Corrosion of Constructional Steels in Marine and Industrial Environment

Engineering Materials
For further volumes:

http://www.springer.com/series/4288
Jayanta Kumar Saha
Corrosion of Constructional
Steels in Marine and
Industrial Environment
Frontier Work in Atmospheric Corrosion
123
Jayanta Kumar Saha

Technical Functions
Institute for Steel Development and Growth
Kolkata
India
ISSN 1612-1317
ISBN 978-81-322-0719-1
DOI 10.1007/978-81-322-0720-7
ISSN 1868-1212 (electronic)

ISBN 978-81-322-0720-7 (eBook)
Springer New Delhi Heidelberg New York Dordrecht London

Library of Congress Control Number: 2012943376
Springer India 2013
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief
excerpts in connection with reviews or scholarly analysis or material supplied specifically for the
purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the
work. Duplication of this publication or parts thereof is permitted only under the provisions of
the Copyright Law of the Publishers location, in its current version, and permission for use must always
be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright
Clearance Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)
Dipanjan and Debosmita

learned many things during the
compilation work
Preface
Mild steel is used as prime constructional steel for buildings, bridges, etc. But this
is very much prone to corrosion in industrial and marine environments in presence
of harmful pollutants and other industrial effluents in addition to normal humid
atmosphere. These corrosion problems are much severe in a tropical country like
India with vast coastline. Mild steel corrodes relatively faster and thus leads to
colossal loss in every year and to reduce this loss some kind of protection in the
form of paints and coatings is always used. Weathering steel considerably
improves its corrosion resistance in industrial and rural atmosphere. Engineers and
scientists all over the world are concerned for much indepth research works at
surface science and engineering of atmospheric degradation of construction steels.
Endeavours have been made in the present work for better understanding of the
degradation process and uses of high performance paints formulation and compare
them through various AC/DC electrochemical test methods and surface characterisation through electron microscopy, XRD and Raman spectroscopy.
The protective systems and their corresponding corrosion data are available in
the relevant overseas codes for reference. These data are not applicable in Indian
specific context as atmospheric corrosion is location specific. Atmospheric corrosion is the frontier research area where limited work has been carried out as it
comprises with three phases (solid/atmospheric/liquid environment). Therefore it
is significant to study the rust on MS and WS as well as coated steels in a given
atmospheric environments. This is important for the selection of more suitable
materials as well as for the safety of structures, including the realisation of
essential economic effect.
Weathering steel is used to protect structures from atmospheric corrosion in
specific environments. Rust layers on mild steel are not protective and are permeable to air and moisture. Protective impervious rust coatings are supposed to
form on weathering steel with time and due to high initial cost and sometimes
unfavourable environments this steel cannot be used. Microscopic observation of
rust layers on weathering steel reveals the two phases in layers parallel to the steel
surface. The inner phase extends up to outer surface if exposure periods are longer
than 5 years, and then it becomes the only component of the rust coating. In mild
vii
viii
Preface
steel, corrosion products form as two different phases, and there are two layers
with a common interface allowing moistures to go in and experiences higher
corrosion rate in outdoor exposure.
Present research work has been carried out in two parts: field exposure tests and
accelerated laboratory tests on unexposed panels. Under field exposure test three
representative sites P1: very close to sea shore and free from SO2, P2: away from
sea with presence of SO2 and P3: industrial environment with presence SO2 were
chosen. Analytical techniques (EDX, XRD and Raman spectroscopy) were used
for rust characterisation and SEM was employed for understanding the morphological state. Corrosion rates were estimated after different periods of exposure at
sites. It is found that though accelerated laboratory test can predict the general
trend of corrosion, actual field test data cannot be created in laboratory. So an
attempt was made to simulate protective rust coating in the laboratory that is
typical of field exposure protective rust formation on weathering steel.
It is my proud privilege to record a deep sense of gratitude and indebtedness to
Prof. P. K. Mitra & Prof. S. Paul, Metallurgical & Material Engineering of Jadavpur University, Kolkata, India and Dr. D. D. N. Singh, Sr. Deputy Director of
National Metallurgical Laboratory, Jamshedpur for their encouragement and
endless support throughout the research steps. I also would like to show my
gratitude to my beloved mother, gorgeous wife Mousumi, son Dipanjan and
daughter Debosmita for infinite emotional support, enduring love, encouragement,
patience through my hard times until this work came to completion. Finally I thank
god for granting me the strength to write this book and thank editorial team of
Springer publication for giving the work final shape as book.
Kolkata, India
Jayanta Kumar Saha
Contents
Theoretical Evidences . . . . . . . . . . . . . . . . . .
1.1 Rusting of Iron . . . . . . . . . . . . . . . . . . . .
1.2 Corrosion of Steel . . . . . . . . . . . . . . . . . .
1.2.1 Mild Steel . . . . . . . . . . . . . . . . . .
1.2.2 Weathering Steel. . . . . . . . . . . . . .
1.3 Atmospheric Corrosion. . . . . . . . . . . . . . .
1.3.1 Corrosion Products . . . . . . . . . . . .
1.3.2 Atmospheric Corrosion Mechanism .
1.3.3 Effect of Acidity of Solution . . . . .
1.3.4 Effect of Alloying Elements . . . . . .
1.3.5 Environmental Factors . . . . . . . . . .
1.4 Corrosion Protection by Coating . . . . . . . .
1.4.1 Passive Rust Coatings . . . . . . . . . .
1.4.2 Organic Coatings . . . . . . . . . . . . .
1.5 Degradation of Organic Coatings. . . . . . . .
1.5.1 Delamination of Coatings. . . . . . . .
1.6 Corrosion Measurement and Analysis . . . .
1.6.1 Corrosion Rate Measurement . . . . .
1.6.2 Electrochemical Methods . . . . . . . .
1.7 Rust Characterisation . . . . . . . . . . . . . . . .
1.7.1 SEM and EDX . . . . . . . . . . . . . . .
1.7.2 X-ray Diffraction . . . . . . . . . . . . .
1.7.3 Raman Spectroscopy . . . . . . . . . . .
1.8 Rust Simulation. . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
1
3
4
4
6
7
10
10
11
14
14
15
16
16
18
19
20
27
27
27
29
31
32
Practical Approach . . . . . . .
2.1 Materials Used . . . . . . .
2.1.1 Structural Steels .
2.1.2 Organic Coatings
2.1.3 Test Electrolytes .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
39
39
39
40
41
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
ix
Contents
2.2
Test Panel Preparations . . . . . . . . . . . . . . . . . . . . . .

2.2.1 Blasting of Panel . . . . . . . . . . . . . . . . . . . . .
2.2.2 Bare Panel . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.3 Coated Panel . . . . . . . . . . . . . . . . . . . . . . . .
2.2.4 Scribed Panel . . . . . . . . . . . . . . . . . . . . . . . .
2.2.5 Reference Plate. . . . . . . . . . . . . . . . . . . . . . .
2.3 Adequacy Test . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1 Weathering Test . . . . . . . . . . . . . . . . . . . . . .
2.3.2 Pull Off Adhesion Test . . . . . . . . . . . . . . . . .
2.3.3 Cross Cut Adhesion Test . . . . . . . . . . . . . . . .
2.4 Exposure Test Set-Up . . . . . . . . . . . . . . . . . . . . . . .
2.4.1 Test Racks . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.2 Exposure Sites . . . . . . . . . . . . . . . . . . . . . . .
2.5 Field Test Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.1 Rust Appearance . . . . . . . . . . . . . . . . . . . . . .
2.5.2 Corrosion Rate . . . . . . . . . . . . . . . . . . . . . . .
2.5.3 DC Corrosion Measurement . . . . . . . . . . . . . .
2.5.4 AC Impedance Spectroscopy . . . . . . . . . . . . .
2.5.5 Rust Characterisation by SEM and EDX . . . . .
2.5.6 Rust Characterisation by X-Ray Diffraction . . .
2.5.7 Rust Characterization by Raman Spectroscopy .
2.6 Controlled Laboratory Tests . . . . . . . . . . . . . . . . . . .
2.6.1 Humidity Test . . . . . . . . . . . . . . . . . . . . . . .
2.6.2 Salt Spray Test . . . . . . . . . . . . . . . . . . . . . . .
2.6.4 AC Impedance Spectroscopy . . . . . . . . . . . . .
2.6.6 Rust Characterisation by Raman Spectroscopy .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
41
41
42
42
43
44
44
45
45
46
46
46
46
48
48
48
49
50
50
50
51
52
52
53
54
55
56
56
56
Experimental Evidence. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1 Rust Appearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Corrosion Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3 DC Corrosion Measurement . . . . . . . . . . . . . . . . . . . . . .
3.4 AC Impedance Spectroscopy. . . . . . . . . . . . . . . . . . . . . .
3.4.1 EIS on Uncoated Panels After 42 Months Exposure
3.4.2 EIS of Uncoated Panels After 48 Months. . . . . . . .
3.4.3 EIS of Coated Panels After 18 Months Exposure . .
3.4.4 EIS of Scribed Coated Panels After
18 Months Exposure . . . . . . . . . . . . . . . . . . . . . .
3.4.5 EIS of Coated Panels After 42 Months Exposure . .
3.4.6 EIS of Coated Panels After Humid SO2 Exposure. .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
57
57
57
57
58
60
61
61
....
....
....
62
63
63
Contents
3.5
Rust Characterisation by SEM and EDX . . . . . . . . . .

3.5.1 Rust Characterisation by XRD . . . . . . . . . . . .
3.5.2 Rust Characterization by Raman Spectroscopy .
3.6 Controlled Laboratory Tests . . . . . . . . . . . . . . . . . . .
3.6.1 Corrosion Rate by Humidity Test . . . . . . . . . .
3.6.2 Corrosion Rate by Salt Spray Test . . . . . . . . .
3.6.4 AC Impedance Measurement . . . . . . . . . . . . .
3.6.6 Rust Characterisation by Raman Spectroscopy .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xi
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
63
64
65
66
67
67
68
70
81
81
109
Field Exposure Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.1 Field Exposure Test . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.1 Rust Appearance . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.2 Corrosion Rate . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.3 DC Measurement . . . . . . . . . . . . . . . . . . . . . . . .
4.1.4 AC Impedance . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.5 Rust Characterisation. . . . . . . . . . . . . . . . . . . . . .
4.1.6 Summary of Field Test Findings . . . . . . . . . . . . . .
4.2 Accelerated Laboratory Test . . . . . . . . . . . . . . . . . . . . . .
4.2.1 Humidity Test . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.2 Salt Spray Test . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.3 Corrosion Rate . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.4 DC Corrosion Measurement . . . . . . . . . . . . . . . . .
4.2.5 AC Impedance . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.6 Rust Characterisation. . . . . . . . . . . . . . . . . . . . . .
4.2.7 Summary of Accelerated Laboratory Test Findings .
4.3 Comparison of Field and Laboratory Tests . . . . . . . . . . . .
4.4 Simulation of Field Test Conditions at Laboratory . . . . . .
4.4.1 Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4.2 Experimental Methods and Results . . . . . . . . . . . .
4.4.3 Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.6 Suggestions for Further Work . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
111
111
111
112
113
113
115
120
122
122
122
123
123
124
125
127
128
129
129
129
134
141
143
143
Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
145
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Abbreviations
AR
MS
WS
ZP
ZR
ZP + MS
ZR + MS
ZP + WS
ZR + WS
MIO
PU
CR
T
RH
R
AC
DC
OCP
PD
CP
EIS
RE
WE
CPE
CPE + D
REAP
SCE
PSCT
Ecorr
Icorr
Mod
As received
Mild steel
Weathering steel
Zinc Phosphate system
Zinc Rich system
Zinc phosphate system on MS
Zinc rich system on MS
Zinc phosphate system on WS
Zinc rich system on WS
Micaceous iron oxide
Polyurethane
Corrosion rate
Temperature
Relative humidity
Rainfall
Alternate current
Direct current
Open circuit potential
Polarisation diagram
Cyclic polarisation
Electrochemical impedance spectroscopy
Reference electrode
Working electrode
Constant phase element
CPE with diffusion Fit
Rapid evaluation of paints
Saturated calomel electrode
Potentiostatic current transient
Corrosion potential
Corrosion current
Modulus
xiii
xiv
Zmod
Zimg
Zreal
h
Freq
Z
Ru
Rp
Rpo
C
Cc
Cf /Cdl
Yo
Wd
a
nm
pA
lA
lm
lm/yr
DFT
h
s
m
SEM
EDX
XRD
Raman
a FeOOH
b FeOOH
c FeOOH
a Fe2O3
c Fe2O3
Fe3O4
Fig
Ref
Abbreviations
Modulus of impedence
Imaginary part of impedence
Real part of impedence
Phase angle
Frequency
Impedance
Solution resistance
Polarization resistance
Pore resistance
Capacitance
Coating capacitance
Double layer capacitance
Admittance
Warburg impedance
CPE exponent
Nanometer
Picoampre
Microampre
Micron
Micro meter per year
Dry film thickness
Hours
Seconds
Months
Scanning electron microscope
Energy dispersive X-ray
X-ray diffraction
Raman spectroscopy
Goethite
Akaganeite
Lepidocrocite
Hematite
Maghemite
Magnetite
Figure
Reference
About the Author
Dr. Jayanta Kumar Saha is serving with Institute for Steel Development & Growth
(INSDAG) as Deputy General Manager spearheading key activities like sourcing,
corrosion, fire protection, tech advisory services of steel structures. He did his
graduation from Bengal Engineering & Science University, Shibpur, Kolkata in
Metallurgical Engineering, Post-graduation from Indian Institute of Technology
Madras in Production Engineering and Ph.D. in Metallurgical & Material
Engineering from Jadavpur University, India. He is an IRCA, UK accredited
lead auditor for ISO 9001. He has authored 11 INSDAG books on steel
applications and has credit of 20 papers published in national and 5 papers in
international journals. He is representing various national bodies/committees. He
had previous experience at Hindustan Aeronautics Ltd, Central Mechanical
Engineering Institute and Philips India Ltd. He has visited several countries like
UK, Netherlands, China, South Korea and Singapore, etc., in connection with the
pioneering work in steel applications.
xv
Introduction
Mild steel (MS) has got versatile application as structural steel for construction of
buildings, bridges, flyovers, pipelines etc. But this is very much prone to corrosion
in industrial and marine environments in presence of harmful pollutants and other
industrial effluents in addition to normal humid atmosphere. These corrosion
problems are much severe in a tropical country like India with vast coastline. MS
corrodes relatively faster and thus leads to colossal loss in every year and to reduce
this loss some kind of protection in the form of paints and coatings is always used.
Painting is an effective means but quite costly amounting 1015 % of the initial
construction cost of superstructures besides cost of repainting at regular interval.
Addition of small percent of minor alloying elements especially Cu, Cr, P etc in
mild steel known as weathering steel (WS) considerably improves its corrosion
resistance in industrial and rural atmosphere. While the organic coatings (paints)
are invariably applied to MS structures, WS is used with or without coatings.
Application of a defect-free paint to the structures improves their life. However,
problems start once coatings develop defects, leading to both uniform and
localized corrosion and ultimately failure of structures. Engineers and scientists all
over the world are concerned for much indepth research works at surface science
and engineering of atmospheric degradation of MS and WS as well as improved
formulation of organic paints. Endeavours have been made in the present
investigation for better understanding of the degradation process and uses of high
performance paints formulation and compare them on MS and WS through various
AC/DC electrochemical test methods and surface characterisation through electron
microscopy, XRD and Raman spectroscopy.
WS is used to protect structures from atmospheric corrosion in specific
environments. Rust layers on MS are not protective and are permeable to air and
moisture. Protective impervious rust coatings are supposed to form on WS with
time and due to high initial cost and sometimes unfavourable environments WS
cannot be used. Studies on atmospheric influences on steels in rural, industrial and
marine environments have indicated that P, Cu, Ni, Cr and Si improve the
resistance to corrosion while Mn does not. In an industrial atmosphere the
weathering process is rapid than rural atmosphere, whereas in coastal environment,
xvii
xviii
Introduction
the protective layer does not form an conventional WS. On MS, major rust phase c
FeOOH transforms to the more stable a FeOOH with time whereas in WS after the
same phase change, a continuous transformation of a FeOOH to either protective c
Fe2O3 or a Fe2O3 along with amorphous d FeOOH with increasing exposure time
takes place. On MS, this does not form in a continuous manner and the amorphous
d FeOOH formed are not protective in nature for this reason. Presence of chlorides
in the atmosphere aggravates corrosion of steels leading to the formation of basic
Fe2+, Fe3+ chlorides and b FeOOH on MS. In case of WS, exposure to coastal
environment causes the formation of a FeOOH, b FeOOH, c FeOOH, Fe3O4 and
amorphous rust. In Industrial environment, formation of a FeOOH is predominant
on steels which remains uniformly distributed throughout the rust layer.
Concentration of a FeOOH is higher and c FeOOH is lower on WS than on
MS. Amorphous rust is found at the bottom of the rust layer on both MS and WS.
On WS, the alloying elements Cu and Cr become enriched in the inner layer of rust
while Si, P and Ni are not found to exhibit any characteristic distribution. The acid
rain triggered atmospheric corrosion of steel and presence of SO2 accelerates
corrosion. Thus in an industrial environment, presence of humid SO2, helps in the
formation of a very thin oxide film composed of an inner layer of Fe3O4 covered
with an outer layer of FeOOH. Investigations on atmospheric corrosion of WS in
the presence of NaCl and SO2 revealed that NaCl can accelerate the corrosion rate
and their combined effect is greater than that caused by each single component. In
the absence of SO2, a FeOOH, b FeOOH, c FeOOH, Fe3O4 and c Fe2O3 are the
dominant corrosion products, while b FeOOH, c FeOOH, Fe3O4 and FeSO4H2O
formed in the presence of SO2. Microscopic observation of rust layers on WS
reveals the two phases in layers parallel to the steel surface. The inner phase
extends up to outer surface if exposure periods are longer than 5 years, and then it
becomes the only component of the rust coating. In MS, corrosion products form
as two different phases, and there are two layers with a common interface allowing
moistures to go in and experiences higher corrosion rate in outdoor exposure.
Organic coatings are used on steels to provide an effective barrier protection.
With time organic coating systems acquire dielectric properties, which cause them
to act as capacitors. Corrosion occurring at the metal surface has a polarisation
resistance related to the corrosion rate, and an electric double layer that also
behaves as a capacitor. Rusting does not occur until the polarisation resistance fall
below 106 ohm cm2.
The performance of protective coatings can be determined through accelerated
laboratory tests but these are not the true representation of actual service
conditions. Hence field tests are carried out to get reliable performance data on
uncoated and coated steels. The protective systems and their corresponding
corrosion data are available in the relevant overseas codes for reference. These
data are not applicable in Indian specific context as atmospheric corrosion is
location specific. Atmospheric corrosion is the frontier research area where limited
work has been carried out as it comprises three phases (solid/atmospheric/liquid
environment). Therefore it is significant to study the rust on MS and WS as well as
coated steels in a given atmospheric environments. This is important for the
Introduction
xix
selection of more suitable materials as well as for the safety of structures,

including the realisation of essential economic effect.
Research work has been carried out in two parts viz., field exposure tests and
accelerated laboratory tests on unexposed panels. Under field exposure test three
representative Indian sites like P1: very close to sea shore and free from SO2, P2:
away from sea with presence of SO2 and P3: industrial environment with presence
of SO2 were chosen. Both uncoated and coated panels were withdrawn from field
exposure sites for evaluation after different time intervals. DC corrosion
measurements were carried out in different test electrolyte including exposure in
humid SO2 environment. AC impedance tests were performed on exposed panels.
Analytical techniques (EDX, XRD and Raman spectroscopy) were used for rust
characterisation and SEM was employed for understanding the morphological
state. Corrosion rates were estimated after different periods of exposure at sites.
Under accelerated laboratory experiments both DC corrosion and AC impedance
tests were carried out in different electrolytes and in humid SO2. The severity of
corrosion rates were evaluated by humidity and salt spray tests. Rust formed were
analysed by SEM, EDX and Raman. Comparative analysis of data generated
through field and accelerated laboratory experiments has been performed. It is
found that though accelerated laboratory test can predict the general trend of
corrosion, actual field test data cannot be created in laboratory. So an attempt was
made to simulate protective rust coating in the laboratory that is typical of field
exposure protective rust formation on weathering steel.
Field exposure test is a very slow oxidation process whereas the accelerated
laboratory test is a very fast oxidation process. Several months/years together are
required to get measurable rust on surface of steel panels in field exposure. In
accelerated laboratory test rate of oxidation is much faster and rust layers/oxides
are formed, very quickly. The rust products/oxides formed in both the cases are
identical but their manifestations are different. As a result corresponding
morphologies and corrosion rates are vastly different.
The studies revealed that WS exhibited more pitting than on MS and
deterioration of coated panels was highest at P1. Chloride ions accelerated
corrosion at P1 and corrosion rate on WS is almost equal with MS. WS showed
compact protective oxide film at P3 and its corrosion rate was found lower than
MS for all the environments. Stable corrosion rate was found at P2 and P3 on WS
and presence of SO2 helped to prevent deterioration of weathering steels at these
two sites. Performance of scribed coated panels with MS substrate was inferior
with respect to WS substrate. Rust morphologies on MS showed lot of voids and
micro cracks at all sites but compact, acicular oxides were formed on WS at P2
and P3 sites.
Rust formed in continuous humidity test on weathering steel was not stable and
protective. Again WS developed more protective oxide than MS when exposed in
humid SO2 for longer period. No damage of coating was noted with initial
exposure of 18 months on coated MS and WS. The data generated with accelerated
laboratory tests on both coated and uncoated MS and WS panels have wide gap
and difficult to predict the performance in field exposure.
xx
Introduction
It was envisaged that the slow oxidation of field test can be simulated in
laboratory by selecting a slow scan rate in a potentiodynamic experiment. Na2SO4
solution was taken as test electrolyte because SO2 was created in situ and this is
also available in industrial environment at P3. Slow potentiostatic oxidation is
possible at a low scan rate in Na2SO4. Passivated oxide layers were found only on
WS but not on MS. Results obtained were very encouraging as the corrosion rate
was comparable to field test and rust morphology of WS passivated at -820 mV is
quite identical to rust formed on WS exposed at P3 for 42 months. Raman
characterisation of passivated WS panels showed the presence of c FeOOH along
with stable Fe3O4 and a Fe2O3, which are very similar to the rust of WS exposed at
P3 after 42 months. It was concluded that laboratory simulation by passivating WS
at -820 mV for 80 min, is almost similar with the oxidation state of WS exposed
for 42 months at P3.
Chapter 1
Theoretical Evidences
1.1 Rusting of Iron

Iron, in its various forms, when exposed to the different facets of environment it
tends to be highly reactive owing to its natural tendency to form iron oxide. This
degradation of iron is known as corrosion, more particularly rusting, when oxidation occurs in presence of moisture. However, if a thin film of iron oxide
develops on its surface which is impervious and tenacious, it protects iron from
further oxidation loss and it is called protective oxide film. This spontaneous
formation of protective oxides which forms only on certain type of alloy steels is
known as passivation. This hard nonreactive surface film (14 nm) inhibits further
corrosion.
Corrosion is an electrochemical phenomenon leads to the generation of very
low electric currents. A mathematical relationship is available between the oxidation rate and the electrical properties. Good resistance to oxidation may generally be expected when the electrical resistance of the oxide formed is high. J.C.
Hudson worked on ferrous metals and established its relative resistance in different
atmospheric conditions [1].
1.2 Corrosion of Steel

Steel corrodes when exposed to myriad conditions including outdoor atmosphere.
It is noteworthy that all types of steel including the low alloy type are prone to rust
in moist atmosphere. Rusting is an electrochemical process characterised by
exchange of electrons. In some cases, the additions of 0.3 % copper to carbon steel
can reduce the rate of rusting to a greater extent. The elements Cu, P, Cr and Ni
have all been shown to improve resistance to atmospheric corrosion. Formation of
a dense, tightly adhering rust scale is responsible in lowering the corrosion rate
J. K. Saha, Corrosion of Constructional Steels in Marine and Industrial Environment,
Engineering Materials, DOI: 10.1007/978-81-322-0720-7_1, Springer India 2013
1 Theoretical Evidences
leading to use without protection and can also extend paint life by decreasing the
amount of corrosion underneath the paint. The rate of rusting is usually higher in
the first year of exposure to atmosphere than in subsequent years, and increase
significantly with the degree of pollution and moisture in the air. Alloying elements contribute to a more compact and less porous corrosion product as surface
film. Adherent, protective films on these steels seal the surface against further
penetration of water, which does not easily wet the oxide surface. Compact surface
oxide films develop more rapidly in industrial atmospheres containing SO2, which
is probably involved in film formation in presence of moisture by forming
sulphurous and sulphuric acids which are very corrosive. This theory was based
largely on the observation that the corrosion products formed on steel when
exposed to industrial atmosphere were usually rich in sulphates. However, the
corrosion rates of weathering steels are not reduced in industrial atmospheres to
levels lower than those in non-corrosive rural or semi rural atmosphere. Periodic
drying is required for the surface film to develop its protective properties [2].
In acidic solution, it involves an oxidation reaction (anodic reaction) where the
metal gets into an ionic state by dissolution and releases electrons. Simultaneously,
a reduction reaction (cathodic reaction) consumes the free electrons released by
the anodic reaction and either a metal gets deposited or more usually the cathodic
reaction. In order to continue the corrosion process in steel, formation of distinct
anodic and cathodic areas is a prerequisite, which are electrically connected. These
anodic and cathodic reactions occur in presence of an electrolyte, which can be a
common aqueous solution, acidic medium or a thin film of moisture present on the
surface, pores and crevices. At higher humidity, corrosion increases due to condensation of moisture film on the metallic surface leading to formation of innumerable galvanic cells. These cells are formed due to generation of electromotive
force between surface film and trapped film in pores and crevices, which act as
cathode and anode. Intrinsically, two important factors influence the corrosion
phenomenon at a fundamental level. These are electromotive forces generated
between the two electrodes and pH of the aqueous media. The electromotive
forcepH relation was first proposed by M. Pourbaix [3] as shown in Fig 1.1 and
these are useful in predicting zones of corrosion, passivity and immunity in metal
aqueous system. Pure iron exhibits formation of protective scale whereas carbon
steel shows formation of incoherent layers of scale which easily flakes off to
expose fresh areas for further attack.
There are different forms of corrosion of which most important one is uniform
corrosion which occurs over the majority of the surface of a metal at a steady and
often predictable rate. Rusting can be slowed or stopped by using paint, controlling
conductivity of solution, by applying current to metals and/or by stopping oxygen
to reach the surface. Other forms of corrosion commonly encountered during
service exposure of iron and steel is localised corrosion which is more severe than
the uniform one as failure occurs without warning in a short period of use or
exposure. Galvanic corrosion can occur when two different metals are placed in
contact with each other and caused by the greater tendency of one of the metal to
give up electrons than the other. Pitting corrosion occurs in materials that have a
1.2 Corrosion of Steel
Fig. 1.1 E-pH (Paurbaix

Diagram) of iron in sulphate
containing aqueous media
protective film such as a corrosion product or when a coating breaks down.

Metallurgical factors that can affect corrosion in steel are crystal imperfections,
grain size and shape, grain heterogeneity, impurity inclusions and residual stress.
1.2.1 Mild Steel

In mild steels, passivation in the stricter sense is not possible. The passive region
of iron is characterised by a thin film of cubic oxide of c Fe2O3/Fe3O4 in neutral
solution. This type of film is formed by the reaction of clean iron with oxygen or
dry air. The composition of the passive film depends on the type of electrochemical reactions and the nature of solution to which it is subjected. In such a
situation, Fe2+ in solution may anodically form on the surface to give an outer c
FeOOH layer. Another passive film on iron is Fe(OH)2, which is a polymeric
layered structure [1]. However, it is reported to change character on removal from
the passivating medium and long-term drying, to a form more closely resembling
to c Fe2O3. It is reported that with the exception of those formed at very low
passivating potentials, passive films do not seem to undergo significant local
structural changes upon drying in the air. It is also reported that the passive film on
iron composed of small particle size of c Fe2O3/Fe3O4 [4]. Cahan and Chen [5]
suggested that the passive film is not a semiconductor but a highly doped film with
Fe2+ and Fe3+ as defects. The oxide film near the iron electrode contains Fe2+ and
Fe3O4 on outer surface. Raman spectroscopy study of the passive film indicates
that the film consists of a layered structure with at least two components. The inner
layer is most likely Fe3O4 and the outer layer primarily Fe3+ species. X-ray diffraction data shows a spinel oxide (c Fe2O3, Fe3O4), which is inconsistent with
other crystalline bulk oxides, hydroxides or oxyhydroxides [6].
1.2.2 Weathering Steel

Weathering steels comprise a group of high strength, low alloy steels containing
alloying elements to give an enhanced resistance to rusting compared with carbon
steels. These steels have 12.5 % of alloying elements (Cr, Cu, Si, and P) and have
a tendency to form rust at a rate depending on the access of oxygen in the presence
of moisture and air. As the process progresses, the rust layer acts as a barrier to the
ingress of oxygen and the rate of rust growth slows down. In mild steels, the rust
layer becomes non-adherent and detaches after specific time to exposure conditions. In weathering steels, the rusting process is initiated in the same way, but the
alloying elements help to produce less porous and more adherent rust film. This
rust system develops with time, becomes protective by impeding further access of
oxygen and moisture to the metal surface and hence reduces considerably the rate
of rust growth. The rust colour and its characteristic of weathering steel depend
upon the nature of the environment and exposure time. In an industrial atmosphere,
the weathering process will generally be more rapid and the final colour becomes
darker. In the rural atmosphere, the oxide formation is usually slower and the
colour becomes lighter. The tightly adherent oxide usually forms over a period of
18 months to 3 years in industrial atmosphere. Weathering steel promotes formation of an adherent rust layer after about 8 years of service and retards the
corrosion by 75 % compared to mild steel [7] and in presence of relatively high
airborne sea salt (coastal environment) the protective layer cannot be formed.
Weathering steel is not advised to be used in bare conditions involving severe
marine and severe industrial environments [8].
Pourbaix [9] showed that the typical behaviour of weathering steel is due to
passivation during drying and lack of activation during wetting. Rust reduction of
weathering steel is slower than mild steel while not much difference has been
found in chemical analysis of the rust films on MS and WS, the morphology is
quite different [10, 11]. Rust formed on weathering steel is rather compact in
comparison to the loose rust found on mild steel. However, favourable atmospheric conditions are required to get stable rust on weathering steel, like air borne
chloride (\0.5 mg/100 cm2/day), average wetness time \60 %, industrial pollutants (SO2 \ 2.1 mg/100 cm2/day) [12, 13].
1.3 Atmospheric Corrosion

Atmospheric corrosion is an electrochemical process with the electrolyte being a
thin layer of moisture on the metal surface. The composition of the electrolyte
depends on the deposition rates of the air pollutants and varies with the wetting
conditions. The factors influencing the corrosivity of atmospheres are gases in the
atmosphere, critical humidity and dust content. Two rural environments can differ
widely in average yearly rainfall and temperature and can have different corrosive
tendencies and corrosion products [14, 15]. During atmospheric exposure, steel
gets a reddish brown corrosion product consisting of different constituents. The
electrochemical reactions at wet surface of steel as proposed by Evans [16] in
neutral alkaline condition are:
Anodic half-cell reaction Fe ! Fe2 2e
1:1
Cathodic half-cell reaction H2 O + 1/2 O2 2e ! 2OH
1:2
The anodic and cathodic reactions are only the first step in the process of
creating rust. Several more stages must occur for rust to form:
Fe2 + 2OH ! FeOH2
1:3
Ferrous hydroxide [Fe(OH)2] and hydrated ferrous oxide (FeO.nH2O) is first

diffusion barrier layer formed on the surface. As the pH of saturated Fe(OH)2 is
about 9.5, the surface of steel corroding in aerated pure water is always alkaline.
Due to incipient oxidation green coloured Fe(OH)2 is formed. Ferrous oxide is
converted to hydrous ferric oxide or ferric hydroxide at the outer rust layer as
dissolved oxygen is available by the following reaction.
2 FeOH2 1=2 O2 ! Fe2 O3 :2H2 O
1:4
In weathering, steel rust formed on atmospheric corrosion in different environments is composed of crystalline compounds like haematite, magnetite and
oxyhydroxides of iron like goethite, akaganeite, lepidocrocite and feroxyhite apart
from amorphous ferric oxyhydroxide rust. These rust constituents transform to one
another during wetdry cycles of atmospheric exposure [17].Various phases of
corrosion products formed in progressive exposure to atmosphere are given in
Table 1.1 [6, 18].
The alloying elements play a major role in modifying the oxyhydroxide rust
layer which inhibits the ingress of oxygen and iron cations. Orange to reddish
brown in colour hydrated ferric oxide formed is called rust and available as non
magnetic a Fe2O3 and magnetic c Fe2O3. Rust layers are not protective because
they are permeable to air and water and steel continues to corrode even after rust
has formed [1].The rusting of steel in the atmosphere is given by:
2 Fe + H2 O + 3/2 O2 ! 2FeOOH
1:5
FeOOH is the main component of the rust formed in presence of water on steel
at room temperature. Misawa [10] has summarised these processes and according
to him metal dissolution is the anodic reaction, while the dominant cathodic
reaction is oxygen reduction. The presence of a thick electrolyte layer on the
surface can limit oxygen reduction rate. In such situations, the following reduction
reaction supports oxidation of steel.
8 FeOOH + Fe2 + 2e- ! 3Fe3 O4 + 4H2 O
1:6
Table 1.1 Phases of corrosion products in atmospheric exposure

Phases
Lattice
Crystal system Density Free
energy
/Habits
(gm/
a
b
c
(DG kJ/
cm3)
mol)
a FeOOH
Goethite
4.60
9.96
3.02 Orthorhombic/ 4.28

Acicular
c FeOOH
Lepidocrocite
3.06
12.51
3.87 Orthorhombic/ 4.09

Lath
d FeOOH
Feroxyhite
a Fe2O3
Haematite
2.94
4.49
c Fe2O3
Maghemite
Fe3O4
Magnetite
b FeOOH
Akaganeite
5.035 13.72
8.33
8.396
10.48
10.48
3.8
Hexagonal/
Plates
5.26 Hexagonal/
Plates
24.99 Cubic/Lath
5.24
4.69
Inverse spinel/ 5.18

Octahedra
3.02 Tetragonal/
3.55
Somatoids,
Features
-490.4 Yellowish brown to

dark brown,
Scaly/fibrous
-471.4 Polymorph of
goethite, platy,
orange colour,
Red Rust
Thin rolled films
-742.4 Reddish brown to
black flaky rust,
characterised by
red streak
-540.2 Black and similar to
magnetite
-1014.2 Black colour, mill
scale
Brown/white colour
Contains Cl- ions
a,b,c: Relative lengths of crystallographic axes DG : Free energy of adsorption
The oxidation of Fe2+ ions to green rust transformed to c FeOOH in wellaerated systems and in turn is transformed to Fe3O4 in oxygen depleted systems.
The phases of change in rust with time in mild steel are c FeOOH transforms to the
more stable a FeOOH. With increasing time, a FeOOH converts to either c Fe2O3
or a Fe2O3, while conversion to a Fe2O3 usually requires higher temperatures.
The rust on weathering steel after 16 years of exposure in a rural environment
was found to be composed of two layers, with the inner dull layer comprising
nano-sized particles of a FeOOH and the outer bright layer, c FeOOH [19]. The
rust on weathering steel after 25 years of exposure in an industrial environment
exhibited similar characteristics [20].
1.3.1 Corrosion Products

Weathering steels develop a compact adherent protective oxide film that protects
the surface against further corrosion with prolonged exposure to the atmosphere.
The first oxyhydroxide form is c FeOOH and part of it begins to transform to a
FeOOH. The remaining part at later stage is composed of both oxyhydroxides.
These hydroxides are less protective against corrosion and they readily crack
allowing for ingress of oxygen and moisture to reach the metal surface and cause
further corrosion. With time, a part of the FeOOH transforms to magnetic oxides
of iron, which are much more protective than these oxyhydroxides. In addition to a
and c FeOOH, there is another oxyhydroxide of amorphous nature called d
FeOOH. In mild steel, this does not form in a continuous manner and amorphous d
FeOOH forms are not protective in nature for this reason. The formation of
amorphous d FeOOH as a continuous layer next to the metal surface is catalysed
by the presence of P, Cu and Cr in the steel. The presence of this amorphous layer
was thought to be the reason for the excellent corrosion resistance of the weathering steels [21]. However, some findings [22, 23] show that the stable rust layer
was not necessarily composed of amorphous rust but densely packed nano-size Cr
substituted goethite. Cr substituted a FeOOH is very fine through which oxygen,
water and corrosive substances are difficult to penetrate. Furthermore, chloride
ions are also difficult to pass through this. However, for the formation of this
protective rust layer, it is necessary that favourable atmospheric conditions exist
for application of steel in bare condition.
After extensive studies it has been found that Cr compounds are effective for
obtaining the protective rust layer in a short period of time [24]. Cr substituted a
FeOOH forms rapidly in presence of Cr2(SO4)3 solution. This accelerates the
dissolution of steel and promotes the formation of goethite. On the other hand,
Cr3+ forms fine particles of Cr substituted goethite and improves the protection
ability of the rust layer.
1.3.2 Atmospheric Corrosion Mechanism

Misawa et al. [10, 25] first investigated the mechanism of formation of constituents
of atmospheric rust in aqueous solution and identified amorphous oxyhydroxide,
FeOx(OH)32x, besides a FeOOH and c FeOOH in atmospheric rust. Again Misawa et al. [11] and Yamashita et al. [26] have reported that the c FeOOH forms at
early stages of rusting and transforms into amorphous rust before converting to a
FeOOH. Both Misawa [10] and Suzuki et al. [27] also concluded that the presence
of Cu favours the formation of amorphous, crack-free uniform rust layer.
It is reported that the formation of a FeOOH and c FeOOH results from water
loss and crystallization of Fe(OH)3, the main corrosion product, with amorphous d
FeOOH as an intermediate phase during drying cycle. In presence of high
humidity, the reduction of c FeOOH results in the formation of Fe(OH)2 and
finally Fe3O4. Under dry and oxidising conditions, when oxygen is easily able to
penetrate into the rust layer, the ferrous layer/Fe3O4 is oxidised to unordered
Fe(OH)3 and/or amorphous FeOOH which again transforms into crystalline a or c
FeOOH by water loss and crystallization [28].
According to Larrabee et al. [22] Cu inhibits the formation of crystalline a
FeOOH and c FeOOH and thus prevents microcracking in rust. This is attributable
to crystallization of a FeOOH and c FeOOH during drying cycle of rust. Stratmann
et al. [29] found significant differences between the rusting of iron and iron with
0.5 % Cu during drying cycle and attributed this to the formation of a dense
corrosion product on Cu containing iron. Kishikawa et al. [30] reported that
weathering steel alloyed with Cr, Cu, P and Ni forms a non-amorphous densely
packed nano-sized Cr substituted a FeOOH, which prevents the permeation of
water, oxygen and corrosive substances. On the other hand, the c FeOOH membrane was found to possess anion selective property. The rust layer formed on the
weathering steel has double-layered structure and protects steel from corrosion
because of the formation of bipolar membrane which suppresses the cathodic
reaction. Thus, the formation of a FeOOH in inner layer and c FeOOH in outer
layer is important for protection ability against rust. The ratio of a FeOOH to c
FeOOH in rust increases gradually as time passes. Corrosion rate decreases to
almost zero when this ratio exceeds 1.4. A value of 2 is considered as rust stability
index for maximum protection [31].
Yamashita et al. [32] observed that atmospheric rusts on weathering steels are
composed of Cr substituted a FeOOH, c FeOOH and a small amount of c Fe2O3
and/or Fe3O4. The dark Cr substituted a FeOOH area was located in the inner layer
while the bright c FeOOH area was in the outer layer. Thus, the innermost Cr
substituted a FeOOH layer may be the final form of the protective rust layer which
suppresses and prevents the transport of corrosive species through the rust layer to
retard further corrosion.
Study conducted in Taiwan by Wei [33] on carbon steels and weathering steel
with high Phosphorous and exposed to rural, urban, coastal and coastal industrial
environments concluded that the characteristics of the protective rust layer and the
corrosion resistance of weathering steels depend on the environment and the test
period. It was observed that c FeOOH formed in the inner rust layer along with
some a FeOOH, a Fe2O3 and Fe3O4 in the initial exposure period and the amount
of a FeOOH gradually increased in weathering steels. The enrichment of crack
free and dense rust layer with Cr, Cu and P is attributed to the corrosion protection
of the substrate steel.
The rust layers on plain carbon and weathering steels exposed to coastal
industrial environment in Japan for 17 years had been characterised by Asami
et al. [34]. They found that the rust was composed of a FeOOH, b FeOOH,
c FeOOH, Fe3O4 and amorphous rust. a FeOOH was predominant on all steels and
appeared uniformly distributed throughout the rust layer. Concentration of a
FeOOH was higher and c FeOOH was lower on weathering steels than on plain
carbon steel. Amorphous rust was located at the bottom of the rust layer irrespective of the steel types. Previous studies by authors on distribution of phases
and alloying elements in the rust layers in weathering and plain carbon steels
reported that the rust oxides consist of three layers: inner, outer and outermost
[35]. The outermost layer was found to be about 3 lm thick and enriched with
atmospheric deposits. The concentration of b FeOOH was reportedly higher on the
skyward surface of both steels. In weathering steel, the alloying elements Cu and
Cr enriched the inner layer of rust, while Si, P and Ni were not found to exhibit any
characteristic distribution.
According to Yamamoto et al. [36], the amorphous and the crystalline constituents were intermingled in the inner rust layer of weathering steels exposed to
rural environment for 35 years. Asami et al. [37] analysed the rust layers formed
after 17 years on weathering steel bridge exposed to coastal industrial environment
reported that c FeOOH and b FeOOH existed in outer layer while the amorphous
rust and a FeOOH with enriched Cr, Ni and Cu were found widespread in the inner
rust. Kihira et al. [38] referring to the work of Sakashita and Sato [39] tried to
explain the difference in protectiveness of weathering steel rust by the phenomenon of ion selectivity of the rust layer. Keiser et al. [40] found d FeOOH in the
weathering steel rust along with about 10 % c FeOOH and a small amount of a
FeOOH. Yamashita et al. [41] investigated the rust layers formed on weathering
steels exposed for 17 years in Japan and reported that the protective properties of
the rust layer on weathering steels containing alloying elements (Cr, Cu etc.)
composed of a FeOOH. The amount of b FeOOH phases increased with the
progressive increasing level of airborne salt. They also suggested that the protective properties of the rust layer were related to the suppression of ion transport
due to its densely packed structure.
Most of the authors have indicated that long-term exposure results are important in terms of protective rust formation on weathering steels and also supported
the view that the early stages of exposure determine the subsequent corrosion rate.
Thus, corrosion rates during the early months of exposure are far more important
than the ultimate rate in the context of a study of the mechanisms of protection
[42].
It is proposed that during rust formation of steel in alkaline solutions the oxidation of Fe2+ proceeds via Fe(OH)2 and yields magnetite as the end product [43].
In another study of transformation of Fe(OH)2 at pH 11 and at temperature 65 C,
it was noticed that initially both Fe3O4 and a FeOOH form but a FeOOH formation
takes place at an early stage of reaction. It was suggested that excess Fe+2 ions
interacted with Fe+3 oxides, resulting in Fe3O4 formation. It is further reported that
Cl- ions retard magnetite formation by binding the neighbouring OH- ions groups
to form FeOFe linkage in alkaline pH 11. Sulphate has a tendency for a FeOOH
formation of 0.1 M concentration of Fe(OH)2 [2]. Moreover, sulphate also plays
role during transformation of green rust to oxides/oxy hydroxides.
Subramanium [44] have shown that the presence of small amount of cations
(Cu, Mn) in rusts may accelerate the oxidation of Fe+2 in solution due to their
compound forming tendency with Fe+3. Thus, magnetite is a major constituent
in the weathering steel grades, which contains 0.30.4 % (wt.) of Cu. This
corroborates the previous studies of Inouye et.al [45]. They indicated the strong
magnetite promoting tendency of Cu during the formation of magnetite from
Fe(OH)2. However, they have specified the maximum limit of Cu (3 % by wt.)
above which it suppresses the magnetite formation. In another study [46], it was
noticed that the presence of traces of Cr decreases the amount of Fe3O4 in oxide
and increase unstable a FeOOH content in the corrosion product. It can be noted
here is that the tendency of Cr ions to form fine particles of a FeOOH which may
increase with passage of time and help in stabilizing the rust layer and protect the
10
surface from corrosion attack. High Si content in weathering steel also gives the
similar protective effect [47].
1.3.3 Effect of Acidity of Solution

Acid solutions (low pH) are more corrosive than neutral or alkaline solutions. In
ordinary iron or steel, the dividing line between rapid corrosion in acid solution
and moderate or low corrosion is nearly neutral or alkaline solution at pH 7.5. In
case of corrosion of iron or steel in aerated water, anodic reaction takes place at all
pH values as per Eq. (1.1), but the corrosion rate varies due to changes in the
cathodic reduction reaction as per Eq. (1.2). In the intermediate pH 410 ranges,
loose, porous, ferrous oxide deposit shelters the surface and maintains the pH at
about 9.5 beneath the deposit. The corrosion rate is nearly constant and is determined by uniform diffusion of dissolved oxygen through deposit in this range of
pH. In more acidic solutions (\pH 4), the oxide is soluble and corrosion increases,
due to availability of H+ ions for reduction by the equation
2H + 2e ! H2
1:7
The absence of the surface deposit also enhances access of dissolved oxygen,
which, if present, further increases corrosion rate. Dissolved oxygen is cathodically reduced in acid according to
O2 + 4H + 4e ! 2H2 O
1:8
Reactions (1.7) and (1.8) occur simultaneously in acid solutions with dissolved
oxygen. Diffusion of dissolved oxygen controls the corrosion rate at a constant
level in the pH range 410. Thus, metallurgical variables affecting the anodic
reaction [1] have no effect on the corrosion rate. Such is not the case for acid,
where the cathodic reaction is under activation control. The carbide phase shows
low overvoltage (higher rate) for reduction of H+ ions. Thus, high carbon steels
have a higher corrosion rate in acid solution than that of the low carbon steels
[48, 49].
1.3.4 Effect of Alloying Elements

Metallurgical factors affect metal loss and tend to corrode at a lower rate with
higher alloy content. Atmospheric corrosion resistance of steel was improved by
alloying with Cu, P or Cr to form passive oxide layer [50]. Studies have shown that
these steels show superior corrosion resistance in particular during atmospheric
exposure but not so much for immersed exposure as in seawater and close to the
coastline in the presence of high chloride concentrations. Alloying elements like
11
Cu, Cr, Ni, Si, P, etc. are added in carbon steel to achieve compact, adherent and
pore free rust layer which in turn provide good corrosion protection to the steel
surface depending on the environment [2730, 51, 52]. Weathering process of steel
with Cu and P promote the formation of a tightly adherent, protective and stable
rust layer to act as a barrier to electrochemical attack under wetdry cycle. Corrosion of weathering steels containing Cu, P and Cr virtually ceases after 3 years
of exposure. It is noteworthy that the formation of protective rust layer does not
depend only on the alloying elements but also on the environment. It is observed
that the formation of protective rust on the steel surfaces containing Cu is easier in
industrial and rural atmospheres and difficult in marine atmospheres containing
chloride ions. In their research work, Larrabee and Coburn [53] have revealed the
effect of alloying elements on the corrosion resistance of steel where Cu and P
additions are most beneficial in improving the resistance of steels. They have
showed that the corrosion rate of plain carbon steels increases progressively with
exposure time whereas same decreases in Cu containing steels in industrial
atmosphere.
Atmospheric attacks on steels have been studied on field exposed steel in
industrial, rural and marine environments and found that P, Cu, Ni, Cr and Si
improve the resistance to corrosion while Mn does not seem to affect it and S
increases nucleation rate. The relative importance reported for marine atmosphere
is P, Si, Cu (up to 0.3 %) and Cr, Ni, Cu (above 0.3 %) [5254].
Horton et al. [55] observed that when steels containing Cu and Ni are exposed
in industrial and marine atmospheres, the Cu and Ni appear in the rust layers both
in the loose outer and adherent inner rust on skyward and ground ward surfaces.
Also it was shown by chemical analysis that Ni, Cu, Cr and Mn from weathering
steel appear in the rust layer and provides protection. Presence of chlorides in the
atmosphere accelerates corrosion of steels leading to the formation of basic Fe2+,
Fe3+ chlorides and b FeOOH. Townsend et al. [56] conducted 8-year atmospheric
corrosion tests on weathering steel in rural, industrial and marine environments
with different heated conditions and indicated that heat treatments have no effect
on the corrosion resistance/performance of weathering steels.
1.3.5 Environmental Factors

The environmental factors that tend to accelerate metal loss include high humidity,
high temperature and proximity to the ocean, extended periods of wetness and the
presence of pollutants in the atmosphere. The small amount of carbon dioxide
normally present in the air neither initiates nor accelerates corrosion. Vernon [57]
was the first to study the corrosion rate of steel coupons in the presence of welldefined atmospheres. Atmospheric gases such as CO2, SO2, NO2, HCl, etc. after
getting dissolved in the moisture layer on the metal surface, these gases result in a
number of ions and ionic species like H+, Cl-, CO32-, NO3-, SO42-, etc. They
measured corrosion rate was as a function of time, relative humidity and
12
atmospheric pollutants like SO2 and showed that corrosion starts in non-contaminated atmospheres only at a relative humidity near to 100 %. Thus, they suggested
that weather resistant steel is not to be used in uncoated condition in severe marine
and severe industrial environments. Weathering steel is 58 times more corrosion
resistant than plain carbon steel in industrial atmosphere and in marine atmosphere, superiority of the weather resistant steel depends on the salt content of the
atmosphere [58, 59].
Effect of Temperature. Temperature plays an important role in atmospheric
corrosion. There is normal increase in corrosion activity which can theoretically
double for each 10 increase in temperature. As the ambient temperature drops
during the evening, metallic surfaces tend to remain warmer than the humid air
surrounding them and do not allow condensation until some time after the dew
point has been reached. As the temperature begins to rise in the surrounding air,
the lagging temperature of the metal structures will tend to make them act as
condensers, maintaining a film of moisture on their surfaces [6063].
Effect of Relative Humidity. Relative humidity is defined as the ratio of the
quantity of water vapour present in the atmosphere to the saturation quantity at a
given temperature. Atmospheric corrosion takes place in presence of a thin film
electrolyte that can form on metallic surfaces when exposed to a critical level of
humidity. While this film is almost invisible, the corrosive contaminants are
known to reach relatively high concentrations, under conditions of alternate
wetting and drying. The critical humidity level is a variable and depends on the
nature of the corroding material, the tendency of corrosion products and surface
deposits to absorb moisture, and the presence of atmospheric pollutants. It has
been shown that, this critical humidity level is 60 % for iron if the environment is
free of pollutants [60]. In the presence of thin film electrolytes, atmospheric
corrosion proceeds by balanced anodic and cathodic reactions. The anodic oxidation reaction involves the corrosion attack of the metal, while the cathodic
reaction is naturally the oxygen reduction reaction. The most important factor in
atmospheric corrosion is moisture, either in the form of rain, dew, condensation or
high relative humidity. In the absence of moisture, most contaminants would have
little or no corrosive effect. Rain also may have a beneficial effect in washing away
atmospheric pollutants that have settled on surfaces. This effect has been particularly noticeable in marine atmospheres. On the other hand, if the rain collects in
pockets or crevices, it may accelerate corrosion by supplying continued wetness in
such areas. Continuous wetting and drying of the surface is required during
atmospheric corrosion which makes distinctly different from the usual corrosion
mechanism under immersed conditions [61].The atmospheric corrosion cannot be
described as a simple oxidation reaction but associated with electrochemical
reaction kinetics [62, 63]. Atmospheric corrosion experiences metal dissolution
and the oxygen reduction during wetdry cycles. In industrial atmosphere having
sulphur bearing compounds, the corrosion product on steel is a basic iron sulphate.
The Cu bearing steels form Cu compound which plugs the pores in the corrosion
product, resulting in better corrosion resistance [6466].
13
Effect of Sulphates. Sulphur dioxide comes from combustion of fuels and is

identified as one of the most important air pollutants which contribute to the
corrosion of metals. In the presence of SO2, corrosion starts at 60 % relative
humidity and the rates are considerably higher than in the absence of any contaminate. There is a close relation between the uptake of SO2 and the corrosion
rate measured during field exposure over a period and it depends considerably on
the acid hydrolysis of SO2 in water and the fast oxidation of SO2 on iron surfaces.
The cycle consists of two reactions, the oxidative hydrolysis of iron sulphate and
the subsequent acid corrosion of iron [67].
4FeSO4 + O2 + 6H2 O ! 4FeOOH + 4H2 SO4
1:9
4H2 SO4 + 4Fe + 2O2 ! 4FeSO4 + 4H2 O
1:10
The acid rain triggered atmospheric corrosion of steel and presence of SO2
accelerate corrosion rate due to formation of sulphuric acid resulting the deposition of iron oxides in zones of decreased pH. Misawa [10] recognised the
importance of amorphous and less crystalline iron oxide phases for the chemistry
involved between iron dissolution and formation of stable FeOOH. Iron ions
initially form green complexes, which are transformed into green rust, magnetite,
FeOOH and finally after long exposure times transformed to a FeOOH. Hence, a
large number of thermodynamically metastable phases exist which are transformed
to stable oxides with the environmental conditions [68]. Stratmann [69] developed
a number of experimental techniques to investigate the electrochemical reaction
mechanism during the atmospheric corrosion even under wet/dry conditions.
Studies indicate the adsorption of gaseous species into the corroding surface and
subsequent chemical reactions occurring in the thin electrolyte layer [66].
Thus, in an industrial environment SO2 in presence of moisture, leads to the
formation of a very thin oxide film composed of an inner layer of Fe3O4 covered
by an outer layer FeOOH. The weak outer layer of FeOOH is penetrated by
fissures and moisture enters into the pores in Fe3O4 layer in form of condensed
water and dissolved O2, SO2 forms H2SO4 in the oxide pores.
2SO2 + 2H2 O + O2 ! 2H2 SO4
1:11
The oxide pores is to permit easier access for the electrolyte solution to the
underlying surface. This H2SO4 partially dissolves the oxide, producing FeSO4 and
thereby opens the oxide pores to permit easier access for the electrolyte to
underlying surface.
Fe3 O4 + 4H2 SO4 ! FeSO4 + Fe2 SO4 3 + 4H2 O
1:12
Fe2 SO4 3 + H2 O ! FeSO4 + O2 + H2 SO4
1:13
Also, FeSO4 is also hygroscopic in nature which enhances the atmospheric

corrosion rate by attracting moisture from the atmosphere.
14
Effect of Chlorides. Marine environments have high percentage of relative

humidity and airborne salt. Studies have found that the thickness of the adsorbed
layer of water on zinc surface increases with % relative humidity and that corrosion rates increase with the thickness of the adsorbed layer. There also seems to
be a finite thickness to the water layer that, when exceeded, can limit the corrosion
reaction due to limited oxygen diffusion. However, when metallic surfaces become
contaminated with hygroscopic salts their surface can be wetted at lower % RH.
The presence of magnesium chloride on a metallic surface can make a surface
apparently wet at 34 % RH while sodium chloride on the same surface requires
77 % RH to create the same effect. Corrosion rate in chloride environment can also
be reduced in special grade of weathering steel as shown by the study. Maghemite
as oxide phase formed on the carbon steel exposed in the marine environment is
responsible for high corrosion rate [70, 71] although this rust phase is stable and
protective in industrial environment.
1.4 Corrosion Protection by Coating

The major protective coatings applied to structural steelwork are paints, metal
coatings and combinations of both. The choice is partly governed by the actual
environmental conditions and partly by economic considerations [72, 73]. The
protection methods can be divided into three categories like applying organic or
inorganic coatings, controlling electrode potential to make metal immune or
passive by adding alloying elements in steels to promote formation of passive layer
and addition of corrosion inhibitors to the environment [74].
1.4.1 Passive Rust Coatings

Passivity is due to the presence of a thin film which isolates the metal surface from
a corrosive aqueous environment. A major impetus for work in neutral solutions
came from the research of Nagayama and Cohen [75]. They considered that in the
passive region, iron is covered by a thin film of cubic oxide of c Fe2O3/Fe3O4.
Other compositions and structures were proposed for the passive film, some
involving the inclusion of hydrogen or the presence of water [76, 77]. In fact, the
composition of the passive film on iron depends on the type of electrochemical
parameters during the formation of the film and the nature of solution in which it is
formed. In passivity study, it is reported that the passive film on iron is composed
of Fe(OH)2, c Fe2O3/Fe3O4. [78, 79].
1.4 Corrosion Protection by Coating
15

Organic coatings used on steel provide an effective barrier protection by isolating
steel from the attacking species. Scientific investigations have determined that the
limiting factor in the protective mechanisms of barrier coatings is their resistance
to the flow of ionic current [80]. Coatings which contain large quantities of
metallic zinc provide corrosion protection by galvanic action. The barrier properties of the coating are improved by increased thickness, by the presence of
pigments and fillers that increase the diffusion path for water and oxygen, and by
the ability to resist degradation. Degradation allows access of reactants to the
coatingsubstrate interface without the necessity for diffusion through the coating
and it is an electrochemical process which follows the same principles as corrosion
of uncoated steel. The total corrosion process comprises of components like
transport of water, oxygen and ions through coating, development of aqueous
phase at the coatingmetal interface, activation of the metal surface for the anodic
and cathodic reactions and deterioration of the coatingmetal interfacial bond [81].
As coatings age in a corrosive environment, the interconnecting network of
pores within the coating eventually become saturated with water, salts, etc.,
exposing the metal substrate to a corrosive environment. The saturation of the
pores also creates paths of lower electrical resistance through the coating. Aged
organic coating systems also possess dielectric properties, which cause them to act
as capacitors to electrical current. Corrosion occurring at a metal surface has a
polarisation resistance related to the corrosion rate, and an electric double layer
that also behaves as a capacitor.
An important property of a coating is its resistance to water penetration and two
related properties are coating dielectric strength and coating resistance to ionic
movement. Water penetration in coating decreases the dielectric strength, resistivity and makes the coating less insulative. Once corrosion has begun, the corrosion products formed can cause undercutting and loss of adhesion of the coating.
Water penetration may swell the coating and produce stresses that eventually lift
the coating from the substrate. The presence of water increases coating deterioration and substrate corrosion, since they can accumulate underneath the coating,
cause delamination by blistering or accelerate corrosion of the substrate. Bacon,
Smith and Rugg [82] determined a direct correlation between resistances and the
ability of the coating to protect the underlying steel from corrosion. All coatings
were found to exhibit an initial decrease in resistance, which varied in terms of rate
and duration. For a good coating, this initial decrease was followed by an abrupt
recovery to around the original value. They found that the resistance of a poor
coating continued to decrease resulting in failure within 60 days. A coating that
maintained a resistance of 108 ohm-cm2 provided good corrosion protection while
those between 106 and 108 ohm-cm2 were fair and resistance less than 106 ohm-cm2
were poor performers [83, 84]. The dissolution of environmental water into the
coating was more important than the uptake of salt from the solution by the coating.
16
The level of dissolved salts of the exposure environment had an effect on the
properties of coatings [85].
Mayne [86] showed that, upon immersion in an aqueous solution, most organic
coatings acquire a negative charge and the acquisition of this charge has the effect
of creating a selectively permeable membrane, which is preferentially permeable
to cations; that is, a film that has gained a negative charge due to immersion.
1.5 Degradation of Organic Coatings

There are several types of corrosion found beneath organic coatings and these are
blistering, filliform corrosion, rusting, anodic undermining and cathodic delamination. Blistering is one of the first signs of breakdown in the protective nature of
the coating. The blisters are local regions where the coating has lost adherence from
the substrate and where water may accumulate and corrosion may begin. The blister
formation occurs by volume expansion due to swelling, gas inclusion and gas
formation [87, 88]. In all the cases, the blister provides a locale for collection of
water at the coatingsubstrate interface. Oxygen penetrates through the coating and
leaching of ionic materials from the interface. All the constituents are available for
electrochemical corrosion and oxygen is necessary for the cathodic reaction:
Again filliform corrosion is encountered on steel underneath organic coatings in
a humid air environment. Corrosion initiates in the presence of soluble ionic
species at defects in the coatings and propagates at the metalcoating interface as
worm like filaments due to differential aeration oxygen concentration. Oxygen
diffuses through the tail and leads to the separation of anodic and cathodic reaction
zones. The primary cathodic region is near the back of the head (at the headtail
boundary) where oxygen is supplied and the primary anodic region is at the front
edge of the head of the filament.
Organic coatings slow the mass transport process of water, oxygen and ionic
species which is necessary for corrosion. Non-sacrificial coatings show good barrier
properties but corrosion phenomena can occur at corrosion defects where the steel
substrate is not protected. Once corrosion starts on steel protected by an organic
coating system, growing of blisters appear and a rapid deterioration occurs. This
leads to the second type of protection that a coating can provide, sacrificial protection. In addition, corrosion products from the sacrificial layers promote pore
blockage in organic coatings preventing environmental intrusion [89].
1.5.1 Delamination of Coatings

Blistering and delamination are the most common forms of failure found in organic
coatings. Factors affecting the performance of a system include surface preparation, coating application, cure regime and film integrity. Soluble salts at the
1.5 Degradation of Organic Coatings
17
interface can form a concentrated salt solution and that acts to draw water through
the coating, which behaves as a semipermeable membrane from the exposure
environment. Anodic blistering mode of failure was addressed by Koehler [90]
who considered liquid filled blisters to be anodic in nature. Cathodic blistering is
the result of an alkaline environment under the coating caused by the cathodic
reaction, associated with corrosion that occurs at a damaged site of the film
[91]. The fault may take the form of mechanical damage to the coating or may be
inherent coating faults like pores/holidays.
Anodic half-cell reaction:
Fe ! Fe2 + 2e
1:14
Fe2 + O2 + H2 O ! Fe2 O3 + H2 O
1:15
Cathodic half-cell reaction:

O2 + 4H + 4e ! 2H2 O reduction in acidic solution
1:16
H2 O + 1/ 2 O2 + 2e ! 2OH reduction in neutral/basic solution
1:17
Some pathways must exist through the film to allow the sodium ions to the
interface in order to produce the alkaline environment. These pathways could be
due to pores. However, an alternative theory is proposed by Leidheiser [92]
suggests that beyond a given concentration, alkali cations may have a deleterious
effect on the coating, which leads to morphological changes and introducing
conductive pathways to the interface.
2Na + 2OH ! 2NaOH in presence of alkaline solution
1:18
Similar to cathodic blistering, cathodic delamination is also the result of

alkalinity at the interface. Again, this alkalinity is the result of cathodic activity
under the coating. It is associated with faults, either inherent or induced, in the
coating. Cathodic polarisation may be a consequence of either corrosion at the
point of damage or the application of cathodic protection. Resulting from experiments carried out by Smith and Dickie [93] on primer failure, it has been shown
that under impressed cathodic conditions, corrosion inhibitive pigments play no
part in the reduction of disbonding.
A number of explanations have been put forward for delamination mechanism
whereby the alkaline environment under the film affects the integrity of the metal
polymer interface, or perhaps more properly the interface between the oxide and
the polymer. Koehler [94] showed that this form of failure only occurs when there
are alkali metal cations available in the environment to act as counter ions to the
cathodically generated OH- ions.
Considering a coated steel substrate, immersed in an electrolyte of neutral or
near neutral pH, the half-cell reaction responsible for the delamination process is
to be oxygen reduction. This reaction generates OH- ions at the cathodic site and
is responsible for the alkaline environment at the delamination front. The elements
18
required for the process to proceed are water, oxygen and free electrons. The
electrons may be generated by either an anodic reaction or through the application
of cathodic protection [95].
Cathodic delamination is a result of a damaged coating; there are two possible
routes that the reactants for the cathodic reaction may take. The two alternatives
are either through the coating or along the polymermetal interface. An extensive
review of the delamination process was carried out by Leidheiser et al. [96] and the
results indicated that water was transported to the reaction zone through the
coating. It was suggested that a certain fraction of this could be in the form of a
cation as the cathodic nature and may favour the transmission of water associated
with an ion possessing a positive charge. The supply of oxygen to the cathodic site
was found to be largely through the coating, with a small contribution from
interfacial transport, in the case of the epoxy coating studied.
Adhesion plays an important role in the protective mechanism of coatings.
When one considers the process of cathodic delamination it is clear that, once the
paint has become detached from the substrate, the underlying metal is exposed to
the environment and is no longer accorded any protection from the coating system. Whilst the loss of adhesion, resulting from the delamination process, effectively reduces the protection afforded by the coating, it is important to consider
whether the original adhesion is the deciding factor in the delamination process.
Gowers and Scantlebury [97] suggested that the beneficial role of the adhesion of a
paint/coating is due to the impairment of the formation of a layer of electrolyte at
the coatingsubstrate interface, preventing ionic current flow and the spread of
corrosion over the surface. Gosselin [98] showed that good surface preparation is
the key to good adhesion but the type, as well as the condition, of the substrate has
been found to have a strong influence upon the initial dry, and the subsequent wet,
adhesion of a metal/coating .
1.6 Corrosion Measurement and Analysis

Laboratory corrosion testing and evaluation of uncoated and coated materials is an
integral part of corrosion studies. This involves immersion, salt spray and electrochemical testing techniques. These simulative tests may prove to be very useful
in generating data for estimation of corrosion performance and subsequent degradation. The testing must consider procedures which either reproduce a service
environment or use an environment with higher severity. Emphasis is placed on
coatings applied to steel surfaces and not many accelerated test methods are
available for predicting reliably the service performance of paints.
The permeability of organic coatings increases with time or the resistance to
penetration decreases with time. The degradation is associated with corrosive ions
and water penetration into the coating, transport of ions through the coating, and
subsequent corrosion reactions at the coatingmetal interface [99]. Standard
coating immersion tests can take hundreds to thousands of hours, whereas
19
electrochemical impedance spectroscopy (EIS) can provide reliable data on performance in a short time. Capacitance and electrical properties of the coating are
measured as a function of time. Since corrosion is an electrochemical process, it
appears logical that the electrical resistance of a coating would be related to its
protective ability. The DC resistance of the coating was essentially considered to
be the internal resistance of the cell metal/coating/aqueous environment. For good
coatings, the resistance changed slowly but for poor coatings the resistance
dropped more rapidly. Rusting generally was not noted on the test panels until the
DC resistance dropped \106 ohm-cm2. Since it is difficult to accelerate evenly all
the various factors involved, an accelerated method of detecting the deterioration,
or the lack of continued protection, of a coating could be more useful and accurate
than method of actually speeding up the deterioration or the corrosion process.
This is one reason why electrical methods for detecting paint breakdown appear to
show a comparatively high degree of correlation with actual breakdown in the
same environment. The extension of electrical methods for measuring the degree
of deterioration to coatings and uncoated steels exposed in atmospheric environments may thus be promising [100, 101].
1.6.1 Corrosion Rate Measurement

Corrosion occurs at a rate determined by equilibrium between opposing electrochemical reactions. The rate of any given electrochemical process depends on the
rates of two conjugate reactions proceeding at the surface of the metal. Transfer of
metal atoms from the lattice to the solution (anodic reaction) with the liberation of
electrons and consumption of these electrons by some depolarisers (cathodic
reaction). When these two reactions are in equilibrium, the flow of electrons from
each reaction of balanced and no net electron flow (current) occurs. Various
methods are available for the determination of dissolution rate of metals in corrosive environments but electrochemical methods employing polarisation techniques are by far most widely used. The corrosion rate (CR) is evaluated by mass
loss method considering uniform corrosion. The Corrosion rate is determined by
the following formula as per standard [102].
CRlm=year
87600W
AT D
1:19
Where, W is weight loss (mg), A is area of the specimen (cm2), D is density of

the specimen (gm/cm3), T is exposure time (hours) and unit lm/year is micrometre/year. Indirect methods of corrosion rate measurement involve anodic/
cathodic reaction, consideration of current potential relationship or polarisation
resistance values. Tafel extrapolation method is the most popular laboratory
methods for measuring corrosion rate of a metal from electrochemical data in a
corrosive medium.
20
1.6.2 Electrochemical Methods

Since kinetics and mechanism of corrosion is controlled by electrochemical
principles, the technique based on electrochemical methods is used to determine
the corrosion rate and understand the mechanism of corrosion process. The testing
methods are based on principle of accelerating the corrosion process without
changing the environment and the corrosion rates can be measured without
removing the test specimens.
These processes require anodes and cathodes in electrical contact and an ionic
conduction path through an electrolyte. The electrochemical process includes
electron flow between the anodic and cathodic areas; the rate of this flow corresponds to the rates of the redox reactions that occur at the surfaces. Monitoring this
electron flow provides the capability of assessing the kinetics of the corrosion
process. This also records the thermodynamic tendencies (potential) with the
accumulated metal loss registered. This is used to manipulate potential of test
specimen beyond its equilibrium value (OCP), a phenomenon called polarisation,
to effect measurements and magnitude of polarisation is called the overvoltage or
overpotential. It can have a plus or minus sign depending on whether it is above or
below the equilibrium potential value. The test electrode polarisation can be
accomplished by either DC/AC based polarisation measurements using a power
supplying equipment called Potentiostat. Three electrode corrosion testing cell is
employed with test electrolyte, specimen, counter and reference electrodes. The
counter electrodes are usually conducting, noble materials such as graphite, platinum, etc. Reference electrode is used to measure and record the potential of the
test electrode during the testing process. Normally, the more negative the potential,
the higher the metal tendency to corrode [103].
Open Circuit Potential. Metal immersed in an aqueous solution develops an
electric potential at its surface called open circuit potential (OCP) which is a
characteristic of the metal solution system. The magnitude of OCP is measured
with respect to reference electrode with the help of high impedance voltmeter and
potentiostat is used to polarise or displace equilibrium potential of specimen in the
negative (cathodic) or positive (anodic) direction with reference to OCP. This is
manipulating the rates (ionic currents) of respective cathodic and anodic half-cell
electrochemical reactions. The electrochemical potential of a metal in a certain
solution is dependant on the type of the metal, the composition of the solution and
its pH, oxygen content and temperature [104, 105].
Polarisation Test Method. This method is used to determine the corrosion rate.
Polarisation resistance (Rp) is the resistance of specimen to oxidation during the
application of an external potential in DC corrosion measurement methods. The
CR and Icorr are related to Rp and can be calculated from equation given below and
polarisation resistance is related to Icorr according to Stern Geary relation [106].
Rp
babc
2:303I corr ba bc
1:20
21
Where, ba and bc are anodic and cathodic Tafel slopes (mV/decade), Icorr is
corrosion current density (A/cm2) and Rp is polarisation resistance (ohm-cm2).
This involves a potential scan 250 mV of Ecorr at a scan rate of 0.11.0 mV/s.
The technique is used to determine the equilibrium corrosion current, potential,
Tafel constants and corrosion rates. The corrosion rate (CR) is determined from
the Faradays law:
CR
0:13IcorrEW
q
1:21
Where, CR is corrosion rate in mpy (1 mpy = 0.054 lm/yr), EW is equivalent

weight, q is density of material in gm/cm3 and Icorr is corrosion current in A/cm2.
Tafel extrapolation is used to determine the equilibrium corrosion current, where
linear extrapolations of anodic and cathodic branches of the plot beyond 50 mV
of OCP are made to intersect at OCP to measure the Icorr. This is a destructive
technique as it can cause some degree of surface roughening on the test specimen
[107]. General corrosion occurs in the active region, little or no corrosion occurs in
the passive region and pitting corrosion can occur in the transpassive region [108].
Cyclic Polarisation. Cyclic polarisation curves are considered as an extension
of potentiodynamic polarisation curves and used to measure the pitting tendencies.
The potential scan begins at Ecorr (OCP) and continues in the positive (anodic)
direction up to the transpassive region, where a large increase in current (corrosion) occurs. At a threshold current density, the scan is reversed and continued in
the negative (cathodic) direction back. The applied potential versus the log values
of the measured current density are plotted. The cyclic polarisation plots can show
positive hysteresis, negative hysteresis, repassivation or protection potential.
Negative hysteresis occurs when reverse scan current density is less than that for
the forward scan and positive hysteresis occurs when reverse scan current density
is greater than that for the forward scan. A passive film is damaged when potential
is raised into the transpassive region and pits can initiate when film damage is at
discrete (localised) locations on the metal surface. Pits will continue to grow when
protection potential (Epp) is greater than Ecorr and pits will not grow when Epp is
less than Ecorr. In cyclic polarisation curve, hysteresis can provide information on
pitting corrosion rates and how readily a passive film repairs itself. Positive hysteresis occurs when passive film damage is not repaired and/or pits initiate; negative hysteresis occurs when a damaged passive film repairs itself and pits do not
initiate. Area of hysteresis is very important as more the area, more aggravated the
corrosion is. Generally, the reverse scan is at a higher current level than the
forward scan. The size of the pitting loop is a rough indication of pitting tendency;
the larger the loop, the greater the tendency to pit [109, 110].
Electrochemical Impedance Spectroscopy. Electrochemical Impedance Spectroscopy (EIS), a non-destructive investigative technique enables an insight into
the corrosion process not obtained by DC techniques. EIS provides information on
reaction parameters, corrosion rates, oxide characteristics and coating integrity,
data on electrode interfacial capacitance and charge transfer resistance. It provides
22
kinetic and mechanistic information on electrochemical systems such as corrosion

processes. The slow electrode kinetics, slow preceding chemical reactions and
diffusion impede electron flow in electrochemical cells much in the same way as
resistors, capacitors and inductors do in AC circuits. The working electrode
interface undergoing an electrochemical reaction is analogous to an electronic
circuit with a specific combination of resistors and capacitors and AC circuit
theory can be used to characterise an electrochemical system in terms of equivalent circuit. The technique broadly involves subjecting an electrochemical system
to a range of small magnitude AC polarising voltage frequencies and corroding the
system response in the form of complex impedance plots. The complex impedance
diagrams are correlated with an equivalent AC circuit model with unique values
for circuit elements. These values can then be used to infer kinetic and mechanistic
information about an electrochemical system [111113]. The response of a corroding metal to small amplitude AC signal (1020 mV) of widely varying frequency (0.001100 kHz) can be analysed by EIS following the absorbance of
electrical energy at a certain frequency at the metal solution interface. On application of a sinusoidal alternating potential signal of the form:
V t V0 Sin xt
1:22
Time dependence current response of the form:

I t I0 Sin xt h
1:23
Where V(t) is applied potential,V0 is amplitude of applied potential, I0 is

amplitude of generating current, electrode surface expressed as an angular frequency (x) and h is phase between V and I. Due to the applied potential frequency
(x) dependent impedance Z(x) may be expressed as:
!

Rp
xR2P
Z x Ru
Zreal j Zimg 1:24
j
1 x2 R2p Cdl2
1 x2 R2P Cdl2
Where Ru is the solution resistance, Rp is the polarisation resistance and Cdl is
the double-layer capacitance. Various electrochemical phenomena at the metal
solution interface causes a time lag and a measurable phase angle h. These processes will be simulated by resistive and/or capacitive electrical networks. The
impedance behaviour of an electrode may be expressed in Nyquis plot of Zimg
(imaginary part of impedance) as a function of Zreal (real part of impedance) or in
Bode plots of mod Impedance and h versus frequency, where x = 2 pf.
To evaluate a coating, along with Ru, Rp, Cdl and Wd (Warburg impedance) two
additional circuit elements, namely coating capacitance (Cc) and resistance of
coating pores (Rpo) come into account. The presences of Cdl or Cc can be idealised
by a constant slope in Zimg versus frequencies plot and peaks in h versus frequencies plots.
For uncoated sample, Zimg versus frequencies plot shows early low impedance
at all frequencies. In coated specimen, Rpo measures the early deterioration at low
23
frequency impedance. The corrosion product accumulation at the coatingmetal

interface can induce coating defects and thereby reduces Rpo. The coating in these
cases is applied to the surface, which is not completely derusted. Rp for corrosion
beneath the coating is apparently quite high, as stated earlier and would require
still lower frequency measurements, which are difficult and time consuming. In
absence coating, the Zimg versus frequency plot measures the low value of Rp at
low frequency resulting from the comparatively high corrosion rate.
Electrified interfaces called electric double layers (Cdl) are set up at metal
electrolyte boundaries during electrochemical process. These interfaces are characterised by impedances to electron flow and ionic movement. The impedance of
an electrified boundary manifests as interfacial capacitance and associated charge
transfer resistance. The electrified interfaces are typified by time constants, which
are given by product of magnitudes of associated capacitances and resistances. The
time constants are noticeable in EIS spectra as semicircles in Nyquist plots,
negative slopes in Bode magnitude plots and negative inflections in Bode phase
plots. Mathematical regression of time constants in EIS spectra with equivalent
electrical AC circuit models leads to quantification of associated resistances and
capacitances. A higher charge transfer (ohmic) resistance implies greater polarisation or corrosion resistance of the metal in a given aqueous environment. The
capacitances themselves can be used to identify the corrosion, coating and diffusion processes with different time constants.
Thus, the metalelectrolyte interface behaves and responds like an AC circuit
with a specific combination of resistors and capacitors under the influence of the
AC polarising voltage frequencies. Figs. 1.2, 1.3, 1.4, 1.5 show metal solution
interface (single time constant system) where Ru is solution resistance, Rp is polarisation resistance.
The EIS spectra for coated metalelectrolyte systems are characterised by two
time constants, two semicircles in Nyquist plots, two negative slopes in Bode
magnitude plots and two negative inflections in Bode phase plots [114, 115].
Figs. 1.6, 1.7, 1.8, 1.9 show coated metal solution interface (two time constant
system) and Cdl is double-layer capacitance.
The coating time constants are smaller and manifest at higher frequency regions
of the impedance spectra, whereas time constants corresponding to metal corrosion
appear at lower frequency regions. An impedance plot obtained can be correlated
with one or more equivalent pore resistance (Rpo), coating capacitance (Cc) and
polarisation resistance (Rp).
CPE is used in a model in place of a capacitor to compensate for non-homogeneity in the system. A rough or porous surface can cause double-layer capacitance to appear as CPE and Warburg element [116, 117].
Kihira et al. [118] applied EIS to investigate the condition of the rust film
formed on the weathering steel, and proposed new corrosion monitoring method
based on rust film resistance. Nishimura et al. [119] measured the electrochemical
impedance of a carbon steel covered with rust film formed in a wet/dry environment containing chloride ions. They reported that the charge transfer resistance
(Rp) increased with the wetdry cycles of exposure.
24
Fig. 1.2 Metal electrolyte

interface of uncoated
corroding steel
Fig. 1.3 EIS spectra for

single time Constant of bode
magnitude plot

single time constant of bode
phase plot
Itagaki et al. [120] used EIS to investigate the electrochemical properties of the
rust film membrane formed on low alloy steels. The electrochemical impedance of
the actual rust film membrane formed by wetdry cycles showed the capacitive
semicircle on Nyquist plot corresponding to a single time constant. The time
constant of the capacitive semicircle was found composed of the rust film resistance and the film capacitance. The value of rust film resistance was shown to
depend on the alloying elements in weathering steel and it was shown that the
25

single time constant nyquist
plot
Fig. 1.6 Electrolyte

interface of corroding
coated steel
Fig. 1.7 EIS spectra for two

time constant of Bode
magnitude plot
addition of alloying elements increases the diameter of the capacitive semicircle.

The result meant the low permeation rate of chloride ions in the rust films of these
alloys. The Nyquist plots were found to diverge from a true semicircle due to the
current distribution in the film [121].
Feliu et al. [122] have applied EIS to study the corrosion and electrochemical
activity at the metalrust interface in connection with the application of protective
26

time constant of Bode phase
plot

time constant nyquist plot
treatments with rust converters to rusted steel. Mild steel with mill scale were
prerusted for 2 years in a rural atmosphere before applying conversion treatments.
They found that the shape of the low frequency areas of the Nyquist plots are
markedly influenced by diffusion processes in the rust layer and/or by the porous
nature of the rusted steel electrode itself.
With alloying and increasing period of exposure to saline atmospheres, the
magnitudes of rust pore resistances are expected to increase and rust capacitive
reactance will decrease since capacitance is inversely proportional to AC impedance. It is also likely that the charge transfer resistance which is indicative of metal
corrosion, itself will undergo an increase with alloying thereby signifying higher
corrosion resistance for alloyed steels. Corrosion, coating and diffusion processes
are not always associated with same frequency ranges. Corrosion resistances are
observed at low frequencies, but coating pore resistances can also be observed at
low frequencies, particularly, when a coating is saturated with electrolyte and
metallic corrosion does not occur. Capacitance values can be used to guide
interpretation as to what type of process is associated with each time constant.
Corrosion time constants have capacitance values (120 lF/cm2), coating time
constants have capacitance values (nF/cm2) and oxides have capacitance values
(1000 lF/cm2). Capacitance values of the order of C100 lF/cm2 are found when
surface adsorption occurs in conjunction with corrosion [103].
27
EIS data are analysed by fitting them to an equivalent electrical circuit model
consisting of resistors, capacitors, and inductors. The working electrode interface
undergoing an electrochemical reaction is analogous to an electronic circuit and
can be characterised as an electrochemical system in terms of equivalent circuit.
Typical circuits are shown in Figs. 1.10, 1.11, 1.12 and 1.13 where Yo is admittance (ohm-cm2),Cf is double-layer capacitance and a is the exponents [114]. (R.E:
Reference Electrode and W.E: Working Electrode)
1.7 Rust Characterisation

Several techniques are used in different stages of rust characterisation for steels
and microscopy related techniques are useful in understanding the topological state
of the corroded layers and in analysing their cross sections.
1.7.1 SEM and EDX

The degree of corrosion, surface morphology, particle size and texture can be
effectively studied by scanning electron microscope (SEM) and energy dispersive
X-ray analysis (EDX). The optical microscope can be used for imaging the surface
but it has limitations of resolution and depth of field at higher magnifications. SEM
can be used for high-resolution imaging of the surface, with a large depth of focus.
Atmospheric corrosion of weathering steel in the presence of NaCl and SO2 was
investigated by A.Q. Qu found that NaCl can accelerate the corrosion [123]. The
relationship between mass loss and amount of NaCl deposition follows the quadratic function both in SO2 free air and in air containing SO2. The combined effect
of NaCl and SO2 on the corrosion of steel is greater than that caused by each single
component [124]. SEM and EDX are used to characterise the corrosion products of
steel. In the absence of SO2, a FeOOH, b FeOOH, c FeOOH, Fe3O4 and c Fe2O3
are the dominant corrosion products, while b FeOOH, c FeOOH, Fe3O4 and
FeSO4.H2O dominate in the presence of SO2 [125].
1.7.2 X-ray Diffraction

X-ray diffraction (XRD) is used for identifying the oxides in rust and sometime
provides incorrect identification of the composition of the rust formed on weathering and carbon steels [126, 127]. Separate identification of Fe3O4 and c Fe2O3 is
not possible as both oxides have cubic structure and nearly identical lattice
parameters at room temperature. Analysis of rust coatings by XRD significantly
underestimates the goethite fraction in the corrosion products, especially for
28
Fig. 1.10 Representative

randle equivalent circuit

CPE equivalent circuit

CPE with diffusion
equivalent circuit

REAP equivalent circuit
weathering steel [128]. This is due to the presence of the nanophase oxides whose
diffraction lines are very broad and are frequently overlooked owing to their
overlapping with sharper peaks for larger particles of the same oxide phases in the
rust and are believed to be incorrectly referred to as amorphous. XRD measurements have lead to general conclusion that weathering steel forms a protective
coating with ratio a FeOOH/c FeOOH [ 2 [129, 130].
In another study, corrosion rates of Mnsteel and CuMn weathering steel in a
simulated coastal environment were measured by wetdry cyclic test. The rust
layer was observed and analysed by SEM and XRD. The experimental results
29
showed that the corrosion rate of CuMn weathering steel was lower than that of
Mnsteel, due to the formation of a denser rust layer. The rusts on the two steels
consisted of Fe3O4, a FeOOH, b FeOOH, c FeOOH and amorphous phases. The
amount of a FeOOH and b FeOOH in the rust of Mnsteel was larger than that of
CuMn weathering steel. The addition of Cu increased the amount of Fe3O4, while
the addition of Mn decreased the amount of c FeOOH in the rusts [131].
1.7.3 Raman Spectroscopy

Raman spectroscopy is used to study the internal structure of molecules and
provides unique information about molecular patterns, spacing, and bonding. This
is based on Raman Effect, which is the inelastic scattering of photons by molecules
as every compound possesses a typical Raman spectrum. In order to be Raman
active a molecular rotation or vibration must cause some change in any component
of molecular polarisability. This is defined as the induced dipole moment set up in
the molecule by applied electric field. Practically, stokes lines (high k, low t & low
t ) are intense in the spectrum than anti stokes lines (low k, high t & high t ) and the
shift are measured with respect to the reference Rayleigh lines (unshifted with
same k, t & t ). Fine structure effects are not considered in practical situation as the
corresponding effects are of less important. It is essentially an emission spectroscopy. The source is a monochromatic (laser) and the instrumentation is simply
a typical visible range (HeNe) spectrometer. Raman Effect can take place for any
frequency of the incident light which is simply a light scattering phenomenon.
Spherical top molecules are completely Raman inactive whereas asymmetrical top
one is Raman active [132134]. This case is corresponding to rotation of Raman
mode (some phases identified by XRD not by Raman due to inactive Raman). For
vibrational Raman mode, symmetrical vibration always produce intense Raman
lines whereas unsymmetrical ones are normally weak and sometimes unobservable
[135, 136]. Table 1.2 provides the important bands of some common corrosion
products of iron and [R] indicates the published references.
Thibeau et al. [143]. have used Raman and infrared spectroscopy to investigate
the structure of the inner rust layer formed on weathering steels exposed to an
industrial environment for 4.5 and 8 years. The inner rust layer on weathering steel
was composed primarily of d FeOOH with 1020 %, c FeOOH and some
a FeOOH irrespective of the exposure period.
Dunnwald and Otto [137] found phase transformation of iron corrosion product
to Fe(OH)3 in the atmosphere containing SO2 with humidity by Raman spectroscopy. Subsequently, Fe(OH)3 gets transformed to crystalline FeOOH with
amorphous FeOOH. It has been shown that the amorphous rust is the primary
product of atmospheric corrosion, which later transforms to crystalline forms in
the absence of copper. Yamashita et al. [144] studied the long-term growth of the
protective rust layer formed on weathering steel under atmospheric corrosion in an
industrial region involving an exposure for 26 years. The outer layer of rust was
226,245,292,411,497,612
381,486,670,718
532,667
Gives strongest peaks
Characterised by 4 peaks
Characterised by 2 peaks
a Fe2O3
Haematite
c Fe2O3
Maghemite
Fe3O4
Magnetite
Bold Strongest Peak in Spectrum, Underlined Next Strongest Peak in Spectrum
297,392,666
219,252,311,349,379,528,648
314,380,549,722
Characterised
by 4 peaks
Charactersed
by 7 peaks
Gives relatively weak peaks
b FeOOH
Akaganeite
c FeOOH
Lepidocrocite
d FeOOH
205,247,300,386,418,481,549
Gives relatively strong peaks
Wave nos (cm-1)
a FeOOH
Goethite
Table 1.2 Important Bands of Corrosion Products

Oxides
Description
245, 300, 390, 420, 480, 550, 685

248, 303, 397, 485, 554, 680, 1002, 1120
245, 300, 390, 485, 550, 675
298, 397, 414, 474, 550
310, 386, 497, 538, 723
310, 385, 415, 480, 535, 615, 675, 725
255, 380, 528, 654, 1054, 1307
252, 380, 660
400, 655
220, 295, 385, 495, 670
27, 245, 293, 298, 414, 501, 612
225, 245, 295, 415, 500, 615, 1320
265, 300, 345, 395, 515, 645, 670, 715, 1440
350, 505, 660, 710, 1425
616, 663
298, 319, 418, 550, 676, 1322
Published Wave nos (cm-1)
[138]
[137]
[139]
[140]
[141]
[139]
[140]
[142]
[141]
[143]
[143]
[138]
[138]
[141]
[143]
[137]
[R]
30
31
composed of c FeOOH while the inner layer was comprised mainly of densely
packed nanoparticles of a FeOOH. Further, a FeOOH was found enriched with Cr
and reported to be the stable and protective uniform rust layer. It was proposed that
the c FeOOH, as an initial rust layer of weathering steel, formed after a few year of
exposure, is transformed eventually into the final stable rust layer consisting of
nano-size a FeOOH after decades with amorphous ferric oxyhydroxide as an
intermediate transition product which is formed after several years of exposure in
atmosphere. The mean diameter of the rust particles was found to be approximately 0.5 lm in the outer loose layer aggregate whereas the inner layer was
composed of densely packed fine particles within the larger secondary particles. In
contrast, the corrosion product formed on mild steel contained number of voids
and microcracks.
Microscopic observation of weathering steel exposed outdoors during stable
protective rust coating development, reveal two phases in layers parallel to the
steel surface. The layer adjacent to the steel is grey and compact and the external is
reddish and porous. The thickness of the inner phase increases up to outdoor
exposure periods longer than 5 years, when it becomes the only component of the
patina. It is responsible for the electrochemical potential increase and low corrosion rate of steel, restricting oxygen and water access as a barrier to elements
controlling further corrosion. In carbon steels corrosion products form also as two
optically different phases, but they are mixed up. They experience lower increase
of electrochemical potential during natural or simulated outdoor exposure. Their
corrosion rate remains up to an order of magnitude above those determined for
weathering steels in the respective atmospheres [145, 146].
1.8 Rust Simulation

Pourbaix diagram [3] maps out possible stable equilibrium phases of an aqueous
electrochemical system and indicates that pure iron is passive at pH values from 9
to 12.5 to form iron hydroxide. Considering the interplay of atmospheric factors
this diagram was used as guide to the steel dissolution process to form passivity on
WS in laboratory. The passive films formed on pure iron are not so stable and
consequently the passivation state of iron is not maintained for prolonged time
periods [147, 148]. The rust layers of steels play a role as a barrier against corrosion, and their growth rate is decreased to a rate similar to that of the passive
films, when suitable elements are added to the steel [149, 150].
Rust on weathering steel changes over time and the final protective rust has a fine
a FeOOH and is dispersed as amorphous rust. It is reported that the addition of seed
rust, which is a stage in rust formation, results in the preferential formation of
homogeneous rust. This phenomenon suggested the possibility that protective rust
will also form preferentially in atmospheric environments when protective rust is
present [151, 152].
32
References
1. Graham, J.: Passivity of iron. Electrochem. Soc. Proc. 13, 112115 (2002)
2. Evans, U.R.: Mechanism of atmospheric rusting. Corros. Sci. 12, 227246 (1972)
3. Pourbaix, M.: Atlas of Electrochemical Equilibria in Aqueous Solutions. Pergamon, New
York (1966)
4. Devenpori, A.J., Oblonsky, L.J., Ryan, M.P., Toney, M.F.: Electrochem. Soc. 174, 2162
(2000)
5. Chen, C.T., Cahan, B.D.: The nature of passive film on iron. J. Electrochem, Soc. 129,
1726 (1982)
6. Matsubara, E., Suzuki, S., Waseda, Y.: Corrosion Mechanism of Iron from an X Ray
Structural Viewpoint, Book on Characterisation of Corrosion Products on Steel Surfaces
7. Shastry, C.R., Friel, J.J., Townsend, H.E.: 16 yr atmospheric corrosion performance of
weathering steels in marine, rural, and industrial environments: degradation of metals in
atmosphere, STP 965, pp. 515. S.W. Dean & T. S. Lee, ASTM, Philadelphia (1988)
8. Townsend, H.E.: Estimating the atmospheric corrosion resistance of weathering steels in
outdoor atmospheric corrosion, STP 1421, pp. 292300. ASTM, West Conshohocken
(2002)
9. Pourbaix, M.: The linear bilogarithmic law for atmospheric corrosion. Wiley, New York
(1982)
10. Misawa, T., Asami, K., Hashimoto, K., Shimodaira, S.: Corros. Sci. 14, 279289 (1974)
11. Misawa, T., Yamashita, M., Miyuki, H., Nagano, H.: Protective rust layer formed on
weathering steel by atmospheric corrosion for a quarter of a century, J. Iron Steel Inst. Jpn.
79(1):6975 (1993)
12. Feliu, S., Morcillo, M., Feliu, Jr.S.: The prediction of atmospheric corrosion from
meteorological and pollution parameters I: annual corrosion. Corro. Sci. 34:403
13. Felik, S., Morcillo, M., Felik, Jr.S.: The prediction of atmospheric corrosion from
meteorological and pollution parameters ii: long term forecast, Corro. Sci. 34:415 (1993)
14. Natesan, M., Venkatachari, G., Palaniswamy, N.: Corrosivity and durability maps of india.
Corro. Prev. Control 52(2), 43 (2005)
15. Natesan, M., Palaniswamy, N., Rengaswamy, N.S.: Atmospheric corrosivity survey of
India. Mater. Perform. 45:52 (2006)
16. Evans, U.R.: The Corrosion and Oxidation of Metals, vol. 11. VCH Publishers, New York
(1992)
17. Veleva, L.: Phase Transformation of Iron Hydroxide in the Corrosion Products Formed in
Humid Tropical Climate, 58 NACE Conferences, Corrosion (2003)
18. Ohtsuka, Toshiaki: Passivation Oxide Films and Rust Layers on Iron Characterization of
Corrosion Products on Steel Surfaces. Springer, Berlin Heidelberg (2006)
19. Oh, S.J., Cook, D.C., Townsend, H.E.: Atmospheric corrosion of different steels in marine
rural & industrial environments. Corro. Sci. 41, 16871702 (1999)
20. Balasubramaniam, R., Rameshkumar, A.V., Dillmann, P.: Curr. Sci. 85, 15461555 (2003)
21. Kihira, H., Ito, S., Murata, T.: The behavior of phosphorous during passivation of
weathering steel by protective patina formation, Corr. Sci. 31:383388 (1990)
22. Morcillo, M., Chiro, B., Otero, E.: Mater. Perform. 38, 72 (1999)
23. Mattson, E.: The atmospheric corrosion properties of some common structural metals: a
comparative study. Mater. Perform. 21, 919 (1982)
24. Zhang, Q.C., Wu, J.S., Wang, J.J., Zheng, W.L., Chen, J.G., Li, A.B.: Corrosion Behavior of
Weathering Steel in Marine Atmosphere 77, 603608 (2003)
25. Misawa, T., Hashimoto, K., Shimodaira, S.: The mechanism of formation of iron oxide and
oxyhydroxides in aqueous solutions at room temperature. Corros. Sci. 14, 131141 (1974)
26. Yamashita, M., Maeda, A., Uchida, H., Kamimura, T., Miyuki, H.: J. Jpn. Inst. Metal 65,
967971 (2001)
27. Suzuki, I., Hisamatsu, Y., Masuko, N.: J. Electrochem. Society 127, 2210 (1980)
References
33
28. Oh, S.J., Cook, D.C., Townsend, H.E.: Characterisation of iron oxides commonly formed as
corrosion products on steel. Hyperfine Interact 112, 5965 (1998)
29. Stratmann, M., Bohnenkamp, K., Ramchandran, T.: Corros. Sci. 27, 905 (1987)
30. Kishikawa, H., Miyuki, H., Hara, S., Kamiya, M., Yamashita, M.: Sumitomo Search 20
(1998)
31. Oh, S.J., Cook, D.C., KWon, S.J., Townsend, H.E.: Hyperfine Interact 4:4954 (1999)
32. Yamashita, M., Nagano, H., Misawa, T., Townsend, H.E.: Structure of protective rust layers
formed on weathering steels in industrial atmospheres of Japan and North America. ISIJ Int.
38(3), 285290 (1998)
33. Wei,F.I.: Atmospheric corrosion of carbon steels and weathering steels in Taiwan. British
Corros. J. 26(3):209214 (1991)
34. Asami, K., Kikuchi, M.: Corros. Sci. 45, 26712688 (2003)
35. Townsend, H.E.: The Effects of alloying elements on the corrosion resistance of steel in
industrial environments. In: Proceeding of 14th. International Corrosion Congress,
Corrosion Institute of South Africa (1999)
36. Yamamoto, M., Katayama, H., Kodama, T.: Current dev. Mater. Process 12, 422 (1999)
37. Asami, K. Kikuchi, M.: In-depth distribution of rusts on a plain carbon steel and weathering
steels exposed to coastal industrial atmosphere for 17 years. Corros. Sci. 45(11):26712688
(2003)
38. Kihira, H.: Electrochemical Phenomena at Interface, p.429 Marcel Dekker Inc. New York,
(1998)
39. Sakashita, M., Sato, N.: Corros. Eng. 28, 450 (1979)
40. Keiser, J.T., Brown, C.W., Heidersbach, R.H.: Raman bands of different iron oxides
compared to the bands seen in spectra of the inner rust on weathering steel. Corros. Sci.
23(2), 251 (1983)
41. Yamashita, M., Asami, K., Ishikawa, T., Ohisuka, T., Tamura, H., Misawa, T.: Zairyo to
Kankyo 50, 521 (2001)
42. Fyfe, D. Shanahan, C.E., Shreir, L.L.: Proceeding 4th International Congress On Metallic
Corrosion. NACE 399 (1972)
43. Burstein, G.T., Marshall, P.I.: Growth of passivating films on steel in alkaline solution.
Corros. Sci. 23, 125137 (1983)
44. Balasubramaniam, R.: The protective rust on low alloy steel. Corros. Sci. 14: 279 (1971)
45. Inouye, K., Ishii, S., Kaneko, K.: J Inorg. Gen. Chem. 391:86 (1972)
46. Townsend, H.E., Simpson, T.C., Johnson, G.: Structure of rust on weathering steel in rural
and industrial environments. In: Proceding 20th International Corrosion Congress, NACE,
Houston, 2:624641 (1993)
47. Ishikawa, T., Nakazaki, H., Yasukawa, A.: Corros. Sci. 41, p1665p1680 (1999)
48. Uhlig, H.H., Revie, R.W.: Corrosion and Corrosion Control, 3rd edn, p. 96. Wiley,
NewYork (1985)
49. Cleary, H.J., Greene, N.D.: Corros. Sci. 3 (1969)
50. Oh, S.J., Cook, D.C., Townsend, H.E.: Hyperfine Interact 3, 8487 (1998)
51. Yamashita, M., Miyuki, H., Matsuda, Y., Nagano, H., Misawa, T.: Corros. Sci. 36, 283299
(1994)
52. Cook, D.C., Vanorden, A.C., Reyes, J., Oh, S.J., Balasubramanian, R., Carpio, J.J.,
Townsend, H.E.: Marine Corrosion In Tropical Environments, ASTM STP 1399, (2000)
53. Larrabee, C.P., Coburn, S.K.: Proceeding 1st International Congress on Metallic Corrosion,
London 276 (1962)
54. Townsend, H.E., Simpson, T.C., Johnson, G.L.: Corrosion 50, 546554 (1994)
55. Horton, J.B., Goldberg, M.M., Watterson, K.F: Proceeding 4th. Internatioanl Congress on
Metallic Corrosion, NACE 385391 (1972)
56. Fontana, M.G.: Corrosion Engineering, 3rd.Edn, Mcgraw Hill Book, New York (1986)
57. Vernon, W.H.J.: Trans. Farady Soc. 31, 1668 (1935)
58. Stratmann, M., Streckel, H.: Corros. Sci. 30, 697714 (1990)
34
59. Rendn, J.L., Valencia, A.: Kinetics of structural rust transformation in environments
containing chloride and SO2. Rev. Met. 914 (2003)
60. Bolvar, F., Meneses, C.A.B., Minotas, J., Grenche, J.M.: Variable temperature mssbauer
study of some rust converters. Hyperfine Interact 148, 219225 (2003)
61. Townsend, H.E.: Atmospheric corrosion performance of quenched and tempered
weathering steel. Corrosion 56, 883886 (2000)
62. Itagaki, M., Tajima, S., Nozue, R., Watanabe, K., Katayama, H., Noda, K.: 204th Meeting,
The Electrochemical Society (2003)
63. Tait, W.S.: An Introduction to Electrochemical Corrosion Testing for Practicing Engineers
and Scientists, Wisconsin, (1994) 79115
64. Atmospheric Corrosion Resistant Steels: Product Catalogue. Nippon Steel Corporation,
Japan (2005)
65. Veleva, L., Perez, G., Costa, M.A.: Statistical analysis of the temperature humidity complex
and time of wetness of a tropical climate in the Yucatan peninsula in Maxico. Atmosphere
Environ. 31, 773776 (1997)
66. Evans, U.R., Taylor, C.A.J.: Corros. Sci. 12, 227246 (1972)
67. Pourbaix, M.: Corros. Sci. 14, 25 (1974)
68. Singh, D.D.N., Yadav, S., Saha, J.K.: Role of climatic conditions on corrosion
characteristics of structural steels, Corros. Sci. 50(1):93110 (2007)
69. Stratmann, M.: The Atmospheric corrosion of iron & steel: historic review and future
perspectives. Corros. Rev. 2024 (2001)
70. Kimura, M. Kihira, H., Otha, N.: Use advance weathering steel to Avoid Cr as alloying
element ti dosed to get chloride corrosion resistance. Corros. Sci. 47 (2005)
71. Leygraf, C., Graedel, T.: Atmospheric corrosion. J. Electrochem. Soc. 1497, p1010 (2000)
72. Environmental information system in India, http://envis.tropmet.res.in dated Jan 2007
73. Rao, K.N.P., Lahiri, A.K.: Corrosion Map of India Corrosion Advisory Bureau. Metal
Research, India (1970)
74. Sato, N.: An overview on passivity of metals. Corros. Sci. 31, p1 (1990)
75. Nagayama, G.M., Cohen, M.: J. Electrochem. Soc. 110: 164 (1963)
76. Mizoguchi, T., Ishii, Y., Okada, T., Kimura, M., Kihira, H.: Corros. Sci. 47:24772491
(2005)
77. Usami, A., Kihira, H., Kusunoki, T.: Nippon Steel Technical Report 87, 17 (2003)
78. Okada, H., Hosoi, Y., Yukawa, K., Naito, H.: Proceeding 4th International Congress on
Materials, Corrosion, NACE, 392398 (1972)
79. Antunes, R.A., Costa, I., Faria, D.L.A.: Characterization of corrosion products formed on
steels in the first months of atmospheric exposure, Mater. Res. 6(3):287293 (2003)
80. David, L. Peterson, P. Rodgers, B.: Evaluation of organic coating with electrochemical
impedance spectroscopy, part 2. Application of EIS to coatings. J. Coat. Technol. 8893
(2004)
81. Kittelberger, W.W., Elm, A.C.: Water immersion testing of metal protective paints. role of
osmosis in water absorption and blistering. Ind Eng Chem 38, 695699 (1946)
82. Bacon, C.R., Smith, J.J., Rugg, F.M.: Electrolytic resistance in evaluating protective merit
of coatings on metals. Ind Eng Chem 40, 161167 (1948)
83. Thomas, N.L.: The barrier properties of paint coatings. Prog. Org. Coat. 19, 101121 (1991)
84. Ravie, R.W., Baker, B.G., Bockris, O.M.: Modern aspects of electrochemistry.
Electrochem. Soc. 122 1460 (1975)
85. Greenfield, D., Scantlebury, D.: The protective action of organic coatings on steel a review.
Corros. Sci. Eng. 3, 5 (2000)
86. Mayne, J.E.O.: Paints for the protection of steel: a review of research into their modes of
action. Corrosion 5, 160111 (1970)
87. Scantlebury, J.D.: Organic coatings systems and their future in corrosion protection.
Proceedings of EUROCORR, Budapest, (1982)
References
35
88. Rosales, B.M., Sarli, A.R.D., Rincn, O.D., Rincn, A., Elsner, C.I., Marchisio, B.: An
evaluation of coil coating formulations in marine environments. Prog. Org. Coat. 50,
105114 (2003)
89. Leidheiser, Jr.H. Simmons, G.W., Kellerman, E.: Electrochem. Acta, 45:257 (1973)
90. Koehler, E.L.: under film corrosion currents as the cause of failure of protective organic
coatings. Org. Coat. 8796 (1981)
91. Schwenk, W.: Adhesion Loss for organic coatings causes and consequences for corrosion
protection by organic coatings. Corrosion, 103110 (1981)
92. Leidheiser, Jr.H.: Alkali metal ions as aggressive agents to polymeric corrosion protective
coatings. Corrosion, 43:296297 (1987)
93. Smith, G., Dickie, R.A.: Adhesion failure mechanisms of primers. Ind. Eng. Chem. 17,
4244 (1978)
94. Koehler, E.L.: The influence of contaminants on failure of protective organic coatings.
Corrosion NACE 33, 209217 (1977)
95. Leidheiser, Jr.H., Wang, W., Igetoft, L.: the mechanism for cathodic delamination of
organic coatings from a metal surface. Prog. Org. Coat. 11:1941 (1983)
96. Leidheiser, H., Funke, W.: Water disbandment and wet adhesion of organic coatings on
metals a review and interpretation. Surf. Coat. Int. 70, 121132 (1987)
97. Gowers, K.R., Scantlebury, J.D.: An electrochemical investigation of the effect of the
adhesion of a lacquer coating on the under film corrosion. Surf. Coat. Int. 4, 114121 (1988)
98. Gosselin, C.A.: Effect of surface preparation on the durability of structural adhesive bonds,
polymeric materials for corrosion control. In: ACS Symposium 322:180193 (1986)
99. Funke, W.: The role of adhesion in corrosion protection by organic coatings. JOCCA 68,
229232 (1985)
100. Princeton Applied Research Application note AC-1. Basics of electrochemical impedance
spectroscopy, USA, (2009)
101. Chico, B., Galvn, J.C., Fuente, D.D.L., Morcillo, M.: Electrochemical impedance
spectroscopy study of the effect of curing time on the early barrier properties of saline
systems applied on steel substrates. Prog. Org. Coat. 60(1):4553 (2007)
102. ASTM G1 Standard practice for preparing, cleaning & evaluating corrosion test specimens
(1990)
103. Scully, J.R.: Electrochemical impedance of organic coated steel correlation of impedance
parameters with long term coating deterioration. Electrochem. Soc. 13(6), 979 (1989)
104. Jones, D.A.: Principles and Prevention of Corrosion. Prentice Hall, USA (1996)
105. Scully, J.R.: Polarization resistance method for determination of instantaneous corrosion
rates. Corrosion 56, p199 (2000)
106. Stern, M. Geary, A.L.: Electrochemical polarization, J. Electrochem. Soc. 104(1):5663
(1957)
107. Harvey, F., Schweinsberg, J., Paul, D.D.: Evaluation of corrosion rate from polarisation
curves not exhibiting a tafel region. Corros. Sci. 47, 30343052 (2005)
108. Pourbaix, M.: Atlas of electrochemical equilibria in aqueous solutions, NACE, (1974)
109. Tait, W.S.: An Introduction to Electrochemical Corrosion Testing for Practicing Engineers
and Scientists. Pair O Docs Professionals E-Book, Madison (1994)
110. Corrosion Basics: An introduction, national association of corrosion engineers, (1984)
111. Scully, J.R.: Electrochemical impedance spectroscopy for evaluation of Organic coating
deterioration & under film corrosion. A state of the art technical review, D. W. Taylor
Research Center, Report SME-86/006, (1986)
112. Kendig, M. Scully, J.R.: Basic aspects of the application of electrochemical impedance for
the life prediction of organic coatings on metals. Corrosion 31(1):22 (1990)
113. Mansfeld, F.: Evaluation of corrosion behaviour of coated metals with ac impedance
measurements. Corros. Sci. 2(3), 317 (1983)
114. Equivalent Circuit Modeling using the Gamry EIS 300 Electrochemical Impedance
Spectroscopy Software: Gamry Application Note, USA, (2001)
36
115. Ke, W.: Chinese Corrosion Survey Report, 1st Edn. Chemical Industry Press, Beijing, ISBN
7-5025-4792-4/TQx1816, (2003) p 13
116. Macdonald, J.R.: J. Electrochemical Society 223, 25 (1987)
117. Scibner Assciates Inc, Southern Pines, North Carolina (1998)
118. Kihira, H., Ito, S., Murata, T.: Corros. Sci. 31, 383388 (1990)
119. Nishimura, T., Katayama, H., Noda, K., Kodama, T.: Corros. Sci. 56, 935 (2000)
120. Itagaki, M., Nozue, R., Watanabe, K., Katayama, H., Noda, K.: Corros. Sci. 46, 13011310
(2004)
121. Konishi, H., Yamashita, M., Uchida, H., Mizuki, J.: Characterization of Rust Layer formed
on Fe Fe-Ni and Fe-Cr alloys Exposed to Cl- Rich Environment. Mater. Trans. 46(2),
329336 (2005)
122. Feliu, S., Galvan, J.C., Bastidas, J.M., Simancas, J., Morcillo, M., Almeida, E.M.: Corros.
Sci. 35, 13511358 (1993)
123. Qu, A.Q., Li, L., Bai, W., Yan, C.: Initial atmospheric corrosion of weathering steel in the
presence of NaCl and SO2, 16th Int Corrosion Congress, Beijing, (2005)
124. Cox, A., Lyon, S.B.: An Electrochemical study of the atmospheric corrosion of mild steel,
part iii: the effect of sulphur dioxide. Corros. Sci. 36(7):11931199 (1994)
125. Cornell, R.M., Schwertmann, U.: The iron oxides: structure, properties, reactions,
occurrences and uses. Wiley, Weinheim (2003)
126. Cullity, B.D., Stock, S.R.: Elements of X-ray diffraction, 3rd Edition Addison Wesley,
Weinheim (2001)
127. Azaroff, L.V.: Elements of X-ray Crystallography, Mcgraw Hill, New York (1986)
128. Cook, D.C., Oh, S.J., Balasubramanian, R., Yamashita, M.: Hyperfine Interact. 122, 5970
(1999)
129. Legault, R.A. Preban, A.G.: Kinetics of atmospheric corrosion of low alloy steels in an
industrial environment. Corrosion, 31(4) 117122 (1975)
130. Baker, E. A.: Long term corrosion behaviour of materials in the marine atmosphere,
degradation of metals in the atmosphere, STP 965, ASTM, New Jersey 125144 (1998)
131. Liu, A., Dong, G.C., Han, J.H., Hou, E., Wei, K.: Influence of Cu and Mn on corrosion
behaviour of low alloy steel in a simulated coastal environment. Corros. Sci. Protect.
Technol. 20(4):235238 (2008)
132. Anderson, A.: The Raman Effect. Marcel Dekker, (1973)
133. Herzberg, G.: Molecular Structure and Molecular Spectra. Van Nostrand Reinhold, New
York (2008)
134. Gilson, T.R., Hendra, P.J.: Laser Raman Spectroscopy. Wiley Inter Science, New York
(1970)
135. Loader, J.: Basic Laser Raman Spectroscopy. Heyden & Son, London (1970)
136. Gui, J. Devine, T.M.: Proceeding of 12th International Corrosion Congress, NACE, 2052
(1993)
137. Dunnwald, J., Otto, A.: Corros. Sci. 29, 11671176 (1989)
138. Ohtsuka, T., Kubo, K., Sato, N.: Corrosion 42, 476481 (1986)
139. Ohtsuka, T.: Materials transctions, JIM 37:67 (1996)
140. Brown, C.W., Heidersbach, R.H.: Applied spectroscopy. 32(6):532532 (1978)
141. Boucherit, N., Delicher, P., Joiret, S., Hugot, A.: Mater. Sci. Forum 51, 44 (1989)
142. Thierry, D., Persson, D., Lyegraf, C., Delichere, D., Joiret, S., Pallotta, C., Hugot, A.: J.
Electrochem. Soc. 135(2):350310 (1988)
143. Thibeau, R.J., Brown, C.W., Heidersbach, R.H.: Appl. Spectrosc. 32, 532 (1978)
144. Yamashita, M., Uchid, H.: recent research & development in solving atmospheric corrosion
problems of steel industries in Japan, Hyperfine Inter. 139(4):153166 (2002)
145. Graedel, T.E., Frankonthal, R.P.: Corrosion mechanism for iron and low alloy steel exposed
to the atmosphere. J. Electrochem. Soc. 137(8):23852394 (1990)
146. Ramana, K.V.S., Kaliappan, S., Ramanathan, N., Kavitha, V.: Characterization of rust
phases formed on low carbon steel exposed to natural marine environment of Chennai
harbor, India. Mater. Corros. 5(8), 873 (2007)
References
37
147. Jan, J.A., Muz, A., Justavino, J., Hernndez, C.: Characterization of initial atmospheric
corrosion of conventional weathering steels and a mild steel in a tropical atmosphere
192(13):5159 (2009)
148. Hirofumi, K., Hideaki, M., Syidu, H., Mitsuaki, K., Masato, M.: Development of surface
treatment technique promoting protective surface formations. Sumitomo Search 60, 2026
(1998)
149. Mattson, E.: The atmospheric corrosion properties of some common structural metals: a
comparative study. Mater. Perform. 21, 919 (1982)
150. Bolivar, F., Morales, A.L., Aramburo, C.: Simulation of a long term atmospheric corrosion
process on plain and weathering steels. Rev. Metal 265269 (2003)
151. Uhlig, H.H.: Passivity in metals and alloys. Corros. Sci. 19:777 (1979)
152. Komori, T., Kyono, K., Kato, C.: New surface treatment technology for promoting
protective rust formation on weathering steel. In:132nd Symposium Corrosion Prevention.
6572 (2001)
Chapter 2
Practical Approach
2.1 Materials Used

To study atmospheric corrosion, different types of structural steels were used both
in uncoated and coated conditions. Some steels provide excellent corrosion protection against certain atmospheric conditions in uncoated state, but most steels are
coated with paints to serve the same purpose. High performance organic coatings
were used for barrier protection.
2.1.1 Structural Steels

Two types of hot rolled structural steel sheets (5 mm thick) sourced from leading
manufacturers were taken for the experiments. These were plain carbon steel (MS)
and weathering steel (WS). MS conforming to standard [1] is the most common
structural steel and WS conforming to standard [2] exhibits increased atmospheric
corrosion resistance compared to MS due to its chemical composition. These
sheets were sheared into standard pieces (150 9 100 mm2) to prepare test panels
as per standard [3]. Both chemical analysis of the steels using optical emission
spectrometer (ARL3460, USA) and mechanical properties using Universal testing
machine (Instron: 1195, UK) were carried out as indicated in Table 2.1.
Metallographic cross sections were prepared from the test specimens which
were ground and polished with successive finer grits. The final polishing was done
with 0.25 mm diamond grit and digital microscope (Olympus, Japan) was used to
take micrographs. Microstructure of MS shows coarse equiaxed ferrite and pearlite
with an average grain size 1516 lm, while WS is having fine equiaxed ferrite and
pearlite with an average grain size 1011 lm. The microstructures of the steels are
shown in Figs. 2.1a, b respectively.

39
40
2 Practical Approach
Table 2.1 Chemical composition and properties of steels

Steels Composition in weight %
MS
WS
Fig. 2.1
Properties
Mn
Cu
Cr
Ni
Si
YS (mpa)
Elongation ( %)
0.17
0.09
0.76
0.38
0.01
0.35
0.03
0.45
0.01
0.27
0.01
0.41
0.01
0.01
0.01
0.11
295
418
27
34
Microstructures at X200 of (a) MS and (b) WS

Two types of paint systems were used for experiments. First system used was zinc
phosphate primer (50 % Zn as compound) with micaceous iron oxide (MIO) as
intermediate coat followed by polyurethane (PU) as top coat denoted as ZP. Other
system used was zinc rich primer (80 % Zn as dust) with MIO as intermediate coat
and PU as top coat denoted as ZR. This system is costlier with respect to ZP and
considered for laboratory evaluations purpose only. ZP is designed with respect to
places where test panels are deployed for exposure based on manufacturers catalogues and standards [4]. These systems possess good resistance from water
permeability, weathering, abrasion and good adhesion to maintain a proper barrier
to the environment. The drying mechanism of the systems is the reaction between
epoxy resin and polyamide. The details of each paint system used comprising
primer, intermediate and top coats with the respective layer of film thickness are
shown in Table 2.2.
2.1 Materials Used
41
Table 2.2 Paint systems with coating thickness

Systems Coat Name
Description
ZP
ZR
Primer coat
Intermediate coat
Top coat
Primer coat
Intermediate coat
Top coat
Epoxy zinc phosphate

High build epoxy MIO
Aliphatic acrylic PU
Epoxy Zn rich
Epoxy MIO
Aliphatic acrylic PU
Coats
1
1
1
1
1
1
9
9
9
9
9
9
DFT
(lm)
75 250
125
50
50 250
150
50
Application
Laboratory and
atmospheric
Laboratory
2.1.3 Test Electrolytes

Aqueous solutions were used to simulate specific conditions to carry out electrochemical tests on both field exposed and fresh panels. The test electrolytes are SAEJ
2334 solution (0.25 % NaHCO3 ? 0.5 % NaCl ? 0.1 % CaCl2, pH * 9.1) to
simulate the atmospheric conditions in the laboratory at 25 3 C, neutral salt
solution (3.5 % NaCl, pH * 6.7) and weakly alkaline solution (0.1 M Na2SO4 ?
0.1 N NaCl, pH * 8.5) to get chloride and sulphate ions in the environment for
carrying out tests. Electrolyte used in SAEJ 2334 test was used to determine corrosion performance for coating system as this solution shows a high degree of
correlation with field service conditions [5].
2.2 Test Panel Preparations

Coated and uncoated steel panels were prepared for field exposure as well as
laboratory tests.
2.2.1 Blasting of Panel

Panels made from as received hot rolled steels were compared and matched with C
grade type of rust as per standard [6] and shot blasted with 1.2 mm steel grits
having seven bars pressure, 600 angle, 10 mm nozzle size, 500 mm nozzle distance. Panel surface were blasted to near white metal as per Sa 2.5 and checked to
ensure free from oil, grease, dirt, dust, mill scale, rust, oxides, etc. The surface
profile of blasted panels was found to be within 5070 lm by digital surface
profile gauge (Elcometer 224) and compared with the standard [7]. Figures 2.2 a,b
42
Fig. 2.2
Macrostructure before blasting (a) MS and (b) WS
showed MS and WS as received panels, whereas Fig. 2.3 showed panel after
blasting.
2.2.2 Bare Panel

Blasted steel panels were washed with soda ash and running water to remove oil,
grease and dirt. After drying, the panels were pickled in 5 % inhibited HCl to
remove the surface hard scales and washed with running water repeatedly to clean
the surface properly. Dried panels were belt polished on 60 grit papers to remove
light brown spot on the surface and to remove the scratches caused in earlier
operations. Panels were then polished on 120 grit belts after turning the panel at
90. The panels were then subjected to hand polishing with the help of emery
papers (from grit 60, 80, 120, 220) and the direction of the panels were changed
with change of grit size. The polished panels were washed with magnesia powder
and rinsed with running water. Area and weight of each bare panel were recorded
after keeping overnight in desiccator with proper identification as per ASTM
standard [3].
2.2.3 Coated Panel

Based on the primers, two sets of coated panels ZP and ZR were taken for the
experiments. Spray gun was used to apply the coatings in a standard paint booth
maintaining humidity and temperature. ZP system was coated with epoxy zinc
2.2 Test Panel Preparations

Fig. 2.3
2.5)
43
Blasted panel (Sa
phosphate as primer; high build epoxy MIO as intermediate coat and PU as topcoat
and ZR system was coated with epoxy zinc rich as primer, epoxy MIO as intermediate coat and PU as top coat. Time lag allowed between blasting and primer
coat was four hours. To ensure the proper curing of each coat, time allowed for
epoxy Zn phosphate was 6 and 8 h for epoxy Zn rich. Finally, MEK double rub
test as per standard [8] was followed to ensure no effect on surface and no Zn on
cloth after 50 double rubs.
Wet film comb gauge (TQO, WG II) was used to measure the wet film thickness
(WFT) of paints immediately after each coating application to ensure the theoretical thickness. The dry film thickness (DFT) of coats were checked (Elcometer
415) and in some cases mist coat was applied to achieve the desired thickness. Top
coat PU was applied after 8 h from the intermediate coat. The paint used for
sealing edges of panels was of two components, low volatile contents, high solids,
modified epoxy, designed to give long term protection in single coat application.
2.2.4 Scribed Panel

Tungsten carbide tip tool having 60 cutting edges (Elcometer 1537) was used to
scribe coated panels diagonally leaving 12 mm intact from edges to obtain V cut
through coatings as per standard [3] as shown in Fig. 2.4 so that substrate steel was
visible to simulate the failure observed when the panels were subjected to abrasion/accidental damage.
44
Fig. 2.4
panel
Painted scribed
2.2.5 Reference Plate

Atmospheric corrosion tests were continued for years with number of panels, and
to have traceability, each uncoated panel was marked by hole with respect to
reference plate made as per standard [3] as shown in Fig 2.5. The reference plate
made of Aluminium has identical size of test panels and a hole is made in left hand
corner marked as R, which stands for reference mark. Just below this hole, nine
holes are made vertically engraving numbers 19 and similarly, in the right hand
side, again nine holes are made vertically giving numbers 1090 from top to
bottom. At the top of the panel, five holes are made horizontally to represent
hundreds.
2.3 Adequacy Test

DFT of coating applied on substrate was measured by coating thickness gauge.
Average of eight readings was taken and thickness was found to be in the range of
240260 lm. Routine physical testing like accelerated weathering test [9], pull off
test [10] and cross cut adhesion tests [11] were carried out as per relevant standard
in order to ensure that the prepared coated panels are free from normal defects.
The summary of all the test results is given in Table 2.3.
2.3 Adequacy Test

Fig. 2.5
45
Reference plate
Table 2.3 Adequacy tests on paint systems

Tests
ZP on MS/WS
ZR on MS/WS
Weathering
Pull off adhesion
Cross cut adhesion
1000 h passed
12.5 mpa
0 (no peeling)
1000 h passed
12 mpa
0 (no peeling)
2.3.1 Weathering Test

This was carried out in weathering chamber (QUV: SE, QLAB, USA) to induce
property changes associated with end use conditions like effects of sunlight and
heat as per standard. The panels were exposed to 8 h in UV lamp and 4 h condensation without lamp. These cycles were repeated continuously for 1000 h and
after exposure the coated panels were found free from fading, discoloration, loss of
gloss and other film defects.
2.3.2 Pull Off Adhesion Test

This was performed using loading fixture (Elcometer 108) perpendicular to the
surface of the coatings with an adhesive. The actual force necessary to fracture the
coating system was measured using 20 mm dollies on the panels at three different
spots and adhesion value recorded was between 1212.5 mpa.
46
2.3.3 Cross Cut Adhesion Test

Knife incisions were made for cross cut adhesion test through the coating system
down to the substrate at three different places. Pressure sensitive adhesive tape was
firmly pressed over the inscribed area. Coating removal along the scribed lines was
examined and no peeling or removal of paints was found.
2.4 Exposure Test Set-Up

Field exposure test racks were designed and deployed at three different sites with
proper protection.
2.4.1 Test Racks

Racks were made as per standard [12, 13] using mild steel angles and channels
with height and width (1.50 9 2.10 m2). The racks were coated with epoxy paint
to prevent rusting. Porcelain washers, brass nuts and bolts were used to fix the test
panels at 45 with respect to base. Three types of test panels uncoated, coated and
scribed coated were fixed in the racks and test racks at three exposure sites are
shown in Figs. 2.6a, b and c.
2.4.2 Exposure Sites

Panels were deployed at three sites located in India having different environments
denoted as P1, P2 and P3. Site P1 is situated on the coast of Bay of Bengal. Here,
heavy rainfall is experienced during May to September/October and the winter
spans from November to February. Site P2 is situated on the Coromandel Coast of
Bay of Bengal and situated on the thermal equator and the weather remains
consistently hot and humid. The hottest months are April to July and monsoon
usually stretches from October to mid December. Site P3 is an industrial city
having both hot summer and severe winter when the temperature drops to a great
extent. The exposure sites selected are thus, normal marine environment P1
characterised by proximity to ocean with salt laden air and less polluted marine
environment, P2 characterised by more SO24 with less salt laden air and a polluted
industrial environment, P3 characterised by high SO24 with little salinity selected
as per standard [14]. It is possible to place each of these sites into one of the five
categories of corrosivity [15, 16] as indicated in Table 2.4. The meteorological and
characteristics data of the sites are given in Table 2.5.
2.4 Exposure Test Set-Up
47
Fig. 2.6 a Field exposure

test rack at site P1. b Field
exposure test rack at site P2.
c Field exposure test rack at
site P3
Table 2.4 Site corrosivity and paint system durability

Environment
Sites Category Corrosivity
Normal coastal
P1
Polluted inland to normal coastal P2
Industrial/Polluted inland
P3
C4
C3
C3
Systems Durability (yrs)
High
ZP
Medium high ZP
Medium
ZP
Medium ([5)
Medium ([5)
Medium ([5)
48
Table 2.5 Geographical and meteorological data of exposure sites

Exposure Site Details
Digha (P1)
Chennai (P2)
Jamshedpur (P3)
Longitude (E)
Latitude (N)
Altitude (Ft)
Dist from sea(km)
Av Temp (C) (minmax)
Av RH ( %) (minmax)
Av rainfall(mm)
Salinity (mg) NaCl/100 cm2/day
SO2 (mg)/100 cm2/day
86.18
22.8
1590
200
27 (847)
80 (3398)
905
Trace
22
87.55
21.86
SL
0.1
25 (2128)
63 (4989)
1782
0.83
Trace
80.18
13.06
20
1.5
32 (3035)
66 (30100)
1084
0.42
16.5
Details of meteorological data such as rainfall, temperature, humidity were

collected month wise for the year 2007 and same is given in annexure Table A 2.1.
Salinity and sulphur dioxide available in atmospheres were collected from pollution control departments. The variations of the average temperature, humidity
and rainfall during the year mentioned are shown in the annexure Fig. A 2.1 where
P1 recorded lowest temperature range, Fig. A 2.2 where again P1 maintained
lowest humidity and Fig. A 2.3 where average rainfall was higher for P2.
2.5 Field Test Plan

Three replicates of coated and uncoated panels were exposed at three test sites P1,
P2 and P3 up to 48 months (m) and the performance evaluation were carried out at
different time interval. Various tests were carried out on test panel and the overview of field exposure tests plans are given in Table 2.6.
2.5.1 Rust Appearance

Uncoated panels were exposed at the sites and were drawn after 18 months (m)
and 42 months (m) to check the surface appearance. The details of the same are
given in chapter 4(results).
2.5.2 Corrosion Rate

The exposed uncoated MS and WS panels were drawn from the sites (P1, P2 and
P3) from time to time and corrosion rate was calculated as per mass loss method
using standard [1].
2.5 Field Test Plan
49
Table 2.6 Test plan for field exposed panels

Tests
P1
P2
Rust appearance
Corrosion rate
Potentiodynamic polarisation
Cyclic polarisation
Electrochemical impedance
Scanning electron microscopy/EDX
X-ray diffraction
Raman spectroscopy
P3
P1
P2
P3
MS/
WS
MS/
WS
MS/
WS
ZP+ ZP+ ZP+

MS WS MS
ZP+ ZP+
WS MS
ZP+
WS
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Ha
H
H
H: Test carried out, P1 MS: MS exposed at P1, P1 MS ? ZP: ZP on MS exposed at P1 etc.

a
Scribed panels
Fig. 2.7 Electrochemical

impedance test set up
2.5.3 DC Corrosion Measurement

Gamry Potentiostat (PC/750, USA) with different electrolytes was used. Test panel
acted as working, graphite rod as auxiliary and saturated calomel (SCE: 241 mV)
as reference electrode. Tafel extrapolation method was used to calculate
Icorr. Panels were placed into cell and a luggin capillary was used to provide
electrolytic contact and test set up as shown in Fig. 2.7. Prior to each experimental
run, the working electrodes were mechanically polished with silicon carbide paper
progressively up to 2000 grit, degreased with ethyl alcohol and washed with
running distilled water twice and dried. The solution was prepared using analytical
grade different salts like NaCl, CaCl2, NaHCO3 and Na2SO4 in twice distilled
water. Test plan for DC corrosion measurement along with Fig. no. and Table no.
are given in Table 2.7.
Potentiodynamic polarisation of the panels with 0.5 mV/s scan rate in electrolytic solution was carried out as per standard [17] and voltage was used in the
range of -250 mV1000 mV with respect to SCE.
50
Table 2.7 Test plan for DC corrosion measurements

Tests
Panels
Exposure Sites Electrolytes
Polarisation
diagram
Cyclic
polarisation
MS and WS
MS and WS
MS and WS
ZP ? MS/
ZP ? WS
42 m
18 m
P1
P2
P3
P3
Table no.
Fig. no.
3.5 % NaCl
solution
3.3
2.5 % NaHCO3
? 0.5 %
NaCl ? 0.1 %
CaCl2 for 4080 h
3.4
3.5
3.6
3.7
3.8
For cyclic polarisation (CP) with 1.0 mV/s scan rate, potential scan was started
at Ecorr and continued in the anodic direction up to the transpassive region, where a
large increase in current occurred. At the threshold of current density (1 mA/cm2),
the scan was reversed and continued in the cathodic direction back to or beyond
the OCP followed as per standard [18].
2.5.4 AC Impedance Spectroscopy

Electrochemical impedance spectroscopy (EIS) studies were carried out using
Gamry EIS 300 Potentiostat (PC/750, USA) and AC 10 mV (rms) was employed
with frequency range of 0.01 Hz100 kHz where test Panels acted as working
electrode, graphite rod as auxiliary electrode and standard calomel as reference
electrode (SCE). OCP was measured between SCE and test electrode in electrolyte
before each scan within 50 s initial delay. The impedance spectrum was analysed
with Echem Analyst software (version 5.30) to determine the operating parameters
and the best curve fitting was done based on lowest goodness value (\10-3). The
exhaustive test plan is given in Table 2.8.
EIS was also carried out on the exposed coated MS and WS panels using
Gamry potentiostat and the overall plan of conducting EIS is given in Table 2.9.
2.5.5 Rust Characterisation by SEM and EDX

Morphology of rusts formed on MS and WS at different stages were studied by
scanning electron microscope (SEM) model S3400, Japan. Elemental analysis was
carried out using energy dispersive X-ray spectroscopy (EDX) through NSS 300
model. The summary of SEM and EDX tests is given in Table 2.10.
2.5.6 Rust Characterisation by X-Ray Diffraction

X-ray diffraction (XRD) was carried out using D8, Bruker with Siemens D 500
diffractometer with Cu-Ka filtered radiation to identify phases in the rust deposits.
2.5 Field Test Plan
51
Table 2.8 EIS test plan of uncoated MS and WS panels

Panels Exposure
Sites Electrolytes
(m)
MS
WS
MS
WS
MS
WS
MS
WS
Table
no.
Fig.
no.
3.9
3.10
3.11
3.12
3.13
3.14
3.15
3.16
42
P1
0.1 M Na2SO4 ? 0.1 N NaCl
3.5
42
P2
3.5
42
P3
3.5
48
P3
2.5 % NaHCO3 ? 0.5 % NaCl ? 0.1 % CaCl2

for 72 h
3.6
Table 2.9 EIS test plan of coated MS and WS panels

Panels
Exposure Sites Electrolytes
(m)
ZP
ZP
ZP
ZP
ZP
ZP
ZP
? MS
? WS
? MS
? WS
? MS
? WS
? MS
scribed
ZP ? WS
scribed
ZP ? MS
ZP ? WS
ZP ? MS
ZP ? WS
ZP ? MS
ZP ? WS
ZP ? MS
ZP ? WS
18
P1
18
P2
18
P3
18
P3
2.5 % NaHCO3 ? 0.5 % NaCl

+ 0.1 % CaCl2 for 216 h
2.5 % NaHCO3 ? 0.5 % NaCl
+ 0.1 % CaCl2 for 216 h
2.5 %NaHCO3 ? 0.5 % NaCl
+ 0.1 % CaCl2 for 216 h
3.5 % Nacl for 120 h
Table
no.
Fig.
no.
3.7
3.17
3.18
3.19
3.20
3.21
3.22
3.23
3.7
3.7
3.8
3.24
42
P1
3.5 % NaCl for 936 h
3.9
42
P2
3.9
42
P3
3.9
42
P3
2.5 %NaHCO3 ? 0.5 %NaCl ? 0.1 % CaCl2

(3000 h) after exposed in humid SO2 for 2 m
3.10
3.25
3.26
3.27
3.28
3.29
3.30
3.31
3.32
For oxide identification, JCPDS database system [19] was used and phases were
identified comparing with 2h and intensities of reflection. Table 2.11 indicates the
test plan of XRD for rust analysis at different stages.
2.5.7 Rust Characterization by Raman Spectroscopy

Raman spectroscopy was used to identify the phases present in rust using Nicolet
Almega dispersive Raman Spectroscope by exciting the laser beam of HeNe of
532 nm focused through Olympus microscope at 509. The power of the laser was
52
Table 2.10 Test plan of rust formed on MS and WS by SEM and EDX
Panels
Sites
Exposure (m)
SEM
EDX
EDX line
Fig. no.
Table no.
Profile
MS/WS
MS/WS
MS/WS
MS/WS
MS/WS
MS/WS
MS line scan
WS line scan
WS line scan
MS line scan
WS line scan
P1
P2
P3
P1
P2
P3
P1
18
P2
P3
42
42
3.33
3.34
3.35
3.36
3.37
3.38
42
42
3.11
3.12
3.39
3.40
3.41
3.42
3.43
Spectra
Plot no.
A 3.1
(WS)
A 3.2
A 3.3
A 3.4
Table 2.11 Test plan of rust formed on MS and WS by XRD

Panels
Sites
Exposure(m)
Table no.
Spectra plot no.
MS/WS
MS/WS
WS
MS/WS
A
A
A
A
P1
P1
P2
P3
18
42
42
42
3.13
3.14
3.15
3.16
3.5
3.6a and A 3.6b
3.7
3.8a and A3.8b
kept at lowest possible level (6 mW) in order to avoid the transformation of rusts
due to heating effect of laser. The instrument was calibrated by using pure silicon
at peak of 522.28 cm-1. Raman peaks were identified according to the spectra for
the various iron oxide and hydroxide compounds from the references summarised
in the literature review section and test plan is given in Table 2.12.
2.6 Controlled Laboratory Tests

Accelerating corrosion tests were carried out on uncoated and coated MS and WS
panels. The laboratory test plan is given in Table 2.13.
2.6.1 Humidity Test

Humidity test was carried out as per standard [20] using SD Instruments and
Equip, India as shown in Fig. 2.8 to determine corrosion rate for uncoated panels.
The humidity level of 85 % was maintained in the chamber at 35 C while carrying out tests.
53
Table 2.12 Test plan of rust formed on MS and WS by raman spectroscopy

Panels
Exposure(m)
Sites
Table no.
Spectra plot no.
MS/WS
18
MS/WS
42
P1
P2
P3
P1
P2
P3
3.17
3.18
Table 2.13 Plan for accelerated laboratory tests of panels

Tests
MS
WS
Salt spray
Humidity
Humid SO2 exposure
Open circuit potential
Cyclic polarisation
Electrochemical impedance spectroscopy
Scanning electron microscopy/EDX
Raman spectroscopy
H: Tests carried out, H: Humidity,
A
A
A
A
A
A
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
3.9a and A 3.9b

3.10a and A 3.10b
3.11a and A 3.11b
3.12a and A 3.12b
3.13a and A 3.13b
3.14a and A 3.14b
ZP+
MS
ZR+
MS
ZP+
WS
ZR+
WS
H
H
H
H
H
H
H
H
H
H
After SO2 exposure
Fig. 2.8 Humidity test

chamber
2.6.2 Salt Spray Test

Salt spray test was carried out as per standard [21] using CMEIPL, India as shown
in Fig. 2.9 to find out corrosion rate of uncoated panel and the degrees of protection afforded by coated panels. Panels were exposed to 5 % NaCl solution
sprayed with compressed air at 68 bars and the fog generated inside at 35 C was
54
Fig. 2.9 Salt spray test on

uncoated panel
Table 2.14 Test plan for DC corrosion measurement

Tests
Panels
Electrolytes
Open circuit
potential
Potentiodynamic
polarisation
Cyclic
polarisation
AR MS/
AR
WS
AR MS/
AR
WS
ARMS/
AR
WS
MS/WS
Table
no.
Fig.
no.
3.5 % NaCl solution for 4.1 h
3.46
3.5 % NaCl solution
3.22
3.47
3.48
2.5 % NaHCO3 ? 0.5 % NaCl ? 0.1 % CaCl2,

exposed in humid SO2 for 9 m
Table 2.15 EIS test plan of uncoated MS and WS panels

Panel
Electrolyte
3.23
3.49
Table no. Fig. no.
As received MS 0.1 M Na2SO4 ? 0.1 N NaCl

3.24
As received WS
MS
2.5 % NaHCO3 ? 0.5 % NaCl ? 0.1 % CaCl2 for 48 h 3.25
WS
3.50
3.51
3.52
condensed on the surface of panels kept at 45. The spraying of brine solution was
continued for 8 h and subsequently discontinued for 16 h.

Corrosion test was carried out on as received panels using Gamry potentiostat (PC/
750) and details test plans are given in Table 2.14.
55
Table 2.16 EIS test plan of coated panels

Panel
Potentiostat
Electrolyte
ZP ? MS
ZP ? MS
ZR ? MS
ZR ? MS
ZP ? WS
ZP ? WS
CH instrument
Gamry
3.5 % Nacl for 24 h

3.5 % Nacl for 24 h
2.5 % NaHCO3 ? 0.5 % NaCl
+ 0.1 % CaCl2 for 1500 h
Table no.
Fig. no.
3.26
3.53
3.54
3.55
3.56
3.57
3.58
3.26
3.27
Table 2.17 Test plan of rust formed on MS and WS by SEM and EDX
Panel
SEM
EDX
Spectra plot no.
AR MS
AR WS
AR MS line scan
AR WS line scan
Upper layer of MS
Upper layer of WS
Inner layer of MS
Inner layer of WS
3.59
3.28
A 3.15a and A 3.15b
3.60
3.61
3.62
3.29
A 3.16a and A 3.16b
3.63
3.29
A 3.17a and A 3.17b
Table 2.18 Test plan of rust formed on MS and WS by raman spectroscopy

Panel
Table no.
Spectra plot no.
AR MS
AR WS
MS after humidity test
WS after humidity test
3.30
A 3.18a and A 3.18b
3.31
MS after humidity test

WS after humidity test
3.32
A
A
A
A
A
A
3.19a
3.20a
3.21a
3.22a
3.23a
3.24a
and
and
and
and
and
and
A
A
A
A
A
A
3.19b
3.20b
3.21b
3.22b
3.23b
3.24b
2.6.4 AC Impedance Spectroscopy

Gamry potentiostat was used on both uncoated and coated MS and WS panels in
different electrolytes. The details of test plan of EIS carried out on as received
panels are given in Table 2.15. Panels were also exposed in SO2 environment for
9 m at 46 ppm generated from 1 % Na2S2O3 (850 ml) with 0.1 N H2SO4
(150 ml) under saturated saline and humid conditions at 30 C [22].
EIS was carried out on unexposed coated panels using potentiostats viz. Gamry
and CH Instrument, USA (model: 600C, Reference electrode: Ag/AgCl2(199 mV),
Auxiliary electrode :Platinum). EIS spectra were acquired through suitable model
fit and AC 10 mV (rms) with varying frequencies 1100 kHz were applied. The
test plan of EIS on coated panels is given in Table 2.16.
56

Rust developed on as received panels at P3 as well as after SO2 exposure were
subjected to SEM and EDX. The test plan is given in Table 2.17.
2.6.6 Rust Characterisation by Raman Spectroscopy

Rust formed on as received panels, after humidity and salt spray tests were analysed by Raman spectroscopy. The detail test plan is given in Table 2.18.
References
1. IS: 2062:2006: Hot Rolled Low, Medium and High Tensile Structural Steel
2. IS: 11587:2001: Structural Weather Resistant Steels
3. ASTM G1: 1990: Standard Practice for Preparing, Cleaning and Evaluating Corrosion Test
Specimens
4. ISO 12944-5:1998: Corrosion Protection of Steel Structures by Protection Paint Systems
5. SAE J 2334: Laboratory Test Procedure to Determine Corrosion Performance of Coating
System
6. ISO 8501-1:1988: Preparation of steel substrates before application of paints and related
productsVisual assessment of surface cleanlinessRust grades and preparation grades of
uncoated steel substrates and of steel substrates after overall removal of previous coatings
7. ISO 8503-2:1988: Preparation of Steel Substrates before Application of Paints and Related
ProductsMethod for Grading of Surface Profile of Abrasive Blast cleaned Steel
8. ASTM D 4752-2003: Measuring MEK Resistance of Ethyl Silicate Zn Rich Primers by
Solvent Rube
9. ASTM D 4587:2005: Standard Practice for Fluorescent UV Condensation Exposures of Paint
and Related Coatings
10. ASTM D 4541:2002: Standard Test Method for Pull-Off Strength of Coating Using Portable
Adhesion Testers
11. ASTM 3359 :2002: Standard Test Methods for Measuring Adhesion by Tape Test
12. ASTM G 50:1997: Standard Practice for Conducting Atmospheric Corrosion Testing Metals
13. IS: 5555:1997: Standard Practice for Conducting Field Studies on Atmospheric Corrosion of
Metals
14. ASTM G 9:1993: Standard Practice for Characterization of Atmospheric Test Sites
15. ISO 9225:1992: Corrosion of Metal and Alloys, Corrosivity of Atmosphere Measurement of
Pollution
16. ISO 9223:1992: Corrosion of Metals and Alloys: Corrosivity of Atmospheres: Classification
17. ASTM 5:2004: Standard Reference Test Method for Making Potentiostatic and Potentiodynamic Anodic Polarization Measurements
18. ASTM G 61:1993: Std. Test Method for Conducting Cyclic Polarization Measurements for Iron
19. Selected Powder Diffraction Data for Metals & Alloys Data Book, 1st. Edn., vol I & II,
JCPDS International Centre for Diffraction Data, USA
20. ASTM G 60:2001: Standard Practice for Humidity Exposures
21. ASTM B117:1997: Standard Practice for Operating Salt Spray (Fog) Apparatus
22. ASTM 87:1984: Standard Practice for Conducting Moist SO2 Tests
Chapter 3
Experimental Evidence
3.1 Rust Appearance

The macrographs of rust formed on MS and WS after 42 months (m) of field
exposure at P1 are shown in Figs. 3.1 and 3.2, where both downward and skyward
sides were considered. The rust macrographs of MS and WS after 42 m exposure
at P2 and P3 are shown in Figs. 3.3 and 3.4, respectively. Rust appearance
observed after 18 m and 42 m of atmospheric exposure has been listed in
Table 3.1 for all the sites.
3.2 Corrosion Rate

The exposed uncoated MS and WS panels were drawn from the sites for corrosion
rate determination. Corrosion products were removed by pickling solution and the
respective corrosion rates calculated by adopting standards [1, 2] and the values
3.3 DC Corrosion Measurement

Polarisation tests were carried on uncoated MS and WS drawn from P1, P2 and P3
after 42 m atmospheric exposure in 3.5 % NaCl solution. The polarisation diagrams are shown in Figs. 3.5, 3.6, 3.7 and the corrosion rate (Icorr) is also given in
Table 3.3.
Cyclic polarisation was carried out in SAE J2334 solution (2.5 %
NaHCO3 ? 0.5 % NaCl ? 0.1 % CaCl2) exposed for 4,080 h (h) on coated
ZP ? MS and ZP ? WS drawn after 18 m atmospheric exposure from P3 site.
57
58
3 Experimental Evidence
Fig. 3.1 Rust appearance exposed for 42 m at P1 downward side (a) MS and (b) WS
Fig. 3.2 Rust appearance exposed for 42 m at P1 skywards side (a) MS and (b) WS
The polarisation diagrams are shown in Fig. 3.8 and corrosion rate data are given
in Table 3.4.
3.4 AC Impedance Spectroscopy

EIS was carried out on both uncoated and coated MS and WS panels in different
test electrolytes with progressive atmospheric exposure.
59
Fig. 3.3 Rust appearance exposed for 42 m at P2 skywards side (a) MS and (b) WS
Fig. 3.4 Rust appearance exposed for 42 m at P3 skywards side (a) MS (b) WS
Table 3.1 Rust appearance of uncoated MS and WS exposed for 18 and 42 m

Site
Panel
After 18 m exposure
After 42 m exposure
P1
P2
P3
MS
WS
MS
WS
MS
WS
More brown with little black spot

Adherent brown and black
Mixed brown and blackish
Adherent brown with little black spot
Both mixture of brown and black
Adherent brown with black
Mainly blackish with pitting marks

Blackish with more pitting marks
Mainly blackish with pitting
Adherent brown with black
Mainly black with pitting
Adherent deep brown and black
60
Table 3.2 Corrosion rate (lm/y) of MS and WS exposed at P1, P2 and P3 Sites
Exposure months
Digha (P1)
Chennai (P2)
Jamshedpur (P3)
18
30
42
MS
WS
MS
WS
MS
WS
24.1
24.2
26.4
20.2
19.1
18.2
19.5
18.3
17.4
13.4
13.2
12.1
13.6
12.4
12.3
10.3
9.4
9.5
Fig. 3.5 Polarisation diagram in 3.5 % NaCl exposed for 42 m at P1 (a) MS and (b) WS
3.4.1 EIS on Uncoated Panels After 42 Months Exposure

EIS was carried out in 0.1 M Na2SO4 ? 0.1 N NaCl solution on uncoated MS and
WS drawn from P3 site after 42 m atmospheric exposure. The results presented are
an average of three tests for each panel. Impedance Plots of MS exposed for 42 m
are given in Fig. 3.9 for P1, Fig. 3.11 for P2 and Fig. 3.13 for P3. Impedance Plots
of WS exposed for 42 m are given in Fig. 3.10 for P1, Fig. 3.12 for P2 and
Fig. 3.14 for P3. The EIS parameters are given in Table 3.5.
61
Fig. 3.6 Polarisation diagram in 3.5 % NaCl exposed for 42 m at P2 (a) MS and (b) WS
3.4.2 EIS of Uncoated Panels After 48 Months

EIS was carried out in SAEJ solution on uncoated MS and WS drawn from P3 site
after 48 m atmospheric exposure. The impedance plots are given in Figs. 3.15 and
3.16, respectively, and the EIS parameters are given in Table 3.6.
3.4.3 EIS of Coated Panels After 18 Months Exposure

EIS was carried out in SAEJ solution exposed for 216 h on coated ZP ? MS and
ZP ? WS drawn from P1, P2 and P3 after 18 m atmospheric exposure. Impedance
plots of coated MS for P1, P2 and P3 are given in Figs. 3.17, 3.19 and 3.21,
respectively. Similarly, the impedance plot of coated WS for P1, P2 and P3 are
given in Figs. 3.18, 3.20 and 3.22, respectively. The EIS parameters are given in
Table 3.7.
62
Fig. 3.7 Polarisation diagram in 3.5 % NaCl exposed for 42 m at P3 (a) MS and (a) WS
Table 3.3 Corrosion rate in 3.5 % NaCl exposed for 42 m of MS and WS
Sites
Panels
Icorr lA/cm2
P1
P2
P3
MS
WS
MS
WS
MS
WS
63.80
103.3
45.32
45.52
30.77
27.89
Ecorr mV
-701.8
-643.9
-725.9
-656.2
-637.5
-524.9
3.4.4 EIS of Scribed Coated Panels After 18 Months Exposure

EIS was carried out in 3.5 % NaCl solution exposed for 120 h on scribed coated
ZP ? MS and ZP ? WS drawn from P3 after 18 m atmospheric exposure.
Impedance plots are given in Figs. 3.23 and 3.24, respectively, and the EIS
parameters are given in Table 3.8.
63
Fig. 3.8 Cyclic polarisation in SAEJ solution of coated MS and WS exposed at P3 for 18 m
Table 3.4 Cyclic polarisation of coated MS and WS after 18 m exposure at P3
Panel
Electrolyte
Icorr lA/cm2
Ecorr mV
ZP ? MS
ZP ? WS
-700.8
-685.5
2.5 % NaHCO3 ? 0.5 %

NaCl ? 0.1 % CaCl2 for 4,080 h
50.25
10.30
3.4.5 EIS of Coated Panels After 42 Months Exposure

EIS was carried out in 3.5 % NaCl solution exposed for 936 h on coated ZP ? MS
and ZP ? WS drawn from P1, P2 and P3 after 42 m of atmospheric exposure.
Impedance plots of coated ZP ? MS for P1, P2 and P3 are given in Figs. 3.25,
3.27 and 3.29, respectively. Similarly, impedance plots of coated ZP ? WS for P1,
P2 and P3 are given in Figs. 3.26, 3.28 and 3.30, respectively. The EIS parameters
3.4.6 EIS of Coated Panels After Humid SO2 Exposure

EIS was carried out in SAEJ solution exposed for 3,000 h on coated MS and WS
drawn after 42 m atmospheric exposure from P3 site. The panels were exposed in
humid SO2 for 2 m after atmospheric exposure and before EIS. Impedance plots are
given in Figs. 3.31 and 3.32, respectively, and EIS parameters are given in Table 3.10.
3.5 Rust Characterisation by SEM and EDX

SEM was used for rust morphology studies on uncoated MS and WS panels with
progressive exposure. Figures 3.33, 3.34 and 3.35 indicate rust micrographs of MS and
WS drawn after 18 m from P1, P2 and P3, respectively. Elemental analysis of WS rusts
is given in Table 3.11 and EDX spectra plot of WS are given in annexure Fig. A 3.1a, b.
64
Fig. 3.9 Impedance plots in 0.1 N NaCl ? 0.1 M Na2SO4 exposed at P1 for 42 m of MS
SEM was used for rust morphology studies on uncoated MS and WS panels
drawn after 42 m of atmospheric exposure from P1, P2 and P3. Figures 3.36a, b,
3.37a, b, and 3.38a, b indicate rust micrographs of MS and WS, respectively.
Elemental analysis of these rusts is given in Table 3.12 and EDX spectra plots are
given in annexure Figs. A 3.2a, b, A 3.3a, b and A 3.4a, b. Corresponding EDX
line scans are given in Figs. 3.39, 3.40, 3.41, 3.42, 3.43.
3.5.1 Rust Characterisation by XRD

In order to identify the rust phases XRD was carried out on uncoated MS and WS
after atmospheric exposure. The details of oxide phases are given in Table 3.13 for
3.5 Rust Characterisation by SEM and EDX
65
Fig. 3.10 Impedance plots in 0.1 N NaCl ? 0.1 M Na2SO4 exposed at P1 for 42 m of WS
MS and WS after 18 m atmospheric exposure at P1 and in Table 3.14 for MS and

WS after 42 m atmospheric exposure at P1.XRD spectra plots are given in annexure Figs. A 3.5a, b, A 3.6a, b, A 3.7 and A 3.8a b.
The details of oxide phases for WS after 42 m atmospheric exposure at P2 are
given in Tables 3.15 and 3.16 indicates the rust phases of MS and WS after 42 m
of atmospheric exposure at P3.
3.5.2 Rust Characterization by Raman Spectroscopy

Rust phases of uncoated MS and WS panels were identified through Raman
spectroscopy after 18 and 42 m atmospheric exposure. Raman shift values are
given in Table 3.17 with Raman spectra in annexure Figs. A 3.9a, b, A 3.10a, b
and A 3.11a, b, for MS and WS after 18 m of atmospheric exposure. Table 3.18
66
gives the Raman shift values along with spectra plots in annexure Figs. A 3.12a, b,
A 3.13a, b and A 3.14a, b, for 42 m atmospheric exposure for MS and WS.

Various accelerated laboratory tests were conducted on both uncoated and coated
MS and WS panels.
67
3.6.1 Corrosion Rate by Humidity Test

Rust collected at different exposure time like 1,000, 2,000, 3,500 and 4,500 h and
corrosion rate was calculated by weight loss method. Table 3.19 shows the corrosion rates of uncoated MS and WS after different exposure time.
3.6.2 Corrosion Rate by Salt Spray Test

The macrograph of MS and WS after 1,000 h of exposure is shown in Fig. 3.44a
and b and after 4,500 h exposure the same is given in Fig. 3.45a and b. The
corrosion deposits were removed by modified Clarke solution (Concentrated
HCl ? 2 % Sb2O3 ? 5 % SnCl2). The corrosion rate was calculated by weight
loss method. The corrosion rate is shown in Table 3.20. For evaluating the degree
68
of rusting, blistering on exposed coated panels relevant standard [3] was used as
given in Table 3.21.

Open circuit potential measurement was carried out in 3.5 % NaCl exposed for
4.1 h on as received MS and WS panels and the plots are given in Fig. 3.46a, b,
respectively.
Polarisation tests were carried out in 3.5 % NaCl solution on as received MS
and WS panels and the diagrams are shown in Fig. 3.47a, b. Similar tests were
carried out in 0.1 M Na2SO4 ? 0.1 N NaCl solution and the diagrams are shown
in Fig. 3.48a, b. The corrosion rates in theses electrolytes are given in Table 3.22.
69
Table 3.5 EIS in 0.1 M Na2SO4 ? 0.1 N NaCl on MS and WS exposed for 42 m
Cf F/cm2
Sites Panel Ru ohm cm2 Rp ohm cm2 Rpo ohm cm2 Cc F/cm2
P1
P2
P3
MS
WS
MS
WS
MS
WS
2.39
4.0
1.11
2.5
6.52
1.44
9
9
9
9
9
9
10-3
10-3
10-3
10-3
10-3
10-3
6.63
2.66
3.95
2.48
6.66
9.56
9
9
9
9
9
9
106
106
105
107
102
105
13.64
19. 21
55.61
1.52
26.75
4.92
9
9
9
9
9
9
103 20.84 9 10-9

102
1.06 9 10-6
103 28.00 9 10-8
103
1.43 9 10-6
4
10 13.91 9 10-5
104 77.73 9 10-12
2.86 9 10-6
50.76 9 10-11
8.30 9 10-6
20.84 9 10-11
42.12 9 10-12
16.55 9 10-11
Fit
REAP
REAP
REAP
REAP
REAP
REAP
Cyclic polarisation test carried out in 2.5 % NaHCO3 ? 0.5 % NaCl ? 0.1 %
CaCl2 solution on MS and WS after exposed in humid SO2 environment for 9 m.
The polarisation plots are given in Fig. 3.49 along with corrosion rate (Icorr) in
Table 3.23.
70
Fig. 3.15 Impedance plots in SAEJ solution for 48 m at P3 of MS
3.6.4 AC Impedance Measurement

EIS was carried out 0.1 M Na2SO4 ? 0.1 N NaCl solution on as received MS and
WS and impedance spectra interpretation was performed after numerically fitting
and using equivalent circuit. The impedance plots are given in Figs. 3.50 and 3.51
and EIS parameters are given in Table 3.24.
EIS was carried out in SAEJ solution exposed for 48 h on MS and WS after
exposing in humid SO2 environment for 9 months (m.) The impedance plots are
given in Fig. 3.52 along with EIS parameters in Table 3.25.
EIS was carried out in 3.5 % NaCl solution exposed for 24 h and 168 h on zinc
phosphate coated (ZP ? MS). The impedance plots are given in Figs. 3.53
71
Fig. 3.16 Impedance plots in SAEJ solution for 48 m at P3 of WS
Table 3.6 EIS in SAEJ solution for 72 h on MS and WS exposed for 48 m at P3

Rp ohm cm2
Rpo ohm cm2
CcF/cm2
Cf F/cm2
Panels
Ru ohm cm2
MS
WS
58.70
62.54
3.24 9 10
8.50 9 105
-2
3.25 9 10
8.09 9 104
-6
0.98 9 10
4.44 9 10-6
0.001
0.003
Fit
REAP
REAP
and 3.54, respectively. Similarly, EIS was also carried out after 24 and 168 h
exposure on zinc-rich coated (ZR ? MS) and impedance plots are given in
Figs. 3.55 and 3.56, respectively. The EIS parameters are given in Table 3.26.
EIS was carried out in SAEJ solution exposed for 1,500 h on coated ZP ? WS
and ZR ? WS and impedance plots are given in Figs. 3.57 and 3.58, respectively
and the EIS parameters are given in Table 3.27.
Fig. 3.18 Impedance plots in SEAJ solution exposed for 18 m at P1 of ZP ? WS
Fig. 3.17 Impedance plots in SEAJ solution exposed for 18 m at P1 of ZP ? MS
72

73
74
ZP
ZP
ZP
ZP
ZP
ZP
?
?
?
?
?
?
MS
WS
MS
WS
MS
WS
2.23
8.67
61.32
8.32
53.38
4.85
9
9
9
9
9
9
10-3
10-3
10-3
10-3
10-3
10-3
7.59
6.15
2.22
3.30
6.61
1.85
9
9
9
9
9
9
108
108
108
109
108
108
41.57 9 10-3
69.14 9 10-3
1.46 9 103
4.77 9 102
39.74 9 10-9
5.76 9 10-9
Fig. 3.23 Impedance plots in SAEJ for 1,000 h exposed for 18 m at P3 of Scribed of ZP ? MS
P3
P2
P1
Table 3.7 EIS of ZP ? MS and ZP ? WS in SAEJ solution for 216 h after exposed for 18 m
Rp ohm cm2
Wd S*s^(-1/2)
Site
Panel
Ru ohm cm2
9
9
9
9
9
9
10-9
10-9
10-9
10-9
10-9
10-9
Yo S*sâ
3.89
3.52
7.29
5.07
2.07
2.57
a
0.88
0.83
0.83
0.81
0.87
0.81
Fit
CPE
CPE
CPE
CPE
CPE
CPE
?
?
?
?
?
?
D
D
D
D
D
D

75
ZP ? MS
ZP ? WS
11.58 9 10-3
12.39 9 10-3
29.87 9 104
27.23 9 106
4.96 9 10-10
2.60 9 10-9
Table 3.8 EIS of scribed ZP ? MS and ZP ? WS in SAEJ after exposed for 18 m at P3

Rp ohms cm2
Yo S*sâ
Panel
Ru ohms cm2
Fig. 3.24 Impedance plots in SAEJ for 1,000 h exposed for 18 m at P3 of Scribed of ZP ? WS
46.34 9 10-6
Wd S*s^(1/2)
0.93
0.78
CPE
CPE ? D
Fit
76
Fig. 3.26 Impedance plots in 3.5 % NaCl for 936 h exposed for 42 m at P1 of ZP ? WS
Fig. 3.25 Impedance plots in 3.5 % NaCl for 936 h after exposed for 42 m at P1 of ZP ? MS

77
Fig. 3.28 Impedance plots of ZP ? WS in 3.5 % NaCl for 936 h after exposed for 42 m at P2
Fig. 3.27 Impedance plots in 3.5 % NaCl exposed for 42 m at P2 for 936 h of ZP ? MS
78
Fig. 3.30 Impedance plots of ZP ? WS in 3.5 % NaCl for 936 h after exposed for 42 m at P3
Fig. 3.29 Impedance plots of ZP ? MS in 3.5 % NaCl for 936 h after exposed for 42 m at P3

79
ZP
ZP
ZP
ZP
ZP
ZP
?
?
?
?
?
?
MS
WS
MS
WS
MS
WS
Cf: double layer capacitance
P3
P2
P1
20.09
10.89
77.35
34.60
96.73
1.02
9
9
9
9
9
9
10
10-3
10-3
10-2
10-3
10-3
-2
46.18
25.76
12.25
80.54
2.59
16.34
9
9
9
9
9
9
10
104
104
106
103
103
9.39 9 10
9.03
23.68 9 104
24.22 9 106
59.42 9 103
31.85 9 107
-8
3.45 9 10
19.17 9 10-8
71.54 9 10-12
22.20 9 10-12
2.11 9 10-9
2.42 9 10-11
Table 3.9 EIS parameters of ZP ? MS and ZP ? WS in 3.5 % NaCl solution after exposed for 42 m
Rp ohm cm2
Rpo ohm
Cc F/cm2
Site
Panel
Ru ohm cm2
10.64 9 10-4
58.33 9 10-3
4.71 9 10-9
16.30 9 10-11
2.58 9 10-6
11.37 9 10-6
Cf F/cm2
REAP
REAP
REAP
REAP
REAP
REAP
Fit
80
81
Fig. 3.31 Impedance plot of ZP ? MS in SAEJ after exposed for 42 m at P3 and in humid SO2

Rust micrograph of as received MS and WS was taken by SEM as shown in
Fig. 3.59a, b and elemental analysis is given in Table 3.28 along with spectra plots
in annexure Fig. A 3.15a, b. The EDX line scans are also given in Figs. 3.60 and
3.61.
SEM was carried out on the rust developed after SO2 exposure on MS and WS
from the top surface as well as from the cross-section of the panels. The micrographs are given in Figs. 3.62a, b and 3.63a, b, respectively, along with spectra
plots in annexure Figs. A 3.16a, b, and A 3.17a, b. The elemental analysis by EDX
is given in Table 3.29.
3.6.6 Rust Characterisation by Raman Spectroscopy

Rust developed on as received panels, panels after humidity, salt spray tests were
analysed by Raman spectroscopy. The Raman peak values of the phases are given
in Tables 3.30, 3.31 and 3.32, respectively. Corresponding Raman spectra plots
0.72
0.78
1.26 9 10
2.54 9 108
EIS carried out in SAEJ electrolyte exposed for 3,000 h
ZP ? MS
ZP ? WS
2.81 9 10
6.37 9 10-10
-10
Table 3.10 EIS parameters of ZP ? MS and ZP ? WS at P3 in SO2 after exposed for 42 m

Rp ohm cm2
Yo S*sâ
Panel
Ru ohm
Fig. 3.32 Impedance of ZP ? WS in SAEJ after exposed for 42 m at P3 and in humid SO2
1.03 9 10
1.86 9 10-3
-4
Wd S*s^(1/2)
0.91
0.89
CPE ? D
CPE ? D
Fit
82
83
Fig. 3.33 SEM of rust exposed for 18 m at P1 (a) MS and (b) WS
Fig. 3.34 SEM of rust exposed at P2 for 18 m (a) MS and (b) WS
Table 3.11 Elemental analysis ( % At) by EDX of WS exposed for 18 m

Sites
Panels
Si
Mn
S
Al
Cl
P1
P2
P3
0.50
WS
WS
WS
0.37
1.24
0.94
0.81
0.69
0.2
0.54
0.67
0.21
0.79
84
85
Table 3.12 Elemental analysis by EDX ( %At) MS and WS of exposed for 42 m

S
P
Mn
Site
Elements
O2
P1
P2
P3
MS
WS
MS
WS
MS
WS
Fig. 3.39 Line scan

elements of MS rust exposed
at P1 for 42 m (a) oxygen and
(b) iron
70.40
70.38
68.52
52.39
59.29
59.32
0.20
0.28
0.24
0.48
0.39
0.54
0.21
0.32
0.22
0.19
Cr: 0.5, Cu: 0.6
Cl
0.50
0.60
86
Fig. 3.40 Line scan
elements of WS rust exposed
(b) iron

Fig. 3.41 Line scan
(b) iron
87
88
Fig. 3.42 Line scan
elements of MS rust exposed
at P3 for 42 m (a) Oxygen
and (b) Iron

Fig. 3.43 Line scan
at P3 for 42 m (a) Oxygen
and (b) Iron
89
90
Table 3.13 Rust phases by XRD exposed at P1 for 18 m of MS and WS

MS
WS
Peak
no
% Relative
intensity I/I0
Inter planer spacing Peak

(d) A0
no
% Relative
intensity I/I0
Inter planer spacing

(d) A0
6
5
16
10
7
18
12
9
13
19
100
53
43
40
34
26
23
21
21
21
2.458
2.542
1.511
1.932
2.359
1.442
1.638
2.081
1.609
1.394
100
81
63
50
48
35
33
31
48
38
37
25
2.462
6.232
3.29
1.939
1.533
1.561
1.509
2.193
2.567
1.726
2.354
1.64
7
1
3
12
16
15
17
10
6
13
8
14

MS
WS
Peak
no
% Relative
intensity I/I0

(d) A0
no
% Relative
intensity I/I0

(d) A0
1
4
8
5
3
2
10
6
7
9
100
22
14
19
27
16
12
18
13
37
2.474
1.936
1.729
3.300
2.578
2.247
3.204
2.775
3.181
2.361
19
100
26
12
24
16
29
9
10
13
2.599
2.479
1.945
1.734
3.304
2.255
2.366
2.195
2.710
2.991
5
1
3
8
4
6
2
10
9
7
Table 3.15 Rust phases by XRD exposed at P2 for 42 m of WS

Peak no
% Relative intensity I/I0
6
1
3
2
5
7
4
17
100
25
26
11
14
19
9
Inter planer spacing (d) A0

3.297
2.470
1.939
2.583
2.978
2.703
2.359
2.318
91

MS
WS
Peak
no
% Relative
intensity I/I0

(d) A0
no
% Relative
intensity I/I0

(d) A0
1
3
6
4
7
9
5
8
100
53
38
51
22
16
44
15
2.448
1.936
3.309
2.695
1.864
2.394
2.023
1.798
100
45
71
66
36
67
39
46
2.478
3.296
2.548
1.846
1.940
2.523
1.691
1.820
1
6
2
4
8
3
7
5
Table 3.17 Rust phases by Raman spectroscopy after 18 m at P1, P2 and P3 of MS and WS
Raman shift cm-1
Panels
P1
MS
1,307.28
251.12
221.48
375.33
695.92
265.37
392.92
241.60
1,292.10
491.01
WS
P2
P3
390.62
298.92
686.51
549.08
252.95
654.26
526.92
374.90
267.71
384.40
210.81
1,292.87
589.62
1,323.19
1,425.66
249.21
654.77
374.49
Table 3.18 Rust phases by Raman Spectroscopy after 42 m at P1, P2 and P3 of MS and WS
Raman shift cm-1
Panels
P1
P2
P3
MS
387.60
387.60
261.22
1,327.83
651.69
398.12
528.32
259.88
1,314.28
253.90
383.57
383.57
656.66
1,310.18
1,305.14
290.63
223.93
414.29
608.43
225.87
484.20
290.81
1,302.04
408.20
WS
92
Table 3.19 Corrosion rate of MS and WS in humidity test

Exposure hours (h) MS
WS
1,000
2,000
3,500
4,500
Wt loss mg
Area cm2
CR lm/y
Wt loss mg
Area cm2
CR lm/y
459
850
1,250
1,538
84.17
84.17
85.46
84.17
461
425
418
382
359
706
1,050
1,337
82.49
82.49
84.49
83.20
374
355
341
334
Fig. 3.44 Exposure in salt spray for 1,000 h (a) MS and (b) WS
Fig. 3.45 Exposure in Salt Spray for 4,500 h (a) MS and (b) WS
93
Table 3.20 Corrosion rate of MS and WS in salt spray test

MS
WS
Hour
Wt loss mg
Area cm2
CR lm/y
Wt loss mg
Area cm2
CR lm/y
1,000
2,000
3,500
4,500
38,770
57,050
88,670
98,480
336.56
327.07
352.17
335.91
992
742
727
617
37,832
43,673
60,425
78,015
312.24
312.05
306.11
312.63
1,042
592
558
546
Table 3.21 Rating of coated panels exposed in salt spray test

Hours
MS
WS
1,000
2,000
3,500
4,500
Coated
Rating
Coated
Rating
No. effect
Few blisters
Many blisters
Blisters increased
10
9.75
9.5
8
No. effect
0.5 mm rust creepage
Few blister
1.5 mm creepage/Blisters increased
9
8
7.5
7
Before exposure Rating = 10, Rating \7 test panels to discard
Fig. 3.47 Polarisation diagram in 3.5 % NaCl (a) AR MS and (b) AR WS
Fig. 3.46 OCP Plots in 3.5 % NaCl solutions of (a) AR MS and (b) AR WS
94
Ecorr mV
-272.7
-260.1
-50.10
-231.7
25.6 9 10-6
4.46 9 10-12
278.8 9 10-6
10.65 9 10-12
MS
WS
MS
WS
3.5 % NaCl
Icorr A/cm2
Table 3.22 Corrosion rate in different electrolytes of AR MS and AR WS

Panels
Electrolyte
Fig. 3.48 Polarisation diagram in 0.1 M Na2SO4 ? 0.1 N NaCl (a) AR MS and (b) AR WS

95
96
Fig. 3.49 Cyclic polarisation in SAEJ of MS and WS after exposed in humid SO2
Table 3.23 CP of MS and WS in SAEJ solution after exposed in humid SO2 for 9 m
Ecorr mV
Panel
Electrolyte
Icorr mA/cm2
MS
WS
2.5 % NaHCO3 ? 0.5 %

NaCl ? 0.1 % CaCl2
55.21
4.32
-1,175.5
-1,195.5
Fig. 3.51 EIS Plots in 0.1 M Na2SO4 ? 0.1 N NaCl of AR WS
Fig. 3.50 EIS Plots in 0.1 M Na2SO4 ? 0.1 N NaCl of AR MS

97
98
Table 3.24 EIS in 0.1 M Na2SO4 ? 0.1 N NaCl for 24 h on AR MS and AR WS

Rp ohm cm2
Y0 S*sâ
a
Panel
Ru ohm cm2
MS
WS
62.47
23.44
16.65
3.15 9 105
646.3 9 10-6
70.72 9 10-12
0.61
0.91
Fit
CPE
CPE
S*sâ S: siemens, s second, a: CPE exponent
Fig. 3.52 Nyquist plots of WS and MS in SAEJ after exposed in humid SO2 for 9 m
Table 3.25 EIS of MS and WS in SAEJ solution for 48 h after exposed in SO2
Panel
Ru ohm cm2
Rp ohm cm2
Cf F/cm2
a
MS
WS
-3
5.87 9 10
4.35 9 10-3
3.37 9 10
6.01 9 104
-8
1.33S 9 10
2.67 9 10-8
0.64
0.63
Fit
Randle
Randle
Fig. 3.53 Impedance plots in 3.5 % NaCl solution for 24 h of ZP ? MS
99
100
Fig. 3.54 Impedance plots in 3.5 % NaCl solution for 168 h of ZP ? MS
Fig. 3.55 Impedance plots in 3.5 % NaCl solution for 24 h of ZR ? MS
101
102
Fig. 3.56 Impedance plots in 3.5 % NaCl solution for 168 h of ZR ? MS
24
168
24
168
Fig. 3.57 Impedance plots SAEJ solution for 1,500 h of ZP ? WS
ZR ? MS
ZP ? MS
1.2
2.1
4.1
3.2
9
9
9
9
10
10-2
10-3
10-3
-3
Table 3.26 EIS in 3.5 % NaCl solution of coated ZP ? MS and ZR ? MS

Panels
Exposure (h)
Ru ohm cm2
36
1.42
27
53.0
9
9
9
9
10
103
103
103
Rp ohm cm2
53 9 10-12
0.15 9 10-9
98 9 10-12
0.12 9 10-9
Cc F/cm2

103
Fig. 3.58 Impedance plots in SAEJ solution for 1,500 h of ZR ? WS
104
105
Table 3.27 EIS in SAEJ solution for 1,500 h on coated ZP ? WS and ZR ? WS

Rp ohm cm2 Yo S*s^ a
Wd S*s^(1/2) a
Panels
Ru ohm cm2
ZP ? WS
ZR ? WS
23.23 9 10-3
7.97 9 10-3
10.51 9 107
1.57 9 109
5.96 9 10-9
4.48 9 10-9
2.38 9 10-7
0.85
0.82
Fit
CPE
CPE ? D
Fig. 3.59 Rust micrograph by SEM of (a) AR MS and (b) AR WS
Table 3.28 Elemental analysis by EDX ( %At) of As received MS and WS

Panels
O
Si
Cr
Mn
MS
WS
0.38
0.31
28.37
14.28
0.84
0.51
106
Fig. 3.60 Line scan of
elements by EDX of AR MS
(a) Oxygen and (b) Iron

Fig. 3.61 Line scan of
elements by EDX of AR WS
(a) oxygen and (b) iron
107
108
Fig. 3.62 SEM of rust upper layer exposed in SO2 (a) MS and (b) WS
Fig. 3.63 SEM of rust inner layer exposed in SO2 (a) MS and (b) WS
Table 3.29 Elemental analysis by EDX of rust of MS and WS after exposed in SO2
Layer
Panel
Mn
Cr
Cu
Upper
Inner
MS
WS
MS
WS
0.68
0.33
0.55
0.62
3.48
Ni
2.32
1.52
Table 3.30 Rust phases by Raman spectroscopy of AR MS and AR WS

Raman shift cm-1
MS
WS
1,321.37
671.52
613.34
1,314.79
658.79
611.79
109
Table 3.31 Rust phases by Raman spectroscopy in humidity of MS and WS

Raman shift cm-1
Panel
2,000 h
3,500 h
4,500 h
MS
1,318.63
296.26
230.32
411.02
612.66
503.87
1,322.58
290.84
226.86
407.91
292.39
227.29
612.30
404.00
1,320.71
299.38
411.27
657.44
293.22
226.41
403.60
1,300.35
254.72
1,305.91
657.20
378.76
WS
Table 3.32 Rust phases by Raman spectroscopy in salt spray of MS and WS

Raman shift cm-1
Panel
2,000 h
3,000 h
4,500 h
MS
1,321.17
233.54
415.69
300.15
617.23
507.09
1,350.80
294.54
294.54
227.46
410.23
297.41
413.02
613.33
500.05
1,310.81
384.34
533.67
648.37
258.85
1,292.5
387.31
584.83
211.72
273.34
387.46
1,298
212.50
274.88
WS
are given in annexure Fig. A 3.18a, b for as received MS and WS, for humidity in
annexure Figs. A 3.19a, b to A 3.21a, b and for salt spray in annexure Figs. A
3.22a, b to A 3.24a, b.
References
1. ASTM G101:1997 Standard guide for estimating the atmospheric corrosion resistance of low
alloy steel
2. ASTM G33:2004 Standard practice for recording data from atmospheric corrosion tests of
metallic-coated steel specimens
3. ASTM 610D:2001 Standard test method for evaluating degree of rusting on painted steel
surfaces
Chapter 4
Field Exposure Test
4.1 Field Exposure Test

4.1.1 Rust Appearance
Both uncoated and coated exposed panels were taken out after 18 m of exposure
from sites. No appreciable changes were noted for coated panels exposed at all the
sites. However, for coated and scribed panels a light attack of rust was visible
along scribed lines. Rust product developed at all sites was brownish, loosely
bound on MS and adherent brownish black on WS (Table 3.1). Uncoated panels
had different pitting attack and WS exhibited less pitting marks than MS. The
variation of pitting densities and distribution at different sites depended on the
material and the environment prevalent at the site.
Another set of panels was withdrawn from the sites after 42 m of exposure.
Corrosion attack along the scribed lines on coated panels at all sites was more in
comparison to the panels drawn after 18 m. Uncoated panels at P1 which is very
near to sea did show appreciable difference in oxide film morphology on MS and
WS. The downward face of the exposed panels at P1 (Figs. 3.1a, b) shows uneven
dark brown coarse granular oxide film on WS and MS. The skyward surface on the
rusted WS exhibited more pitting with blackish appearance. MS rust showed less
pitting mark (Figs. 3.2a, b). This is due to the presence of high amount of chloride
ions at P1. At P2 which is relatively far from sea and considerable presence of SO2
in the environment, the skyward surface showed loosely blackish oxides on MS
and adherent brown oxide on WS (Figs. 3.3a, b respectively). The appearance of
WS at P2 was different with respect to P1, whereas not much difference was
noticed for MS at P1 and P2. In former case, the rust was adherent than MS and
The results obtained from experiments as recorded in Chapter 3 are discussed in three parts viz.
field tests, laboratory tests and simulation tests.

111
112
4 Field Exposure Test
this indicates the role of minor alloying elements present in WS and shows a
tendency to form stable protective layer on its surface in presence of SO2. The
skyward surface on MS at P3, shows formation of blackish loosely attached oxide
film (Fig. 3.4a), whereas WS showed compact deep brown fine oxide film
(Fig. 3.4b). Again this is due to the presence of SO2 at P3.
Thus, there is a significant difference in the appearance of oxide film/rust
products on the uncoated exposed panels with respect to site and exposure time.
Uncoated panels exposed are having different rust colour and varying degree of
adherence to surface. The corrosion attack was severe at P1 followed by P2
and P3.

During the first 18 m of exposure, the corrosion rate of uncoated MS and WS
panels (Table 2.8) at P1 is higher with respect to P2 and P3. This is due to the
variation of relative humidity is maximum at P1 as shown in annexure (Fig. A.2.2)
and most importantly is due to the presence of high content of chloride ions at P1.
For WS at P2 and P3 corrosion rate progressively decreases with time due to the
availability of SO2 compared to P1. Figure 3.1 shows the trend of rate of progress
of penetration with time at all sites for WS and MS. At all sites, corrosion of MS
was found to be higher than WS. MS recorded highest corrosion rate at P1 followed by P2 and P3. Higher corrosion rate of MS is due to the formation of nonprotective rust layer developed. Rusting on MS is proceeding at a comparatively
higher rate (Fig. 4.1).
Lower corrosion rate on WS at P2 and P3 appears primarily due to availability
of SO2 in the environment. Formation of continuous rust product causes a
reduction in oxygen supply to diffuse through rust deposits and reduces the corrosion rate.
After 42 m exposure, the corrosion rate is stabilised at P2 and P3 but it continued at P1 on WS. These phenomena may be primarily due to the atmospheric
pollution and meteorological conditions (annexure: Figs. A 2.1 to A 2.3 and Table
A 2.1). The corrosion rate on WS is lower due to presence of SO2 at P2 and P3 and
the steady-state corrosion rate on WS (\12 lm/y) is achieved at SO2 (2050 lg/
100 cm2) which is in accordance with the findings of Knotkova [1] who reported
that SO2 played an important role for lowering corrosion rate on WS. WS does not
perform well at P1 due to presence of Chloride ions and recorded higher corrosion
rate. The alloying elements perhaps make it more vulnerable. Similar findings for
WS have been reported by Madlangbayan et al. [2]. The corrosion rate of WS is
found lower than MS for all the environments and shows stable corrosion rate at
P2 and P3.

30
Corrosion Rate (m/yr)
Fig. 4.1 Atmospheric

corrosion trend of MS and
WS at P1, P2, P3
113
P1 MS
25
20
P1 WS
P2 MS
15
P2 WS
P3 MS
10
P3 WS
5
15
20
25
30
35
40
45
Exposure Time (month)
4.1.3 DC Measurement
Corrosion rate of exposed panels after 42 m exposure, obtained from the polarisation diagrams (Figs. 3.5, 3.6, and 3.7) in 3.5 % NaCl solution, was found to be
of the order of 10-6 A/cm2 for MS and WS at all sites (Table 3.3a). Such low Icorr
may be due to the formation of compact rust layers which are not allowing
electrolytes to go in except the holidays exposed on the surface. Ecorr values for
both the materials were greater at P3.
To study the effectiveness of coatings, the coated panels after 18 m exposure at
P3 were dipped in SAEJ solution for 4,080 h followed by cyclic polarisation in
same medium. Interestingly, both the polarisation diagram (Fig. 3.8) showed a
tendency to passivate at higher potential and during reverse cycle Zcp (zero current
potential) changed to greater value. Icorr was also lower (Table 3.4).
4.1.4 AC Impedance
EIS carried out in 0.1 N NaCl ? 0.1 M Na2SO4 solution on uncoated WS panels
exposed for 42 m at P2 recorded lower coating capacitance (Cc: 28.00 9 10-8)
and double layer capacitance (Cf: 20.8.4 9 10-11) and higher polarisation resistance (Rp: 2.48 9 107) and pore resistance (Rpo: 1.52 9 103) as per Fig. 3.12.
Identical trend is found at P3 with lower Cc (77.73 9 10-12) and Cf
(16.55 9 10-11) and higher Rp (9.56 9 105) and Rpo (4.92 9 104).with respect to
exposure at P1 with higher Cc (1.06 9 10-6) and Cf (50.76 9 10-11) and lower Rp
(2.66 9 106) and Rpo(19.21 9 102) as per Figs. 3.14 and 3.10, respectively.
Similar trend was also noticed for MS (Table 3.5 and Figs. 3.9, 3.11 and 3.13).
Rust coatings formed showed REAP model fit and typical electrical equivalent
representative circuit as shown in Fig. 1.13 in the literature review section.
Warburg diffusion (Wd) is visible on WS for all the sites as per Nyquist plots and
predominant at P3. Phase angles are varying on WS at P1 (-35), P2 (-21) and
114
higher at P3 (-55). For MS phase angles are at P1 (-31), P2 (-18) and P3

(-18) which indicates none of the interfaces showing ideal capacitive behaviour.
It is interesting to note that with increase in exposure time to 48 m, corrosion
resistance is enhanced on WS as per Table 3.6 where test electrolyte was changed
to SAEJ solution. REAP model fit remained with improved Rpo (8.09 9 104) due
to the formation of rust having higher Rp (8.50 9 105) as per Figs 3.15 and 3.16.
Both Nyquist and Bode plots fit well at low to medium frequencies. In Nyquist plot
of WS, Zimg increases at lower frequencies with respect to MS and Wd is found in
WS. Presence of Wd indicates diffusion barrier to the flow of corrosive ions. Kihira
et al. [3] and Nishimura et.al [4] earlier reported same findings under such condition where the impedance increased and capacitance decreased on WS. Formation of the compact adherent corrosion products seems to be the reason for
higher values on WS at P2 and P3.
EIS was carried out on coated ZP ? MS and ZP ? WS, exposed at P1, P2 and
P3 after 18 m of exposure. The Bode and Nyquist plots are shown in Fig. 3.17
3.22. The only differences are found in the values of Rp and admittance (Yo) as per
Table 3.7. The order of Rp ([108) and Yo ([10-9) indicates very good, non-porous
coating surface as reported by Macdonald [5]. It is to be noted here that, the nature
of plots is identical and irrespective type of materials and environments. Nyquist
plot is present with Wd representing diffusion barrier against corrosive ions. P2
showed higher Wd (4.77 9 102) for both MS and WS as per Figs. 3.19 and 3.20e
Bode phase angles indicate capacitive behaviour as phase angles of ZP ? MS are
in the range of P1 (60), P2 (75) and P3 (80) and same is for ZP ? WS with
phase angles (70) for P1, P2 and P3. The electrical circuit elements as per CPE
with diffusion model fit are shown in Fig. 1.12 in the literature section. Thus, after
18 m exposure very little moisture and pollutants could penetrate into the coating
resulting in higher Rp and low Yo values.
The effect of scribed holidays on coated WS panels has also been studied by
EIS. Bode and Nyquist plots are given in Figs. 3.23 and 3.24. Although the order
of Rp value has decreased to some extent due to creation of these holidays, but is
still sufficiently higher (27.23 9 106) to produce high corrosion resistant surface
which is better than Rp (29.87 9 104) of ZP ? MS (Table 3.8). Phase angle (80)
of WS shows capacitive behaviour. CPE and CPE ? D model fit were used for
ZP ? MS and ZP ? WS, respectively.
EIS of coated panels exposed to 42 m was done in slightly stronger corrosive
medium (3.5 % NaCl) with respect to earlier as deterioration after 18 m was
negligible. At P1 as per Fig. 3.25 and Table 3.9 ZP ? MS show higher Rp
(46.18 9 106) and Rpo (9.39 9 102) as compared to ZP ? WS as per Fig. 3.26
which shows lower Rp (25.76 9 104) and Rpo (9.03) indicating higher degradation
rate, although double layer capacitance (Cf : [103) developed at P1 and Cc
([108) reduced resulting in higher corrosion rate. Rpo drops at P1 due to the
penetration of electrolyte through the coating to the substrate. ZP ? WS at both
P2, P3 are having higher Rp and Rpo ([106) (Figs.3.28 and 3.30, respectively).This
is possible due to the formation of protective passive layer on WS substrate. The
appearance of second time constant at P2 and P3 indicates the damage of barrier
115
properties of paint film by the formation of double layer capacitance. REAP model
fit is found suitable. All these corroborate the corrosion rate data as obtained in
Table 3.9 and Figs. 3.27 and 3.29 show the impedance plots of ZP ? MS for P2
and P3, respectively. Compared to exposure of 18 m, after 42 m exposure, Rp has
come down to 103 from 108 due to the degradation of surface with prolonged
exposures.
The effect of presence of SO2 on the coated panels was investigated by subjecting panels withdrawn after 42 m from P3 to SO2 atmosphere for two months.
The EIS behaviour in SAEJ solution is given in (Figs. 3.31 and 3.32). It is
interesting to note that, SO2 being a corrosive agent has quite unlikely improved
the corrosion resistance properties of ZP ? WS indicated by comparatively higher
Rp (2.54 9 108) as per Table 3.10 as compared to the values given in Table 3.9.
These values of ZP ? WS are superior to Rp (1.26 9 106) of ZP ? MS. Diffusion
phenomenon is prominent from the Nyquist plot of ZP ? WS (Fig. 3.32).
4.1.5 Rust Characterisation

Characterisation by SEM and EDX. Rust morphologies of MS and WS exposed at
P1, P2 and P3 were studied by SEM after 18 m exposure. MS micrograph (Fig.
3.33a) shows lot of voids at P1, MS at P2 (Fig. 3.34a) shows void/open structure of
oxide layer and at P3 (Fig. 3.35a) shows flowery structures having voids. At P1,
the rust appearance of WS is globular, non-compact type which is distinctly different with respect to P2 and P3 (Figs. 3.33b, 3.34b and 3.35b, respectively).
Acicular morphology is observed on WS at P2 and P3. The oxide layer on MS and
WS at P1 is discontinuous due to closeness to sea with high Chloride ions, permitting easy access of corrosive species to steel substrate. EDX of WS at P1 (Table
3.11) shows chloride build up on surface.
The presence of Mn, S, Si and Al is found at P2 and P3, whereas Mn and Cl
were found at P1. Chloride ions diffused through rust and substrate interface and
played an important role in accelerating corrosion on WS at P1. This corroborates
the corrosion rate (Table 3.2) where corrosion rate on WS is observed almost equal
with MS. In case of other two sites P2 and P3 due to the presence of higher amount
of S, MnS is formed and retard the corrosion rate as found in Table 3.2.
After 42 m of exposure at P1, morphology of MS is layered and chunky as per
Fig. 3.36a. For WS it is an agglomeration of smaller oxides and shows relief effect
with very fine network of cracks on the oxide film as per Fig. 3.36b. Line scans as
shown in Figs. 3.40a and b for WS and Figs. 3.39a and b for MS show identical O2
count. SEM and EDX line scan, thus, justifies the identical corrosion rate of both
MS and WS as reported in Table 2.8.
At P2, rust morphology on MS shows as per Fig. 3.37a a continuous chain of
chunky oxides, whereas WS contains sheet of oxides layer as per Fig. 3.37b. Line
scan on WS surface (Fig. 3.42a, b shows uniform distribution of O2 count. Because
116
of the continuous rust cover on WS corrosion rate is expected to be lower as

confirmed by Table 3.2.
At P3, MS oxides (Fig. 3.38a) look to be comprised of punctured globular
zones. Line scans pattern (Figs. 3.43a, b) confirms this where O2 level is slightly
low at punctured sites with lower O2 count between two globules. In case of WS
rust cover (Fig. 3.38b) appears to be agglomeration of small globular regions.
There are smaller as well as chunky rust products. Line scan (Figs. 3.44a, b) shows
considerable depletion of O2 at certain points. Presence of humidity and SO2 at P2
and P3 helps in the formation of stable and protective rust which have positive
effect in arresting the corrosion during the longer duration of exposures.
Table 3.12 gives elemental analysis of O2, S and Cl by EDX of MS and WS
after 42 m exposure at three sites. It can be seen that S pick up of WS, due to the
presence of alloying elements is more compare to MS; Cl- pick up at P1 is more;
S pick up is maximum at P3, followed by P2 and minimum at P1. In general, O2
levels in both MS and WS are at par. However, O2 pick up for WS was maximum
at P1 and minimum at P2. EDX analysis highlights the role of S on WS exposed at
P3, whereas O2 levels and Cl contents at P1 explain why corrosion rate of MS and
WS is same.
The SEM micrographs of exposed panels also help in differentiating in rust
morphology of MS and WS from site-to-site and effect of increase of exposure
time on the rust formation. A good correlation between corrosion rate and SEM
and EDX has also been established.
Characterisation by XRD and Raman Spectroscopy. Rust formed on uncoated
panels after 18 and 42 m of exposure was analysed by XRD and Raman spectroscopy. The oxide phases were identified after matching with JCPDS data
available for XRD [6, 7] and are given in Tables 4.1, 4.2 and 4.3. Phases identified
by Raman spectroscopy were matched with the published data [814, 15] and is
given in Tables 4.4, 4.5, 4.6 and 4.7.
Substantial difference in the diffraction peak intensities was noted with respect
to steel type, exposure site and time. XRD results are limited due to relatively
small amounts of these compounds, amorphous nature of some rust products and
interference of the base metal (Figs. A 3.9a, b). Low intensity peaks are therefore
difficult to identify. In case of Raman spectroscopy, direct rusted uncoated panels
were taken for study and both crystal and amorphous phases could be identified
very accurately.
After 18 m exposure at P1, XRD detected mostly b FeOOH, a FeOOH and c
FeOOH phases on MS and c FeOOH and b FeOOH on WS. With further exposure
to 42 m, phases found are b FeOOH and c FeOOH on MS and a Fe2O3, c FeOOH,
b FeOOH on WS as shown in Table 4.1 and spectra in Fig. A 3.8.
A few interesting observation can be made from Table 4.1. For the same period
of exposure compared to MS, stable oxides formed on WS and with increase of
exposure time further oxidation takes place. b FeOOH is a porous product and
detected by XRD. Though lots of peaks can be seen in Figs. A 3.8a and b for 18 m
at P1, all these peaks are at very low intensity level and hence no definite inferences can be derived from these. Interestingly, only one or two sharp peaks are
Spacing
d(A)
2.458
2.542
1.511
1.932
1.638
2.474
1.936
3.300
2.361
Intensity
% I/I0(peak)
100(6)
53(5)
43(16)
40(10)
23(12)
100(1)
22(4)
19(5)
37(9)
Experimental data
50
80
40
72
100
76
72
100
20
c FeOOH
c FeOOH
b FeOOH
b FeOOH
Intensity
% I/I0
(peak)
a FeOOH
b FeOOH
b FeOOH
c FeOOH
b FeOOH
Phase
Published data
Intensity
% I/I0 (peak)
18 m Exposure
2.450
100(7)
2.543
81(1)
1.515
63(3)
1.935
50(12)
2.343
48(16)
1.445
25(14)
42 m Exposure
2.473
19(5)
1.935
100(1)
1.735
26(3)
3.311
24(4)
Spacing d(A)
Experimental data
Table 4.1 Rust phases by XRD exposed at P1 for 18 and 42 m of MS and WS

MS
WS
2.599
2.479
1.945
3.304
2.462
6.232
3.29
1.939
1.533
1.64
Spacing
d(A)
a Fe2O3
c eOOH
c eOOH
b FeOOH
c FeOOH
c FeOOH
c FeOOH
c FeOOH
c FeOOH
b FeOOH
Phase
Published data
70
76
53
100
76
61
100
72
33
100
Intensity
% I/I0(peak)
2.591
2.473
1.940
3.311
2.473
6.27
3.294
1.935
1.534
1.635
Spacing
d(A)

117
118
Table 4.2 Rust phases by XRD exposed at P2 for 42 m of WS

Experimental data
Published data
Intensity
% I/I0
(peak)
Spacing
d(A)
100(1)
25(3)
26(2)
14(7)
2.470
1.939
2.583
2.703
Phase
c FeOOH
c FeOOH
a FeOOH
a Fe2O3
Intensity
% I/I0
(peak)
Spacing
d(A)
76
72
12
100
2.473
1.935
2.583
2.700
obtained in Figs. A 3.9a and b for 42 m exposure, the absolute intensities of these
peaks are rather higher around 3,000 counts. For P2 and P3 sites XRD for 18 m
exposure were not done since for P1, it was seen that peaks were of low intensities.
After 42 m exposure, both oxides and hydroxides phase are detected on WS at
P2 and same phases also detected at P3 along with b FeOOH (Tables 4.2 and 4.3)
After 18 m exposure at P1 as per Table 4.4 Raman spectroscopy detected major
peaks c FeOOH (1307, 252, 379) and b FeOOH (675) on MS and c Fe2O3, d
FeOOH (392), c FeOOH (245,1,307) and a Fe2O3 on WS. After 42 m exposure a
FeOOH (386), b FeOOH (386), c FeOOH and c Fe2O3 were formed on MS and c
FeOOH (255, 528), a Fe2O3 and a FeOOH were found on WS (Table 4.4 and
spectra Figs. A 3.12 and A 3.14). Non-appreciable changes in the molecular dipole
moment explain the absence of b FeOOH on WS [15].
After 18 m exposure at P2 as per Table 4.5, Raman spectroscopy could detect a
FeOOH, a Fe2O3 and Fe3O4 on MS and c Fe2O3, a FeOOH and d FeOOH, on WS
and with further exposure to 42 m Phases detected were c FeOOH (255,1307), d
FeOOH (655),a FeOOH (386) and Fe3O4 on WS. The phases detected after 18 m
of exposure on MS by Raman spectroscopy were unstable,a FeOOH and Fe3O4
and the trend continued even with higher exposure. Presence of stable amorphous
d FeOOH and absence of porous b FeOOH on WS shows less corrosion rate than
MS (Table 3.2).
Raman spectroscopy after 18 m exposure at P3 as per Table 4.6 detected only c
FeOOH on MS and a FeOOH,c FeOOH, c Fe2O3 and Fe3O4 on WS. With further
exposure to 42 m phases transformed to c FeOOH, a Fe2O3 (292, 225, 612) and d
FeOOH on MS and a FeOOH, c FeOOH, a Fe2O3 (292, 225, 386) and d FeOOH
(400) detected on WS.
Thus, WS shows better corrosion resistance than MS due to the presence of d
FeOOH phases at P2 and P3. From the foregoing results and discussion on various
types of phases present, it is rather difficult to relate the rate of corrosion on the
panels with the characteristics of the various phases present on panels. But still
there is some evidence on the presence of c FeOOH phases as the major phases
particularly on WS. As the time of exposure is increased as well as the severity of
corrosive environment at P1 to P2 and P3. There is a tendency of b FeOOH get
converted to c FeOOH with time and reduction of severity of corrosive environment. It is to be noted from corrosion rate data (Table 3.2) that, the corrosion rate
a FeOOH
c FeOOH
c FeOOH
a FeOOH
b FeOOH
2.448
1.936
3.309
2.695
2.394
100(1)
53(3)
38(6)
51(4)
16(9)
50
72
100
35
20
2.450
1.935
3.294
2.693
2.343
100(1)
45(6)
71(2)
66(4)
36(8)
Intensity
% I/I0
Phase
Spacing
d(A)
Intensity
% I/I0
Spacing
d(A)
Published data
Experimental data
Published data
Intensity
% I/I0
WS

MS
2.478
3.296
2.548
1.846
1.940
Spacing
d(A)
c FeOOH
c FeOOH
b FeOOH
a Fe2O3
c FeOOH
Phase
76
100
80
40
53
Intensity
% I/I0
Experimental data
2.473
3.294
2.543
1.840
1.944
Spacing
d(A)

119
120
Table 4.4 Phases by Raman Exposed after 18 and 42 m at P1 of MS and WS

18 m
42 m
Panel Raman
shift
(cm-1)
MS
WS
1307.28
251.12
221.48
375.33
695.92
265.37
392.92
241.60
1292.10
491.01
Phase
c FeOOH
c FeOOH
d FeOOH
c FeOOH
b FeOOH
c Fe2O3
d FeOOH
c FeOOH
c FeOOH
a Fe2O3
Published
Raman shift
(cm-1)
1307
252
220
379
675
265
392
245
1307
497
Ref Raman
shift
(cm-1)
Phase
[11]
[11]
[10]
[11]
[12]
[9]
[12]
[24]
[11]
[24]
c Fe2O3
Fe3O4
a FeOOH
b FeOOH
c FeOOH
c FeOOH
a Fe2O3
a FeOOH
c FeOOH
261.22
1327.83
387.60
387.60
651.69
259.88
1314.28
398.12
528.32
Published
Raman shift
(cm-1)
265
1322
386
386
654
255
1320
397
528
Ref.
[9]
[8]
[9]
[12]
[11]
[11]
[9]
[8]
[11]
Bold strongest peak in spectrum, Underline next strongest peek in spectrum
of WS is much less than that of MS under similar condition. The corrosion rate is
decreased with time and decreased in severity of environments due to the presence
of high corrosion resistance phases. In this case, it is the c FeOOH which may be
responsible for decrease in corrosion rate. c FeOOH is a top layer of passive film
as claimed by most investigators [16, 17]. Hence, the formation of stable high
corrosion resistant passive layer may be the reason for substantial mitigation of
corrosion rate when WS is exposed for prolonged time period in corrosive environment of P2 and P3.
4.1.6 Summary of Field Test Findings

The corrosion attack was severest at P1 followed by P2 and P3. Corrosion rate
decreased with time and the rate of corrosion was lowest for WS at sites for all
times. DC measurement on uncoated panels after 42 m exposure showed that
corrosion was lowest at P3 on WS and also corrosion potential was greater. In fact,
corrosion potential for all the cases was greater for WS than MS. Corrosion rates
of MS and WS at P2 were comparable, whereas at P1 corrosion rate of MS was
lower than WS. Apparently, WS is more vulnerable to chloride ions and the
product/rust of WS is also more vulnerable than MS. Again corrosion rate of
coated WS after 18 m exposure was much lower than that of MS at P3, even
the Ecorr was also greater. AC impedance data of uncoated 42 m exposure
corroborated the DC test data and the rust product acquired good resemblance of
coatings which can be compared with paint model.
Effectiveness of rust coatings was tested by EIS and the trend was same i.e., rust
coating developed on WS was more protective than MS. However, panels became
121
Table 4.5 Phases by Raman exposed after 18 and 42 m at P2 of MS and WS

18 m
42 m
Panel Raman
shift
(cm-1)
MS
WS
390.62
298.92
686.51
549.08
267.71
384.40
210.81
1292.87
589.62
Phase
a FeOOH
a Fe2O3
a FeOOH
Fe3O4
c Fe2O3
d FeOOH
a FeOOH
c FeOOH
d FeOOH
Published
Raman
shift
(cm-1)
390
298
685
550
265
385
205
1307
580
Ref. Raman
shift
(cm-1)
Phase
[9]
[14]
[9]
[8]
[9]
[10]
[9]
[11]
c FeOOH
c FeOOH
d FeOOH
a FeOOH
Fe3O4
259.00
1313.83
657.21
387.19
534.80
Published
Raman shift
(cm-1)
Ref.
255
1307
655
386
532
[13]
[13]
[14]
[11]
[24]
Published
Raman shift
(cm-1)
Ref.
Table 4.6 Phases by Raman exposed after 18 and 42 m at P3 of MS and WS

18 m
42 m
Panel Raman
shift
(cm-1)
MS
WS
252.95
654.26
526.92
374.90
1323.19
1425.66
249.21
654.77
374.49
Phase
c
c
c
c
FeOOH
FeOOH
FeOOH
FeOOH
Fe3O4
c Fe2O3
a FeOOH
c FeOOH
c FeOOH
Published
Raman shift
(cm-1)
252
654
528
379
1322
1425
248
654
379
Ref. Raman
shift
(cm-1)
[11]
[11]
[11]
[11]
[8]
[12]
[8]
[11]
[11]
Phase
1305.14 c FeOOH
290.63
a Fe2O3
414.29 d FeOOH
223.93
a Fe2O3
608.43
a Fe2O3
1302.04 c FeOOH
290.81
a Fe2O3
225.87
a Fe2O3
408.20 d FeOOH
484.20 a FeOOH
1307
292
400
225
612
1307
292
225
400
485
[11]
[14]
[14]
[9]
[14]
[11]
[14]
[9]
[12]
[8]
less corrosion resistant after 42 m exposure. After 18 m of exposure coated panels

exhibited almost identical behaviour. The models approximated to CPE were in
diffusion. Scribed coating panel was found inferior with respect to unscribed
coated but here again WS provided better protection than MS substrate. SO2 dosed
panels showed improvement in corrosion resistant property of both MS and WS.
However, the improvement was more for ZP ? WS (marginally)
Presence of humidity and SO2 at P2 and P3 helped in the formation of stable
and protective rust on WS which was not the case for WS at P1. EDX analysis
after 42 m indicated the role of S on WS exposed at P3, whereas oxygen and
chloride contents (Table 3.12) at P1 made corrosion rate of MS and WS same.
XRD of WS and MS after 42 m exposure at P2 and P3 revealed no differentiating or clinching informations. All these showed maximum intensities for c
122
Table 4.7 Rust phases by Raman spectroscopy of as received MS and WS

MS
WS
Experimental data
Published data
Raman shift
cm-1
Raman shift
cm-1
1321.37
671.52
613.34
Phase
Fe3O4
c Fe2O3
a Fe2O3
1322
670
612
Experimental data
Ref. Raman shift
cm-1
Phase
[8]
[9]
[14]
a Fe2O3
c FeOOH
a Fe2O3
1314.79
658.79
611.79
Published data
Raman shift
cm-1
1320
660
612
Ref.
[9]
[13]
[14]
FeOOH line for WS panels and maximum intensity for a FeOOH for MS, while
XRD of WS at P2 showed only one strong line, XRDs at P3 showed b FeOOH line
as second strongest line for WS and c FeOOH as second strongest line for MS.
From XRD plots of peaks could be seen at P1 after 18 m exposure, but these
were at very low intensity level. XRD of WS and MS after 42 m exposure at P2
and P3 revealed no informations. Raman spectroscopy on rust products at P2 and
P3 after 18 and 42 m exposure showed the increase of higher oxidation state with
increased in exposure time and from MS to WS as was found at P1. Another
interesting observation that could be seen was that after 42 m exposure two lines
for a Fe2O3 and one line of a FeOOH were found which showed oxidation was
highest at P3 on WS in the presence of SO2. In fact, for MS also three a Fe2O3
lines were found. The difference between these two was that of sharp peaks that
are noticed for WS (Figs. A 3.17a, b).
4.2 Accelerated Laboratory Test

4.2.1 Humidity Test
In humidity test, rust strain appeared on the scribed line on coated panels only after
2,000 h of exposure while uncoated panels showed appearance of rusting after
1,000 h on both MS and WS. Finally after 4,500 h, the corrosion rate for WS
(334 lm/y) was found lower than MS (382 lm/y) as per Table 3.19. Due to the
presence of minor alloying elements, the rust on WS seemed to be more compact in
nature than MS. Corrosion rate of WS at all exposures was lower with respect to MS.
4.2.2 Salt Spray Test

In salt spray test rust appeared on the scribed lines of coated panels with progressive exposure of 1,0004,500 h. After 2,000 h, rust appeared on the scribed
lines (\0.5 mm) on coated WS and increased ([1.5 mm) with 4,500 h exposure.
123
Fig. 4.2 Corrosion rate in

humidity and salt spray test
Humidity MS
Humidity WS
SaltSpray MS
Salt Spray WS
450
925
400
825
375
725
350
625
325
500
CR Salt Spray (m/yr)
CR Humidity (m/yr)
1025
425
525
1500
2500
3500
4500
Exposure Time (Hour)
In case of MS less rust developed with respect to earlier (Table 3.20). On uncoated
MS and WS, the nature of rust developed after 1,000and 4,500 h exposures is
shown in Figs. 3.44a and b and Figs. 3.45a and b, respectively. From the figures it
can be seen that the severity of pitting of surface is more in WS. Thus, the rust
spreading across the scribe lines started developing faster for coated WS than MS
and after 4,500 h exposure the corrosion rate of WS (546 lm/y) was found marginally less than MS (617 lm/y) as shown in Table 3.20. Here also the alloying
elements in WS helped in forming compact and adhering oxides which eventually
helped to retard the rate of corrosion.

The combined trends of corrosion loss in both tests are shown (Fig. 4.2). The
penetration rate in humidity tests for MS was higher for entire exposure due to the
formation of unstable and non-adherent phases which allowed continuous exposure of fresh surface. In case of WS, compact, adherent stable phases were
developed with time which retarded the corrosion.
In salt spray test, the corrosion rate was again higher for MS but the difference
with WS was not much. Perhaps this was due to continuous formation of nonprotective rust in NaCl on both MS and WS.

Initial variation of open circuit potential (OCP) developed at steel/electrolyte
interface on as received uncoated panel of WS in 3.5 % NaCl showed more
negative potential with respect to MS indicating higher corrosion tendency.
However, with further exposure the trend got reversed. Corrosion potential moved
to greater side as the oxides of these elements formed sparingly soluble protective
124
corrosion product as shown (Figs. 3.46a, b). This may be due to the fact that WS
starts forming oxides which fall in the passive region above pH 9 [18] and may be
due to the presence of Cr, Mn as is evidenced from polarisation data (Table 3.22).
Polarisation diagram of as received panels in 3.5 % NaCl indicated higher Icorr
and Ecorr for MS compared to WS with no indication of passivation (Figs. 3.47a, b)
and data Table 2.36. Same trend was also found in 0.1 M Na2SO4 ? 0.1 N NaCl,
where Icorr was higher for MS and lower for WS (Figs. 3.48a, b). Here, WS
showed signs of passivation but could not be stabilised due to high scan rate. The
lower corrosion rate of WS is due to formation of protective and stable oxides in
presence of alloying elements.
Cyclic polarisation was carried out in SAEJ solution on MS and WS panels that
were exposed in humid SO2 for 9 m (Fig. 3.49 and data Table 3.23). Rust formed
on MS and WS exhibited almost identical anodic and cathodic plots but their Icorr
values were quite different. The trend of WS and MS in NaCl and Na2SO4 +NaCl
solution was seen in SAEJ solution also. Interestingly, the passivity was seen in
both the panels which were corroborated by the reverse cycle where current
density surges by an order of 10. This proves that humid SO2 exposure for longer
period helped in developing protective oxide.
4.2.5 AC Impedance
EIS was carried out on uncoated as received panels.WS exhibited different
behaviour than MS at lower frequency ranges in 0.1 M Na2SO4 ? 0.1 N NaCl.
WS showed higher Rp ([105) and lower Yo (10-12) resulting much higher
impedance than MS (Figs. 3.50, 3.51). While both MS and WS exhibited wide
variation in phase angles (-1 to -50), CPE model was found suitable for both
the panels. The electrical parameters are given in Table 3.24 and typical equivalent circuit for this model is given in Fig. 1.11 in the literature review section. This
higher impedance on WS is due to the effect of electrolyte and formation of stable
oxides as was already found in polarisation diagram.
EIS was carried out on uncoated panels exposed in SO2 for 9 m and the Nyquist
plot of MS and WS is given in Fig. 3.52 and parameters are given in Table 3.25.
WS showed higher (Cf: 2.67 9 10-8, Rp: 6.01 9 104) than MS (Cf: 1.33 9 10-8,
Rp: 3.37 9 103). This indicates the formation of protective and compact rust on
WS. Randle model was used to get the electrical elements and equivalent. The
depressed semicircle diameter of WS was found higher than MS. This behaviour
indicates the presence of diffusion resistant layer at the electrolyte/steel interface.
In case of MS, the real component of resistance was found to reduce with decrease
in frequency, a phenomenon attributed to inductive behaviour of the electrolyte/
steel interface. The availability of SO2 reduced dissolution in WS. Presence of
alloying elements in WS may be attributed as the reason.
EIS was carried out on different coating systems on MS substrate (ZP ? MS and
ZR ? MS) in 3.5 % NaCl and after 24 h exposure, Rp([103) values were almost
125
identical with same phase angle (90) as seen from Bode Plot. Nyquist plot
indicated two time constant for ZP ? MS (Fig. 3.53) and one time constant for
ZR ? MS (Fig. 3.55) and parameters are given in Table 3.27. After 168 h exposure it showed higher Rp(53 9 103) of ZR ? MS than ZP ? MS(Rp:1.42 9 103)
indicating comparative better resistance against charge transfer through coating
pores (Figs. 3.54 and 3.56). In Bode plot high Rp in the low frequency limit and the
phase angle remained close to 010 representing inductive behaviour.
EIS was carried out on different coating system on WS substrate (ZP ? WS and
ZR ? WS) in SAEJ solution with higher pH. No damage of coating was found
with initial exposure and after 1,500 h of exposure in electrolyte, ZR ? WS
revealed higher Rp(1.57 9 109). CPE with diffusion model fit was found suitable
and admittance (Y0: [109) was almost equal with ZP ? WS as per Figs. 3.57 and
3.58. The corresponding parameters are given in Table 3.28. Probably, the accumulation of corrosion products in the pores of coatings on ZR ? WS favoured the
appearance of straight line diffusion tail at low frequency of about 45 slope. In
this case due to WS substrate, Rp increased in both the cases and passive/protective
oxides formed on WS.
4.2.6 Rust Characterisation

Characterisation by SEM and EDX. The SEM micrograph of as received MS and
WS is given (Figs. 3.59 a and b). Formation of oxides in the shape of lumps is seen
on MS, whereas thin and uniform oxides were found for WS. Surface of MS rust
layer was rougher and contained more cavities compared with that of WS rust
layer. Line scanning pattern (Figs. 3.60a, b) for MS and (Figs. 3.61a, b) for WS
indicated higher oxygen level for MS than WS (elemental analysis by EDX
Table 3.29). The line scan pattern on WS showed uniform distribution of Fe and
O2 and corroborated the findings that a thin uniform oxide had formed on it.
In MS, there were regions of high O2 and Fe concentration indicating formation
of lumpy oxides at certain regions and regions of Fe rich zones on the surface.
Based on the above results, higher corrosion rate of MS can be expected and same
was found from polarisation diagram.
Rust formed on MS and WS panels was exposed in humid SO2 for 9 m and were
analysed by SEM and EDX on the exposed surface and cross section of the panels
(Figs. 3.62 and 3.63 and Table 3.30). Rust/steel interface of WS was found to be continuous and uniform than MS. The latter showed abrupt crystalline layer at the interface.
In upper layer on WS rust, Mn and in inner layer Cr, Cu, Ni were found. The alloying
elements are wined away from the surface. However, these were intact at the interior
layer and made rust more compact and impervious to corroding electrolyte to penetrate
further. No traces of Cu, Cr, etc. were recorded in the upper layer of this rust.
SEM micrographs revealed that because of presence of minor alloying element,
O2 residence on WS was more, uniform and continuous as compared to MS. The
effect was even more when the panels were exposed to SO2 atmosphere.
126
Table 4.8 Rust phases by Raman spectroscopy in humidity of MS and WS

Exposed MS
Hour
Experimental data
Raman
shift
(cm-1)
2000
3500
4500
WS
1318.63
296.26
411.02
612.66
503.87
292.39
227.29
612.30
404.00
293.22
226.41
403.60
1300.35
Phase
a Fe2O3
d FeOOH
a Fe2O3
a Fe2O3
c Fe2O3
a Fe2O3
a Fe2O3
a Fe2O3
d FeOOH
a Fe2O3
a Fe2O3
d FeOOH
c FeOOH
Published data
Raman
shift
(cm-1)
1320
297
411
612
505
292
227
612
400
293
226
400
1307
Experimental data
Published data
Ref. Raman
shift
(cm-1)
Phase
[9]
[12]
[14]
[14]
[12]
[14]
[14]
[14]
[12]
1322.58
290.84
226.86
407.91
1320.71
299.38
411.27
657.44
Fe3O4
a Fe2O3
a Fe2O3
a Fe2O3
a Fe2O3
a Fe2O3
a Fe2O3
d FeOOH
1322
292
226
411
1320
298
411
655
[8]
[14]
[14]
[14]
[9]
[14]
[14]
[12]
254.72
1305.91
657.20
378.76
c
c
d
c
255
1307
655
379
[11]
[11]
[12]
[11]
[14]
[14]
[12]
[11]
FeOOH
FeOOH
FeOOH
FeOOH
Raman
shift
(cm-1)
Ref.
Characterisation by Raman Spectroscopy. Rust morphology was analysed

with Raman shift obtained from experiments on as received MS and WS as per
Table 3.31. The experimental shift values were compared with those available
in the literature and given in Table 4.7. While phases found are metastable c
Fe2O3, stable Fe3O4 and a Fe2O3 on MS, less stable c FeOOH and stable
compact a Fe2O3 were found on WS. The finding is in accordance with as
reported by Sam [19].
Rusts formed in humidity test (Table 3.32) was analysed by Raman spectroscopy. The corresponding literature matching is also given in Table 4.8. Higher
amount of stable and protective oxide phases like a Fe2O3 and c Fe2O3 were
detected after 2,000 h exposure on MS. With further exposure amorphous d
FeOOH, metastable c FeOOH and a Fe2O3 were detected. On WS for same set of
experiments, the presence of stable and protective phase a Fe2O3 along with stable
Fe3O4 were detected. With increase in exposure time, with continuous humidity,
c FeOOH and d FeOOH were formed. Hence, the stability of rust diminishes when
exposure time is too long along with continuous high humidity, leading to formation of FeOOH. No oxide phases are observed on WS as they reduced to
hydroxides for long time continuous exposure in presence of high humidity.
Rust formed in Salt Spray test (Table 3.33) was also subjected to Raman
spectroscopy and phases are identified as given in Table 4.9. Stable a Fe2O3 along
with metastable c Fe2O3 were detected as major phases on MS after 2,000 h
exposure. With further exposure stable a Fe2O3 disappeared and metastable c
Fe2O3 and a FeOOH were detected. The oxide phases are more stable and
127
Table 4.9 Rust phases by Raman spectroscopy in salt spray of MS and WS

Exposed MS
WS
Hour
2000
3500
4500
Experimental data
Published data
Raman shift
(cm-1)
Raman
shift
(cm-1)
1321.17
233.54
415.69
300.15
617.23
507.09
297.41
413.02
613.33
500.05
1310.81
387.31
584.83
211.72
273.34
Phase
a Fe2O3
a Fe2O3
a Fe2O3
c Fe2O3
Fe3O4
c Fe2O3
d FeOOH
a Fe2O3
a Fe2O3
a Fe2O3
c FeOOH
a FeOOH
a FeOOH
c Fe2O3
1320
227
415
300
616
505
297
414
612
500
1307
386
205
265
Experimental data
Published
data
Ref.
Raman shift
(cm-1)
Raman
shift
(cm-1)
Ref.
[9]
[14]
[11]
[9]
[14]
[12]
[12]
[14]
[612]
[9]
[11]
[9]
[9]
[9]
1350.80
294.54
294.54
227.46
410.23
384.34
533.67
648.37
258.85
1292.5
387.46
1298
212.50
274.88
1322
295
295
227
411
385
532
648
255
1307
386
1307
219
265
[10]
[10]
[9]
14]
[14]
[10]
[14]
[11]
[11]
[11]
[9]
[13]
[11]
[9]
Phase
Fe3O4
d FeOOH
a Fe2O3
a Fe2O3
a Fe2O3
b FeOOH
Fe3O4
c FeOOH
c FeOOH
c FeOOH
a FeOOH
c FeOOH
c FeOOH
c Fe2O3
protective than a FeOOH [20, 21]. On WS presence of stable oxide phases a Fe2O3
along with Fe3O4 were detected after 2,000 h exposure. Additionally unstable and
non-protective, porous b FeOOH was detected which is less dense than a FeOOH
and c FeOOH was found with further exposure. With further increase in exposure
time to 4,500 h metastable c FeOOH, c Fe2O3 and a FeOOH were detected.
Thus, the corrosion product is mostly a Fe2O3 and c Fe2O3 on MS and does not
change with exposure time, whereas a Fe2O3, a FeOOH and c FeOOH formed on
WS. Overall the percentage of oxy hydroxide phases increased with time.
Raman spectroscopy, thus, revealed that stable oxides formed on WS as
compared to MS when testing was done on as received panels and panels exposed
to extended salt spray test. However, in continuous humidity tests oxides
decomposed to form less stable FeOOH.
4.2.7 Summary of Accelerated Laboratory Test Findings

Corrosion rate of WS at all exposures was lower with respect to MS in continuous
humidity conditions and a trend toward lower corrosion rate with time was
noticed. In salt spray environment, WS is better initially and after longer exposure
MS and WS both gave almost same corrosion rate. Pitting was more on WS, which
128
healed up with exposure. In DC corrosion test, WS showed signs of passivation

and also lower corrosion rate. WS exposed in humid SO2 exposure helped in
developing protective oxide. WS exhibited different behaviour than MS in 0.1 M
Na2SO4 ? 0.1 N NaCl in EIS and higher impedance on WS was observed. EIS on
WS after exposure in SO2 indicated higher impedance than MS. EIS on ZP ? MS
and ZR ? MS showed almost identical impedance and 90 phase angle. SEM
micrographs reveal, due to alloying element, oxygen residence on WS was more
uniform and continuous as compared to MS. The effect was even more when the
panels were exposed to SO2 atmosphere. Raman spectroscopy revealed that
compared to MS stable oxides formed on WS when testing was done on as
received panels and salt spray test.
4.3 Comparison of Field and Laboratory Tests

It was tried to recreate the parameters of field tests at the accelerated laboratory
tests where the effect of humidity, chloride ions and presence of SO2 were considered. Comparison of Table 3.22 (lab test) and 3.3 (field test) reveals that the
corrosion rate of as received MS and WS in 3.5 % NaCl is lower than the exposed
panels. Ecorr of as received panels are also greater. However, corrosion rate in
cyclic polarisation of MS and WS after SO2 exposure is comparable to cyclic
polarisation of coated panels after 18 m exposed at P3 as can be seen from
Tables 3.23 and 3.4 and Figs. 3.39 and 3.8. Both plots show passivity. In general,
the trend WS is more corrosion resistant than MS as seen in both field and laboratory tests.
The corrosion rate of MS and WS as obtained from humidity and salt spray tests
(Table 3.19 and 3.20 respectively) when compared with atmospheric corrosion
rate at three sites with different exposures (Table 3.2) show wide gap. This is
expected, because at laboratory the exposure is more severe and continuous
compared to atmospheric conditions that prevail at the three sites.
Comparison of Tables 3.24 and 3.25 (lab test) with Table 3.5 (field test) and
Tables 3.27 and 3.28 (lab test) with Table 3.7 (field test) clearly demonstrates that
laboratory test data no way compares with field test data.
However, Raman data of rust formed on field test somewhat matches with lab
test data. The corrosion products i.e. oxides/hydroxides formed in both the cases are
somewhat similar. Since the other corrosion properties of field and laboratory test
are widely different as discussed earlier, it means the extent of oxides/hydroxides
formed during the field tests and their morphology are widely different than that
obtained from laboratory test. That the morphology of the rust product of field tests
are widely different than as received (lab test) is clearly seen when Figs. 3.333.38
(field test) are compared with Fig. 3.59 (lab test). Presence of SO2 at P3 was found
to improve the corrosion resistance of WS. It was also demonstrated that SO2 helps
in formation of passive layer on WS (Figs. 3.8 and 3.49). Hence, as received panels
were subjected to SO2 atmosphere (Figs. 3.62 and 3.63) and consequently SEM of
4.3 Comparison of Field and Laboratory Tests
129
cross section of exposed panels was done. There was distinct resemblance of these
SEM with Fig. 3.38 of field test panels exposed at P3.
In summary, it can be said that the data generated with laboratory tests, though
predict the trend of field tests, are grossly insufficient to predict field behaviour of
WS and MS in absolute terms. Since SO2 is a factor which gives some convergence at some points of two types of tests, separate set of experiments was contemplated to simulate similar field conditions at laboratory.
4.4 Simulation of Field Test Conditions at Laboratory

A few simulative studies addressing the processes occurring on the steel surface
that cause atmospheric corrosion at sites have been carried out and mechanism of
rusting process of WS and MS has been proposed. The aim is to develop an early
formation of protective rust on WS by applying various surface treatments using
electrolytes and suggesting ways to improve the weathering characteristics.
4.4.1 Hypothesis
E-pH diagram [22] of iron in Na2SO4 shows some passive regions. Raman
spectroscopy of 42 m exposed WS panels revealed a few common oxide and
hydroxide phases as reported in E-pH diagram. It was thought that long field
exposures of WS in some sites lead to the formation of these oxides/hydroxides
phases due to slow interaction with environments. It was envisaged to simulate the
slow oxidation process in laboratory by subjecting WS panels to very slow
potentiodynamic scanning rate. Also laboratory test on WS exposed in humid SO2
for long duration indicated passivation behaviour in cyclic polarisation. Some
experiments were designed to investigate the behaviour by slow polarisation and to
see whether such oxides/hydroxides phases at all formed or not.
4.4.2 Experimental Methods and Results

Atmospheric corrosion of steels is very complex and is a function of interplay of
various factors like electrolytes; pH and potentiodynamic scan rates and so on. In
order to emulate these factors to form protective rust a test plan is given in
Table 4.10.
Potentiodynamic tests were initially carried out on WS in 0.1 M Na2SO4
+0.1 N NaCl having pH 9.5 at scan rate 0.005 mV/s but no sign of passive region
was found as per Fig. 4.3 where corrosion rate is very low (Icorr : 32 qA/cm2).
130
While polarising WS panels in 0.1 M Na2SO4 at pH 8.8 with scan rate 0.5 mV/
s, it was noticed that a passive region forms with current density of about 10 mA/
cm2 as per Fig. 4.4a where Icorr is found 20 lm/cm2. Repeating the experiment
with scan rate 0.01 mV/s as per Fig. 4.4b, revealed passivity at 1 mA/cm2 with a
tendency to passivate even at lower current density. The experiment was repeated
with even slower scan rate of 0.005 mV/s and it was found that the polarisation
curve yielded a well defined passive region like stainless steel as shown in
Fig. 4.4c.
Encouraged with this result, polarisation of WS was done in 0.1 M Na2SO4 at
pH 9.2 with a scan rate of 0.005 mV/sec. Passivity was obtained and with a current
density of about 500 lm/cm2 and the range continued beyond 850 mV as shown
in Fig. 4.5. Since it has been reported in the literature [23] that very small addition
of Cl- ions enhance passive range, the polarisation experiment of WS was repeated in 0.1 M Na2SO4 ? 0.001 M NaCl at pH 8.6 with 0.005 mV/sec scan rate.
Passivity range was indeed found to be enhanced as indicated in Fig. 4.6.
From Fig. 4.4c passive potentials of -530 and -520 mV, respectively, were
selected and at these potentials in 0.1 M Na2SO4 at pH 8.6, WS panels were
passivated by applying these potentials instantaneously. The respective current
transient plots are shown in Figs. 4.7a, b.
It was noticed that, although in the polarisation diagram passivity is maintained
at 530 and 520 mV, respectively, potentiostatic current transient plots show
very high positive current densities of 21 mA/cm2 and 6.5 mA/cm2, respectively.
However, when the panels were passivated in same solutions by applying
instantaneously potentials of 820 and 800 mV, current densities of 1.6 and
6.25 mA/cm2 were, respectively, obtained as shown in Figs. 4.8a, b.
These curves imply that the panels got into the cathodic regions and got stabilised. The experiments of instantaneous passivity of WS were repeated in 0.1 M
Na2SO4 ? 0.001 N NaCl at pH 8.6 solution at 1100 mV. Here also the panels got
into cathodic region and stabilised at 5.6 mA/cm2 as per Fig. 4.9.
WS passivated as shown in Figs. 4.8a, b and Fig. 4.9 were tested for corrosion
resistance in three types of electrolytes viz. neutral, alkaline and acidic medium.
The details of results are given in Table 4.11 and polarisation diagrams are shown
in Figs. 4.10, 4.11, 4.12, 4.13, 4.14, 4.15.
It is found that, the corrosion rate of WS in different electrolytes of varied pH
level is of lA/cm2 order.
SEM of panels, passivated at -820 and -800 mV, are shown in Figs. 4.16 and
4.17 respectively, at two magnifications. Elemental analysis of WS passivated at
820 mV is shown in Table 4.12. Similarly, Raman spectroscopy of these two
panels passivated at -820 and -800 mV was also done. It was found that
820 mV passivated panels compared very well with rust phases found at P3 while
800 mV passivated did not compare well including Raman Characteristics at P3.
Raman spectroscopy at two spots is shown in Fig. A 4.1 and A 4.2 and phases
identified are given in Table 4.13 where peak values of Fe3O4 and a Fe2O3 were
matched with published data. Raman shift obtained at passivated potential
800 mV with published data were also found matching. In case of 820 mV
131
Table 4.10 Test plan for rust simulation on polished WS

0.1 M Na2SO4
Electrolyte
0.1 M
Na2SO4
+ 0.1 N
NaCl
pH#
Scan mV/s
Potentiodynamic
Polarisation
Potentiostatic Current
Transient(PSCT)
Scanning electron microscopy
Raman spectroscopy
after Passivation (PDP)
9.5
0.005
H
8.8 8.8 8.8

0.5 0.01 0.005
H H
H
0.1 M
Na2SO4
0.1 M
Na2SO4 ? 0.001 N
NaCl
9.2
0.005
H
8.6
0.005
H
H
H
H
H
H
H
H Test performed, # pH adjusted with NaOH, * PDP PD conducted after passivation
Fig. 4.3 Polarisation

diagram of WS in 0.1 M
Na2SO4 ? 0.1 N NaCl
solution (pH 9.5, scan
0.005 mV/s)
stable oxide phase a Fe2O3 is very near to the strongest peak 292 and next to
strongest peak 226 [24]. Hence, at -820 mV stable and protective phases are
formed which are absent at -800 mV.
Similar passivation experiments at slow scan rate were carried out on MS
panels in identical solutions also as per test plan given in Table 4.11. Result
compared with corresponding field exposed panels. Polarisation curves are given
in Figs. 4.18, 4.19, 4.20, 4.21, 4.22.
Stable passivation is not found at pH 9.5 while polarisation carried out in 0.1 M
Na2SO4 ? 0.1 N NaCl gave very low Icorr (2.1 lm/cm2) as shown in Fig. 4.18.
However, passivation was found at pH 9.2 at higher potential with high current
density (approx 5 mA/cm2) as shown in both Figs. 4.19 and 4.20. Passivation did
not sustain for longer period at lower scan rate.
Potentiostatic current transient (PSCT) plots as shown in Figs. 4.21 and 4.22
showed higher current densities of 9 mA/cm2 and 40 mA/cm2 when passivated at
1146 and 1610 mV, respectively.
132
Na2SO4 solution. (pH 8.8,
scan (a) 0.5,(b) 0.01 and
(c) 0.005 mV/s)

Na2SO4 solution (pH 9.2,
scan rate 0.005 mV/s)
133

Na2SO4 +0.001 N. NaCl
solution (pH 8.6, scan e
0.005 mV/s)
Fig. 4.7 PSCT of WS in

0.1 M Na2SO4 solution (pH
9.2). Passivation potential at
(a) -530 (b) -520 mV
The selected potentials for instantaneous passivation were within the passivation range and in the low passive current density region. SEM of MS as per
Figs. 4.23 and 4.24 shows many cracks and voids in the rust layer.
Raman spectroscopy was carried out at two spots and unstable a FeOOH and
porous b. FeOOH phases were detected at instantaneous passivation potential
1146 mV. Strongest peak of 252 was found for metastable c FeOOH at passivation
potential 1609 mV. Raman spectra are given in Figs. A 4.3 and A 4.4 and detail
of phases is given in Table 3.12. Thus, it is seen that results with MS were not at
all encouraging.
134

0.1 M Na2SO4 solution (pH
9.2). Passivation Potential at
(a) -820 (b) -800 mV

0.1 M Na2SO4 +0.001 N
NaCl Solution. (pH 8.6)
Passivation Potential at
-1100 mV
4.4.3 Discussions
The corrosion rate of instantaneously passivated WS panels is shown in Table 4.12
compares well with corrosion rate of 42 m field exposure on WS panels in 3.5 %
NaCl as given in Table 3.2. Corrosion rate on WS panels was also almost similar
range and order when tested in SAEJ and as well as NACE solution. It was
interesting to note that corrosion rate was low in SAEJ solution when passivated in
0.1 M Na2SO4 +0.001 N NaCl solution. Very low amount of chloride ions in the
solution perhaps consolidate the passive layer resulting in lower corrosion rate. A
fact which agrees well with similar data that reported chloride ions leads to a shift
of the repassivation potential toward negative side and the sulphates in 0.1 M
135
Table 4.11 Test plan for rust simulation on polished MS

Electrolyte
0.1 M Na2SO4 0.1 M
+
Na2SO4
0.1 N NaCl
pH#
Scan mV/s
Potentiodynamic
Polarisation
Potentiostatic current transient(PSCT)
Scanning electron microscopy
Raman spectroscopy
Potentiodynamic polarisation after passivation
(PDP)
0.1 M Na2SO4
+
0.001 N NaCl
9.5
0.005
H
9.2
0.005
H
8.6
0.005
H
H
H
H
H
H
H
H
H
# pH adjusted with NaOH, H Test performed

diagram in 3.5 % NaCl
solution (pH 6.6) of
passivated. WS in 0.1 M
Na2SO4 at potential (a) -820
(b) -800 mV
Na2SO4 with chlorides broaches the passive region and leads to lower Icorr. This is
explained by the fact that sulphate ions are adsorbed preferentially on the surface
of the steels, promoting retention of the passive condition [25]. Again the corrosion rate was in the same range as with three field exposure sites viz. P1, P2 and
P3, SEM of instantaneously passivated WS panels as given in Figs. 4.16 and 4.17
also compared well with the SEM on WS exposed at P3 for 42 m as per Fig. 3.38.
There is no comparison of similar SEM of either P1 or P2 as given in Figs. 3.36
and 3.37, respectively.
136
diagram in 3.5 % NaCl
solution (pH 6.6) of
passivated WS. in 0.1 M
Na2SO4 ? 0.001 N NaCl
at potential -1100 mV

diagram in SAEJ solution
(pH 9.1) of passivated WS. in
0.1 M Na2SO4 at Potential
(a) 820 and (b) 800 mV

diagram in SAEJ solution
(pH 9.1) of passivated WS. in
0.1 M Na2SO4 ? 0.001 N
NaCl at Potential -1100 mV
137

diagram in NACE solution
(pH 3.5) of passivated. WS in
0.1 M Na2SO4 at Potential
-800 mV

diagram of WS in NACE
solution after passivated at
potential -1,100 mV in
0.1 M Na2SO4 ? 0.001 N
NaCl
Fig. 4.16
SEM of WS passivation at -820 mV in 0.1 M Na2SO4
The elemental values of rust formed as per Table 4.13 and Fig. A 4.5 on WS
passivated at 820 mV are very much similar to rust formed on WS exposed at P3
for 42 m Table 3.12 where Cr and Cu are also detected in 42 m. The EDX analysis
of exposed WS showed presence of alloying elements in the rust layer.
Since the morphology of the simulated rust compared well with P3, 42 m
exposure panels, corresponding comparison was done with Raman spectroscopy as
well. Raman characterisation on WS panels passivated at 820 mV showed the
presence of c FeOOH along with Fe3O4 and a Fe2O3 as shown in Table 4.14,
whereas for the panels passivated 800 mV only Fe3O4 and a Fe2O3 phases were
138
Fig. 4.17
SEM of WS passivation at -800 mV in 0.1 M Na2SO4
Table 4.12 Corrosion rate (Icorr) and Ecorr values of passivated WS in different electrolyte
Ecorr
Electrolyte
pH Passivation details
Passivation potential, Icorr
lA/
mV
mV
cm2
3.5 % NaCl
solution
SAEJ solution
NACE solutiona
6.6
9.1
3.5
0.1 M Na2SO4
-820
64.81
-800
53.28
-1100
75.15
-820
65.99
-800
81.91
0.1 M Na2SO4 ? 0.001 N

NaCl
0.1 M Na2SO4
-1100
11.92
-800
57.75
0.1 M Na2SO4 ? 0.001 N

NaCl
-1100
207.5
0.1 M Na2SO4 ? 0.001 N

NaCl
0.1 M Na2SO4
845.2
822.3
537.5
723.7
759.7
572.7
713.0
584.0
(25 mg NaCl/lit ? 2.4 ml glacial CH3COOH ? Buffer pH 7 tablet 6 nos.)
Table 4.13 Elemental analysis of surface treated WS passivated at 820 mV

Element
At %
Element
O
S
P
52.25
0.86
1.15
Cr
Cu
At %
0.51
1.71
found. When experimental Raman spectroscopy as reported in Table 4.14 is

compared with WS exposed at P3 for 42 m the similarities are remarkable for
panels passivated at 820 mV. Since not only c FeOOH and a Fe2O3 are common,
139
Fig. 4.18 Polarisation diagram of MS in 0.1 M Na2SO4 ? 0.1 N NaCl. solution (pH 9.5, scan
0.005 mV/s)
diagram of MS in 0.1 M
Na2SO4 ? 0.1 N NaCl
solution (pH 9.5, scan
0.005 mV/s)
Fig. 4.20 PD in 0.1 M

Na2SO4 ? 0.001 N NaCl
solution (pH 8.6) of MS at
Scan 0.005 mV/s
the experimental Raman shifts of 1311.11 for c FeOOH and 292.65, 224,39 for
a Fe2O3 are in agreement with 1302.04 (c FeOOH), 290.81 (a Fe2O3) and 225.87
(a Fe2O3), respectively, Raman shift of WS exposed for 42 m. The oxide phases
are more stable than the oxy hydroxide, which lowers Icorr, at 820 mV than that at
800 mV and panel becomes greater when passivated at 820 mV. It is also due to
the stable oxide phases (a Fe2O3, Fe3O4) than the less stable oxy-hydroxide phases
(c FeOOH), (Table 4.15).
140
Fig. 4.21 PSCT plot of MS

in 0.1 M Na2SO4 solution
(pH 8.8) at passivation
potential -1146 mV
Fig. 4.22 PSCT plot of MS

in 0.1 M Na2SO4 ? 0.001 N
NaCl solution (pH 8.6)
passivation at potential
-1610 mV
Interestingly, Raman shift of 800 mV passivated panels do not compare well

with 42 m field test panels although c FeOOH and a Fe2O3 phases are common. It,
thus, can be conclusively said that laboratory simulation by passivating WS at
820 mV for 80 min, is almost similar with the oxidation state of 42 m field
exposed panels at P3. It needs to be explained here why P3, compares well with
laboratory simulation rather than P1 and P2. Laboratory simulation done in 0.1 M
Na2SO4 which yields SO2 as available in the environment of Jamshedpur (P3).
Similar exercise with simulation on MS and comparison with P1, P2, and P3
drew blank since neither the corrosion rate nor SEM compared with each other.
Obviously, there was no similarity between Raman spectroscopy of passivated MS
panels with those of field exposed panels at P1, P2 and P3 (Figs. 3.36, 3.37 and
3.38). Rust layers formed on MS have shown voids, cracks and irregularities which
are acting as the transferring path for the electrolyte and has loose structure like
MS. High current density of 1040 mA/cm2 promotes the rapid oxidation. For MS,
Raman spectroscopy confirms the presence of metastable and porous phases.
Field exposure is a slow oxidation exposure process which has been imitated at
the laboratory by slow scanning. During instantaneous passivation current gradually decreased and reached a steady-state value within 2 kilosec, but passivation
was continued till 5 kilosec before taking it to corrosion rate measurement in three
different media (820 mV) as given in Fig. 4.8a. The panel passivated at 800 mV
reached steady state with 500 s and passivation was continuous till 1.8 ksec. The
difference in Raman shift of 820 mV passivated and 800 mV passivated can
Fig. 4.23
SEM of MS passivation at -1,160 mV in 0.1 M Na2SO4 ? 0.001 N NaCl
Fig. 4.24
SEM of MS passivation at -1,610 mV in 0.1 M Na2SO4 ? 0.1 N NaCl
141
perhaps be explained by the difference of time of passivation other than the oxidation potential obtainable at 820 mV and 800 mV, respectively. This opens up
further scope of work i.e. by varying the time of passivation at 820 mV, perhaps
it is possible to form more impervious, compact and stable oxides which will even
give much lower corrosion rate. This idea has tremendous industrial potential.
4.5 Conclusion
From the foregoing series of experimental results and analysis the following
statements can be made.
Among the three field exposure sites, the corrosion rate was found to be highest
at P1 for both weathering as well as mild steels due to the proximity of sea
having high chloride ions in the corrosive environment. The corrosion rate of
mild steels is higher than weathering steels for all the sites.
142
Table 4.14 Rust phases by Raman at pH 8.8 on WS passivated at 820 and 800 mV
820 mV
800 mV
Spot Experimental
Raman
shift
(cm-1)
1
2
1323.64
225.51
1311.11
Published
Phase
Raman shift
(cm-1)
Experimental
Published
Ref. Raman shift

(cm-1)
Phase
Raman
shift
(cm-1)
Ref.
1322
226
1307
[12]
[17]
[15]
1321.75
415.21
1320.36
Fe3O4
a Fe2O3
a Fe2O3
1322
415
1320
[12]
[13]
[12]
293.65
Fe3O4
a Fe2O3
c
FeOOH
a Fe2O3
293
[17]
655.67
654
[15]
224.39
a Fe2O3
225
[13]
c
FeOOH
Table 4.15 Rust phases by Raman in 0.1 N Na2SO4 & Na2SO4 ? 0.001 N NaCl on MS
-1146 mV
-1609 mV
Spot Experimental
Raman shift
(cm-1)
1323.80
306.52
725.68
307.52
Published
Phase
Fe3O4
a
FeOOH
b
FeOOH
b
FeOOH
Experimental
Raman shift Ref. Raman shift

(cm-1)
(cm-1)
1322
[12]
1323.26
303
[12]
250.26
725
[14]
1320.82
310
[16]
254.22
Published
Phase
Fe3O4
c
FeOOH
a
Fe2O3
c
FeOOH
Raman shift Ref.

(cm-1)
1322
[12]
252
[15]
1320
[13]
255
[15]
For weathering steels corrosion product accumulated with time leading to the
retardation of corrosion rate with time, whereas for mild steel corrosion rate
increased with time.
Corrosion potential of weathering steels is greater than mild steels for all the
sites as found by DC corrosion measurement. EIS on exposed weathering steel
also supported the fact.
Rust coatings formed on weathering steels are compact except at P1, whereas
micro cracks/voids are seen on mild steels at all the sites as per SEM. Rust
coating developed on weathering steel is more protective than mild steel.
With reference to performance of scribed coated panels, rate of deterioration of
weathering steel substrate was lesser than mild steel substrate as per EIS.
Deterioration of coated panels was highest at P1 for both steels, and also higher
for mild steels at P2 and P3 as found by EIS. Presence of SO2 helped to prevent
deterioration of coated weathering steels at these sites.
More stable and protective phase/oxides (d FeOOH, a Fe2O3 & Fe3O4) formed
on weathering steels at Chennai and Jamshedpur with prolonged exposure as
4.6 Suggestions for Further Work
143
identified by Raman spectroscopy. The oxidation state is lowest at P3 for

weathering steel.
Corrosion rate of weathering steels was found lower with respect to mild steels
in accelerated laboratory tests. They corrode most rapidly with respect to field
exposure test.
Weathering steels exposed in humid SO2 developed protective oxides and
became greater than mild steel confirmed by SEM and EIS.
Performance of coated panels showed higher resistance for all as per EIS in
different electrolytes. This indicated that the paint systems applied are very
durable.
Rust formed in continuous humidity on weathering steel was not stable and
protective confirmed by Raman spectroscopy. Weathering steels developed
passive layers in Na2SO4 but mild steel did not.
The data generated with accelerated laboratory tests on both coated and
uncoated mild steels and weathering steels are grossly insufficient to predict
field behaviour.
Potentiostatic polarisation in Na2SO4 with slow scan rate showed prominent
active passive transition on weathering steel which could not be formed on
mild steel.
Rust morphology and analysis of simulated weathering process on weathering
steels compared fairly similar to field exposed weathering steels at P3.
Laboratory simulation by passivating weathering steel at 820 mV for 80 min
was almost similar with the oxidation state of weathering steel exposed for
42 months at P3.
4.6 Suggestions for Further Work

Suggestion to undertake further work:
Exhaustive mapping of corrosion rate of different locations in India

Recreate WS rust for P1 and P2
Develop improved, economic, durable paint system
Modeling of atmospheric Corrosion Rate with respect to locations
References
1. Madlangbayan, M., Otsuki, N., Saito, T., Min, A.K.: Corrosion behaviour of coated steel
plates with defects exposed to marine environment. J. Soc. Mat. Sc. 58(5), 437442 (2009)
2. Knotkova, D.: Atmospheric corrosion of metals, American Society for Testing and Materials
767, Philadelphia, p. 7 (1982)
3. Kihira, H., Ito, S., Murata, T.: Passive rust layer on low alloy steel exposed in atmospheric
environment. Corros. Sci. 31, 383388 (1990)
144
4. Nishimura, T., Katayama, H., Noda, K., Kodama, T.: Electrochemical behaviour of rust
formed on carbon steel in a wet/dry environment containing chloride ions. Corros. Sci. 56,
935 (2000)
5. Macdonald, J.R.: Impedance spectroscopy and its use in analysing the steady state AC
restorage of solid and liquid electrolytes. J. Electrochem. Soc. 223, 25 (1987)
6. Selected Powder Diffraction Data for Metals & Alloys Data Book, 1st. Edn., vol I & II,
JCPDS International Centre for Diffraction Data, USA
7. International Table for X ray Crystallography, Kyonoch Press, Birmingham, vol. 3, p. 161
(1962)
8. Loader, J.: Basic Laser Raman Spectroscopy. Heyden and Son, London (1970)
9. Gui, J., Devine, T.M.: Proceedings of 12th International Corrosion Congress, NACE, p 2052
(1993)
10. Dunnwald, J., Otto, A.: An investigation of phase transition in rust layers using Raman
spectroscopy. Corros. Sci. 29, 11671176 (1989)
11. Ohtsuka, T., Kubo, K., Sato, N.: Raman spectroscopy of thin corrosion films on iron at 100 to
150 C in air. Corrosion 42, 476481 (1986)
12. Ohtsuka, T.: Raman spectra of passive films of iron in neutral borate solution. Mat. Trans.
JIM. 37, 67(1996)
13. Brown, C.W., Heidersbach, R.H.: Raman spectra of possible corrosion products of iron.
Appl. Spectrosc. 32(6), 532535 (1978)
14. Boucherit, N., Delicher, P., Joiret, S., Hugot, A.: Mater. Sci. Forum 51, 44 (1989)
15. Faria, D.L.A., Silva, S.V., Oliveira, M.T.: Raman spectroscopy of some iron oxides and
oxyhydroxides. J. Raman. Spectrosc. 28, 873878 (1997)
16. Sato, N.: An overview on passivity of metals. Corros. Sci. 31, p1 (1990)
17. Jones, D.A.: Principles and Prevention of Corrosion. Prentice Hall, USA (1996)
18. Goidanich, S., Lazzari, L., Ormellese, M., Pedeferri, M.: Influence of AC on corrosion
kinetics for carbon steel, zinc and copper, Corrosion 5 and 2 (2005)
19. Sam, C.S., Balaya, P.: Investigation of structural disorder in electrode materials. http://
www.nus.edu.sg/nurop/2009/FoE/U066456W.PDF (2010) Accessed Jan 2010
20. Matsubara, E., Suzuki. S., Waseda. Y.: Corrosion mechanism of iron from an X ray structural
viewpoint. Book on Characterisation of Corrosion Products on Steel Surfaces
21. Ohtsuka, T.: Passivation oxide films and rust layers on iron. Characterization of Corrosion
Products on Steel Surfaces, Springer, Heidelberg (2006)
22. Pourbaix, M.: Atlas of Electrochemical Equilibria in Aqueous Solutions. Pergamon, New
York (1966)
23. Leygraf, C., Graedel, T.: Atmospheric corrosion. J. Electrochem. Soc. 1497, 1010 (2000)
24. Thibeau, R.J., Brown, C.W., Heidersbach, R.H.: Raman spectra of possible corrosion
products of iron. Appl. Spectrosc. 32, 532 (1978)
25. Dong, J.H., Han, E.H., Ke, W.: Progress in research on rust layer of weathering steel. Corros.
Sci. 18, 414 (2006)
Appendix
Fig. A 2.1 Average temperature of exposure sites (2007)
P2
P1
P3
Temperature (C)
45
30
15
0
0
10
11
12
Months (Jan --Dec 2007)
P2
Relative Humidity(%)
100
P1
P3
90
80
70
60
50
10
11
12
Months (Jan--Dec 2007)

Engineering Materials, DOI: 10.1007/978-81-322-0720-7, Springer India 2013
145
146
Appendix
Fig. A 2.3 Average rainfall

of exposure sites (2007)
P2
600
P1
P3
Rainfall (mm)
500
400
300
200
100
0
0
10
11
12
Months (Jan - Dec 2007)
Table A 2.1 Average Temperature, Relative Humidity and Rainfall of Exposure Site 2007
Site
P1
P2
P3
Month
RH
RH
RH
J
F
M
A
M
J
J
A
S
O
N
D
16.96
28.78
29.80
33.12
33.45
33.65
31.22
23.14
31.60
30.31
29.21
26.63
92.38
78.07
79.94
77.07
80.03
81.47
85.96
84.61
83.06
85.74
76.77
73.67
0.00
0.00
1.16
0.43
4.38
5.65
8.03
10.94
23.29
25.01
0.03
0.00
31.70
36.00
37.20
37.00
41.00
41.00
40.00
38.60
36.20
35.40
34.00
30.60
89.00
93.00
83.00
98.00
81.00
81.00
92.00
92.00
93.00
98.00
96.00
98.00
0.30
6.70
1.60
162.50
52.40
14.30
90.90
66.90
140.8
499.7
450.2
462.8
32.90
33.80
37.20
43.00
40.60
34.40
33.80
34.20
34.00
32.00
31.40
26.40
86.00
92.00
89.00
76.00
90.00
92.00
94.00
95.00
89.00
95.00
95.00
86.00
0.00
153.30
22.30
1.90
102.70
133.40
240.20
198.80
92.60
127.70
11.00
0.00
J Jan, F Feb etc, T Temperature (C), RH Relative humidity (%), R Rainfall (mm)
Appendix
Fig. A 3.1 EDX Plots of WS exposed for 18 m at a P1, b P2 and c P3
147
148
Fig. A 3.2 EDX plots after exposed for 42 m at P1 a MS and b WS
Appendix
Appendix
149
150
Appendix
Appendix
Fig. A 3.5 XRD spectra of rust exposed at P1 for 18 m a MS and b WS
151
152
Appendix
Fig. A 3.6 a XRD spectra of MS rust exposed at P1 for 42 m. b XRD spectra of WS rust
exposed at P1 for 42 m
Appendix
Fig. A 3.7 XRD Spectra of WS rust exposed at P2 for 42 m
153
154
Appendix
Fig. A 3.8 a XRD spectra of MS rust exposed at P3 for 42 m.b XRD spectra of WS rust exposed
at P3 for 42 m
Appendix
Fig. A 3.9 Raman spectra of rust exposed at P1 for 18 m a MS and b WS
155
156
Fig. A 3.10 Raman spectra of rust at P2 for 18 m a MS and b WS
Appendix
Appendix
157
158
Fig. A 3.13 Raman spectra of rust exposed at P2 for 42 m of WS
Appendix
Appendix
Fig. A 3.14 Raman spectra of rust exposed at P3 for 42 m a MS and b WS
159
160
Fig. A 3.15 EDX plots a AR MS and b AR WS
Appendix
Appendix
Fig. A 3.16 EDX plots of upper rust layer of a MS and b WS in humid SO
161
162
Fig. A 3.17 EDX plots of inner rust layer of a MS and b) WS in humid SO2
Appendix
Appendix
Fig. A 3.18 Raman spectra of as received a MS and b WS Rust
163
164
Appendix
Fig. A 3.19 Raman spectra of a MS and b WS rust exposed in humidity for 2,000 h
Appendix
165
166
Appendix
Appendix
Fig. A 3.22 Raman spectra of a MS and b WS rust exposed in salt spray for 2,000 h
167
168
Appendix
Fig. A 3.23 Raman spectra of a MS and b WS rust exposed in salt spray for 3,500 h
Appendix
Fig. A 3.24 Raman spectra of a MS and b WS rust exposed in saly spray for 4,500 h
169
170
Appendix
Fig. A 4.1 Raman spectra of WS after passivated at -820 mV in 0.1 M Na2SO4 (pH 9.5) at 2
spots
Appendix
171
Fig. A 4.2 Raman spectra of WS after pssivated at -800 mV in 0.1 M Na2SO4 (pH 9.5) at 2
spots
172
Appendix
Fig. A 4.3 Raman spectra of MS after passivated at -1,146 mV in 0.1 M Na2SO4 (pH 9.5) at 2
spots
Appendix
173
Fig. A 4.4 Raman spectra of MS after passivated at -1,609 mV in 0.1 M Na2SO4 (pH 9.5) at 2
spots
Fig. A 4.5 EDX plot of rust simulated on WS passivated at -820 mV

Corrosion of Constructional Steels in Marine and Industrial Environment

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Corrosion of Constructional Steels in Marine and Industrial Environment

Încărcat de

Drepturi de autor:

Formate disponibile

Engineering Materials

For further volumes:

Jayanta Kumar Saha

Jayanta Kumar Saha

ISSN 1868-1212 (electronic)

Springer New Delhi Heidelberg New York Dordrecht London

Dipanjan and Debosmita

Jayanta Kumar Saha

Test Panel Preparations . . . . . . . . . . . . . . . . . . . . . .

Rust Characterisation by SEM and EDX . . . . . . . . . .

Field Exposure Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

About the Author

selection of more suitable materials as well as for the safety of structures,

1.1 Rusting of Iron

1.2 Corrosion of Steel

1.2 Corrosion of Steel

Fig. 1.1 E-pH (Paurbaix

protective film such as a corrosion product or when a coating breaks down.

1.2.1 Mild Steel

1.2.2 Weathering Steel

1.3 Atmospheric Corrosion

1.3 Atmospheric Corrosion

Cathodic half-cell reaction H2 O + 1/2 O2 2e ! 2OH

Ferrous hydroxide [Fe(OH)2] and hydrated ferrous oxide (FeO.nH2O) is first

Table 1.1 Phases of corrosion products in atmospheric exposure

3.02 Orthorhombic/ 4.28

3.87 Orthorhombic/ 4.09

Inverse spinel/ 5.18

-490.4 Yellowish brown to

a,b,c: Relative lengths of crystallographic axes DG : Free energy of adsorption

1.3.1 Corrosion Products

1.3 Atmospheric Corrosion

1.3.2 Atmospheric Corrosion Mechanism

1.3 Atmospheric Corrosion

1.3.3 Effect of Acidity of Solution

1.3.4 Effect of Alloying Elements

1.3 Atmospheric Corrosion

1.3.5 Environmental Factors

1.3 Atmospheric Corrosion

Effect of Sulphates. Sulphur dioxide comes from combustion of fuels and is

4H2 SO4 + 4Fe + 2O2 ! 4FeSO4 + 4H2 O

Fe2 SO4 3 + H2 O ! FeSO4 + O2 + H2 SO4

Also, FeSO4 is also hygroscopic in nature which enhances the atmospheric

Effect of Chlorides. Marine environments have high percentage of relative

1.4 Corrosion Protection by Coating

1.4.1 Passive Rust Coatings

1.4 Corrosion Protection by Coating

1.4.2 Organic Coatings

1.5 Degradation of Organic Coatings

1.5.1 Delamination of Coatings

1.5 Degradation of Organic Coatings

Cathodic half-cell reaction:

H2 O + 1/ 2 O2 + 2e ! 2OH reduction in neutral/basic solution

Similar to cathodic blistering, cathodic delamination is also the result of

1.6 Corrosion Measurement and Analysis

1.6 Corrosion Measurement and Analysis

1.6.1 Corrosion Rate Measurement

Where, W is weight loss (mg), A is area of the specimen (cm2), D is density of

1.6.2 Electrochemical Methods

1.6 Corrosion Measurement and Analysis

Where, CR is corrosion rate in mpy (1 mpy = 0.054 lm/yr), EW is equivalent

kinetic and mechanistic information on electrochemical systems such as corrosion

Time dependence current response of the form: