple equipment trains.
Table 2 illustrates the benefits of
packaged modular vs. field (or stick) built
processing units based upon a number of
project criteria.
Individual packaged process units can
be integrated to ensure overall system
flexibility is optimised. This capability played a vital role
in the rapid development and success of the US wet
shale gas industry and is likely to contribute to the
success of the development of sub quality and
unconventional gas elsewhere. An example is the
Thomas Russell Company, now majority owned by
Honeywell and sold under the UOP Russell product
family. The Thomas Russell story parallels the US shale
gas revolution. As a result of the perseverance of
visionary entrepreneurs, dry shale gas production
increased by a factor of 10 in just six years and now
makes up 30% of total US gas production. The rapid
increase in supplies outstripped demand and the market
price of dry natural gas fell to the point that it is no longer
economic to drill for dry gas in most shale gas reservoirs
in the US. Natural gas production continues to rise
because drilling activity shifted to wet shale gas and oil
producing significant volumes of associated gas. This
shift to wet shale gas and shale oil was enabled by the
NGL recovery packaged equipment solutions provided
by Thomas Russell. Without these gas processing
solutions, these vital resources could not be developed. The
Thomas Russell fast gas business model was key to enabling
quick resource monetisation. Its pre-fabricated modular process
units could be quickly installed and had much lower lead-times
than customised or stick-built plants with no loss of quality. The
UOP Russell product line is now ready to enable shale gas, shale
oil, and wet conventional gas resource development globally.
Case study: sub quality gas
An example feed gas high in H
2
S and CO
2
is shown in Table 3 to
serve as a basis for this case study.
The UOP modeled process flow diagram presented in
Figure 3 incorporates acid gas, mercury, mercaptan, water, and
NGL removal via an integrated process flow scheme that utilises
gas-phase separation of RSH and COS within the sulfur plant.
An integrated flow scheme will provide higher overall
facility operating efficiency, increased hydrocarbon recovery,
and increased sulfur recovery compared to a conventional flow
scheme that utilises liquid phase desulfurisation. Additionally,
Table 2. Comparison of packaged modular vs. field built process units
Capability Packaged
modular offer
Field built
alternatives
Ability to minimise or eliminate FEED
Availability of major equipment from suppliers
Degree of cost control
Reduced field labour cost
Flexibility to process a wide range of natural gas
streams
Ability to relocate plant
Fast cycle delivery performance
High level of field performance
Ability to deliver the plant globally
Key: filled circle represents higher capability; unfilled circle represents lower
capability
Figure 4. Relative change in cycle time and regeneration flow for the MolSiv unit as a
function of mercaptan feed percentage increase.
LNG_MayJune_2013_39-44.indd 42 21/05/2013 11:33
eliminate handoffs
With integrated natural gas processing technology from one
supplier, UOP is the right connection.
Integration isnt just the key to a successful contaminant-removal system.
Its also fundamental to saving money as you monetize your resources. As
your single-source supplier for a wide range of gas processing technologies,
UOP reduces engineering and execution schedules so that your time to rst
gas is reduced. UOP technologies are designed for true integration, delivering
energy savings. For removal of acid gas, water and trace sulfur compounds
plus recovery of natural gas liquids, dont risk handoffs. Go with the
integrated processes of UOP.
For more information, visit www.uop.com.
2012 Honeywell International, Inc. All rights reserved.
LNG_MayJune_2013_39-44.indd 43 21/05/2013 11:33
when an integrated gas plant project is executed via
prefabricated modular supply of key process units, including
the AGRU, MolSiv, and NGL, significant lead-time
acceleration and reduced start-up complexity will be
achieved vs. field (stick) built process units. For LNG
applications, an added modular liquefaction unit can be
integrated into the system.
The integrated process can adjust to changing
mercaptan feed conditions with a minimal impact to
operating costs. As the mercaptan content in the feed
increases, the integrated scheme has the flexibility required
to treat this new feed with minimum revamp of the units.
The operating strategy involves changing the cycle time of
the MolSiv unit and rerouting the regeneration gas streams.
No incremental MolSiv vessels or Selexol columns are
required. Figure 4 shows the impact of the increasing
mercaptan contents in the feed on the MolSiv unit cycle time
and the regeneration flow. As the regeneration flow is
increased, the associated equipment in the regeneration
loop, such as heater, cooler and compressor, will need to be
increased in their duty or size. Depending on the mercaptan
increase, the regeneration equipment may already have the
extra capacity. As the regeneration flow increases, the
pressure drop across the Selexol unit also needs to be
managed carefully. Additionally, increasing mercaptan feed
will have a minor impact on operating costs. A 50% increase
in mercaptan feed has been modeled to result in only a 3 to
4% increase in total operating costs.
In cases of high or increasing H
2
S and CO
2
feed
composition, a packaged membrane system can be
incorporated upstream of the AGRU, as shown in Figure 5.
Incorporating a membrane system prior to the AGRU for
bulk CO
2
and H
2
S removal can offer the following
advantages for acid gas removal vs. a stand-alone AGRU:
Operating flexibility: ability to respond to changing acid
gas feed composition via high turndown capability.
Investment flexibility: membrane capacity can be easily
expanded by incorporating additional elements to
existing membrane modules or by installing incremental
membrane modules.
Capital cost reduction potential: can lower capital cost
and size of the AGRU at the expense of incremental
membrane capital.
Operating cost reduction potential: can provide
operating cost advantages vs. stand-alone AGRU in
situations where electricity is cost-advantaged vs. steam.
Furthermore, membranes can be particularly
advantageous for remote locations where labour or solvent
replenishment capabilities are a concern or in situations
where plant plot space is limited.
Summary
UOP believes an integrated design approach with an
emphasis on flexibility for both operations and investment
decisions will prove critical to successfully monetising
new gas resources. Flexibility is important due to the
complex interrelationship among input upstream factors,
interactions among the processing plant functional
blocks, and overall delivered project economics and/or
downstream commitments. Furthermore, incorporation of
modularised process units into an integrated package can
accelerate project schedules and help stage investment
decisions.
Table 3. Feed gas conditions for case study
Feed gas conditions Mol %
H
2
S 23%
CO
2
10%
N
2
<0.1%
Methane 58%
C2+ hydrocarbons 9%
H
2
O Saturated
COS 0.04%
Mercaptans ~ 150 ppmv
Molecular weight 25
Flowrate (million ft
3
/d) 500
Temperature (C) 50
Pressure (bara) 67
Figure 5. Integrated flow scheme incorporating a membrane prior to the AGRU.
LNG_MayJune_2013_39-44.indd 44 21/05/2013 11:33
O
ne of the most neglected areas in gas
treating is the correct selection of
tower internals in both absorbers and
regenerators. The literature is replete with
data on vapour-liquid equilibrium in a host of
aqueous amine systems containing both single
amines and mixtures. Considerable data also
exist on a variety of physical, thermal, and
transport properties such as solution density,
viscosity, heat of reaction, heat capacity,
and diffusion coefficients. The kinetics of the
reaction between dissolved CO
2
and various
amines has been measured and reported in
terms of Arrhenius parameters numerous times
for all the commonly used amines. There is no
question that such parameters are important
in interpreting laboratory measurements of
absorption rates and all of them have been used
in that setting. Far fewer of these parameters
have been applied in the design of commercial
equipment using traditional methods such as
equilibrium stage calculations, for the simple
reason that ideal stages know nothing about
what, if anything at all, is in the column.
However, acid gas absorption and solvent
regeneration are carried out commercially in
columns with real internals. Apart from hydraulic
considerations, the mass transfer performance
of tower internals unfortunately has tended to
be almost completely ignored. This is despite
the fact that the mass transfer characteristics
of the internals play a central role, and are at
least as important in setting tower performance
as phase equilibrium and reaction heat. The
translation from numbers of ideal stages
to actual tray counts, and to the required
depths of structured or random packing, has
traditionally been done solely on the basis of
experience. This approach works quite well in
light hydrocarbon separations, for example,
PACKING FOR
CO
2
REMOVAL
Ralph H. Weiland and Nathan A. Hatcher,
Optimized Gas Treating, Inc. and
Jenny Seagraves, INEOS Oxide, USA, discuss
the benefits of choosing the right packing
for absorbers and regenerators in a gas
processing facility.
LNG_MayJune_2013_45-52.indd 45 21/05/2013 11:43
where tray efficiencies are fairly constant and well known
(except for very high purity) and where vendor data exist
on HETPs and HTUs for various packings for the systems
of interest. Amine treating, however, is undoubtedly one of
the processes least amenable to extrapolation into areas
where experience is lacking. When parts-per-million (ppm)
specifications must be met on product gases, selecting the
wrong packing size or bed depth can result in a failed design.
For many years, packing has had a bad reputation in
absorption and distillation at high pressure. Part of the
reason is inattention to proper distributor design. Another is
the persistent and still unresolved difficulty in translating
ideal stages to actual packed bed depths and the selection of
a particular commercial packing. However, packing is now
being used increasingly in gas processing for several reasons.
For a given size, packed columns tend to permit higher
throughput than trays. And in offshore operations, such as
FPSO and FLNG, columns using structured packing are much
less susceptible to the effects of rocking motion caused by
wave action. It is particularly critical to select the right
packing type and size in FLNG because of the high cost
associated with weight and footprint, so the need for
reliability and accuracy is even greater.
It is no longer necessary to engage in any form of
guesswork when it comes to designing towers containing
packings from a large array of possible packing types and
sizes. The key is to simulate the column as the mass transfer
device it really is. This goes well beyond hydraulic or capacity
calculations and entails a mass transfer rate approach to the
process itself. Although rate calculations require knowledge
of the mass transfer characteristics of the internals (mass
transfer coefficients for both phases, and the interfacial area),
this kind of information is available in the literature as well as
within the ProTreat
amine treating simulator. The ProTreat

simulation makes packing performance just as easy to
predict as trays. The emphasis here is on spot-on prediction.
The simulation is out-of-the-box reliable and accurate.
In what follows, attention is on the performance of tower
packings, first by comparing simulation with performance
measured in an LNG plant, and then by considering the effect
of packing size for a particular family (brand) of structured
packing used in the same service.
Predicting performance of
random and structure packing
The selection of tower internals for distillation has been well
reviewed.
1
In CO
2
removal, packing has tended to be used
less frequently, but there are several reasons why packing
may be preferable to trays:
Pressure drop is usually lower.
Tower capacity is often higher/smaller diameter towers.
Foaming tends to be less severe, if it exists at all.
In applications subject to rocking motion, such as FPSO
and FLNG, structured packing offers better resistance to
upsets caused by periodic tower tilt.
In low liquid load applications where the trays become
dry and operate in the spray regime.
In columns of less than 750 mm diameter, where trays
are impractical.
A cautionary note is that random and structured packing
should not be used in applications where solids are present
or if fouling is a concern.
For the low to modest liquid loads typical of LNG
applications where the CO
2
content of the inlet gas is not
extremely high, structured packing has several advantages.
Increased gas handling capacity can be a significant factor in
high pressure towers, in situations where excessive weight
and footprint have severe cost penalties. This is particularly
the case in FPSO and FLNG where tightening a design can
yield large cost advantages. The smaller column diameter
offered by packing, coupled with a high capacity specialty
amine solvent, is certainly beneficial, but so is using only the
packing depth actually required. The trick is determining this
depth reliably. Tower diameter is determined by hydraulics,
and hydraulic performance is well documented and well
understood. The same cannot be said for packed bed depth.
This is determined by mass transfer, and the mass transfer
performance of structured packings with chemical reactions
is still an uncertain and uncomfortable area for many
practitioners.
Packed columns always seem to have presented a
challenge to designers. This is possibly due to the fact that
there are so many varieties, types, and sizes of packing, and
perhaps even more because the experience base is so small,
especially in gas treating with amines. However, today there
is no reason why packed columns cannot be designed with
just as much certainty and confidence as trays. The processes
taking place in absorption and regeneration towers are just
normal mass transfer processes. As long as one has access
to the basic mass transfer characteristics, as embodied in
mass transfer coefficient correlations for the particular
internals of interest, packed columns are no harder to specify
and design than their trayed counterparts. The fundamental
correlations contained within the ProTreat simulators
information base have been developed from literature,
vendor and research data. They have been shown repeatedly
to allow accurate and reliable predictions of column
performance without recourse to estimating artificial
parameters, such as tray efficiencies or fictitious residence
times on theoretical stages or tray thermal efficiencies.
Before embarking on a case study to show the effect of the
Figure 1. Structure packing bricks used in revamp.
LNG_MayJune_2013_45-52.indd 46 21/05/2013 11:43
chart-ec.com 1-281-364-8700
Chart Modular LNG...
Building Her
Energy Future
Standard modular liquefaction plants for the
transportation and energy industries
- C100N 100,000 gpd Nitrogen Cycle
- C250IMR 250,000 gpd IPSMR
- C450IMR 450,000 gpd IPSMR

- Optlmum llquefactlon solutlons for
monetlzlng small and mld-scale gas
reserves
- Standard deslgns and maxlmlzed
shop bulld for reduced costs and
mlnlmum pro[ect schedule
- 8razed alumlnum heat
exchanger technology for
superlor performance and
lmproved plant nexlblllty

Delivering a clean-burning,
safe fuel alternative to diesel
for her future and yours
LNG_MayJune_2013_45-52.indd 47 23/05/2013 11:06
size of the structured packings within a particular vendors
portfolio, it may lend conviction to the analysis to
demonstrate first ProTreats ability to predict performance in
a commercial unit.
Comparison with commercial
data
The demonstration case involves the revamp of the amine
section of an LNG plant using GAS/SPEC* CS-2020* solvent
to treat a feed gas with 2.24% CO
2
at approximately 4 MPa.
The unit was originally built with trayed columns but was
unable to achieve more than 70% of the nameplate capacity,
purportedly because of foaming. The process flowsheet was
completely conventional with the usual absorber-regenerator
combination connected to each other through a flash tank,
cross exchanger, trim cooler and pumps. Analysis of the
treated gas (27.7 C) showed 21.2 ppmv CO
2
. Using all
the known plant conditions plus the tray details, ProTreat
predicted 18.2 ppmv at 27.7 C. This was what one might
call an out-of-the-box prediction in which absolutely no
parameters were guessed or estimated. The simulation data
consisted of tray and column vendor drawings along with
process flow, temperature and pressure measurements taken
directly from the plants DCS. A true prediction to within a
couple of parts per million of the measured value is certainly
encouraging.
The cause of foaming (if indeed there really was foaming)
was never determined. However, even at only 70% of the
design capacity, the regenerator was running close to the
recommended 85% hydraulic flood limit, so there was a
question as to whether the plant was being prevented from
operating at full capacity because the regenerator was
perhaps undersized, or if maybe there were other problems
with its trays. Because both absorber and regenerator were
designed with the trays inaccessible for inspection, the real
cause was never identified. Regardless of the root cause of
the capacity bottleneck, the decision was taken to replace
both towers with somewhat larger diameter versions, and to
replace the trays with structured packing. The packing was of
local manufacture, but from photographs (Figure 1) its
appearance is similar to several well-known commercial
brands. The crimp size had to be estimated from photographs
and it was found to be between 18 and 25 mm from peak to
peak. Full-capacity plant performance data were taken in
mid-2012, again using GAS/SPEC CS-2020 solvent. The
treated gas was reported to be 7 ppmv CO
2
. Out-of-the-box
ProTreat simulation indicated the unit should have been
producing 1.2ppmv CO
2
with the absorber and regenerator
running at 23% and 36% of flood, respectively, and at actual
design conditions. This was more discrepancy than expected
based on past experience (usually predictions are within
1 4 ppm of measured performance), so the company set
out to assess the possible cause(s). The CO
2
absorber was
running in the completely mass transfer rate controlled
regime.
*
When mass transfer rate controls absorption, and all
the process parameters such as flow rates, compositions and
temperatures have been verified, the reason for discrepancies
must be sought in either mass transfer or sensitivity to some
process condition. All that could be determined for this
particular packing was a very rough estimate of the crimp
size (18 mm) and that the packing sheets were perforated and
embossed, a treatment that promotes liquid spreading. It
was surmised that the packing surface area, which is
calculated from the estimated crimp size, was probably
responsible for the discrepancy. In addition, scatter in wetted
surface areas measured under carefully-monitored laboratory
conditions is known to be within only +/-12 to 15%.
2
In
commercial installations, variance is likely to be greater still.
Surface area for gas to liquid contact has a large effect on the
treated gas purity.
Another factor that pertains to the packing (one that is
impossible to quantify) is the efficacy of the distributor used
in the column. Imperfect liquid distribution will never
generate higher interfacial area or cause better performance.
A perfect match between reported and simulated CO
2

content of the treated gas (and a perfect match with all other
measured data) can be achieved by reducing the wetted area
of a roughly equivalent commercial packing by
approximately 20%.
In light of these factors and the inherent uncertainty in
process plant measurements, the 6 ppm discrepancy
between simulation and measurement is certainly within the
bounds of plant measurement accuracy and the inability to
do more than roughly estimate the packing size. Simulation
yields satisfactory performance from a gas quality point of
view. These simulated results are pure predictions, and are
certainly close enough to actual measured performance to
promote confidence in ProTreats ability to predict the effect
of packing type and size, for example, on CO
2
removal using
reactive amines.
Effect of packing size on CO
2

removal in an LNG facility
The effect of packing size on treating for CO
2
removal can be
seen by considering a range of structured packings within a
particular brand series, in this case Koch-Glitsch FLEXIPAC 1X,
1.4X, 2X, 3X and 4X (X is the common designation for a crimp
angle of 60). The FLEXIPAC brand is merely representative
of structured packing in general, and it was selected
arbitrarily and without prejudice. In the order written above,
this packing series is in increasing crimp size and decreasing
dry specific surface area. For the same gas and liquid flow
rates, different sizes also have somewhat different liquid-film
Table 1. Raw gas used in case study
Composition (mol%)
H
2
S 0.0001 iC4 0.020
CO
2
2.000 nC5 0.006
C1 94.331 iC5 0.010
C2 1.900 nC6 0.030
C3 0.170 CH
3
SH 0.0007
nC4 0.030 N
2
1.500
Conditions
Temperature (C) 37.2
Pressure (bara) 64.4
Flow (Nm
3
/d) 6.7 million
LNG_MayJune_2013_45-52.indd 48 21/05/2013 11:44
weatherford.com
Improve
Process Control
Shorten project schedules and reduce reworking losses by choosing
a single-source provider for all of your process pipe needs.

2
0
1
3

W
e
a
t
h
e
r
f
o
r
d
.

A
l
l

r
i
g
h
t
s

r
e
s
e
r
v
e
d
.

I
n
c
o
r
p
o
r
a
t
e
s

p
r
o
p
r
i
e
t
a
r
y

a
n
d

p
a
t
e
n
t
e
d

W
e
a
t
h
e
r
f
o
r
d

t
e
c
h
n
o
l
o
g
y
.
Drilling
Evaluation
Completion
Production
Intervention
Pipeline & Specialty Services
Precommissioning
Commissioning
Ha|ntenance
- Pigging
- Cleaning
- Inspection
- Surveying
- Repairs
- Clamps
Shutdowns
Decommissioning
Turn to Weatherford for all of your process pipe services, including:
Bund|ed services save you money, whi|e sing|e-source efciency
shortens project duration and avoids costly reworking.
Enjoy the convenience and advantages of having a single point of
contact and accountabilityalong with a single contract and invoice.
Weatherford offers complete solutions for process pipe service
projects of any size, anywhere in the world. A recent success in
Yemen reduced the original duration projection by 14 days.
For more information on this success or any of our services, email
pss@weatherford.com. To view our full line of Pipeline & Specialty
Services offerings, visit us online at weatherford.com/pss.
Bo|ting
Va|ve testing
Chemica| c|eaning
F|ange management
Nitrogen services
Hydrostatic testing
Pneumatic testing
He|ium |eak detection
Delivering Quality Under Pressure
SM
LNG_MayJune_2013_45-52.indd 49 21/05/2013 11:44
mass transfer coefficients and, therefore, the same chemical
reactions affect mass transfer rates to different extents.
Table 1 describes the raw gas common to all the cases in
the study. The solvent was 50 wt% GAS/SPEC* CS-1160*
solvent flowing at 204 m
3
/h and 50 C. The regenerator
contained 13.7 m of type FLEXIPAK 3X structured packing in
all cases. The absorber contained 15.25 m of various crimp
sizes of FLEXIPAC structured packing as described above. In
all cases, both columns were sized for 70% flood regardless
of the packing. The regenerator was 1875 mm diameter
while the absorber ranged from 2825 mm diameter with the
finest packing (1X) to 1885 mm diameter with the coarsest
(4X). Reboiler duty was held constant at 13.5 MW and the
molar stripping ratio was typically 1.2 at the regenerator
overhead.
Table 2 is a synopsis of the simulated treating
performance with 15.25 m of each packing size. The lowest
size designation has the largest specific (dry) area and also
treats the gas to the lowest CO
2
level. Note that although
there is an inverse relationship between area and treated gas
CO
2
content (just as one should expect), it is anything but
inversely linear, treated gas quality is a very strong function
of dry area, i.e., packing size. As ever coarser packing is used,
the temperature bulge moves closer to the top of the
column. What may be surprising is that it also grows in size
until it starts to fall again at very large packing sizes. It may
also be interesting to note that the ratio of wetted area to dry
area of the packing grows with packing size, and can exceed
unity by a considerable fraction. The same liquid flow has a
much larger area to spread across for small packings, thus
leaving more of the packing in the dry state. But for large
packings, the available area for spreading is more restricted.
The wetted area can exceed the dry area because all liquid
flow is not restricted just to the mechanical surface of the
packing when the film becomes thick. A thick film becomes
quite disrupted, large waves form and a certain amount of
sparging of the gas through the liquid probably occurs.
In many ways, tray performance is close to the
performance of the smallest packing, but this is definitely not
because of higher surface area for mass transfer. Indeed, in
this case, equivalent wetted surface area in the biphase
based on the volume between any two trays is
approximately 130m
-1
whereas the column-average wetted
area for 1X packing is twice as high at about 270 m
-1
. Instead,
the intense agitation on the trays gives a mass transfer
coefficient almost four times larger than for the 1X packing
(this also turns out to be true for all the packing sizes in the
FLEXIPAC family).
Several points emerge from this. First, structured (and
random) packings are no
more challenging to a true
mass transfer rate based
simulation than trays.
Secondly, one cannot
simply scale up the effect
of packing size in any
simple or logical
arithmetic way. Packing
performance is a complex
function of packing size,
hydraulics, and chemical
reaction kinetics. Third, the
effective wetted area
active on a structured
packing is not limited to
the packings dry area, so
scale-up based on dry area
Table 2. Simulated treating performance, bulge temperature and bulge position
FLEXIPAK
Dry area Treated gas CO

2
Bulge
temperature
Bulge position
from top
Wetted/dry
area ratio
Packing size (m
2
/m
3
) (ppmv) (C) (m or tray No.) (unit less)
1X 440 0.45 88.9 13.4 0.614
1.4X 340 0.46 96.9 13.1 0.741
2X 220 1.79 114.8 9.0 0.973
3X 110 108 110.3 7.0 1.332
4X 55 516 105.4 7.0 1.691
25 trays 0.93 88.9 Tray 20
Figure 2. Absorber CO
2
profiles for various packing sizes and
for trays.
Figure 3. Absorber temperature profiles for various packing
sizes and for trays.
LNG_MayJune_2013_45-52.indd 50 21/05/2013 11:44
could be wrong by several fold. In short, scale-up based on
random and structured packing size is simply not feasible; for
example, one cannot predict the performance of 2X packing
from knowing how 1X performs. One must inevitably
conclude that the only way forward is with a mass transfer
rate-based simulation capability, soundly based on principles
of mass transfer.
The response in performance that accompanies various
packing sizes is described in greater detail in the plots of
Figures 2 and 3. In Figure 2, the gas-phase CO
2
profiles for 1X
and 1.4X packing are closely coincident. Thus, for small-crimp
packing the final treating level is determined by the partial
pressure of CO
2
in equilibrium with the entering lean amine,
not by mass transfer rates. In other words, the absorber with
very fine packing is completely lean-end pinched. The treated
gas CO
2
content is almost identical between 25 trays on
600mm tray spacing and 15 m of fine packing, but the trayed
absorber does not exhibit quite the same degree of lean-end
pinching as the packing (although it is still pinched because
the top six trays reduce the CO
2
content by less than 1 ppmv).
However, with 2X and larger packing, treating ceases to be
lean-end pinched altogether.
As Figure 3 shows, coarser packing results in a larger
temperature bulge, which moves up towards the centre of
the column. The largest packing simply does not have
sufficient area to allow rapid enough CO
2
absorption to
achieve satisfactory treating. The temperature bulge is much
larger though, because a lot of CO
2
is being absorbed locally
into a relatively small volume of liquid. Insufficient physical
supporting structure (i.e., dry packing surface area) for the
liquid results in lower holdup volumes, therefore
temperatures are higher, backpressure of CO
2
goes up and
absorption rates suffer. Actually, the absorber with 2X
packing is bulge pinched. This can be seen from Figure 4,
where the actual CO
2
concentration in the gas at various
positions almost coincides with the equilibrium values along
the lower half of the absorber. Figure 5 is a similar plot for a
much larger packing size. In both cases, there is almost no
driving force for absorption. Once past the bottom metre or
Figure 4. Actual and equilibrium CO
2
concentrations in the gas
at various positions in the absorber. Packing is FLEXIPAK
2X.
Taking the Guesswork Out of CO
2
Removal
ProTreat Simulation Software
Get Real
Dont Guess!
SYNGAS, LNG AND CARBON CAPTURE
ProTreat accurately predicts column performance.
100% Mass-Transfer-Rate based no ideal stages.
Grassroots? Revamp? Designing for Packing?
Using MDEA with piperazine? HotPot?
Build a virtual plant with the industrys
most advanced gas treating simulator.
www.ogtrt.com +1 281 970 2700 info@ogtrt.com
Learn how state-of-the-art simulation science can benet you.
LNG_MayJune_2013_45-52.indd 51 21/05/2013 11:44
so of packing, the gas has to wait until it is nearly half way
up the column before significant absorption rates can
resume. It is the intensified temperature bulge that is
responsible for this type of pinch.
Using structured packing in
gas treating
In the last few years, engineers have shown increasing
interest in using structured packing coupled with specialty
solvents in amine-based gas treating applications.
Consequently, structured packing is being applied more and
more widely. Nevertheless, the idea is still new enough that
questions are often asked as to whether structured packing
should or should not be considered in a given application.
One of the most obvious application areas in which the
use of structured packing is almost mandatory is in columns
subject to periodic tilting motion, as on floating structures
such as FPSO and FLNG platforms. An advantage of using a
specialty solvent such as GAS/SPEC CS-1160 and CS-2020 is
the lower circulation requirement. This can result in a smaller
plant and lower operating costs. Structured packing resists
liquid maldistribution brought about by rocking motion better
than random packing. Trays have very poor resistance to the
sloshing and seiching induced by lateral back and forth
motion (exceptions are trays that rely on confined centrifugal
motion during contacting, e.g., ConSep and ULTRAFRAC

trays, although these kinds of trays tend to be more
expensive). Structured packing should always be considered
in offshore applications, especially in the context of periodic
tilting motion. However, care must be taken to use the right
kind of liquid distributor. The distributor should have no free
liquid surfaces and should be high pressure drop type, not a
gravity flow device such as a trough distributor. This also
mandates that solid amine hygiene be practiced (i.e.,
filtration and corrosion management).
There are very few reasons to exclude structured packing
from consideration. One such reason is fouling. If the system
is a fouling one, the deposits that will inevitably occur on the
surfaces of the packing will be almost impossible to remove
and operations may soon become plagued by plugging
problems. On the other hand, unless structured packing is
used too close to the flood point where liquid holdup
becomes high, it is naturally resistant to foaming. It should
be recognised, however, that if the system is a bad foamer,
structured packing may not be the answer. Instead, the root
cause of the foaming should be determined and alleviated.
Even a good design is no match for poor amine hygiene.
In a revamp for higher capacity, the naturally higher
vapour handling ability of structured packing may
recommend it as a way to achieve higher capacity in the
same shell. Except in tail gas treating and acid gas
enrichment, pressure drop is not usually an issue. However, if
it is, structured packing can almost always be made to work
at lower pressure drop. To that end, the largest possible
crimp consistent with being able to achieve the target
separation within the height of the existing tower shell
should be used. Again, finding out what that crimp size is can
be facilitated greatly by using mass transfer rate based
simulation.
Summary
Packed absorption columns can operate in various modes
of pinching or be mass transfer rate controlled throughout.
There is no way to tell beforehand which mode will prevail,
and with what packing type and size. This makes it almost
impossible to develop a design that is truly as optimised
as it should be in FPSO and FLNG applications using any
approach other than one with a mass transfer rate basis.
A real mass transfer rate model is constructed from
components that are soundly based in fundamental sciences
and engineering, and not on approximations made to avoid
what used to be impossibly complex and arduous hand
computations. The enormous power of desktop and laptop
computers has turned computer time into a non-issue. Rigour
can be achieved for the meager cost of computing times
measured in tens of seconds. The computer models that
result can be accurately described as virtual plants in which it
is quite easy to investigate very involved what-if scenarios.
Because of their wide range of sizes and the somewhat
laterally compartmentalised flows, structured packings offer
a great deal of flexibility in gas treating, which allows them
to be used where conventional trays are very difficult, if not
impossible, to apply. Possibly two of the greatest barriers to
using structured packing in gas treating have been the very
small experience base and the resulting difficulty in
translating the results of more conventional, approximate
calculations into real internals.
References
1. Pilling, M., and Holden, B. S., Choosing Packings
and Trays for Distillation, Chem. Eng. Prog., p. 44,
September 2009.
2. Lewis, J.C., Seibert, A.F., and Fair, J.R., et al.,
Interfacial Area in Irrigated Packings, paper presented
at Annual Meeting of AIChE, San Francisco, CA,
14 November 2006.
Notes
*CO
2
in the treated gas was not limited to a value set by the
lean solvent CO
2
loading (lean end pinched) and it was not
limited to a value set by too low a solvent flow, i.e., limited
by the solvent capacity (rich end pinched).
Figure 5. Actual and equilibrium CO
2
concentrations in the gas
at various positions in the absorber. Packing is FLEXIPAK
4X.
LNG_MayJune_2013_45-52.indd 52 21/05/2013 11:44
M
ercury, which is present
in natural gas almost
everywhere, causes corrosion
damage to aluminium heat exchangers
and cryogenic aluminium cold boxes
in LPG recovery and LNG plants. This
leads to unexpected shutdowns
and requires expensive repairs to
equipment. For safe and reliable
operation of the plant, mercury must
be removed from the gas. There is a
wide variety of commercially available
mercury adsorbent products for use
in fixed-bed mercury removal vessels.
Sulfur-impregnated activated carbon
offers reliable, high-efficiency mercury
removal upstream or downstream of
Mercury rising
John Markovs,
Adsorption
Solutions LLC,
USA, and
Robert W. Soffel,
Selective Adsorption
Associates Inc.,
USA, examine
solutions to
mercury problems
in LNG plants.
the gas driers. It also offers the highest
mercury capacity per unit volume of the
non-regenerable mercury adsorbents.
Regenerable mercury adsorbents offer
additional options to plant designers.
Several reliable mercury analysers are
commercially available, combined with
good gas sampling techniques, these
allow plant operators to monitor the
performance of the plants mercury
removal unit (MRU).
Early approaches
Mercury is present in many geological
formations and thus can be present in
natural gas. It had not posed a serious
problem to natural gas operations until
LNG_MayJune_2013_53-56.indd 53 23/05/2013 10:32
the early 1970s when the LNG plant at Skikda, Algeria, was
built. Early in its operation, the aluminium heat exchangers were
destroyed,
1,2
due to corrosion induced by the presence of mercury
in the natural gas. Therefore, the LNG industry became aware that
it needed to remove mercury from the natural gas. In addition, the
industry also realised that it needed the capability to analyse for
mercury and to be able to measure low mercury concentrations.
Soon after the Skikda experience, two new LNG plants were
built in Indonesia at Badak, Kalimantan and Arun, Sumatra. The
management for each of these plants wanted to have protection
from mercury. However, there was no industry precedent on how
best to accomplish the desired removal of mercury. Calgon
Carbon Corp. developed a mercury adsorbent that consisted of
activated carbon granules that were infused with elemental
sulfur. A similar product was developed by Union Carbide Corp.,
except that this product was a zeolite molecular sieve pellet
(extrudate) that was infused with elemental sulfur.
3
Thus, mercury
could be removed from the gas by fixed bed adsorption in much
the same way as water is removed from gas in desiccant beds.
The gas flows through a packed bed of adsorbent particles. The
mercury reacts with the sulfur and is thus fixed inside the
adsorbent particle. The mercury is then retained within the
adsorption vessel. So, there were two commercially available
choices of mercury removal adsorbent. The question was, where
in the LNG process should the mercury removal bed be installed?
Management at the Badak plant decided to use the Calgon
sulfur-impregnated activated carbon mercury adsorbent. The
MRU was installed downstream of the molecular sieve gas driers
and just ahead of the cryogenic unit. The sulfur-impregnated
molecular sieve product was used at the Arun plant. The MRU at
Arun was installed on the plant inlet gas, upstream of the acid gas
removal unit and the gas driers. The MRU at Badak operated
without problems, whereas the MRU at Arun had a number of
problems. The sulfur/molecular sieve adsorbent experienced
problems during start-up and the upstream MRU did not have
any protection from liquid carry-over; it experienced liquid
entrainment resulting in inadequate mercury removal. The
sulfur/molecular sieve adsorbent was replaced with
the sulfur-impregnated activated carbon, and a second MRU was
later installed downstream of the gas driers, loaded with
sulfur-impregnated activated carbon. Since that time, the LNG
plants have generally followed the example of Badak, with the
MRU located downstream of the gas driers.
Subsequent developments
Other mercury removal adsorbents have been made available,
based on metal oxide or metal sulfide as the active component.
Metal oxide or metal sulfide mercury adsorbents are available
in three types, with very different amounts of internal surface
area: agglomerated metal oxide (40 m
2
/g), activated alumina
that is impregnated with metal oxide (250m
2
/g), and activated
carbon that is impregnated with metal oxide (1000 m
2
/g). The
oxide form can be used when there is enough H
2
S in the gas
to convert the oxide to sulfide, in situ. When there is no H
2
S
present, the adsorbent must be installed in the presulfided
form. The metal sulfide, unlike the elemental sulfur, does not
dissolve in liquid hydrocarbons. Improvements have also been
made in the dispersion of elemental sulfur within the activated
carbon particle, resulting in better performance. Maximising the
surface area of sulfur exposed to the mercury greatly improves
the adsorbent performance. In all cases with these products, the
mercury stays and accumulates on the adsorbent. The adsorbent
has to be replaced when it reaches its maximum capacity for
mercury.
In the late 1980s, a new concept in mercury removal was
introduced: regenerative mercury removal. This concept originated
from the observation that when natural gas containing mercury is
dried using conventional, type 4A molecular sieve, there is a peak
of mercury concentration in the spent regeneration gas. Thus, the
molecular sieve adsorbs some mercury. A new, silver-containing
molecular sieve was developed and commercialised as a
regenerable mercury adsorbent. By using some of the silver-
containing molecular sieve in the natural gas drier, both moisture
and mercury are removed from the gas in one unit. During the
drying step, the mercury amalgamates with the silver and stays on
the adsorbent. During the normal regeneration step of the gas
drier, when the adsorbent is heated, the mercury is released and
leaves the vessel with the spent regeneration gas. Mercury has a
fairly high vapour pressure and it condenses to a liquid, so it
behaves similarly to water. If the mercury concentration is high
enough and if the gas gets cool enough, mercury condensation
will take place. Since the regeneration is always carried out under
reducing conditions, there is no oxidation of mercury, so it is easy
to separate the condensed mercury from the condensed water.
One plant operating with some 50 g of Hg/Nm
3
in its feed gas,
was drawing off shiny mercury from the bottom of its existing
regeneration gas water knock-out drum. With regenerative
mercury removal it is possible to remove and recover the mercury.
A non-regenerative MRU can be used to permanently remove the
mercury that remains in the spent regeneration gas that is sent to
fuel or to the sales gas.
Analytical issues
The mercury that damaged the exchangers at Skikda was
elemental mercury. However, mercury can exist in various forms:
elemental, organic (e.g. methyl-mercury), inorganic (e.g. HgCl), or
it can be attached to a solid particle. To date, the form of mercury
that has been detected in natural gas has been elemental
mercury. When natural gas liquids (NGLs) have been analysed, all
forms of mercury have been found.
All of the mercury needs to be removed to maximise
protection of the heat exchangers. This means removal to below
an industry standard mercury concentration that can be measured
in the plant with commercially available instrumentation. The
industry standard concentration of mercury has been < 0.01 g of
mercury/Nm
3
. This equals 10nanograms Hg/Nm
3
and
corresponds to about one part per trillion by volume. The mercury
content of raw natural gas has ranged from non-detectable to
several hundred g/Nm
3
.
Great improvements have been made in mercury analytical
equipment and techniques. There are numerous analysers
available that use a variety of detection mechanisms, including
atomic fluorescence and cold vapour atomic absorbance (CVAA).
However, no analyser can measure the low concentrations of
mercury directly. All the analysers require that a sample gas be
taken and passed through an analytical trap containing a gold
surface. The trap is then desorbed into the analyser and the
mercury peak sent to the detector, which then quantifies the
mercury. The accuracy of any in-plant mercury measurement is
seldom limited by the sensitivity of the detector itself. The design
of the sampling system and the analytical procedure greatly
affects any readout generated by the analyser. All of the gas
analysers measure total mercury, not just elemental mercury,
since gold adsorbs all forms of mercury.
4
LNG_MayJune_2013_53-56.indd 54 21/05/2013 11:51
GAZTRANSPORT & TECHNIGAZ
1 route de Versailles, 78470 Saint-Rmy-ls-Chevreuse - France
Tel: +33 (O)1 30 234 789 - E-mail: commercial@gtt.fr
www.gtt.fr
at the heart of LNG
T
r
a
n
s
p
a
r
e
n
c
y
I
n
n
o
v
a
t
i
o
n
E
x
c
e
l
l
e
n
c
e
T
e
a
m
w
o
r
k
S
a
f
e
t
y
SOLUTIONS
In the short term, the release of polluting
exhaust gases (SO
x
, NO
x
, CO
2
) will be
restricted.
A set of new international regulations are
coming progressively into force, which
will impose the use of cleaner fuels.
LNG, being by far the cleanest and probably the cheapest
of all fuels comes to mind as the most obvious alternative.
GTT, through its experience in the design of LNG
handling and storage systems, both at sea, and on land,
is ideally positioned as the preferred partner for the
implementation of clean, efcient and reliable solutions
at each step of the supply chain.
GTT has solutions for land storage tanks, and bunker
tanks for all types of vessel.
LNG_MayJune_2013_53-56.indd 55 21/05/2013 11:51
Adsorbent characteristics,
performance and process design
Adsorption is a surface-active process. Therefore, to maximise
performance, it is important to maximise the surface area in the
adsorption unit. Sulfur-impregnated activated carbon pellets have
1000 m
2
/g of internal surface area and a bulk density of
560 kg/m
3
, therefore they deliver 560 million m
2
per m
3
of
adsorbent installed in the adsorption vessel. Activated
alumina-based mercury adsorbents have an internal surface
area of 250 m
2
/g and a bulk density of 900 kg/m
3
, thus they
deliver 225 million m
2
per m
3
. Agglomerated metal oxide has an
internal surface area of 40 m
2
/g and a bulk density of 1000kg/m
3
,
thus they deliver 40 million m
2
perm
3
of vessel volume. Not
all sulfur-impregnated activated carbons are equal; the best
have high internal dispersion and retention of sulfur to provide
maximum active surface area of sulfur.
The pellet (extrudate) form of sulfur-impregnated activated
carbon is physically strong and is not affected by oxygen in the
air, so it can be loaded in air the same way that molecular sieves
are loaded into their vessels. The pellets retain their physical
strength even when hit with liquid entrainment. However, the
performance of this adsorbent will be permanently reduced when
hit with liquid hydrocarbons because the elemental sulfur
dissolves in liquid hydrocarbons. Loss of sulfur will result in
reduction in mercury removal capacity and efficiency.
All adsorbents require single-phase flow for high efficiency
contaminant removal. Some manufacturers claim that only their
product can be used in an upstream MRU. However, there is no
magic mercury adsorbent product that provides high
performance when subjected to liquid mists and liquid carry-over.
Agglomerated metal oxides do not have many, if any,
micropores, that is why their surface area is low. Activated
aluminas have micropores, so they have over five times more
surface area per unit of volume as metal oxide. The
sulfur-impregnated activated carbon has 14 times the surface area
compared to the metal oxide.
When an upstream MRU is well-protected from liquids with a
knock-out drum and a filter separator, then the upstream MRU is
reliable and delivers high mercury removal efficiency and high
mercury capacity using a sulfur-impregnated mercury adsorbent.
An additional advantage of an upstream MRU is that it prevents
mercury from getting to the rest of the plant. It also avoids the
necessity of dry-out during start-up that is required when the
MRU is downstream of the gas drier. Placing the MRU upstream
does not really require additional equipment vs. a downstream
MRU, as a knock-out drum and filter separator are usually used to
protect the gas drier.
The metal sulfide adsorbents require loading under a nitrogen
blanket. When exposed to air, the sulfide is converted to the
sulfate, which is inactive for mercury removal.
5
Some metal
sulfide adsorbents have had problems retaining particle strength
when exposed to liquid entrainment. The adsorbent beads
compact into a solid mass, which results in a loss of performance
and makes it difficult to remove the adsorbent from the vessel.
Activated carbon pellets do not lose their strength in water or
liquid hydrocarbons. They are used in water, in amine solution,
and in other liquid streams without any loss of strength. Activated
carbon pellets are the only adsorbents that do not lose strength
when wet.
Metal oxides and activated aluminas are more hygroscopic
than activated carbons. When an MRU is located downstream of
the gas drier, the MRU bed must be dried out during start-up, to
prevent freeze-up of the cryogenic unit. It has been reported that
metal-oxide type product took much longer to dry out than the
sulfur-impregnated activated carbon.
6
Summary
LNG plants have been protecting their aluminium process
equipment from mercury corrosion for over 33 years. In that time,
mercury analysers and analytical techniques have improved to
the point where nanograms of Hg/Nm
3
can be reliably measured
in natural gas. Also, the market now offers several types of
non-regenerable mercury adsorbents and at least one
regenerable mercury adsorbent. Sulfur-impregnated activated
carbon still offers the highest performance of the non-regenerable
products. With proper process design, a MRU can be installed
upstream or downstream of the gas driers and provide high-
efficiency mercury removal, or a regenerable mercury adsorbent
can be used to remove mercury in the gas drying unit itself.
Which option is best depends on the operating philosophy of
plant management.
References
1. Dolle, J., and Gilbourne, D., LNG: Startup of the Skikda
LNG Plant, Chem. Eng. Progress, Vol. 72, p. 39,
January 1976.
2. Leeper, J. E., Mercury LNGs Problem, Hydrocarbon
Processing, November, 1980.
3. Ambrosini R. F., et al., Selective Adsorption of Mercury
from Gas Streams, US Patent 4,101,631.
4. Dumarey, R., Dams, R., and Hoste, J., Comparison of the
Collection and Desorption Efficiency of Activated Charcoal,
Silver, and Gold for the Determination of Vapor-Phase
Atmospheric Mercury, Analytical Chemistry, Vol. 57,
No. 13, November 1985.
5. Baageel, O. M., Gas Plant Solves Hg Problems with Copper-
sulfate Adsorbent, Oil and Gas Journal, 24 May 2010.
6. Mock, J., Hahn, P., Ramani, R., and Messersmith, D.,
Experiences in the Operation of Dehydration and Mercury
Removal Systems in LNG Trains, presented at the 2008
Laurance Reid Gas Conditioning Conference, Norman OK, US.
LNG_MayJune_2013_53-56.indd 56 21/05/2013 11:51
Handling
cryogenic fluid is
inevitably a hazardous
activity, particularly when the
fluid is as vulnerable and flammable as LNG. The
volatility of the refrigeration phase of the LNG
train demands that all equipment meets exacting
technical standards. This includes the valves used
for process control.
The rise of LNG megaprojects has brought a
corresponding rise in high capacity liquefaction
phase compressors. Consequently, larger
diameter valves are needed typical contracts
can include cryogenic control valves of 42 in. and
over. This brings a host of challenges and is driving
new developments in the engineering and testing
of valves for cryogenic applications.
The cryogenic challenge
The -160 C liquefaction phase of LNG production
is particularly challenging for control valves. They
are required to provide repeatable bubble-tight
shut-off in a controlled manner, whilst handling
significant differential thermal dynamics. This
needs to be achieved with minimum heat
leakage, minimum cool-down mass and cold
THE
CRYOGENIC
CHALLENGE
Barry Wilder,
Severn Glocon,
UK, looks at
how to maximise
cryogenic control
valve performance.
LNG_MayJune_2013_57-60.indd 57 21/05/2013 12:02
impact strength materials. Valves are designed to operate at
temperatures ranging from +80 C to -196 C, so maximising
their performance and prolonging their life requires dedicated
attention from the brightest engineering minds.
Clearly, when dealing with volatile media and extreme
temperatures, safety is a prime consideration. For LNG
operators, safety and productivity have a symbiotic relationship.
It is widely acknowledged that valves can be a major culprit
when it comes to leakage, both internally and to the
atmosphere. Atmospheric leakage of natural gas is considered a
serious environmental pollutant, so minimising the risk of
fugitive emissions is a key priority in the design of the valve seat,
as well as body and stem sealing systems.
If leakage does occur, the problem can escalate very quickly.
It is not unheard of for a valve-related emissions incident to lead
to 48 hours of plant downtime, bringing serious consequences
for production and supply chain management.
Industry standards and
specifications
In the absence of a dedicated international standard for
cryogenic LNG valves, there are many end-user technical
specifications for valves in cryogenic service. These generally
involve a hybrid of traditional British Standards such as
BS6364:1984 valves for cryogenic service and additional US
or international standards. This amalgamation of standards is
quite appropriate. The UK has a long track record of dealing with
cryogenic processes associated with air separation, whilst most
modern control valves for LNG facilities are constructed from US
materials.
The harsh demands of cryogenic applications mean that
each and every valve needs to receive a robust, intelligence-led
specification. Severe service valves linked to the compressors,
such as anti-surge, hot gas bypass or Joule Thompson gas
expansion (associated with plant blow-down requirements),
should be precisely tailored in line with the specific operating
conditions they will face. Advanced engineering surrounding
features such as trim design can bring significant operational and
reliability advantages. Engaging all parties, from engineering
contractors to operators to suppliers, at a projects FEED phase
can facilitate the development of innovative solutions for
seemingly insurmountable technical difficulties.
Avoiding liquid lock
One of the special challenges facing valves handling cryogenic
fluids is the liquid lock phenomenon. When a valve closes,
some liquid can become trapped in the valve body. With
cryogenic fluids, as the temperature rises, the trapped liquid
expands and regasification can occur. This can exert enough
pressure to deform or even rupture internal components,
causing a major incident.
A core design feature of valves that may be at risk of liquid
lock is the inclusion of an internal relief path to prevent
over-pressurisation.
Extended bonnet and stem for
LNG valves
Due to low operating temperatures, the typical bonnet and
stem design for LNG valves is the extended bonnet type,
allowing for a vapour space. Calculations usually take account
of the valve size and the operating temperature range. Draft
BS ISO/DIS 28921-1 for industrial valves outlines a minimum
vapour space and lagging length (bonnet extension).
The extended bonnet provides a temperature gradient to
minimise the risk of moisture freezing at the gland sealing
system. It has a minimum wall thickness section, normally
manufactured from tubular section stainless steel due to its heat
insulation properties.
The valve is designed to allow a very small amount of the
cryogenic liquid or vapour to enter the lower part of the
extension bonnet, and this evaporates inside the extended
bonnet. The vapour has a higher temperature than the cryogenic
fluid, which allows the temperature at the gland sealing area to
rise above freezing point. Since control valves are used outside,
the ambient air conditions also have to be considered. If the
gland area is below freezing, any ice that forms may be dragged
down through the gland packing, causing a leak at the gland.
This will allow cryogenic fluid to fill the extended bonnet,
leading to an ice ball forming around the gland, which may stop
the valve from working.
To compensate for this risk, extension bonnets can be
supplied with an insulation collar/drip plate.
This collar/drip plate is either welded or
clamped to the extension bonnet with any
bolting on the upper side to enable easy
adjustment. The gap between the collar/
drip plates is sealed to prevent
condensation entering the insulated area.
Draft ISO/DIS 28921-1 outlines positioning
and fixing details for collar/drip plates.
It is essential that no product leakage
enters the valves thermal insulation or the
atmosphere, so stem and body sealing
undergoes a fugitive emission test. This
involves the use of a mass spectrometer
and helium gas to comply with the fugitive
emission requirements of ISO 15848
and/or the customers own specific
requirements. Non-destructive examination
and pressure tests are also carried out in
accordance with ASME V Appendix IV and
ASME B16.34 and/or end-user
specifications.
Figure 1. The extended bonnet and stem design reduces the risk of moisture freezing
at the gland sealing system.
LNG_MayJune_2013_57-60.indd 58 23/05/2013 10:35
Find out through our reader survey
measuring the NOV / DEC issue of
FOR MORE INFORMATION PLEASE CONTACT
JOHN BAUGHEN or CHRIS ATKIN:
E: john.baughen@lngindustry.com
E: chris.atkin@lngindustry.com
T: +44 (0)1252 718999
Is your advertising
campaign ticking
all the right boxes?
LNG_MayJune_2013_57-60.indd 59 23/05/2013 16:13
The importance of trim design
Modifying trim design can have a significant impact on a
valves ability to perform well in a given application. If the trim
arrangement is not correctly specified it can increase the risk of
leakage and poor process control, whereas adapting it to suit the
operating conditions will enhance performance.
High-output LNG plants tend to have high-capacity trains,
which require larger diameter pipes, thereby increasing the flow
speed of the cryogenic fluid. Any high-velocity process medium
can cause cavitation, erosion and abrasion of internal valve
components, which can result in poor control and premature
valve failure. Simply using harder trims or pipe lagging cannot
solve the problem, it needs to be dealt with at source.
Equipping valves with multi-labyrinth trims (MLTs) can
significantly reduce the velocity of the cryogenic fluid. This is
achieved by splitting the flow into many smaller streams via
multiple passageways, with each flow stream made up of many
right angle flow turns. Every flow turn achieves a pressure or
velocity reduction. This tortuous path removes the kinetic energy
from the fluid whilst lowering pressure in a controlled manner.
The number of required turns is calculated precisely by taking
into account the expected fluid velocity at the inlet, and the ideal
velocity at the outlet. This enables the flow speed to be
accurately controlled in each flow passage so the valves
operation can be well managed over its full service range.
Unlike other severe service trim options, the MLT is fully
pressure-balanced within the stack. This helps hold the valve
plug in the centre of the stack, giving good stability and avoiding
plug vibration, which extends the service life of the trim. As well
as prolonging valve life and improving control performance,
using an MLT ensures the valve produces lower noise levels than
valves using alternative trims, such as multi-cage devices. It also
offers simple, quick change-out, which reduces maintenance
costs.
Valve trim seat sealing and pressure balance seal design at
cryogenic temperatures are critical to the performance of LNG
valves. Attention to detail in the design, manufacture and
machining methods, together with the correct actuator selection,
are vital to provide the degree of repeatable seat leak tightness
required by the industry, typically ANSI/FCI 70-2 Class V or VI.
Application spotlight:
compressor anti-surge
The MLT solution is especially suited to compressor anti-surge
valves. This demanding application is greatly enhanced by a
reduction in the velocity of cryogenic fluid. It facilitates
control and decreases the vibration and noise typically
associated with compressors.
Improving reliability and performance of the
anti-surge valves is cost efficient on multiple levels. Even
minimal leakage of anti-surge valves results in significant
wasted energy, as the fluid returns to the suction side of
the compressor, meaning it has to work harder. Achieving
repeatable bubble-tight shut-off prevents this leakage. In
addition, the precise flow control means there is no need
for additional by-pass valves, and the reduced vibration
extends the life of the trim.
Testing and installing
Valves destined for LNG applications require test
temperatures of -196 C to ensure they can operate
safely and effectively at temperatures as low as
-160 C in the field. Selecting a provider who can conduct
cryogenic testing on even the largest valves in-house enhances
turnaround time and facilitates better project management.
Since many LNG megaprojects now employ time-sensitive
modular construction methods, this capability is advantageous
for all parties.
Typically, LNG valves are supplied with a two-year
warranty, but conducting a detailed pre-commissioning site
survey, then cross-checking the eventual installation, can
underpin an extended warranty of up to four years. This
provides peace of mind and can help reduce the total lifetime
cost of each valve. It also demonstrates confidence on the
suppliers part in terms of the quality and durability of the
valves, as well as the high standard of the specification, design
and manufacture.
Key areas of concern for engineers checking
post-construction valve installations on greenfield sites include:
pressure testing, hydrostatic line tests and calibration checks.
They also verify that the valves have not been damaged in
transit, have been reassembled correctly and are installed at the
correct upstream/downstream orientation, which is of particular
importance for valves that have an extended bonnet design to
overcome liquid lock. Cleanliness is also vital throughout the
installation phase and during ongoing operations:
BSEN12300:1999 and in-house standards and procedures are
followed to ensure product safety and functionality.
Insisting on a thorough best-practice approach to testing and
installation can reduce start-up time and enhance long-term
performance.
Ongoing maintenance
Once a valve is operational, the big question is how often it
requires attention. In many processing industries, a regular
annual or bi-annual assessment is adequate for most valves.
However, a timetabled procedure for testing and management
across the valve population is not best practice for LNG
processing, where closer control is needed. A more dynamic,
intelligence-led approach optimises safety and performance, and
can ultimately be far more cost-effective than static methods
alone.
Whilst valves are a relatively small component in
themselves, they are associated with all the major assets in the
liquefaction process. Cryogenic valves are never just a
commodity. They are a highly-engineered component that has
considerable bearing on plant performance.
Figure 2. In-house cryogenic testing enables better project management.
LNG_MayJune_2013_57-60.indd 60 23/05/2013 10:36
T
he current situation in the LNG industry can be brought back to one simple equation: Weight reduction + size reduction +
power reduction = operational cost reduction. For the production of LNG there are four main licensed processes. The most
commonly used is probably the Cascade liquefaction process. Figure 1 shows an overview of the process and the required
equipment.
Liquefaction
This article focuses on the cooling side of the process. For cooling of the refrigerants, very large coolers are needed, both air cooled
and/or water cooled. These are major cost factors. However, by taking a closer look at cooling and utilising new equipment, large
cost savings can be realised.
In the present overall design, a large number of coolers are needed for the various refrigerant cycles; very heavy and large
equipment is used, especially for offshore applications. Hence, huge plot spaces are needed, with the high power consumption of
Femke Schaefer,
Bronswerk Heat Transfer,
the Netherlands, looks at size,
weight and power reduction of process
equipment in LNG liquefaction and regasification.
LNG_MayJune_2013_61-63.indd 61 22/05/2013 09:39
the air coolers having a major influence on the CAPEX.
Onshore, the inter stage compression cooling of the
refrigerants is conventionally done by air coolers, and
offshore by means of water cooled shell and tube heat
exchangers.
High efficiency air coolers
Geothermal plants also use a huge number of air cooled
coolers. Although a different industry, a similar challenge is
posed and solved by using the latest available technology;
the high efficiency air cooled cooler. For a geothermal site in
Germany, the onshore air cooled condensers were reviewed.
The installed turbine had a power of 25 MW, the refrigerant
used was R134a and the capacity of the air cooled coolers
was 68 MW. The original design of this geothermal plant
was based on standard, conventional air cooled coolers with
an absorbed electrical power of 1143 KW, a required plot
space of 4224 m
2
and a resulting noise power emission of
101B(A).
The advanced design was based on the high efficiency air
cooled coolers with an absorbed electrical power of 528 KW,
and a required plot space of 2974 m
2
, resulting in noise power
emissions of only 94 dB(A).
The benefits of the advanced design are an area reduction
of 1250 m
2
, equaling 29%. Additionally, a noise reduction of
7dB(A) is realised and the power consumption is reduced by
615KW, equaling 53%. For the OPEX, this results in a saving
of 492000/y (based on an electrical power price of
0.10/KWh), excluding the savings on maintenance, CAPEX
reduction due to reduced E&I, foundations and auxiliaries.
Water cooling
It has been demonstrated that air cooled coolers can help
reduce costs. An additional example of cost/space savings
for advanced compressor water cooling with compact
shell and tube heat exchangers for offshore applications
is used, especially on a FPSO. Two examples of high
pressure offshore gas compressor coolers for an FLNG
facility off the coast of Australia are detailed here. The
mediums are acidic gas and cooling water. For the fourth
stage compressor after cooler the design for pressure gas
is 215 bar and the design for temperature is 160C. For the
final reinjection compressor after
cooler the design for gas pressure
is 295 bar and the design for
temperature is 160 C.
The original design was based
on conventional B type high
pressure shell and tube heat
exchangers, where the fourth stage
compressor after cooler has a
weight of 17000kg and the required
space is approximately 10 m
2
. The
final reinjection compressor after
cooler has a weight of 28000 kg
and the required space is
approximately 15 m
2
. In addition to
the conventional types, Bronswerk
has also made an advanced design
for the same process conditions,
based on a new high pressure shell
and tube heat exchanger
technology called the compact header design. In the
advanced design, the fourth stage compressor after cooler
has a weight of 10 500 kg and the required space is
approximately 8.5 m. The final reinjection compressor after
cooler has a weight of 14 000 kg and the required space is
approximately 13 m
2
.
The advanced design reduces the size and weight of the
equipment. The benefits of these solutions are a volume
reduction of approximately 3.5 m
2
(15%) and a weight
reduction of 20 500 kg (46%). A well known EPC contractor
indicated that an equipment weight saving of 1000 kg will
save 5000 kg of supporting steel structure. An estimate of at
least ten high pressure heat exchangers on each ship, each
costing US$ 500 000, results in an equivalent of
US$5million less CAPEX.
The new Compact Header
high pressure heat exchanger

design is capable of up to 60% weight reduction, which also
reduces the required plot space and the weight of the
supporting steel structure. It also results in easier maintenance.
Regasification
LNG regasification is a relatively simple process compared
to liquefaction. Still, there are opportunities to save on
operational cost. In the LNG process there are a lot of high
pressure booster pumps to bring the LNG up to the right
level of gas pipeline pressure (80 barg).
Innovative high pressure LNG booster pumps can be
used. In the current situation there are high pressure
multistage centrifugal pumps; the construction is large,
robust and heavy; and the bearings are oil lubricated. There is
also limited working space, surge limitation and risk of
cavitation. Valves and bypass circuits control the flow.
Bronswerk set out to reinvent the pump concept.
A new design
The demonstration setup of the companys newly designed
Radiax
pump is situated in the Netherlands. The medium

is water, the flow is >1000 l/min and the pressure is
18barg. The rotor of the pump is capable of purely axial
in and out flow. The stator diffusor is equipped with
tangential in and axial outlet and a special 3D shaped
exhaust. The benefits of this reinvented pump are a size
Figure 1. Overview of the LNG Cascade liquefaction process (source: Report Sunrise LNG in
Timor-Leste by L. Hamutuk).
LNG_MayJune_2013_61-63.indd 62 22/05/2013 09:39
reduction of 1 m in length (61%), a weight reduction of
275 kg (76%), and a 20% power consumption reduction.
There are additional benefits of an oil free operation and
an advanced flat performance curve. This will result in
OPEX savings for each MW of installed pump power
of 160 000/y. The OPEX is based on maximum power
consumption, so the benefit will be much larger when the
conventional pump is not working at the optimal working
point. The CAPEX will also be much lower if no oil cooling
and removal system is required.
Another option
An alternative is to redesign the air heated vaporiser with
Whizz Wheels
. The benefits of this advanced design are

more air flow (double), improved flow distribution inside the
vaporiser, noise reduction, and active condensation/frosting
control. Additionally, half the number of vaporisers and half
the weight (offshore) for the same power consumption is
required. There are also the additional cost savings on piping,
cabling, maintenance and auxiliaries.
Summary
There is a multitude of opportunities for decisive cost
savings due to size, weight and power reduction in LNG
liquefaction and regasification. In the liquefaction case,
the high efficiency air cooler can realise size reductions up
to 50%, and reduced energy consumption of over 40%,
resulting in cost savings on power, civil works, infrastructure
and maintenance.
The high pressure Compact Header heat exchanger can
realise a size reduction up to 50% and a weight reduction up to
60%, resulting in cost savings on the supporting steel structure
and maintenance.
In regasification, the high pressure Radiax booster pumps
can realise a size reduction of up to 90%, a weight reduction up
to 80% and reduced energy consumption of over 20%,
resulting in cost savings on power, controls, and oil treatment
equipment. The Whizz Wheel can also help optimise the air
heater.
A jump in the economy of operational costs is at hand,
using advanced, already available, and elsewhere proven
equipment.
Figure 2. Advanced design using high efficiency air coolers.
Founded in 1962, Temati began by manufacturing, promoting and selling Foster
products in the Benelux. Nowadays Temati is Europes sole licensed manufacturer
and supplier of Foster and Childers products, with now five office- and warehouse-
locations in different countries and an international distributors- network. Today we sell
an extensive range of specialized materials to customers throughout Europe, as well as
overseas. As we continue to expand our services, you can continue to rely on us for a wide
range of products and the best solutions.
Temati B.V. Tel: (+31) (0)251 229 172 Fax: (+31) (0)251 212 380 E-mail: temati@temati.com www.temati.com
Industrial processes come with a
multitude of challenges, which require
reliable and innovative products. At
Temati, we have decades of experience
and expertise. We are your ideal
partner when seeking solutions that
make a real difference.
Unique solutions for industrial processes
LNG_MayJune_2013_61-63.indd 63 24/05/2013 09:59
T
he principal of an LNG tank is similar to that of a coffee
thermos insulation is what makes it work.
Due to the volume of LNG contained in these
tanks, it is impossible to maintain LNG in a liquid phase
through pressure alone. This task is completed using premium
high quality insulation. With the external wall and dome of
reinforced concrete and the inner tank assembled and welded
with 9% Ni cryogenic steel, the space between both tanks is
filled with expanded perlite.
Perlite
Perlite is an amorphous, hydrated glassy volcanic rock of
rhyolitic composition, consisting primarily of fused sodium
potassium aluminium silicate. When the perlite ore is properly
dried, crushed and screened, it is expanded in a vertical
or horizontal furnace at approximately 900 C. Under this
process, the perlite ore pops and solidifies very much like
popcorn. Through this phase change, the perlite changes from
a bulk density of around 1000 kg/m
3
to 40 kg/m
3
with a white
powder appearance.
As a natural mineral, and being completely sterile
and inert, this very low density product serves perfectly as
an insulant and meets all international environmental
standards.
There are a number of specifications that expanded perlite
used in cryogenic applications must meet.
The primary specification is ASTM C549, which sets the
technical properties required for cryogenic insulation. This
standard concludes that the expanded perlite insulation factor
is completely connected with the density and thermal
conductivity of the environment in which it is emplaced
(shown in Figure 1 as per ASTM C549).
The quantity of perlite ore required to fill the annular
space (around the circumference and in the upper reserve
area) for an LNG tank is approximately 550 t. Once the ore
has expanded to 40 kg/m
3
, an area of more than 13 000 m
3

can be filled with expanded perlite.
Due to delivery times, economics of transportation,
quality control and the desire to reduce the carbon footprint
of the process, perlite ore is expanded on site beside each
vibrations
LNG_MayJune_2013_64-68.indd 64 23/05/2013 10:36
Israel Ferreres,
Cryoperl Australia
Pty Ltd, Australia,
outlines the benefits
of vibration prior
to operation.
LNG_MayJune_2013_64-68.indd 65 22/05/2013 10:24
LNG tank using mobile perlite expansion furnaces. The ore,
once expanded, is then pumped with air into the space
between the inner and outer LNG tank.
Perlite vibration
Once emplaced in the annual space around the tank, the
expanded perlite is vibrated into place. Through this vibration
the expanded perlite settles by a further 15%.
The use of vibration is quite a new process in LNG tank
construction and not all companies that carry out perlite
expansion works are prepared to undertake it.
Unless vibrated during installation, the perlite will
naturally settle once the LNG tank is placed into operation.
This settlement leaves a non-insulated space around the top
of the tank, which reduces the thermal efficiency of the tank,
and therefore increases the cost of operation to the tank
operator.
Vibrating the perlite during filling removes the need for
expensive and dangerous refill at a later stage.
From an economic standpoint, a refill requires a boil-off
over the maximum calculated (meaning extra gas loss), and
during refill works the tank must be out of operation for at least
two weeks.
The accepted loose bulk density of the expanded perlite
once poured into the annular space and vibrated is between
48kg/m
3
and 75 kg/m
3
. This will insulate the tank to around
0.044 W/ mk at 0 C.
The maximum compaction rate for expanded perlite is
determined by the Perlite Institute standard PI201-77. The
maximum compaction ratio for the cryogenic perlite grade is
between 18% and 23% by volume.
For the largest LNG tanks, the typical volume of the annular
space around the circumference of the tank to be filled with
expanded perlite is around 6000 m
3
.
The vibration system developed by Cryoperl Australia Pty
Ltd results in a compaction ratio of approximately 16% in
volume. This means that after the tank is commissioned and in
a cold down phase (tank movements plus resilient fibre
blanket compression plus inner tank contraction), the perlite
will usually undergo a further 7% settlement, requiring an extra
450 m
3
to 500 m
3
of expanded perlite with the additional
volume for the compression of the resilient fibre blanket
(depending on the tank design, this volume can be between
300 m
3
and 700 m
3
).
Taking into account the angle of repose of the expanded
perlite and the minimum height of the suspended deck
insulation, the approximate effective reserve of perlite is
around 1000 m
3
.
Therefore, where the expanded perlite emplaced in the
annular space around the circumference of an LNG tank is not
compacted, a further 1400 m
3
of perlite will ultimately need to
be installed at a later stage (on the basis of a maximum
compaction ratio of 23%). Figure 2 shows these details in full.
Vibration advances the settlement and minimises the use
of the perlite reserve, ensuring that there will be enough perlite
in the reservoir to undertake operation for several years.
In conclusion, without vibration there is almost a
guaranteed requirement to undertake a refill operation at a
later stage, causing problems for all parties concerned.
Further to this, there is the critical decision of when to
undertake the refill. This can either be carried out as soon as
insulation problems are detected (resulting in the risk that
additional refilling will be required at a later date), or
postponed for two to four years to ensure definitive refilling
(during which time the LNG tank will operate inefficiently).
Vibration process
The typical perlite vibration process is to undertake visual
inspection at each vibration point. This means that one
worker must follow the vibration tool at the same level
around the tank.
Suspended from a chair or a basket, this person descends
with the vibrator tool to the bottom of the tank, and then rises
up as the tank is filled with expanded perlite and vibrated to
the top of the tank, which can be 30 35 m above the ground.
The annular space within which this person works is
approximately 900 mm.
Cryoperl has developed a procedure to control the
vibration work without the requirement of a person having to
enter this very tight and dangerous annular space. The
company uses two intellectually protected devices that
follow the perlite as it is installed. This mechanism ensures
correct vibration of the expanded perlite in the minimum
possible time.
This process can control the settlement and confirm the
uniformity of the perlite at the final level. A reference point
system is maintained inside and outside the tank to ensure
that on each rotation the tank is properly filled and the
perlite vibrated.
Each vibration lap of
the LNG tank is premised
on approximately four
continuous hours of work
inside the annular space.
Combined with
production, quality and
environmental
requirements, this is busy
and exacting work.
Through the adaption
of some of the traditional
mechanical vibration tool
devices, the companys
system is able to settle
the expanded perlite to
the maximum required. It
Figure 1. Thermal conductivity vs. density depending on temperature.
LNG_MayJune_2013_64-68.indd 66 23/05/2013 10:37
G|oba| Leader |n Deve|opment, Lng|neer|ng and 1esnng of Cryogen|c
umps
Iu|| kange of roducts for
Cryogen|c umps for most
||quehed gases:
llxed ln-Lank pumps for
use on LnC carrlers.
Sucuon vessel mounLed
pumps.
8emovable ln-Lank
pumps.
Cngo|ng serv|ces and sup-
port aher |nsta||anon:
1echnlcal asslsLance aL
malnLenance ume.
unlque condluon monl-
Lorlng servlce.
State-of-the-art tesnng fac|||nes :
erform wlLnessed LesLs on
Ln2, LnC and LC.
ConducL compuLer-conLrolled
performance and nSP LesLs
wlLh llve Lracklng of more Lhan
200 parameLers.
hup:]]www.n|kk|socryo.com
G|oba| Leader |n Deve|opment, Lng|neer|ng and 1esnng of Cryogen|c
umps
Iu|| kange of roducts for
Cryogen|c umps for most
||quehed gases:
llxed ln-Lank pumps for
use on LnC carrlers.
Sucuon vessel mounLed
pumps.
8emovable ln-Lank
pumps.
Cngo|ng serv|ces and sup-
port aher |nsta||anon:
1echnlcal asslsLance aL
malnLenance ume.
unlque condluon monl-
Lorlng servlce.
State-of-the-art tesnng fac|||nes :
erform wlLnessed LesLs on
Ln2, LnC and LC.
ConducL compuLer-conLrolled
performance and nSP LesLs
wlLh llve Lracklng of more Lhan
200 parameLers.
hup:]]www.n|kk|socryo.com
LNG_MayJune_2013_64-68.indd 67 22/05/2013 10:24
can also control the exact position of the perlite at any time,
monitor settlement ratio and take material samples for quality
tests, all from the suspended deck (top of the inner tank). This
avoids the process of entering the annular space and the
inherent dangers involved.
Safety during vibration
The process of vibrating the perlite requires a high level of skill.
There are a number of health and safety issues to contend with,
including operating inside the tank in a confined space, oxygen
restrictions, low visibility with floating perlite
dust and no easy manner of evacuation.
In addition, the vibration occurs at height;
normally more than 30 m.
All of Cryoperls team members are
continuously trained and kept updated on
work and safety procedures and
improvements. The company prepares a
safety plan and work procedure for each
individual project. This is based on its general
safety work procedure and plan, which is
approved by its safety manager, site
manager and general manager.
Once the final safety documents are
approved by the main EPC contractor and the
end client, Cryoperl retrains all team
members on the vibration process required
for that particular project.
The companys aim is to complete each
project on time and in specification with no
injuries or issues, and accordingly its main
focus is on quality, safety and the environment in which it
operates.
Cryoperl Australia has installed and vibrated over
120000 m
3
expanded perlite for projects such as Sonatrachs
Skikda project (2012) and Royal Dutch Shells Gate project
(2011), as well as other projects for Enags and REN Atlntico
(2010). The company believes that vibration is a key factor for
insulation performance and it continuoulsy carries out
development work to improve production, quality, safety and
environmental performance.
Figure 2. Detailed LNG tanks annular space.
Follow us on Twitter
@energy_global
Keep up to date with
the latest news from the
global energy industry.
LNG_MayJune_2013_64-68.indd 68 22/05/2013 10:24
F
low induced and acoustic
induced vibrations are
recognised as a problem in
process plants, as they can induce
piping and supports fatigue and
rupture in a very short time (from
minutes to days, depending on
vibration frequency).
The above broad family of
phenomena include low
frequency and high frequency
pulsations that can be generated
by a variety of sources. Among
those falling in the low to
medium frequency range, there is
the so called flexible riser
singing or whistling, which was
observed in several floating
production plants, in some cases
leading to piping failures at small
bore dead leg branches, due to
resonance between the flexible riser
noise source and the piping. The fatigue
can arise either in the subsea or the
topside piping of the floating facility by
fluid dynamics to acoustic coupling, as a
first step, and by an acoustic to
mechanical coupling, as a second step.
PLANTS
Attilio Brighenti,
Systems and
Advanced
Technologies
Engineering S.r.l.
(S.A.T.E.), Italy,
and Luigino Vitali,
SAIPEM, Italy,
discuss how riser
singing generated by
inner wall turbulence
of flexible risers can
be a severe problem for
LNG offloading and
regasification plants.
DETUNING
LNG_MayJune_2013_69-73.indd 69 22/05/2013 10:20
The origin of this phenomenon is the tonal noise
generated at the corrugated inner wall of the risers
(Figure1), which creates turbulence and, under certain
conditions, self sustained shear layer instabilities due to
the above mentioned fluid dynamic/acoustic coupling.
1 - 5
Summary of knowledge
The study of the flow through corrugated pipes
has several aspects similar to that of side branches
resonances. Indeed, some R&D approached a parallel
methodology addressing both corrugated and multiple
side branch pipes, in order to understand the basic physics
of these phenomena.
Some aspects are still debated by specialists, in
particular as regards the estimate of the amplitude and the
localisation of the self excited acoustic sources along a
corrugated pipe. However, it is clear that a corrugated
pipe like a flexible riser can sing and when it does it
emits a tonal noise, whose dominant frequency increases
with the average fluid flow velocity in the pipe.
The noise frequency locks on to the acoustic natural
frequencies of the piping nearest to the fluid dynamic
frequency of the vortexes generated at the corrugation
cavities, due to a positive energy feedback between the
two phenomena, up to a given amplitude saturation level.
This lock on and self excitation generally occurs when the
velocity exceeds a minimum value depending on several
factors: corrugation geometry scale and shape, fluid
density, absolute viscosity and velocity (thus varying with
the fluid properties, pressure, temperature and flow rate).
The self excited singing onset seems to be
determined
6
by the rate of increase of the momentum
thickness of the boundary layer, which is created and
disturbed when passing over each cavity. If it has time to
develop up to a stable value before reaching the next
cavity it does not collapse. This occurs at low fluid speeds,
relative to the pipe corrugation geometry, i.e. at low
Reynolds number. Conversely, if the momentum thickness
of the boundary layer at the wall surface curvature
upstream the next cavity, in the flow direction, has not yet
reached a stable value the boundary layer detaches and
collapses into the cavity, generating trailing vortexes. This
occurs at relatively high speed. Given the regular gap
between the subsequent cavities in corrugated pipes of
standard production, vortexes are regularly spaced in
length and time, thus generating a tonal noise depending
on the speed and on the corrugation cavities gap.
The above condition is necessary but not sufficient by
itself, as self excited positive feedback (i.e. lock on) arises
only if acoustic resonances also fall close to the vortexes
frequency range. In relatively short corrugated pipes (e.g.
having L/D ratio lower than 100) the minimum fluid
velocity for singing onset depends both on the pipe length
and on the boundary conditions at the pipe ends, with the
related acoustic reflection coefficients, which determine
the pipe acoustic resonant modes. In long tubes, such as
the floating facilities risers, having high L/D ratios (e.g.
higher than 100), the acoustic modes have such a low
frequency gap between each other (e.g. below 1 Hz) that
the resonant modes frequencies increase linearly, with
practically no steps or gaps in between, regardless of the
boundary conditions effects.
An acoustic lock on may then occur with the acoustic
resonances of the piping systems connected to the riser
ends, if they fall within the vortexes excitation range.
The vortex frequency of the most likely vortex mode
being excited prior to lock on (the 2
nd
fluid dynamic
mode
5
), i.e. that at which the riser can sing, can be well
predicted
3
, as a function of the corrugation geometry and
the fluid velocity, by considering a Strouhal number within
the interval 0.32 0.42, where:

f
v
Vortex frequency (Hz).
L Modified gap width of the corrugated pipe (m)
(preferred definition among others in the literature).
W Corrugation gap width (m).
R
up
Radius of curvature of the upstream side of the
gap (m).
U Fluid average velocity through the pipe (m/s).
The locked on frequency for a stand alone corrugated
riser of length L
p
can be calculated by its acoustic modes
considering its ends acoustically open.

The effective speed of sound is lower than that of the
unconfined fluid due to the cavities acoustic compliance,
i.e. it must be corrected as per the following equation
3,5
:

Figure 1. Riser singing source: turbulent vortex generation
at the corrugated wall cavities (thin arrow shows flow direction;
thick arrows show acoustic wave propagation).
LNG_MayJune_2013_69-73.indd 70 22/05/2013 10:20
c
0
Speed of sound in the unconfined fluid (m/s).
c
eff
Effective speed of sound through the pipe (m/s).
V Volume of the corrugation beyond the inner
diameter in a pitch length (m
3
).
S Pipe flowing cross section (at minimum inner
diameter) (m
2
).
P Pitch of the corrugations (m).
The mechanism of riser singing lock on to complex
topside and subsea piping cannot be simply predicted by
the acoustic natural frequency calculation of individual
piping spools (e.g. dead leg branches taken as stand
alone), because the lock on mechanism depends on the
overall acoustic impedance at the riser/plant interface, i.e.
on the acoustic reflection coefficient therein, which varies
with the frequency.
While the above prediction criteria for the onset and
frequency of the phenomenon are quite well agreed by the
scientific community, the accurate prediction of the
amplitude of the tonal noise source is still an open issue.
5

An accepted physical criterion, derived from energy
balance of the self excited instability, is represented by
the following equation:

Where:
p Average acoustic pressure amplitude (Pa rms).
Fluid density (kg/m
3
).
c Speed of sound, i.e. c
eff
for consistency (m/s).
U Fluid average velocity through the pipe (m/s).
k Constant dependent on the corrugation geometry.
The still uncertain aspect is the value to be given to the
dimensionless parameter k, which depends largely on
corrugation geometry details. For preliminary calculations
a value of 0.1 can be considered conservative enough, for
the source strength itself (i.e. considering the riser as a
stand alone resonant tube), as this is the order of
magnitude of saturation values that have been found in
extensive experiments
1,6
and which depend neither on the
fluid average pressure nor on the pipe length.
6
Much higher values could however be generated under
severe resonant conditions with adjacent piping.
5 - 7
However, the published criteria tend to overestimate
the pressure amplitude by two to five times, compared to
actual installation cases.
8
Despite some debates still pending on this aspect,
theory and experiments confirm that the noise is
generated at nodal positions of the acoustic pressure field
along the corrugated pipe, where acoustic velocity grazing
on the cavity edges maximises the vortex shedding.
5

These positions are often close to the ends of open ended
pipes and in the intermediate velocity antinodes of the
excited acoustic standing wave in the corrugated pipe.
The worst riser singing situation arises when a triple
lock on occurs, i.e. when one mechanical natural frequency
m
is close to the self sustained fluid acoustic resonance.
However, this occurrence is not strictly necessary for a
fatigue rupture, as this can occur even with only the above
singing lock on.
Finally, the mechanical ring frequency of the riser
carcass may interact with the system self excitation, but it
is not the primary cause of excitation of the fluid acoustic
resonance, which also arises in infinitely rigid pipes.
7
Onset velocity
The onset velocity of the fluid to excite riser singing
depends on the ratio between the momentum thickness of
the boundary layer and the cavity gap of the corrugation
pattern. This yields the following equation
6
, derived from
the work of S. Belfroid
6
and others on side cavities and
dead branches, with parameters referring to Figure 2.

Where:
a Boundary layer growth coefficient (--) (equal to 5,
for a flat plate
9
).
c
1
Ratio of momentum boundary layer thickness to
boundary layer thickness (--) ( 0.13 for flat plate
6,9
).
c
crit
Critical ratio of momentum boundary layer thickness
to cavity gap (--) = 0.0005.
Fluid absolute viscosity (Pa s).
Fluid density (kg/m
3
).
U Fluid velocity (m/s).
W Cavity gap (m).
x Boundary layer growth length (m) given by:
Figure 2. Definition of riser carcass geometry patterns.
Table 1. Data for the real riser base case design example
Minimum inner diameter D 356 mm
Corrugation pitch P 39 mm
Cavity depth H 12 mm
Cavity width W 8 mm
Cavity upstream radius R
up
6 mm
Cavity downstream radius R
down
6 mm
Gas density 50 kg/m
3
Gas absolute viscosity
1.16 x 10
-5
Pa s
Gas flowing velocity U 8 20 m/s
LNG_MayJune_2013_69-73.indd 71 22/05/2013 10:20

a
1
Constant (--) (found to be equal to 0 from
experiments).
a
2
Constant (--) (found to be equal to 0.4 from
experiments).
R
up
Cavity upstream radius (m).
R
down
Cavity downstream radius (m).
L Cavity plateau length (m).
For the very long corrugated lengths found in flexible
riser systems these boundary layer equation conditions are
alone sufficient to determine and sustain the phenomenon.
It is only for much shorter tubes, as typically used in
experimental research works, that the boundary conditions
at the pipe ends are also important and can determine
much lower onset velocities.
Example of findings
As an example of the above procedure, the flexible riser
of a real case considered by the authors had the data
summarised in Table 1 (see also Figure 2 and profile a in
Figure 3).
The resulting cavity plateau length L is given by:

The boundary layer growth length x then is:

The riser singing onset velocity U
min
is:
y
min
g q

It is clear that no riser singing can occur with this
carcass profile under these conditions, since the onset
velocity exceeds by far the maximum fluid velocity.
Instead, if the riser carcass geometry were like profile b in
Figure 3, e.g. with a shorter pitch (28 mm) but broader
cavity width (10 mm) and radii (9 mm) and no plateau, yet
with the same depth, the onset velocity would be 22 m/s,
under the same gas state. At higher gas pressure, with a
gas density of 75 kg/m
3
, the onset velocity would
decrease to 15 m/s, well within the operational range.
The minimum and maximum riser singing frequencies
would be, in this latter case and with a gas velocity right at
the onset value of 15 m/s:

The first of the two values would be very close to a
resonating mode of the turret piping of the floating facility
considered. In such a case, acoustic lock on would occur
with consequent excitation of high pressure pulsations
around the frequency of 250 Hz (Figure 4).
The absolute dimensions or the scale factor for a same
carcass shape also have an important effect on the onset
velocity, as can be understood from the equations and
shown in Figure 3, which means that with larger diameter
risers the likelihood of riser singing increases.
Conclusion
The phenomenon of riser singing should be carefully
addressed in floating storage and offloading systems
design, particularly of LNG offloading and regasification
facilities, as the consequences of the generated pressure
pulsations may be severe. The phenomenon is very
sensitive to the corrugation geometry, scale factor and
fluid properties such that even apparently marginal
changes of these quantities may yield completely different
results and change the onset possibilities.
References
1. Ltveit, S. A., et al., Flow induced vibrations in
flexible pipes, proceedings of OMAE 2003, the 22
nd

International Conference on Offshore Mechanics and
Arctic Engineering Cancun, Mexico, OMAE2003-
37194, 2003.
2. Nakamura, Y., and Fukamachi, N., Sound generation
in corrugated tubes, Fluid Dynamics Research 7,
pp. 255 261, 1991.
3. Nakiboglu, G., et al., Whistling behavior of periodic
systems: Corrugated pipes and multiple side branch
system, International Journal of Mechanical Sciences
52, pp. 1458 1470, 2010.
4. Swindel, R, and Belfroid, S. P. C., Internal flow
induced pulsation of flexible risers, Offshore
Technology Conference, Houston, Texas, Ref.
OTC18895, 30 April 3 May 2007.
5. Tonon D., et al., Whistling of a pipe system with
multiple side branches: Comparison with corrugated
Figure 3. Sensitivity of the singing onset velocity to riser
carcass shape, scale factor and density.
LNG_MayJune_2013_69-73.indd 72 22/05/2013 10:20
pipes, Journal of Sound and Vibration 329,
pp. 1007 1024, 2010.
6. Belfroid, P.C., et al., Flow induced pulsations
caused by corrugated tubes, Proc. of
PVP2007, ASME Pressure Vessels and Piping
Division Conference, San Antonio, Texas,
22 26 July 2007.
7. Belfroid, S. P. C., et al., Flow induced pulsation
due to flexible risers, Offshore Technology
Conference, Houston, Texas, Ref. OTC19904,
4 7 May 2009.
8. Bruggeman, J.C., et al, Self-sustained aero-
acoustic pulsations in gas transport systems:
experimental study of the influence of closed
branches, Journal of Sound and Vibration 150(3),
pp. 371 393, 1991.
9. Batchelor, G.K., An introduction to fluid
dynamics, Cambridge Mathematical Library,
Cambridge University Press, 1967 2000,
11
th
printing.
10. Brighenti, A., and Osti, P., ACUSYS - Application
of MATLAB-SIMULINK for the simulation of
acoustic pulsation in plants (Italian), 1
st
Italian
MATLAB Conference, Bologna, 14 October 1994.
11. Brighenti, A., and Pavan, A., ACUSCOMP and
ACUSYS A powerful hybrid linear/non linear
simulation suite to analyze pressure pulsations
in piping, ISMA 2002 International Conference
on Noise and Vibration Engineering, Leuven,
Belgium, 16 18 September 2002.
Figure 4. Acoustic transfer function of the turret piping from
the riser to the connection with the topside plant at 250 Hz under
one of the operating cases considered (by ACUSYS
10, 11
).
www. ener gygl obal . com/ news
READ about the latest
developments in the LNG
industry on Energy Global
LNG_MayJune_2013_69-73.indd 73 23/05/2013 11:21
D
uring the middle of the previous decade, Chiles energy
matrix was highly dependant on the supply of Argentinian
natural gas, which was supplied through three pipelines that
crossed the Andean mountains.
This supply was suddenly interrupted by the Argentinian
government due to a shortage of natural gas availability, creating a
big problem for Chilean electricity production, which had to be
replaced by thermolectrical electricity generation fed by huge diesel
oil imports. This affected the cost of energy to the Chilean industry
and households.
At the same time, the country has developed quickly and needed
a reliable and independent source of energy at market prices that
could successfully compete with other fuels.
Quintero LNG facility
As a result of this situation, in May 2004 the Chilean government
decided to push for the construction of an LNG reception facility
to be built at Quintero, a port located 180km away from Santiago,
where demand for LNG is greatest. The state owned oil company
Enap was entrusted to lead the design and construction of the LNG
regasification terminal.
The main
energy operators
in the country were
invited to participate
in the construction of the
Quintero LNG facility,
which finally led to the
creation of a consortium formed
by Enap, Endesa Chile and
Metrogas, each with an equal share.
This partnership was later expanded to
include BG Group, one of the largest
producers and distributors of natural gas in
the world, giving birth to GNL Quintero. Its
mission was to build the LNG reception, storage
and regasification terminal in time to start bringing
gas into Chile in the winter of 2009.
The project was initially expected to start operations
in mid-2007 at an investment cost of US$ 700 million. The
final cost of this project was almost double the initial estimate,
and the start-up date of this facility was April 2009.
Figure 1. LNG carrier Celestine River approaching the Quintero terminal.
LNG_MayJune_2013_74-78.indd 74 22/05/2013 10:26
PULLING OUT
THE STOPS
Alfred Hbner,
Ultratug, Chile,
explains the beginning
of LNG terminal
services in Chile and
how this has impacted
the local market.
LNG_MayJune_2013_74-78.indd 75 22/05/2013 10:26
The initial throughput capacity of the terminal is
10millionm
3
/d of natural gas, enough to supply 40% of the
countrys natural gas demand. Furthermore, the terminal has
been designed to double this capacity without interrupting its
operations. The jetty is 1878 m long and can unload up to
12000m
3
/h of LNG from the carriers. The offloaded LNG is stored
in three tanks, one small 14 000 m
3
tank and two larger ones with
a capacity of 160 000 m
3
each.
Berthing carriers
As soon as BG Group was integrated into the consortium, it
started to look for reliable companies that could adequately
handle its LNG carriers, which would bring the required LNG to
the Quintero terminal.
Considering the hazardous nature of LNG handling, safety
was considered the most important issue for BG. The project
required careful design and preparation of the berthing and
unberthing manoeuvers for the incoming LNG carriers. The
required services for the ships included the port agency and the
other usual port services, such as pilotage and the timely
availability of tug services. Based on BGs agreement with
GAC Network, Ultramar was the chosen agency in Chile.
Once that aspect had been resolved, BGs main concerns
were for the simulation of a safe berthing and unberthing
manoeuver, the contracting of the necessary tugboats that would
assist these manoeuvers, and the training and preparation of the
port pilots and tug masters that would perform these
manoeuvers.
Considering the risky nature of the operations, and the huge
investment cost of LNG carriers, BG has strict and demanding
procedures for its ship handling. In this case, these included the
training of pilots and tug masters at Wallingford, England, where
manoeuvers were simulated at specifically created replications of
the real sea, wind and swell conditions and forces that the LNG
carriers would face at Quintero.
Prior to that, as these operations also required the timely
availability of an adequate number of powerful and duly fitted
tugboats, BGs international tender included highly demanding
conditions for the bidding parties, both in terms of the required
technical specifications of the tugboats, as well as the
implementation and control of a strict safety, quality and
environment (SQE) system by the performing parties.
Requirements
The specific technical features demanded by BG for the tugboats
included the following:
Four tractor or azimuth type tugs of minimum 50 t bollard
pull. Two of these needed to be fitted with Firefighting (FiFi1)
systems.
Special tube-type fendering for the tugboats that improve
distribution on the ships hull during pushing mode.
Special winches with automatic tension and release meters.
Gas detecting systems and automatic closing devices in the
whole tug accommodation and engine room.
Curtain type of water sprinklers around the bridge.
Use of Dyneema type Kevlar shiplines.
It is important to note that several of these features were
unknown or at least not regularly implemented in the Chilean tug
fleet until that date, but have meanwhile become mandatory in
several ports where the larger container ships are calling.
In addition, BG required that:
All manoeuvers should be directed in technical English
language, so that the ship masters are able to understand any
instructions given by the pilots. This, however, demanded
that the Chilean pilots and tug masters understood English,
which finally proved to be unrealistic and this requirement
was eventually discarded.
Pilots and tugmasters should undergo regular retrainings
either at Wallingford and/or in Quintero, under the
supervision of an international expert designated by BG.
Figure 2. A birds eye view of the Quintero LNG terminal.
LNG_MayJune_2013_74-78.indd 76 22/05/2013 10:26
These latter checks have been carried out twice at Quintero
since the beginning of the operations, whilst the pilots and
tug masters are being retrained at the Humboldt Marine
Training Center in Valparaiso.
Berthing manoeuvers should be carried out in accordance
with the norms and practices of BG, which implies
the incorporation of a computer system that controls
the distances by GPS so as to allow a slow and exact
positioning of the vessels at the jetty, to safeguard the port
installations.
Regular updating and monitoring of the tug operators SQE
system on shore and on board by a BG representative, so as
to make sure that it is properly working. This has also been
carried out in regular intervals with satisfactory results.
Quintero is a port located some 20 miles north of
Valparaiso, which handles most of the liquid, gaseous and dry
bulk cargoes in central Chile. As such, it is an intensely used
port that handles crude oil, petroleum products, LPG,
methanol, coal, copper concentrate and other bulk cargoes. It
is located in a relatively small bay that is opened to the
northwesterly winds during the winter months, and is exposed
to permanent swell. As such, the operation for LNG carriers is
not easy, due to the exposed location of the port, the
permanent swell, and the increasing number of other ships
calling at that port.
Utilisation rates
For many years, Ultratug has held a relevant position in
Quintero, attending a larger share of the shipcalls at the
different terminals located in the bay. The company has had
two tugs permanently stationed in this port, in addition to
another two tugboats in Valparaiso, which occasionally assist
its manoeuvers from that port.
The number of LNG carrier calls was initially projected to
start at 15 calls per year, which may gradually increase up to
24 calls per year.
Due to the high investment that would be needed for the
four required tugboats, and the low utilisation rate that these
tugs would have, it made no sense to use exclusively dedicated
tugboats for the LNG Quintero project. As a result, Ultratug
offered to build two new tugboats of 65 t bollard pull for the
BG tender, which would meet all the technical features required
in the tender specifications. To meet the requirement of four
tugs for each manoeuver, an alliance was formed with two
other local tug operators, providing similar tugs on a per call
basis.
This offer implied that BGs manoeuvers and stand-by
operations would have preference with the required number of
tugs whenever needed, without having to pay for the idle time
of the tugs. This offered a cheaper operation and a more efficient
use of the tugs.
Over time, the LNG Quintero project has expanded to move
LNG regularly to the Concepcion area by truck, which lead to an
increase to 40 LNG calls in 2012.
A key component of the Ultratug strategy is its
comprehensive SQE management system. The companys SQE
standards meet and exceed the strictest international
requirements, including those of the oil majors.
Ultratug is part of the Ultramar Group of companies, which
through their holding company Ultranav operates some
200 vessels: tankers, bulk carriers and gas carriers regularly
undergo vetting procedures by oil majors. Humboldt
Shipmanagement in Valparaiso, which is part of the same
Ultramar Group, provides the technical management, crewing
and SQE management for all vessels and applies the same
standards for the Ultratug harbour tugs.
This was crucial during the tendering process, when BG
evaluated Ultratugs offer and reviewed its SQE system on site,
which included physical visits to the tugs.
Furthermore, Humboldt has also developed a marine
training centre in its Valparaiso facilities, which is oriented to
train pilots and ship masters/engineers in several maritime
matters and for the renewal of several ship certificates. This
centre includes a simulator for the training of tug operations,
which has been approved by the Chilean Maritime Authority and
by BG for the training of LNG tug masters and pilots.
Conclusion
The BG operation has been a successful venture, which has
helped the terminal community, Ultratug and its crews, as
well as the local market to raise its working standards. It
also provides the company with a distinctive and unique
knowledge about harbour towage at LNG terminal operations
in South America. This experience has been key to supporting
other LNG projects in Chile, such as the Suez Energy LNG
terminal in Mejillones, as well as providing services to existing
facilities in Puerto Escobar and Baha Blanca, Argentina,
or sharing knowledge and experience with projects under
development, such as Gas Sayago in Uruguay.
Ult t i dd 1 13/05/2013 09 58
LNG_MayJune_2013_74-78.indd 77 22/05/2013 10:26
SUBSCRIBE ONLINE
A GLOBAL INDUSTRY
REQUIRES GLOBAL
PUBLICATIONS
LNG_MayJune_2013_74-78.indd 78 24/05/2013 10:52
TENSION
Blaine Dempke, Markey Machinery
Company Inc., USA, discusses new
winch technology that helps to
improve tanker transit safety.
Figure 1. The SMBC tug Mexicali, one of four
purpose-built tugs working the Energia Costa
Azul LNG terminal in Ensenada, Mexico.
A
utomated winch technology brings a new level of safety to the world of
tanker escort and emergency towing, a vital part of the LNG supply chain.
Escort tugs generally run with tankers once they reach port, prepared
to assist the large ships if necessary. Tankers also rely on tugs to perform
emergency towing on the rare occasions when ships systems fail and they are
unable to get safely to port under their own power.
It has been said that a tug is essentially just a platform for a winch. Winches,
wires and ropes are indeed fundamental to escort and emergency towing
operations. The winch on the deck of the tug controls the towline connected to a
Maintaining
LNG_MayJune_2013_79-82.indd 79 22/05/2013 10:28
ship. In this way, a tug can control a tanker by direct mode,
pulling the tanker behind it in an in-line formation, or by
indirect mode, steering a tanker by pulling the ships stern to
one side or the other (acting as a giant rudder). One of the tug
masters jobs is to maintain even tension on the towline to
avoid line failure due to excessive or spiking payloads. A
parting of the lines means loss of control of a tanker with
potentially disastrous outcomes.
The 1989 Exxon Valdez oil spill in Alaskas Prince William
Sound, and the public outcry that followed, spurred
advancements in environmental and human safety at sea. An
example, specifically benefiting escort and towing operations,
is improvements in electronics that allowed for measuring
tension in the line and gave tug captains the ability to make
far more accurate adjustments.
In the last decade, a second evolution in escort and
towing technology has been prompted by the construction of
LNG terminals in unprotected and challenging ports. Oil
terminals have long been in place, built in ideal locations with
better conditions for ship traffic, often coinciding with
population centres. However, due to current public opinion
and political influences, the more recently built LNG terminals
have had to settle for remote sites, sometimes in exposed
ports with challenging currents and seas. This brought about
the need for tugs and towing systems capable of working in
heavier conditions and the advent of a new class of winches,
generally referred to as automatic line tension winches. These
machines have control systems that automatically adjust line
tension, further reducing the chances of a line failure, allowing
captains to concentrate on driving the tug.
A survey of the latest winch
technology
In August 2012, a report authored by Robert G. Allan and
commissioned by the Prince William Sound Regional
Citizens Advisory Council surveyed the latest technologies in
ship escort and emergency towing systems around the world.
This Escort Winch, Towline, and Tether System Analysis was
used to compare the systems aboard the high-powered tugs
serving the Valdez, Alaska oil terminal to the current best
available technology. The study is equally applicable to LNG
tanker escort and emergency towing, given that there is little
significant difference in technique or equipment whether a tug
is serving an LNG tanker or an oil tanker.
Indeed, because many LNG terminals have been built in
ports characterised by heavier seas, this report examining
tugs in the challenging waters of Prince William Sound is very
relevant. In it, Allan noted: [...] very few places in the world
have such structured escort services as the system that exists
in Prince William Sound. Comparable examples would be
North Puget Sound, Southampton, UK, Milford Haven, UK,
various Norwegian oil terminals (Sture, Mongstad, etc.), and
some Middle East gas terminals. However it is fair to say that
Valdez/Prince William Sound is certainly one of the longest
escort operations in terms of distance, and presents the most
demanding environmental challenges of all of these, save the
fact that the European terminals are generally much busier.
Among others, the report surveyed tug operators
Bukser og Berging AS of Norway, KOTUG of the Netherlands,
IRSHAD of the UAE and Seaspan Marine Corporation of
Canada, as well as winch manufacturers Markey Machinery
Company, Inc. of the US and Rolls-Royce of Norway/Finland.
It found that escort winches have changed more than
anything else in escort technology in the past 10 to 15 years.
Most critically, Allan said, winch braking systems have
evolved to the point where they are the required virtual fuse
in the system, and line tension can be set and controlled quite
accurately.
According to the study, automatic line tension winches at
the appropriate work load rating are a vital part of a modern
escort tugs capabilities, especially one operating in higher
sea-states.
Allan writes: Dynamic loads on a hawser can exceed ten
times the rated Bollard Pull of a tug in extreme sea-states,
and that will far exceed the rated breaking strength of the
towline.
Without automatic line tension capabilities, he continued,
the system is entirely reliant on the skill of the Master to try
to avoid incurring high snatch loads on the towline, but even
the most talented skipper cannot preclude all such events.
Trying to manage towline loads by manoeuvring a 1000 t
vessel is not practical and ultimately leads to less efficient
force application to the attended ship.
Figure 2. A typical big escort winch type DESS-52 on the
Crowley tug Response, which escorts tankers in the Straits of Juan
de Fuca off Washington state in the US.
Figure 3. A tug assists the disabled LNG tanker, Catalunya
Spirit, 33 miles east of Massachusetts, US, in February 2008.
The tankers operator, Teekay Corp., contracted four tugboats
to assist the ship after it lost propulsion and became disabled.
One of the tugs was equipped with a Markey Render/Recover

Hawser Winch and Towing Winch. (Source: US Coast Guard
photo/PA3 Connie Terrell.)
LNG_MayJune_2013_79-82.indd 80 22/05/2013 10:28
Captain Greg Brooks, Principal at Towing Solutions
Inc., described the benefit of this new technology: If an
emergency were to happen and the tug turns to produce
250000 lbs of force, or if it suddenly hits a wave, the
tow line force is not going to be steady. The automatic
line tension feature allows the tug to have a spring
effect.
Brooks worked for Exxon as a tug captain and shore
side manager for 33 years before retiring and becoming a
consultant dedicated to improving the safety and
operational effectiveness of escort systems and general
towing operations. Brooks said that captains cannot
adjust quickly enough in these situations, but the
automated winch controls can. With this new
technology, the captain can increase or decrease the
sensitivity of the winch controls to match the tugs
movement with the period of the sea. For example, when
waves are high and the period between them is short, a
captain could dial up the winchs sensitivity to react more
quickly.
Origins of the automatic line
tension winch
In the mid 1990s, Markey Machinery Company developed
the first winch with automatic line tension in partnership
with Crowley Marine Services. The first Markey
Render/Recover
winch system was installed in 1996 on

Crowleys harbour class of tugs in Long Beach, California.
In 2002, Markey Machinery and Crowley established
another industry milestone when a Markey Asymmetric
Render/Recover
system was installed on the 7000hp

Crowley tug Response. This marked the first winch
system with the capability to exert more force than
the tug itself. Asymmetric refers to the separate water
cooled dynamic braking system allowing the winch to
brake harder than it pulls.
The company has created a system where, even if a
tug loses power, it still keeps the tug connected to the
tanker. A separate system keeps the brake connected
so the line can slip and prevent an overload and line
failure.
In his analysis, Allan found that very few winch
designer/manufacturers worldwide have developed the
very high-performance winches that can provide the
dynamic response necessary for a winch to pay
out/recover a towline under high load at high speed as is
required for many escorts. The electric-driven Markey
Render/Recover
winch does just that.

Although Markeys new winch technology can be
applied to many tug operations, every tanker escort
requires a tug with a winch custom designed to perform
in its particular environment. The company has a
methodology it goes through to analyse the sea-states
and other environmental conditions in which a tug
operates.
Towing systems will continue to advance to meet the
needs of tanker and tug operators. Most LNG companies
are very serious about preventing incidents. It is this
proactive attitude that will further drive improvement in
towing equipment and create more partnerships
between operators and manufacturers like the one that
brought about the automatic line tension winch.
The Aragon
Optimised Dual
N
2
Expander Cycle
www.kanfagroup.com
Developed & Matured Liquefaction Technology
Designed for FLNG
Optimised Cycle Efficiency & Availability
Safe and Reliable
Maximising availability
Maximising NPV
Made for Offshore
LNG_MayJune_2013_79-82.indd 81 22/05/2013 10:28
SPE Offshore Europe is where the E&P community meets to nd the
solutions required to keep up with an accelerating technology race,
and understand the demands of a changing industry committed to
ensuring security of supply in an increasingly complex world.
The Offshore Europe Partnership,
a joint venture between:
Oil and Gas Conference
and Exhibition
connect
3rd-6th September 2013
Aberdeen, UK
CONNECT
3rd6th September 2013
AECC, Aberdeen, UK
SPE OFFSHORE
EUROPE THE
NEXT 50 YEARS
R
E
G
IS
T
E
R

N
O
W
W
W
W
.OFFSHORE-
EUROPE.CO.UK
REGISTER FOR THE FREE CONFERENCE &
EXHIBITION: WWW.OFFSHORE-EUROPE.CO.UK
LNG_MayJune_2013_79-82.indd 82 24/05/2013 10:52
I
n 2006, the state of Cameroon decided to monetise its gas reserves by allocating a significant portion of its
existing gas discoveries to a single LNG export project, the Cameroon LNG project. GDF SUEZ was selected
in 2008 as a strategic partner to jointly work with Cameroon NOC Socit Nationale des Hydrocarbures.
The preliminary studies are close to completion and the front end engineering phase is under preparation. A
robust and reliable design will be important for this new phase.
The determination of extreme weather conditions is crucial for the design of maritime and ground
infrastructures for the plant, which is located 25 km south of Kribi.
Even if benign weather conditions are the rule for the area, the Mesoscale Convective System (MCS) can
generate squalls and wind gusts. Although limited data exists, approximately 100 150 storms/year are recorded
in the area, with wind speeds of up to 40 knots (80 km/h). This poor estimation of weather conditions can result
in low design margins, with an impact on CAPEX and plant availability, as well as LNG production losses. An
accurate estimation can help optimise the net present value (NPV) of the project.
Olivier Gorieu and Max Nussbaum,
GDF SUEZ LNG, France, and Dr Christophe
Messager, Extreme Weather Expertises, France,
look at extreme weather conditions generated
by Mesoscale Convective Systems over the
Cameroon LNG production site.
LNG_MayJune_2013_83-88.indd 83 22/05/2013 10:38
Data acquisition requires important and expensive
measurement campaigns to define the climatology over the
region and to extract the extreme values. However, this
region has limited instruments in place, and data recorded
over a long period of time are currently not available. The
project recently installed a meteorological station on site.
However, the field campaign can miss the most powerful
event because of its limited duration and/or due to the use of
a singular station.
The project is thus conducting a study to assess whether
it is possible to use a long-term modeling approach instead.
For the first phase of this study, two levels of investigation are
being considered: the production of a long-term scenario
using high resolution modeling, and the computation of gusts
based on this long-term scenario.
Objective
Intense MCS activity over West Africa and intensive
convective diurnal cycles
1
induce a strong vertical transfer of
turbulent kinetic energy (TKE).
2
The downward transfer branch
is performed through a powerful release process called wind
gust.
By definition, wind gust (or gust) is the maximum wind
speed measured during a specified time period. The American
Meteorological Society defines a wind gust as a sudden brief
increase in the speed of the wind. More specifically, the
National Digital Forecast Database defines a wind gust as the
maximum 3 second wind speed (in knots) occurring within a
2minute interval at a height of 10 m (~32.8ft).
Classical methods based on a simple empirical ratio
applied to the wind field from the regional climate model
(RCM) failed to quantify gusts, despite the use of high
resolution models.
In order to determine these extremes, the Cameroon LNG
project is performing a 20 year climatic high resolution
hindcast based on regional non-hydrostatic modeling. The
methodology and the first results are presented in this article.
Development
Global to regional
Global climate information is available both for ocean and
atmosphere states during the past 20 years. These data
sets are nevertheless relevant for regional and local studies
where complex physical/chemical interaction processes
occur, inducing modification, enhancement, and inhibition of
regional and local weather events.
All the processes occurring at horizontal resolution under
the global scale are thus not reproduced or are parameterised
for the global scale. This explains the low intensity of the
extreme events within the Global Reanalysis product. For
instance, the higher horizontal resolution currently reached by
the reanalysis products is 0.5 (~ 55km x 55 km), which is too
coarse to reproduce a convective event that presents a lower
horizontal extension or meteorological phenomena occurring
at a lower scale (precipitation, gust).
Regional climate model
The proposed method is based on a RCM known as WRF
(Weather Research and Forecasting), nested within a
20 year global atmospheric reanalysis produced by a world
meteorological centre, and using a high resolution nested
domain. RCM has already demonstrated its ability to simulate
the West African Monsoon cycle with all its seasonal
phases.
3
A numerical simulation method is used to produce
long simulation at high spatial and time resolutions using
supercomputer facilities.
However, the current regional simulations fail to correctly
represent the extreme wind gust associated to a synoptic or
regional meteorological event producing strong wind gusts.
The extreme wind is usually determined using a simple ratio
method applied to the sustained winds (measured or
simulated), at the regional and local model scale outputs.
The computation of gusts
In order to quantify these gusts at the regional and local
model scale outputs, a simple empirical ratio method is
generally used. This simple empirical ratio method does not
Figure 1. Scheme of the regional climate model (RCM)
computation architecture, its inputs (global reanalysis and surface
data) and its outputs.
Figure 2. Monthly wind field intensity and direction (wind
barbs) for the RCM wind (red) vs. Satellite QuickSCAT surface
wind (black arrows).
LNG_MayJune_2013_83-88.indd 84 23/05/2013 10:38
TECHNI CAL PROGRAM | EXHI BI T F LOOR | NETWORKI NG OPPORTUNI TI ES

42
nd
Turbomachinery
29
th
Pump SYMPOSIA
/TurbolabatTAMU @TurboPump
turbolab.tamu.edu
GEORGE R. BROWN CONVENTION CENTER
9.30 10.3.2013
LNG_MayJune_2013_83-88.indd 85 24/05/2013 10:53
take into account the physical processes associated with
gusts.
Consequently, even with high time and space
resolution model outputs, these methods failed to
accurately estimate the strong and rapid variability of the
gusts due to TKE transfer towards the surface. Considering
the MCSs activity over West Africa, this misrepresentation is
critical and recent studies on the gust generation underlined
that these empirical and statistical methods missed the
extreme gusts.
A gust algorithm based on TKE consideration was used
for this work, using a realistic and physically based method.
This method is based on the fact that a gust is the result of
the air parcel vertical deflection from an altitude level to the
surface.
1
A deterministic approach for gusts can occur when
the TKE is high enough to permit this air parcel to move
downwards, against the buoyancy gradient between the
surface and higher levels of the planetary boundary layer.
Several studies during the past few years showed that
the deterministic method based on the TKE transfer
produces an accurate result for intensity and time
variability compared to empirical/statistical methods.
The TKE method approach is currently implemented in
several national meteorological centres, in order to
correctly forecast the extreme winds associated with
gusts.
A dedicated off-line physical tool, named gust module,
was consequently developed for computation of the gusts
from the scenario provided by a RCM (Figure 1). The TKE
method requires 3D information of the main atmospheric
components (wind, temperature and humidity), as well as a
3D TKE field. The use of a RCM with a high level of turbulent
closure scheme is thus necessary.
Results
Validation of the RCM
Preliminary results for a simulation conducted during
June 2006 are presented here for the RCM assessment, as
well as the gust TKE method assessment (Figure 2). The
monthly wind field intensity and direction (wind barbs)
for the RCM wind are shown in red and fit with Satellite
QuickSCAT surface wind, represented by the black arrows.
The QuickSCAT satellite product can be used to assess
the wind produced by the model. Figure 2 presents the
simulated mean monthly wind field compared to the
QuickSCAT monthly wind field. The spatial variability of the
model output is greater than the remote QuickSCAT data.
This variability is most noticeable over the continent, but
there is no data to assess it. The mean wind seems to be
correctly simulated over the ocean, but the QuickSCAT
resolution is poor compared to the model. The Figure thus
perfectly illustrates the difficulties faced when assessing a
model at high spatial resolution.
Figure 3 presents standard skew T plots (temperature,
dew-point and winds with height) of radiosoundings
performed at Douala as the red marks and red curves,
compared to model outputs at the same location and time
represented in black and blue marks for D01 (~17 km x 17 km;
coarse resolution) and D02 (~6 km x 6 km; medium
resolution) domains, respectively. The dashed lines indicate
the water vapour mixing ratio (dew point), while the solid lines
show the potential temperature. The vertical atmospheric
in-situ radio-sounding in red is successfully compared to the
RCM simulation (black, blue) for two situations.
The overall analysis of the RCM outputs revealed that
the model produces an accurate vertical potential
temperature profile either for coarse or medium resolution.
Figure 3. Vertical atmospheric in-situ radiosounding (red) compared to the RCM simulation (black, blue) for the 20060603-00 UTC and
20060616-12 UTC situation.
LNG_MayJune_2013_83-88.indd 86 22/05/2013 10:38
This remarkable accordance between model outputs and
observation is likely to be related to the assimilation of
observed temperature radiosonde data within the NCEP
reanalysis system.
However, the simulated dew point (dash lines) is not as
accurate as the potential temperature and some discrepancies
can be exhibited all along the vertical. The bias in humidity
can be systematic, positive or negative.
The vertical wind profile is well reproduced aloft the
atmospheric boundary layer. The observation revealed an
important variability within the boundary layer, mostly in
direction, while the wind intensity stays under 10 knots. Close
to the ground, the wind intensity is well captured by the
simulation.
Validation of the gust post-processing
method
Remarkable events at Douala can be witnessed in both
meteosat imagery brightness temperature (Figure 4) or in
standardised meteorological bulletins (Table 1).
The satellite pictures in Figure 4 correspond to the 71
st
and
398
th
hours of the time period from 1 June 2006.
All of these stations are part of the World Meteorological
Organization (WMO) network and provide several
standardised meteorological bulletins per day, called METAR
(Meteorological Aerodrome Report) or SPECI (special).
METARs typically come from airports or permanent weather
observation stations. Reports are generated once an hour, but
if conditions change significantly, a report known as a SPECI
may be issued.
Analysis of the METAR and SPECI identified some
particular events with strong winds. The cases listed in Table1
have been selected because specific activities are noted in the
METAR or SPECI data. However, this is not exclusive
information. Some events may occur without being
mentioned in METARs and SPECIs.
The June 2006 time series of RCM 10 m wind (black), gust
ratio method (red), and gust TKE method (blue) are compared
in Figure 5 with remarkable events from bulletins (black
circles).
Figure 4. The meteostat imagery showing intense convective activity over West Africa, by brightness temperature.
Table 1. METARs/SPECIs events at Douala
Date Time Wind dir/int (/KT) T/Td (C) Miscellaneous Hours after run
start
Note
2 June 2006 23 UTC 000/00 27/26 VRB20G 30 KT 71
st
7 June 2006 16 UTC 280/08 30/25 VRB15 KT 184
th
7 June 2006 17 UTC 280/04 28/25 VRB15 KT 185
th
10 June 2006 12 UTC VRB/06G16 29/25 - 252
nd
11 June 2006 17 UTC 000/00 28/25 VRB15 KT 281
st
11 June 2006 18 UTC 000/00 28/25 VRB15 KT 282
nd
16 June 2006 15 UTC 130/21 - 260V090G 31 KT 399
th
18 June 2006 14 UTC 280/31 30/23 - 450
th
Sea breeze
27 June 2006 01 UTC 000/00 25/23 VRB10 KT 649
th
Key: VRB (variable), G (gust), KT (knots), UTC (coordinated universal time), dir/int (direction/intensity)
LNG_MayJune_2013_83-88.indd 87 22/05/2013 10:39
The gust TKE method confirms its ability to capture gust
observations and evaluate their intensity at around +/- 20% in
standard deviation.
Conclusion
For the specific case of Cameroons coast, it has been shown
that:
A RCM model is a relevant tool for downscaling the
atmospheric dynamics from global scale towards the
regional and the local scales. It is not able to correctly
reproduce extreme wind events.
A TKE algorithm is necessary to improve the simulation
of gust events by post-processing 3D outputs of the
RCM. It produces much better results than wind gust ratio
methods.
A combined method (RCM plus TKE) using higher
resolution will be used over a 20 year period in the Kribi area,
and simulations will be further validated against observations:
satellite, meteorological bulletins and data from the on-site
meteorological station. This climatic hindcast will provide a
consideration of local weather conditions, including gusts, in
the design of the Cameroon LNG plant, as well as a basis for a
predictive tool to support maritime operations while the
Cameroon LNG plant is in operation.
References
1. Brasseur, O., Development and application of a physical
approach to estimating wind gusts, Mon. Weather Rev.
129, 5 25, 2001.
2. Messager, C.; Douglas, J.; Parker, O.; Reitebuch, A.;
Agusti-Panareda, C.; Taylor, M.; and Cuesta, J. Structure
and dynamics of the Saharan atmospheric boundary
layer during the West African monsoon onset:
Observations and analyses from the research flights of
14 and 17 July 2006, Quarterly Journal of the Royal
Meteorological Society, DOI: 10.1002/qj.469, 2010.
3. Galle, H.; Moufouma-Okia, W.; Brasseur, O.; Dupays, I.;
Marbaix, P.; Messager, C.; Ramel, R.; and Lebel, T., A
high resolution simulation of a West African rainy
season using a regional climate model, J Geophys Res
109, DOI 10.1029/2003JD004020, 2004.
Figure 5. June 2006 time series of RCM 10 m wind (black),
gust ratio method (red), and gust TKE method (blue).
www.energyglobal.com
ADINDEX
energgyglobal
A
D
V
E
R
T
I
S
E
R
S

I
N
D
E
X
LNG Industry is audited by the Audit Bureau of Circulations (ABC).
An audit certificate is available on request from our sales department.
Air Liquide Global E&C Solutions 02
Air Products 27
Burckhardt Compression 13
Cameron 09
CB&I 07
Chart Energy & Chemicals Inc. 47
ConocoPhillips 31
Energy Global 68 & 73
Eni 04
FMC Technologies IFC
GTT 55
International Registries 37
Kanfa Aragon 81
Lydall 35
Markey Machinery 81
Nikkiso Cryo, Inc. 67
Optimized Gas Treating Inc. 51
Palladian Publications Ltd 28 & 78
Pittsburgh Corning OBC
Pump/Turbomachinery Symposia 85
Qatargas OFC & 25
Quietex IBC
Reader Survey 59
Selective Adsorption Associates 56
SPE Offshore Europe 82
Temati B.V. 63
TGE Marine 15
TNO 21
UltraTug 77
UOP, A Honeywell Company 43
Weatherford 49
LNG_MayJune_2013_83-88.indd 88 24/05/2013 12:32
The Quietex Advantage
INCREASE TENSILE STRENGTH Ideal for applications which require the glass to support its own weight
IMPROVE THERMAL EFFICIENCY Reduces heat transfer, decreases energy consumption
MAXIMIZE FLEXIBILITY Fully Customizable
MINIMIZE TRANSPORTATION COSTS Gulf Coast based
^'o E|||uoo|o |ouu * Hous|o|, oxus ..0^ * ||o|o .oo^'2co * o..c'^occ' * vvvuo||oxcou
RESILIENT BLANKET
and
DECK INSULATION
FOR THE LNG MARKETPLACE
theEXPERTSin
LNG_MayJune_2013_IBC.indd 1 21/05/2013 14:25
For over 65 years, FOAMGLAS
insulation has been used to support and insulate tanks with

proven longevity. The unique qualities of Pittsburgh Cornings original cellular glass technology,
provides unequalled compressive strength and thermal performance. These qualities make
FOAMGLAS
insulation the preferred choice of engineers worldwide.

SUPPORTING THE
WORLDS TANKS
FIND OUT MORE
www.foamglas.com/industry
+1-724-327-6100 l 800-545-5001
Protecting Companies and their People Worldwide
LNG_MayJune_2013_OBC.indd 1 21/05/2013 14:27

LNG Industry May 2013

Încărcat de

Informații document

Descriere originală:

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

LNG Industry May 2013

Încărcat de

Drepturi de autor:

Formate disponibile

L

UNIQUE SOLUTION FOR YOUR

amine treating simulator. The ProTreat

Dry area Treated gas CO

high pressure heat exchanger

pump is situated in the Netherlands. The medium

. The benefits of this advanced design are

winch system was installed in 1996 on

system was installed on the 7000hp

winch does just that.

insulation has been used to support and insulate tanks with

insulation the preferred choice of engineers worldwide.

LNG_MayJune_2013_OBC.indd 1 21/05/2013 14:27

S-ar putea să vă placă și