Investigation of Materials For Use in Exhaust Gas Condensate Environment With Focus On EGR Systems PDF

Investigation of Materials for
Use in Exhaust Gas Condensate

Environment with Focus on
EGR Systems
Andreas Olofsson
August 3, 2012
Master’s Thesis in Energy Engineering, 30 credits
Supervisor at UmU: Britta Sethson
Supervisor at Scania: Baohua Zhu
Examiner: Robert Eklund
Umeå University
Department of Applied Physics and Electronics
SE-901 87 UMEÅ
SWEDEN
Abstract
EGR (Exhaust Gas Recirculation) is a method for reducing N Ox emissions for heavy-duty
diesel engines. EGR works by introducing part of the exhaust gases back to the engine
cylinders. Exhaust gases consists mainly of CO2 , N Ox , SO2 and H2 O. As the temper-
ature decreases, these gases form a corrosive condensate. The EGR components which
are exposed to the condensate environment must therefore be of corrosion resistant mate-
rials. The objective of this Master’s Thesis is to investigate suitable materials for use in
exhaust condensate environment. The goal is to evaluate the pitting corrosion resistance for
eight different commercial stainless steels and two commercial aluminium alloys in exhaust
gas condensate environment. Furthermore, nitriding surface treatments on one martensitic
stainless steel and anodising treatments on one aluminium alloy, were also included in this
study.
Five different exhaust gas condensates with different concentrations of sulphuric acid,
nitric acid and chloride were chosen to perform electrochemical measurements. Two pH
values 2.5 and 1.5; three chloride concentrations, 32 ppm, 200 ppm, 3300 ppm were in-
cluded in the environmental parameters. The testing temperature was 60o C, since it is
the temperature which can still be expected to produce substantial amount of exhaust gas
condensate in the EGR system. The electrochemical method used, was anodic polarisation
measurements. This is a useful method to evaluate the pitting resistance for stainless steels
in chloride containing solutions.
The results show that the two aluminium alloys and the martensitic stainless steel were
subjected to both general and pitting corrosion in a normal condensate solution at pH 2.5.
The anodised film on the aluminium surface was not stable in condensate environments
with low pH value. After twelve hours of exposure to a condensate at pH 2.5 at 60o C, the
protective effect of the film became negligible.
The austenitic, ferritic and duplex stainless steels show, however good resistance against
both corrosion types. Increasing condensate acidity from pH 2.5 to 1.5 could not be observed
to increase risk of pitting corrosion for the austenitic, ferric and duplex steel stainless steels.
High concentrations of sulphuric acid, low pH value, but low chloride content (200 ppm)
do not increase the risk for pitting corrosion for austenitic steels 1.4404 and 1.4301, du-
plex 2304 and ferritic 1.4521. However, chloride concentration of 3300 ppm, significantly
increased risk of pitting corrosion, especially for the austenitic stainless steels. Duplex stain-
less steel show better pitting resistance in high chloride environments, in addition to the
good general corrosion resistance in low pH value environments.
There is no difference in corrosion resistance between the nitride coated 1.4112 steel and
the steel without coatings. No differences can be observed between the plasma and gas
nitrided samples. Further investigation in less corrosive environment is recommended, since
anodic polarisation is not a suitable method to study general corrosion behavior.
The pitting corrosion resistance in condensates with high chloride concentrations at

60o C follows the sequence 1.4301<1.4521<1.4404<duplex 2304<duplex LDX2404<duplex
2205. Clearly, duplex stainless steels have better pitting corrosion resistance in low pH
ii
environment when chloride concentration is increased. Considering the operating conditions

of the EGR components, the element prices, it is probably more beneficial to consider the
duplex stainless steels for use in the EGR system.
iii
Kartläggning av material för användning i

avgaskondensatmiljö med fokus på EGR teknik
Sammanfattning
EGR (Exhaust Gas Recirculation) är en teknik som används för dieselmotorer, för
att möta de hårt satta utsläppskraven för kväveoxider. EGR fungerar genom att en del
av avgaserna återförs till cylindrarna. Avgaserna gör så att den maximala förbrännings-
temperaturen sänks, vilket kraftigt reducerar bildandet av kväveoxider. Dieselavgaser in-
nehåller främst CO2 , N Ox , SO2 och H2 O och innan avgaserna återförs till cylindrarna kyls
de ner. Detta leder till att det bildas ett korrosivt kondensat, eftersom de ämnen som finns
i avgaserna bland annat kan bilda svavel- och salpetersyra.
På grund av detta korrosiva kondensat måste material i EGR systemet vara korrosions-
beständigt och materialkraven förväntas öka i takt med utsläppskraven. I framtiden kan
det innebära att mer korrosionsbeständigt material måste användas.
Detta examensarbete undersöker gropfrätningsmotståndet för åtta olika rostfria stål och
för två aluminiumlegeringar i syntetiskt avgaskondensat. Dessutom inkluderades två ytbe-
handlingar; anodisering av en aluminiumlegering samt nitrering av ett martensitiskt rostfritt
stål. Målet är att kartlägga vilka material som är lämpliga att använda i avgaskondensat-
miljö.
Fem olika syntetiska avgaskondensat med olika halter av svavelsyra, salpetersyra och
klorider, valdes ut för elektrokemisk mätning. Två olika pH-nivåer, 2.5, 1.5, inkluderades
samt tre olika halter, 32 ppm, 200 ppm, 3300 ppm, av klorider. Testtemperaturen valdes
till 60o C, eftersom det är den temperaturen som fortfarande kan förväntas producera be-
tydlig mängd avgaskondensat i EGR systemet. Anodisk polarisation ansågs som den mest
lämpade elektrokemiska metoden för att nå examensarbetets mål.
Erhållna resultat visar att de två aluminiumlegeringarna och det martensitiska rostfria
stålet är utsatta för generell korrosion och gropfrätning vid pH 2.5. Medan de austenitiska,
ferritiska och duplexa rostfria stålen uppvisar bra korrosionsbeständighet mot båda dessa
korrosionstyper. Att sänka pH från 2.5 till 1.5 kunde inte ses öka risken för gropfrätning för
de austenitiska, ferritiska och duplex rostfria stålen.
En kloridhalt av 0.33 vikt% ökade markant risken för gropfrätning, speciellt för de
austenitiska rostfria stålen. Vid pH 1.5 och höga halter av klorider klarar sig duplexstålen
bäst.
Anodisering av aluminium ses initialt ge ett förbättrat korrosionsskydd vid pH 2.5, men
det observerades att det anodiserande oxidskiktet utsattes för kontinuerlig upplösning. Efter
12 timmar var oxidskiktet så kraftigt upplöst att dess skyddande effekt var försumbar. Alu-
minium kan därför inte rekommenderas till användning i EGR kondensat miljö.
Nitrering av det martensitiska stålet kunde inte ses påverka materialens korrosionsegen-
skaper. Men anodisk polarisation är inte en lämplig metod att använda för att studera
generell korrosion, därför rekommenderas fortsatta undersökningar vid antingen en högre
pH-nivå eller med en annan lämplig metod.
iv
Gropfrätningsmotståndet för de testade rostfria stålen i kondensat med höga kloridhalter

vid 60o C följer sekvensen 1.4301<1.4521<1.4404<duplex 2304<duplex LDX2404<duplex
2205. Det är tydligt att duplex stålen har bättre motstånd mot gropfrätning vid låga
pH-nivåer och höga kloridhalter. Med tanke på driftförhållandena i EGR-systemet och på
legeringskostnader är det lämpligt att överväga duplexa stål för användning i EGR-systemet.
v
Acknowledgment
This Master’s Thesis was performed at Scania CV AB, Materials Technology, UTMC. I
would like to thank UTMC for an encouraging and enjoyable working environment.
I want to thank Outokumpu for providing me with the necessary materials, thanks also
to Bodycote for providing me with the surface treatments.
Also thanks to my supervisor at UmU Britta Sethson for your input and guidance.
Above all i want to thank my supervisor at Scania Baohua Zhu for your invaluable help,
support and encouragement throughout the work of this Master’s Thesis.
vi
Contents
1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Diesel Exhaust Gas Condensation 5

2.1 Condensation of Sulphuric Acid . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Condensation of Nitric Acid . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 Condensation of Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.4 Effect of Fuel-sulfur Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.5 Exhaust Gas Condensate from Biodiesel . . . . . . . . . . . . . . . . . . . . . 7
2.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3 Stainless Steels and Aluminium 9

3.1 Stainless Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.1.1 Austenitic Stainless Steels . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.1.2 Ferritic Stainless Steels . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1.3 Duplex Stainless Steels . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1.4 Martensitic Stainless Steels . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2 Aluminium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4 Corrosion & Electrochemistry 15

4.1 Corrosion of Stainless Steels and Aluminium Alloys . . . . . . . . . . . . . . . 15
4.1.1 General Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.1.2 Localised Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.2 Electrochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.3 Anodic Polarisation Sweep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.4 Corrosion Protection Techniques . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.4.1 Corrosion Inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.4.2 Anodic and Cathodic Protection . . . . . . . . . . . . . . . . . . . . . 22
4.5 Automotive Corrosion Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.5.1 Corrosion Study of Different Stainless steels in Synthetic Exhaust Gas
Condensate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.5.2 Corrosion Study for Automotive Mufflers . . . . . . . . . . . . . . . . 23
4.6 Engineers Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
vii
viii CONTENTS
5 Experimental 27
5.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2 Synthetic Exhaust Gas Condensate . . . . . . . . . . . . . . . . . . . . . . . . 29
5.3 Sample Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.4 Potentiodynamic Polarisation Measurements . . . . . . . . . . . . . . . . . . . 30
6 Results 31
6.1 Condensate 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
6.1.1 Condensate 1 – Austenitic, Ferritic and Duplex Grades . . . . . . . . 31
6.1.2 Condensate 1 - Grade 4112 & Effect of Nitriding . . . . . . . . . . . . 33
6.1.3 Condensate 1 – Aluminium Grades & Effect of Anodising . . . . . . . 34
6.2 Condensate 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.2.1 Condensate 2a – Austenitic, Ferritic and Duplex Grades . . . . . . . . 35
6.2.2 Condensate 2b – Austenitic, Ferritic and Duplex Grades . . . . . . . . 37
6.3 Condensate 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
6.3.1 Condensate 3a – Austenitic, Ferritic and Duplex Grades . . . . . . . . 39
6.3.2 Condensate 3b – Austenitic, Ferritic and Duplex Grades . . . . . . . . 39
7 Discussion and Summary 43

7.1 Pitting Corrosion in the EGR System . . . . . . . . . . . . . . . . . . . . . . 43
7.2 Material Choice for EGR Components . . . . . . . . . . . . . . . . . . . . . . 44
8 Conclusions 47
9 Recommendations 49
Bibliography 51
Appendix 53
List of Figures
1.1 N Ox and PM legislation for Euro 3, Euro 4, Euro 5 and Euro 6. . . . . . . . 1

1.2 Schematic of a HP-EGR system [2]. . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Schematic of a LP-EGR system [2]. . . . . . . . . . . . . . . . . . . . . . . . . 2
2.1 Formation mechanisms of exhaust gas condensate summarised from literature

review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.1 Pourbaix diagram for a Al/H2 O system at 25o C. . . . . . . . . . . . . . . . . 13
4.1 Schematic of the autocatalytic process of pitting corrosion. . . . . . . . . . . 16

4.2 Schematic of crevice corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.3 Synergistic effect of SCC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.4 Simplified illustration of the double layer at a metal aqueous interface. . . . . 20
4.5 Schematic of an anodic polarisation sweep for stainless steel. . . . . . . . . . 21
4.6 Isocorrosion curves, 0.1 mm/year, in sulphuric acid [24]. . . . . . . . . . . . 25
4.7 Isocorrosion curves, 0.1 mm/year, in sulphuric acid containing 2000 ppm
chloride ions [24]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.1 Experimental Setup: 1) Solartron 1287 potentiostat 2) Ag/AgCl reference

electrode (3M KCl 3) Platinum plate counter electrode 4) Working Electrode
5) Corrosion Cell 6) B.I.A climatic climate chamber 7) Corrware R Software
Package. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.1 Anodic polarisation plot for stainless steel grades Duplex 2205, Duplex 2304,
1.4301, 1.4404, 1.4509, 1.4521. . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
6.2 Plot from anodic polarisation measurement of nitride coated and uncoated
1.4112 martensitic stainless steel in condensate 1. . . . . . . . . . . . . . . . . 33
6.3 Anodic polarisation plot for aluminium grades AC43000KF and AW-3003 in
condensate 1. AW-3003 and AC-43000KF-anodised also tested after 12 hour
exposure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
6.4 Open circuit potential (OCP) for Al AC-43000KF in condensate 1. . . . . . . 35
6.5 Anodic polarisation plot for stainless stainless steel 2304, 1.4301, 1.4404,
1.4521 in Condensate 2a. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
6.6 Anodic polarisation plot for stainless steel grades 2205, LDX2404, 2304,
1.4404, 1.4301, 1.4521 in condensate 2b. . . . . . . . . . . . . . . . . . . . . . 37
6.7 Anodic polarisation obtained for stainless steels 2304, 1.4404, 1.4301, 1.4521
in condensate 3a. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
ix
x LIST OF FIGURES
6.8 Anodic polarisation for stainless steel grades 2205, 2304, LDX2404, 1.4301,
1.4404, 1.4521 for condensate 3b. . . . . . . . . . . . . . . . . . . . . . . . . . 40
1 Photo of 4112 grade test samples: 1) Before polarisation. 2) After polarisa-

tion. 3) Gas nitrided after polarisation. 4) Plasma nitrided after polarisation. 53
2 Photo of aluminium grade test samples: 1) AC-43000KF before polarisation
2) AC-43000KF after polarisation 3) AC-43000KF anodised before polarisa-
tion 4) AC-43000KF anodised after polarisation 5) AW-3003 before polarisa-
tion 6) AW-3003 after polarisation . . . . . . . . . . . . . . . . . . . . . . . . 53
3 Photo of test samples after polarisation in condensate 2b. Pitts visible by
visual inspection have been marked. . . . . . . . . . . . . . . . . . . . . . . . 54
4 Photo of test samples after polarisation in condensate 3b. Pitts visible by
visual inspection have been marked. . . . . . . . . . . . . . . . . . . . . . . . 54
5 Microscopic examinations of plasma nitride film before anodic polarisation. . 55
6 Microscopic examination of gas nitride film before anodic polarisation. . . . . 56
7 Microscopic examinations of plasma nitride film after anodic polarisation in
condensate 1. Marks indicate exposed area. . . . . . . . . . . . . . . . . . . . 57
8 Microscopic examinations of plasma nitride film after anodic polarisation in
condensate 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
9 Microscopic examination of gas nitride film after anodic polarisation in con-
densate 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
10 Microscopic examination of gas nitride film after anodic polarisation in con-
densate 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
11 Test coupon documentation for duplex steel 2205. . . . . . . . . . . . . . . . . 61
12 Test coupon documentation for duplex steel LDX2404. . . . . . . . . . . . . . 62
13 Test coupon documentation for duplex steel 2304. . . . . . . . . . . . . . . . . 63
14 Test coupon documentation for steel grade 1.4521. . . . . . . . . . . . . . . . 64
15 Test coupon documentation for steel grade 1.4509. . . . . . . . . . . . . . . . 65
List of Tables
3.1 Composition ranges for different stainless steels [13]. . . . . . . . . . . . . . . 9

3.2 Austenitic stainless steels and composition. Source: Outokumpu. . . . . . . . 10
3.3 Ferritic stainless steels and composition. Source: Outokumpu. . . . . . . . . . 11
3.4 Duplex stainless steels and composition. Source: Outokumpu. . . . . . . . . . 11
3.5 Martensitic stainless steels. Source: Outokumpu & Metal Ravne. . . . . . . . 12
3.6 Classification and description for aluminium alloys. . . . . . . . . . . . . . . . 13
4.1 Test materials and steel composition (wt%) [22]. . . . . . . . . . . . . . . . . 23

4.2 Test materials and steel composition (wt%) [23]. . . . . . . . . . . . . . . . . 24
5.1 Stainless steels grades selected for electrochemical evaluation. . . . . . . . . . 27

5.2 Aluminium grades selected for electrochemical evaluation. . . . . . . . . . . . 28
5.3 Chemical composition of synthetic exhaust gas condensates (C) used for an-
odic polarisation measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.4 Grades and condensates (C) experimentally tested in this study. . . . . . . . 29
6.1 Eb , Ep and Imax obtained from anodic polarisation of austenitic, ferritic and
duplex grades in condensate 1. . . . . . . . . . . . . . . . . . . . . . . . . . . 32
6.2 Eb , Ep and Imax obtained from anodic polarisation scan in condensate 2a. . . 36
6.3 Eb , Ep and Imax obtained from anodic polarisation of austenitic, ferritic and
duplex grades in condensate 2b. . . . . . . . . . . . . . . . . . . . . . . . . . . 38
6.4 Eb , Ep and Imax for tested austenitic, ferritic and duplex grades in condensate
3b. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
xi
xii LIST OF TABLES
Chapter 1
Introduction
This chapter provides an initial background to the work of this Master’s Thesis. Also de-
scribed here is the thesis objectives and limitations, followed by a brief explanation of the
solution methodology used.
1.1 Background
Emission legislations for heavy-duty diesel trucks have become increasingly stringent the last
decade, see figure 1.1. Current European emission legislation, Euro 5, will at the beginning
of 2013 be replaced by Euro 6 and require a 77 % reduction in nitrogen oxide emissions
(N Ox ) and 50 % reduction in particulate matter (P M ) [1]. Two frequently used techniques
for N Ox reduction are EGR (Exhaust Gas Recirculation) and SCR (Selective Catalytic
Reduction). EGR works by recirculating a part of the exhaust gases back to cylinders. The
recirculated exhaust gases act as an inert combustion gas, which absorbs heat and dilutes
oxygen concentrations in the combustion zone. This lowers combustion temperature and
as a result significantly reduce N Ox emissions. SCR works by injection of an additive,
called AddBlue, into exhaust gases. AddBlue absorbs heat from exhaust gases and gaseous
ammonia (N H3 ) is produced. N H3 and N Ox is then catalytically transformed into H2 O
and N2 .
Figure 1.1: N Ox and PM legislation for Euro 3, Euro 4, Euro 5 and Euro 6.
1
2 Chapter 1. Introduction
EGR and SCR have their unique advantages and disadvantages. SCR is efficient at high
speeds and high loads, while EGR has advantages at cold starts and low loads. For Euro 5 it
has generally been sufficient to use one of these N Ox reduction techniques. A combination
of EGR and SCR is necessary for fulfillment of the requirements in Euro 6 for most heavy-
duty engine manufacturers [2]. Euro emission legislation must be met during the engines
first seven years of operation or during its first 70 0000 km, whatever comes first [1].
There are different EGR designs, with their respective advantages and disadvantages.
The most notable design difference is usually the location from where exhaust gases are
extracted and returned. The design with most commercial success is High Pressure EGR
(HP-EGR); exhaust gases is extracted before decompression and reintroduced after the air-
to-air (charge-air) cooler, see figure 1.2. Another design with some commercial success is
Low Pressure EGR (LP-EGR); exhaust gases is extracted after decompression and reintro-
duced to the charge-air before compression, see figure 1.3.
Figure 1.2: Schematic of a HP-EGR system Figure 1.3: Schematic of a LP-EGR system
[2]. [2].
LP-EGR has from a fuel and emission perspective more potential than HP-EGR. For
LP-EGR all exhaust gases pass through the turbine, having the potential to increase tur-
bocharger performance. LP-EGR is also known as long route EGR (LR-EGR), since route
for exhaust gases are longer compared to HP-EGR. This longer route enhances mixing of
EGR gases and intake-air, resulting in a more homogeneous mixture which improves N Ox
reduction [2]. Furthermore, exhaust gases are extracted after the diesel particulate filter
(DPF), see figure 1.3. Cleaner exhaust gases are thus reaching the cylinders, which better
preserve engine durability [2].
– Why is then HP-EGR more common than LP-EGR?
There are some challenges that must be overcome to make LP-EGR more attractive for
commercial use, some are listed below:
1) Even though exhaust gases for LP-EGR pass through the aftertreatment system, a
certain amount of particles are still remaining in the exhaust stream, its impact on the
compressor wheel as it turns at high speed. May potentially erode the wheel [2].
2) Charge-air-coolers (CAC) have very narrow cooler-passages, which may be problem-

atic if soot accumulates for a prolonged time. Since this may potentially cause high pressure
losses, resulting in decreased engine performance and fuel efficiency [2].
1.2. Objectives 3
3) Most commercial charge-air-coolers are made of aluminium, because of the materials

high thermal conductivity and low weight. Corrosion resistance of aluminium is known to
be heavily dependent on pH. This may potentially be problematic if a mixture of exhaust
gases and charge-air is to be cooled.
Exhaust gases consist mainly of CO2 , H2 O, N Ox and small amounts of SO2 depending
on the diesel quality. When temperature is decreased nitric acid, sulphuric acid and organic
acids will be formed. Nitric and sulphuric acid are highly corrosive, while the organic acids
are less corrosive. The concentration of these acids in exhaust gas condensate can vary
significantly, since formation/condensation of acids in the EGR system depend on several
factors, such as fuel quality, EGR-rate and temperature. The EGR components which are
exposed to this condensate environment must therefore be of corrosion resistant materials.
Acidity for condensate collected from a 2.0L passenger car engine fulfilling EU4 has been
found to be below pH 3.7 [3]. A more acidic condensate can be expected from a heavy-duty
EU5 engine and shortly after engine start-up, since wall temperatures then are below the
dew point of water, resulting in increased condensation.
1.2 Objectives
The objective of this Master’s Thesis was to investigate suitable materials for use in ex-
haust condensate environment. The goal was to evaluate the pitting corrosion resistance for
eight different commercial stainless steels and two commercial aluminium alloys in exhaust
gas condensate environment. Furthermore, nitriding surface treatments on one martensitic
stainless steel and anodising treatments on one aluminium alloy, were also included in this
study
1.3 Limitations
A material survey for the EGR system is comprehensive work; several material factors needs
to be thoroughly evaluated before making material decisions, e.g. formability, weldability,
heat conductivity, etc. Today, there are many different materials, grades and surface treat-
ments to choose from. Restrictions were therefore needed, since time was limited. Focus for
this Master’s Thesis was subsequently set on the wet section of the EGR system, i.e. where
there is risk of aqueous corrosion. Furthermore, focus was set to primarily consider stainless
steels, aluminium alloys, two surface treatments and to evaluate their tendency for pitting
corrosion in exhaust gas condensate.
1.4 Methodology
Materials surveys are an interdisciplinary work. A literature review was therefore carried
out in the following topics: Exhaust Gas Condensation, Stainless Steels and Aluminium
Alloys, Corrosion and Electrochemistry. These topics differ significantly from each other
and were therefore given separate chapters in this report.
4 Chapter 1. Introduction
In addition to the literature review, electrochemical methods is used to study the pit-
ting corrosion of selected materials. Furthermore, it is motivated to investigate which levels
of acidic substances is required to cause pitting corrosion in the EGR system. Corrosion
of stainless steels and aluminium is mostly of localised type, due to its ability to become
passivated (explained in chapter 3).
Anodic polarisation is an electrochemical method for testing iron- and nickel- base alloys
tendency for pitting corrosion. This method was deemed most suitable, since it provides
sufficiently fast and valuable data for the work of this Master Thesis. Below is a step-by-step
methodology for how the work was done.
Step-by-Step Methodology
Step One: Literature Review – Drastically narrow down the large quantitative of com-
mercially available grades. Ten grades were selected for electrochemical evaluation in
synthetic exhaust gas condensates. Two surface treatments were included.
Step Two: Experimental – Anodic polarisation measurements in different synthetic ex-
haust gas condensates. Five different exhaust gas condensates were selected, with
different concentrations of corrosive species.
Step Three: Result – Evaluation of results, conclusions, material guidelines and recom-
mendations for further work.
Chapter 2
Diesel Exhaust Gas

Condensation
Condensation of exhaust gases has important implications on the design and material choices
of the EGR system. This chapter will therefore address theories and research associated with
the condensation of exhaust gases.
The mechanisms of acidic formation in the EGR system is divided into the following ar-
eas: Condensation of Sulphuric Acid, Condensation of Nitric Acid, Condensation of Water.
Formation of nitric acid is normally not addressed as a specific issue, since it is associated
with the many problems derived from condensation of water, but it will still be addressed in
this study. However, studies regarding condensation of nitric acid was found to be sparse.
2.1 Condensation of Sulphuric Acid

Combustion cause fuel-bound sulfur to oxidize and form sulfur oxides (SOx ) and when
cooled, sulfur oxides may form sulphuric acid. Formation mechanism of sulphuric acid can
be divided in three steps:
1) Oxidation of fuel sulfur, which takes place during combustion, equation 2.1.1.
S(g) + O2 (g) → SO2 (g) (2.1.1)
2) Further oxidation of sulfur may take place, as exhaust gases are cooled, equation 2.1.2.
Formation mechanisms of sulfur oxides are complex, and involve several intermediate steps.
Thus, overall formation reactions are commonly simplified to [4]:
SO2 (g) +1 /2 O2 (g) → SO3 (g) (2.1.2)
3) SO3 is considered very reactive. All formed SO3 is therefore thought to almost
instantaneously convert to sulphuric acid if water is present (vapor or liquid), equation 2.1.3
[4, 5, 6]:
SO3 (g) + H2 O(l/g) → H2 SO4 (l/g) (2.1.3)
5
6 Chapter 2. Diesel Exhaust Gas Condensation
If liquid water is present, sulphuric acid can also form through dissolution of SO2 ,
equation 2.1.4 - 2.1.5 [5, 7]:
SO2 + H2 O → H2 SO3 (2.1.4)
H2 SO3 +1 /2 O2 → H2 SO4 (2.1.5)
Mckinley [6] has developed a model for dew point prediction of sulphuric acid in EGR-
coolers. The model predicts condensation of sulphuric acid between 63o C to 156o C, depend-
ing on fuel-sulfur level and engine operating conditions. Mckinley’s model shows that the
dew point for sulphuric acid is well above the dew point of water. However, required input
to this model is the conversion rate of SO2 to SO3 . Mckinley based his calculations on the
estimation that 3 to 8 % of the gaseous SO2 will convert to SO3 . Reaction times involved
in diesel engines are limited. Therefore, formation of SO3 is expected to be kinetically
limited and thus also formation of vapor-phase sulphuric acid, since formation of SO3 is
expected to be kinetically limited, its conversion rate is difficult to calculate. Studies have
been conducted to try and experimentally determine the conversion rate, but such studies
have shown it difficult to measure this conversion rate.
In a study conducted by M. Mosburger et al. [5] no condensation of sulphuric acid could

be measured, using either low- or high-sulfur fuels. The likely explanation given, was that
the coolant temperature (∼87o C) and timescales, do not allow sufficient kinetics-controlled
oxidation of SO2 to SO3 . Their conclusion was that the main corrosion risk for EGR-coolers
were at operation temperatures below the dew point of water, e.g. during cold-starts and
after engine shutdown.
In a study by A. M. Kreso et al. [4] small condensation of H2 SO4 could be measured

in the EGR system at a collector temperature of 65+ o
− 4 C. However, their results also show
that condensation through vapor-phase sulphuric acid is limited, and that primary forma-
tion mechanisms of sulphuric acid is from dissolution of SO2 in condensed water (Equation
2.1.4 - 2.1.5). More fuel-sulfur was found in particulate filters than as vapor-phase sulphuric
acid. They recommend engine designers to be more concerned about the warp-up operation,
and brief time when EGR-cooler operated below the dew point of water, since the possibility
of forming significant amounts of sulphuric acid is significantly higher when operating near
or below the water dew point.
M. D. Kass et al. [8] evaluated corrosion rate of mild steel in the intake manifold as
a function of fuel-sulfur level, EGR fraction, water dewpoint margin and humidity. For
both tested fuels (15 ppm S, 350 ppm S) it was observed that no significant corrosion
occurred, until onset of water condensation. No significant corrosion was observed for the
low sulfur fuel, even when temperatures where below the dew point of water. However, for
the high-sulfur-fuel, significant increase in corrosion rate was observed when temperature in
the intake manifold was kept below the dew point of water. No clear correlation was found
for corrosion rate with respect to humidity and EGR fraction.
2.2 Condensation of Nitric Acid

Combustion causes the formation of nitrogen oxides, through a reaction with atmospheric
nitrogen and oxygen, equation 2.2.1 - 2.2.2.
N2 (g) + 02 (g) → 2 N O(g) (2.2.1)

2.3. Condensation of Water 7
1
N O(g) + /2 O2 (g) → N O2 (g) (2.2.2)
Nitrogen dioxides may then later react with water and form nitric acid [9]. However, the
dew point of nitric acid is typically below the dew point of water [7]. Therefore it can be
expected that the main formation mechanism of nitric acid in the EGR system is through
dissolution of nitrogen dioxides in condensed water, equation 2.2.3.
1
2 N O + H2 0 + /2 O2 → 2 HN O3 (2.2.3)
2.3 Condensation of Water

The dew point of water in the EGR system is primarily governed by lambda (air/fuel-ratio),
charge pressure and EGR-rate. It is important to take in consideration if EGR gases has
been mixed with charge-air or not, since it heavily affects the humidity.
If condensed water is present, formation of acidic elements will significantly increase,

since both nitrogen oxides and sulfur dioxides may dissolve in water and form nitric and
sulphuric acids.
2.4 Effect of Fuel-sulfur Quality

Diesel fuels can have a wide range of different sulfur levels, e.g. European diesels have a
maximum allowed sulfur content of 10 ppm accordingly to EN-590. However, allowed limits
can vary significantly between certain countries, from 10 ppm up to 2000 ppm [10]. Several
studies have shown that level of sulphuric acid in exhaust gas condensate vary heavily
depending on diesel quality [4, 5, 6, 11].
2.5 Exhaust Gas Condensate from Biodiesel

Biodiesels consists of long chain mono-alkyl esters, rather than long chains of hydro-carbons
as for diesel. Mono-alkyl esters are also known as fatty acid methyl esters (FAME). Physical
and chemical properties for biodiesels are greatly influenced by the feedstock. Most com-
mon feedstocks are vegetable oils, such as rapeseed, soy and palm. European biodiesels are
mostly produced from rapeseed, while in USA most biodiesels are produced from soybeans.
In tropical countries, palm oil is a commonly used biodiesel feedstock, because of its high
yield per cultivated land – 5 times higher than rapeseed, and 10 times higher than soybeans
[9].
S. Moroz et. al [9] investigated the effect of three different biodiesels (rapeseed, soy, palm
oil) with respect to condensate acidity, condensate collection rate, smoke opacity and N Ox
emissions. Test engine was a 2.0L direct injected diesel engine for passenger car application
fulfilling Euro 4. The test engine was equipped with both a HP-EGR and LP-EGR system.
A low sulfur diesel acted as reference (7.4 ppm S), while both 100% biodiesels (B100) and
fifty/fifty (B50) biodiesel and diesel fuel blends were tested. Condensate were collected from
the LP-EGR water-cooled charge air cooler (WCAC) at three different load points.
8 Chapter 2. Diesel Exhaust Gas Condensation
All biodiesels showed a significantly lower opacity compared to diesel, between 64 %

and 87 % lower, which is expected due to the high oxygen content in biodiesels (∼12 %).
Sulfur content in biodiesel is very low, varied between 1.7-2.8 ppm. This very low sulfur con-
tent should promote formation of less aggressive condensate, but the higher BSFC (Brake
Specific Fuel Consumption) of biodiesels is expected to cause more condensation. However,
results from this study show that tested biodiesel fuels, produced similar amounts of conden-
sate. Analyses of collected condensate, show relatively small difference in acidity between
biodiesel and low-sulfur diesel fuels, for all tested fuels and blends pH varied between 3.5
and 4.
A study by G. Bourgoin et. al [3], follows the work of S. Moroz, but with focus on
collecting biodiesel condensate from a HP-EGR system. Same type of 2.0L engine and
biodiesel fuels were used as in the previous study. An additional reference fuel was included
(490 ppm S). Engine test points selected were similar as for the previous study. Acidity
of collected biodiesel condensate varied between pH 2.5-3.2, while pH of reference fuels
varied between 3.2-3.7. Results in this study show larger differences in acidity for biodiesel
condensate compare to low sulfur-diesel, but the differences is relatively small.
2.6 Summary
1) Condensation of sulphuric acid above water dew point is limited.
2) Diesel and biodiesel produce similar amounts of condensates and similar levels of acidity,
but studies are only available for EU4 engines for passenger car applications.
Formation mechanisms of exhaust gas condensate in the EGR system are summarised
in the figure below:
Figure 2.1: Formation mechanisms of exhaust gas condensate summarised from literature
review.
Chapter 3
Stainless Steels and Aluminium
This chapter provides an brief literature review about the many different stainless steels and
aluminum grades.
3.1 Stainless Steels

Stainless steels are high-alloyed steels with more than 10.5 % chromium (Cr). When
chromium is oxidized it will form an thin oxide layer on the metal surface which will protect
the underlying metal from corrosion. With a thin passive film that totally covers the metal
surface the metal is said to be passivated and corrosion is greatly retarded. For passivity to
happen a chromium content of about 10-11 % is required [12, 13, 14]. Stainless steels are
categorised in four groups austenitic, ferritic, duplex, martensitic, depending on the crystal
structure. Table 3.1 lists characteristic composition of these different stainless steels.
Table 3.1: Composition ranges for different stainless steels [13].
Crystal structure Composition (wt%)

C Cr Ni Mo Other
Austenite <0.08 16-30 8-35 0-7 N, Cu, Ti, Nb
Ferrite <0.08 12-19 0-5 <5 Ti
Martensite >0.10 11-14 0-1 - V
Duplex <0.05 18-27 4-7 1-4 N,W
3.1.1 Austenitic Stainless Steels

Austenitic stainless steels usually have very good corrosion resistance, combined with good
form- and weldability. They can be distinguished by their high chromium and nickel (Ni)
content, see table 3.1. Nickel is a strong austenite former and also enhances the repassiva-
tion, especially in reducing environments [13, 14]. Although austenitic stainless steels have
a good resistance to most corrosive environments, they have relatively high susceptibility for
chloride-induced stress corrosion cracking (SCC) [15]. Molybdenum (Mo) increases the resis-
tance against reducing acids and pitting corrosion. Chromium increases corrosion resistance
in oxidizing environments. This combination gives Cr-Ni-Mo alloys very good corrosion re-
sistance in a broad range of environments [14, 16]. Molybdenum promotes ferritic structure
9
10 Chapter 3. Stainless Steels and Aluminium
and for austenitic stainless steels, this needs to be counterbalanced by addition of austenite
formers such as nickel. Nitrogen (N) is a strong austenitic former and it increases resistance
against pitting corrosion [14]. Niobium (Nb) and titanium (Ti) are efficient carbon binders.
Carbon can precipitate as detrimental carbides, this increase susceptibility for intergranular
corrosion. Addition of Nb and/or Ti will therefore have a positive effect against intergranu-
lar corrosion [13]. Manganese (Mn) in small quantity has a similar effect as N, but in large
quantities it can interact with sulfur and form sulfides, which increases corrosion resistance,
especially against pitting corrosion [14, 17].
Austenitic stainless steels have usually higher alloy costs compared to other stainless
steels, due to the high addition of nickel. Table 3.2 lists composition of some commercially
available austenitic stainless steel grades.
Table 3.2: Austenitic stainless steels and composition. Source: Outokumpu.
Steel: EN (ASTM/UNS) C N Cr Ni Mo Other

1.4301 (304) 0.04 - 18.1 8.3 - -
1.4401(316) 0.04 - 17.2 10.2 2.1 -
1.4004 (316L) 0.02 - 17.2 10.1 2.1 -
1.4436 (316) 0.04 - 16.9 10.7 2.6 -
904L (904L) 0.01 - 20 25 4.3 1.5Cu
1.4652 (654SMO) 0.01 0.5 24 22 7.3 3Mn, Cu
3.1.2 Ferritic Stainless Steels

Ferritic stainless steels have high chromium content and low or no nickel content. The fer-
ritic structure result in inferior formability and weldability, compared to austenitic stainless
steels. However, alloying with niobium and titanium increases weldability and toughness
[13]. Molybdenum improves corrosion resistance, both general and localised, as for all stain-
less steel.
Ferritic stainless steels were initially developed to withstand corrosion and oxidation,
while having low susceptibility for SCC [17]. The ferritic structure also gives 30–35% lower
thermal expansion in comparison with austenite. This make ferritic grades suitable for use
in high temperature applications and were thermal cycles are frequent [17]. The chromium
oxide film has lower thermal expansion compared to metal bulk. For austenitic stainless
steels this can become a problem, if exposed to large temperature gradients. Ferritic stain-
less steels with their lower thermal expansion are thus less susceptible to this problem.
Silicon gives higher resistance against oxidations. However, it also results in increased
brittleness. High silicon-ferritic alloys (>14 %) have an exceptional corrosion resistance,
e.g. they can withstand any concentration and temperature of sulphuric acid, but the high
silicon content also increases brittleness, as a result high silicon grades must be casted [18].
The low nickel content makes ferritic grades in general the least expensive stainless steel
type. Table 3.3 lists composition of some commercially available ferritic grades.
3.1. Stainless Steels 11
Table 3.3: Ferritic stainless steels and composition. Source: Outokumpu.

1.4512 (409) 0.02 - 11.5 0.2 - Ti
1.4003 (S40977) 0.02 - 11.5 0.5 - -
1.4000 (410S) 0.04 - 12.5 - - -
1.4016 (430) 0.04 - 16.5 - - -
1.4509 (A43932) 0.02 - 18 - - Nb,Ti
1.4521 (444) 0.02 - 18 - 2.1 -
3.1.3 Duplex Stainless Steels

Duplex stainless steels have a balanced crystal structure of both ferrite and austenite. The
proportion of ferrite is usually between 40-50% [16]. This combination results in high me-
chanical strength; roughly double that of austenitic stainless steels [19]. The increased
strength makes it possible to reduce section thickness compared to austenitic grades. Du-
plex alloys have better ductility and toughness compare to ferritc grades, but lower values
compared to austenitic grades.
The relative low nickel content of duplex alloys makes them to a cost-effective alternative
to austenitic grades. Nickel is used to improve stability of the austenitic phase, but also to
increase toughness and improve repassivation [19]. Nitrogen also improves the stability of
austenite phase, but in addition improves strength, weldability and resistance against pitting
corrosion. Molybdenum increases corrosion resistance, especially against pitting corrosion
[13, 17, 19].
Duplex stainless steels are generally considered to have better corrosion resistance against
chlorides compared to similar austenitic grades [14]. However, the mixed crystal structure of
dulplex limits the operating temperature range. Outokumpu recommends that the working
temperature of their duplex steels should be between -40o C and 250-325o C. Usage out-
side this interval increases risk of embrittlement [19]. Table 3.4 lists composition of some
commercially available duplex grades.
Table 3.4: Duplex stainless steels and composition. Source: Outokumpu.

1.4162 (S32101) 0.03 0.22 21.5 1.5 0.3 5Mn
1.4362 (S32304) 0.02 0.1 23 4.8 0.3 -
1.4662 (S82441) 0.02 0.1 22 5.7 3.1 -
1.4462 (S32205) 0.02 0.17 22 5.7 3.1 -
1.4501 (S32760) 0.02 0.27 25.4 - 3.8 W, Cu
1.4410 (S32750) 0.02 0.27 25 - 4 Al
3.1.4 Martensitic Stainless Steels

Martensitic stainless steels have great hardness, due to its high alloying with carbon. How-
ever, alloying with carbon also increase brittlement and decrease weldability. Futhermore,
high carbon content also reduce corrosion resistance, especially against pitting and crevice
corrosion [13].
Carbon contributes to the formation of chromium carbides, which cause inhomoge-

neous distribution of chromium. This may result in localised areas with significantly lower
chromium content compared to the bulk. These Chromium diluted areas then act as ini-
tiation sites for corrosion, referred as intergranular corrosion (IGC). Just as for the other
stainless steels types, corrosion resistance can be increased by increasing chromium and
molybdenum alloying. Sulfur and selenium can be used to increase machinability [18].
New martensitic stainless steels alloyed with N, Ni, Mo in combination with a slight
reduction in carbon content results in improved corrosion resistance and toughness [14].
However, martensitic stainless steels are generally to be considered as the least corrosion re-
sistant stainless steel grade. Martensitic grades are therefore used for applications were high
hardness and moderate corrosion resistance is needed e.g. knife blades, turbine blades, water
valves, piston rings. Table 3.5 lists composition of some commercially available martensitic
stainless steel grades.
Table 3.5: Martensitic stainless steels. Source: Outokumpu & Metal Ravne.

1.4006 (410) 0.12 0.04 12 - - 5Mn
1.4005 (416) 0.1 0.04 13 - - -
1.4021 (420) 0.2 - 22 13 - 3Mn
1.4028 (420) 0.3 - 22 12.5 - -
1.4112 (440B) 0.9 - 25.4 18 1.1 Si, 0.1V
1.4313 (S41500) 0.03 0.04 12.5 4.1 0.6 W, Cu
1.4548 (-) 0.05 0.07 15.5 4.2 - -
3.2 Aluminium
Today, aluminium and aluminium alloys is the preferred engineering material in a wide
range of applications, because of its many attractive material properties, e.g. combination
of low density and high strength, high thermal and electrical conductivity, and good corro-
sion resistance.
Aluminium gets its corrosion protection, just as stainless steels, from an oxide film,
which for aluminium is relative thick compared to stainless steels. Corrosion resistance
of aluminium is strictly dependent on its ability to form Al2 O3 , since other aluminium
oxides do not passivate the metal. Beyond its passive range, aluminium corrodes in aqueous
solutions, due to the passive films solubility in acids and bases, yielding Al3+ ions in the
former and AlO− 2 in the latter [14], see figure 3.1. Below a certain pH-level (usually below
pH 4) corrosion rate starts to accelerate exponentially [16]. Corrosion is also fast above a
certain pH.
3.2. Aluminium 13
Figure 3.1: Pourbaix diagram for a Al/H2 O system at 25o C.
Alloying elements used depend on desired material properties. The European classifica-
tion system for wrought aluminium alloys (EN AW) categorise aluminium in eight different
groups. The classification system for cast aluminium alloys (EN AC) is similar. Table 3.6
shows the EN AW classification system.
Table 3.6: Classification and description for aluminium alloys.
Series Classification Description [12, 16, 18]

1xxx Al > 99 % Pure aluminium, with some impurities, mostly iron and
silica.
2xxx Al-Cu alloys Copper considerably increases strength, but also decreases
corrosion resistance and weldability.
3xxx Al-Mn alloys Manganese increases strength and formability, while corro-
sions resistance is maintained.
4xxx Al-Si alloys Silica lowers the melting point of aluminium and do not
have a general negative effect on corrosion resistance.
5xxx Al-Mg alloys Magnesium increases strength while good corrosion resis-
tance is maintained. High contend of magnesium (3-6 %)
gives exceptional corrosion resistance against seawater.
6xxx Al-Mg-Si alloys Addition of both magnesium and silica increases strength,
but decreases corrosion resistance compared to Al-Mg al-
loys.
7xxx Al-Zi alloys Zinc and manganese significantly increases strength and
are therefor considered as a high-strength aluminium al-
loy. Corrosion resistance is slightly lower than Al-Mg-Si
alloys
8xxxx Misc. alloys Miscellaneous alloys, for example aluminium-lithium alloys.
Most relevant for this study are the AW-3xxx- and AC-4xxx grades, since they are
frequently used engineering materials for charge-air-coolers and intake manifolds.
Chapter 4
Corrosion & Electrochemistry
Corrosion in aqueous environment is of electrochemical nature, and the cause for corrosion
can depend on several different factors. Thus, basic knowledge of corrosion and electro-
chemistry is vital for understanding why corrosion occurs and how it can be prevented. This
chapter will therefore provide a brief literature review about corrosion and electrochemistry,
followed by how corrosion can be studied by electrochemical methods. Furthermore, corro-
sion protection techniques and results from earlier corrosion studies in synthetic exhaust gas
condensates will be addressed.
4.1 Corrosion of Stainless Steels and Aluminium Alloys

Corrosion of stainless steel and aluminium can be categorised as either general or localised.
General corrosion is defined as a corrosion attack, which uniformly thins the metal. Localised
corrosion is defined to be highly concentrated on local areas, or zones, often resulting in
formation of pits and holes [16]. Localised corrosion can further be divided into: pitting
corrosion, crevice corrosion, stress corrosion, intergranular corrosion, differential aeration
cell, depending on initiation origin.
4.1.1 General Corrosion

General corrosion is rarely a problem for stainless steels and aluminium, because it is usually
a slow process, which often can be managed by increasing metal thickness. However, in very
acidic environments formation of the protective oxide layer can become thermodynamically
unfavorable, leading to loss of passivation and rapid acceleration of uniform corrosion.
For stainless steels addition of molybdenum greatly enhances the ability of passivation
in acidic environments. For aluminium this passivation ability can not be as manipulated
as for stainless steels. Studies performed by ASM, shown that below a certain acidic level
(usually around pH 4) corrosion rate of aluminium starts to increase exponentially [16].
This can become a severe issue, especially for aluminium EGR-coolers, since pH of exhaust
gas condensate has been show to be well below pH 4.
Uniform corrosion rate may be summarised to accelerate with increase in temperature,

fluid velocity, oxidising power (potential) and concentration (pH) [14]. The corrosion of
aluminium is also severe at high pH, as is clear in the Pourbaix diagram, figure 3.1.
15
16 Chapter 4. Corrosion & Electrochemistry
4.1.2 Localised Corrosion

Pitting Corrosion
Pitting Corrosion is for automotive applications a more severe issue than uniform corrosion,
because it is more difficult to detect, predict and design against [16]. Small material pene-
tration can be enough to cause severe system failure, such as EGR-cooler leakage. Chloride
ions are known to significantly increase susceptibility of pitting corrosion and causing pits to
become autocatalytic; metal ions, M 2+ dissolves in the pit and the increased concentration
of positively charged metal ions will start to attract nearby chloride ions, resulting in the
autocatalytic mechanism of pitting corrosion, see figure 4.1. This type of corrosion is very
localised and can result in very fast and unexpected metal penetration.
Figure 4.1: Schematic of the autocatalytic process of pitting corrosion.
Hydration of dissolved metal ions, M e(OH)n , lowers pH drastically in the pit, H2 O →

OH − + H + . Susceptibility for pitting corrosion increase with temperature and by presence
of oxidizing agents, such as oxygen. [14]. Fluid velocity is also has an important factor,
affecting pitting corrosion. Generally stagnant solutions and low fluid velocities is necessary
for pitting corrosion to occur [16]. This may especially be an issue for EGR systems, be-
cause of the cyclic operation of an automotive engine. The materials in the EGR system will
frequently by exposed by wet and dry cycles, which will frequently replenish condensate in
formed pits and cavity’s. Thus, special care should be given, when designing EGR systems,
so that local zones with possibility of an accumulation of condensate is minimized.
From section 3, it is evident that some alloying elements for stainless steels give better
resistance against pitting corrosions than others. This is well known and several corrosion
studies in the area has lead to the development of the Pitting Resistance Equivalent Num-
ber (PREN) = [%Cr+3.3x%Mo+16x%N] [12, 14, 16]. Resistance against pitting and crevice
corrosion is considered to increase with PREN. However, this is not a precise tool, special
consideration needs to be taken for the EGR system, due to the cyclic nature of operation.
4.1. Corrosion of Stainless Steels and Aluminium Alloys 17
The repassivation effect of nickel should for example be given note, because high nickel
content is expected to result in faster repassivation.
Duplex stainless steels are considered to have a better resistance against chloride-induced
pitting corrosion. Aluminium have a tendency for pitting in solutions containing chlorides,
e.g. aluminium brasses are known to be sensible to pitting in pollutant waters [14].
Crevice Corrosion
Small gaps between joints can trap con-
densate, resulting in similar autocat-
alytic corrosion mechanisms as for pit-
ting corrosion, see figure 4.2. Metals
with high PREN-numbers will therefore
also have good resistance against crevice
corrosion. Crevice corrosion appears
mainly due to reduced oxygen concen-
tration. To maintain passivity sufficient
oxygen is usually needed. In crevices,
oxygen concentrations can become sig-
nificantly reduced, which may result in
loss of passivity. Thus, resulting in the Figure 4.2: Schematic of crevice corrosion
anodic behavior seen in figure 4.2.
Crevice corrosion is promoted with increase of tightness, depth, chloride and acid concen-
trations [14]. The primary prevention measure for crevice corrosion is design [16]. Design
strategy should not be to minimize gaps, since this will promote more aggressive crevice
chemistry. Design strategy should instead be to maximize the gap, and to minimize the
length of the gap. The reason is that extremely tight crevices, that are not water tight,
exhibit tremendous capillary action [16].
Differential Aeration Cells

Most solutions are in contact with atmospheric oxygen, but situations can arise where pres-
ence of oxygen can differ from one local part of the metal to another. This circumstance
can lead to localised attack, and is referred to as differential aeration cells [12, 16]. The
local part with the higher oxygen concentration acts as the cathode, while the lower oxygen
concentrated part, acts as the anode. Soot deposits in the EGR system can be expected to
create such differential aeration cells, since the soot prevent the underlaying metal to come
in direct contact with the exhaust gases (oxygen).
Studies in this regarded for EGR systems has not been found, but it is known that dirt
and soot deposits reduce oxygen concentration for the underlaying metal. A combination
of elevated levels of corrosive elements in soot and reduced oxygen concentrations, have the
potential to form a very corrosive environment, similar to what can be found in a crevice,
i.e. soot can be expected to have the ability to form artificial crevices.
Stress Corrosion Cracking (SCC)

Stress corrosion cracking is a term describing the
synergistic interaction between corrosion and me-
chanical stress, see figure 4.3. A specific metal al-
loy has a critical stress concentration factor, Klc ,
when stress load reaches past Klc , material fail-
ure will occur by cracking. If the metal is exposed
to a corrosive solution in combination of stress,
then cracking can occur at much lower stress lev-
els. The chemical environment needs only to be
mildly corrosive for SCC to become a problem.
Therefore, SCC can cause sudden and unexpected
material failure. The beneficial alloying elements Figure 4.3: Synergistic effect of SCC
against SCC are highly dependent on the chemical
environment. Thus, it is difficult to give any generalised advice, but the popular 304 and
316 austenitic stainless steels has shown to be especially sensitive towards chloride-induced
SCC [12, 16, 17].
The relative merits against SCC for different steels grades, depends on specific solutions,
but generally ferritic and duplex grades show much better resistance against SCC than
austenitic grades. The 2205 duplex grade can for example, basically be considered immune
against SCC up to a temperature of 150 o C [17].
For aluminium, SCC occurs for certain alloying grades, often those which have been
developed for medium and high strength (Al-Mg, Al-Cu, etc.). SCC has not been observed
for pure aluminium. Furthermore, SCC for casted aluminium is not common, but happens
time to time [12]. However, aluminium should in general, just as austenitic stainless steels,
be considered sensitive against SCC in chloride containing environments.
Intergranular Corrosion (IGC)

Metals are composed of crystals, i.e. grains surrounded by grains boundaries. These grain
boundaries can have significantly lower chromium alloying content relative to the metal bulk.
These chromium diluted areas may therefore act as initiation sites for localised corrosion.
This corrosion phenomena is referred as Intergranular Corrosion (IGC). Dilution of alloying
elements is primarily an issue at heat affected zones (HAZ).
HAZ is a area which have had its microstructure altered by welding or other intense heat
operations, i.g. cutting. Steel grades with high carbon contents have increased suscepti-
ble for intergranular corrosion, because of carbons ability to form chromium carbides along
grain boundaries. This results in zones near the grain boundaries with reduced chromium
content and thus acting as initiation sites for IGC [16].
Formation of chromium carbides is today often considered as a non-issue, because car-

bon content in modern stainless steel grades is generally very low (<0.03 %). Martensitic
stainless steels are an exception, since their high hardness properties is due to the relatively
high carbon content.
Duplex grades require greater care in welding compared to austenitic grades, because
4.2. Electrochemistry 19
of their mixed microstructure. Too high welding temperatures will cause increased ferritic
structure, which can result in loss of corrosion resistance and thus increased susceptibility
for IGC. Welding techniques for duplex grades should therefore be selected with minimised
heat dispersion [20].
4.2 Electrochemistry
An electrochemical reaction, is a reaction followed with transport of electrons from one
electron conducting material to another. These two conducting materials are called elec-
trodes. The electron source is referred as the cathode and the electron sink referred as the
anode. Oxidation is the loss of electrons and occurs at the anode and reduction is the gain
of electrons from the electron source, i.e. cathode, see example below.
F e(s) → F e2+ (aq) + 2 e− (4.2.1)
1
/2 O2 + H2 O + 2 e− → 2 OH − (4.2.2)
Reaction 4.2.1 shows the anodic reaction (this is where metal loss through anodic oxi-
dation occurs). Reaction 4.2.2 shows the cathodic reaction, in this case oxygen reduction.
Reaction 4.2.1 and 4.2.2 are known as redox reactions (oxidation and reduction) and the
overall equation is:
F e(s) +1 /2 O2 + H2 O → F e2+ (aq) + 2 OH − (4.2.3)

In acidic solutions, hydrogen evaluation is the dominating reduction reaction, see equa-
tion 4.2.4.
2H + (aq) + 2e− → H2 (g) (4.2.4)

An electron does not exist freely in a solution. Therefore oxidation must always be
accompanied with reduction. If a metal, M(s), is immersed in a electrolyte containing
M n+ (aq) ions, electrochemical reactions will occur until equilibrium is reached, see equation
4.2.5.
M n+ (aq) + ne
M (s) (4.2.5)
These reactions will create a charge on the metal surface, either positive or negative. An
electric double layer is formed subsequently, since anions and cations with their opposite
charge attracts and bounds electrostatically to the metal surface, see figure 4.4.
The charge separation creates a potential difference between the metal and electrolyte,
figure 4.4. This creates a potential difference near the metal surface. This is commonly
referred to as the electrode potential, E. The charged interface, can because of it is struc-
ture be simplified as a capacitor. The significance of the double layer is that it provides a
barrier for transfer of electrons. Thus, the double layer act as an energy barrier, that must
be overcome for electrochemical reactions to occur [14].
Metals with high electrode potentials are therefore more inert, compared to metals with
low electrode potentials. The absolute electrode potential cannot be measured directly, but
it is possible to measure a relative value with an reference electrode. Noble metals such
as gold and silver have large positive electrode potentials, while zinc and magnesium have
Figure 4.4: Simplified illustration of the double layer at a metal aqueous interface.
large negative electrode potentials. Zinc and magnesium will because of their lower electrode
potential be more thermodynamically reactive than gold and silver.
Electrode potential gives information about the thermodynamic tendency for an elec-
trochemical reaction to occur, but gives no information about reaction rate i.e. corrosion
rate. However, accordingly to Faraday’s law the electron transport (current flow) is directly
proportional to the material loss, see equation 4.2.5.
I nF m
Q= = (4.2.6)
T M
where,
Q = Charge (C), I = Current (C/s), t = time (s), F = Faraday’s constant, n = Number

of transfered electrons, m = mass of oxidised metal (g), M = Atomic Weight (g/mol).
Thus, by measuring both potential and current flow, information of both thermodynam-
ics and corrosion rate can be obtained. Electrochemical methods such as anodic polarisation,
is because of this a powerful tool for studying passivated metals susceptibility for localised
corrosion.
4.3 Anodic Polarisation Sweep

Anodic polarisation is a method to predict the tendency of an alloy to suffer localised
corrosion in the form of pitting and crevice corrosion. The method was designed for use
with iron- or nickel- base alloys in chloride environments [14]. This experiment requires
a potentiostat, working electrode (WE), counter electrode (CE), reference electrode (RE)
and a corrosion cell. The potentistat provides the necessary potential control for the WE
and CE. The reference electrode, RE, provides the means to observe this experiment. The
potentiodynamic scan is performed at a fixed voltage scan rate (mV /s), during which current
and potential is being measured and registered. Potential is increased until the current
density, i, reach a certain value, scan is then reversed, see figure 4.5. The forward scan gives
information about initiation of pitting, while the reverse scan provides information about
alloys repassivation behavior.
4.3. Anodic Polarisation Sweep 21
Figure 4.5: Schematic of an anodic polarisation sweep for stainless steel.
Figure 4.5 illustrates which information can be obtained from a anodic polarisation
curve. However, it should be stressed that this is a schematic figure illustrating some of the
possible regions present on a anodic scan. Depending on the metal and environment com-
bination, some or all of these features may be present. The polarisation curve is presented
with potential, E, on one axis and the logarithmic current density, log i, on the other axis.
If no polarisation potential is applied, cathodic (reduction of oxygen) and anodic reaction
(oxidation of metal) will be in equilibrium. The size of the anodic and cathodic currents will
be identical so that the net current trough the interface is zero. This potential is referred
as the corrosion potential Ecorr or mixed potential Emix , see point b). If polarisation scan
is performed in negative direction cathodic reactions will dominate, point a), resulting in
reduction of water with hydrogen evolution as consequence. However, potential is generally
applied in the positive (anodic) direction, since this give information about the anodic re-
action, i.e. corrosion of the metal.
Metals are usually not fully passivated when the scan is initiated. Therefore, some oxi-
dation is needed for passivity to occur, point c). Maximum current density obtained before
passivation, is referred as icc , and varies between grades. Ease of passivation, increases with
decrease in icc . If potential is increased from icc current density decrease, until reaching the
metals passive state, point d). A passive layer has now formed on the metal surface, and it
is in this region we want the metal. If the potential is increased, the passive layer begins to
periodically breakdown, see point e). This is referred as transpassive behavior.
If potential is further increased, the passive layer will eventually breakdown, see point f).
This results in significant increase of current density and the potential for when this occur,
is referred as the breakdown potential, Eb . If the metal has very good corrosion resistance,
other reaction can cause this significant increase in current density e.g. oxygen evolution.
After reaching a certain potential or current density, scan is then reversed, until the metal
intersects with the anodic scan, point g). This potential is referred as the protection poten-
tial, Ep . This potential gives information about how easily the metal will re-passivate. If
the loop, referred as hysteresis loop, closes at or above Eb , it indicates very low likeliness for
pitting and crevice corrosion [16, 21]. The maximum obtained current density, Imax , will
vary between different grades and can therefore also be considered when evaluating metals.
Consequently, when evaluating materials relative susceptibility for pitting corrosion focus
is on these three features – Eb , Ep and Imax . In a specific solution and temperature; the
metals with highest Eb and Ep are less likely to be susceptibility to localised corrosion.
Between these to values Ep should be regarded as more conservative [14]. Meaning, if metal
A show higher Eb than metal B, but metal B show higher Ep than metal A. Then metal B
should be regarded superior.
4.4 Corrosion Protection Techniques

There exist several different corrosion protection techniques, but the easiest way to avoid
corrosion is generally by selecting sufficiently corrosion resistant alloy. However, since there
are other ways of protecting components against corrosion a short review will follow, see
below.
4.4.1 Corrosion Inhibitors

Inhibitors suppress the electrochemical reactions that take place during corrosion. Corrosion
can only occur if both a cathodic and anodic reaction takes place. Hence, if inhibitors
successfully suppress one of these reactions corrosion will be prevented. Inhibitors can
be divided as anodic, cathodic or mixed, depending on which corrosion reactions is being
suppressed [14]. A mixed inhibitor suppresses both anodic and cathodic reactions. Inhibitors
are frequently used to prevent corrosion for close-looped water recirculation systems, such
as engine coolant systems.
4.4.2 Anodic and Cathodic Protection

In a system with different metals, the metal with lowest electrode potential will corrode
(act as anode). Thus, thermodynamic driving force of corrosion can be used as a corrosion
protection technique. By sacrificing the anode, the cathode can be protected. This corro-
sion control technique is referred as cathodic protection or galvanic protection. Sacrificial
anodes are frequently used to protect ship hulls to seawater corrosion.
Anodic protection works by shifting the electrode potential into the metals passivated
region, see point d) in figure 4.5. This is a technique used for a limited number of systems
were passivation do not occur naturally. Most frequent use of this technique can be found
in chemical storage tanks. Anodic protection is used to a lesser degree compared to other
corrosion control, since it is only in a few combinations of grades and solutions, this protec-
tion technique is advantageous. If anodic protection is used improperly corrosion rate can
instead accelerate [14].
4.5. Automotive Corrosion Studies 23
4.5 Automotive Corrosion Studies

This section will address exhaust gas condensate corrosion studies for automotive applica-
tions.
4.5.1 Corrosion Study of Different Stainless steels in Synthetic Ex-

haust Gas Condensate
In a study conducted by C. Hoffman et .al.[22] corrosion resistance in synthetic exhaust-gas
condensate was investigated for six different stainless steels. Table 4.1, list grades and com-
positions for stainless steels included in this study.
Table 4.1: Test materials and steel composition (wt%) [22].
Steel: EN Cr Ni Mo Mn
1.4512 11.54 - - -
1.4509 17.59 - - -
1.4526 16.86 - - 0.983
1.4376 18.93 3.33 - 7.77
1.4301 17.88 9.08 - -
1.4404 16.36 11.88 1.84 -
10 ml of their synthetic condensate were comprised of: 11 g acetic buffer solution, 3.3
g sodium chloride, 1 g active carbon, which resulted in solution with pH 4. Before testing;
samples were cleaned and weighted. The test samples were then immersed in the synthetic
condensate and placed in a climate chamber, which had the initial temperature 85o C and
50 % relative humidity. After 12 hours the temperature was lowered to 23o C and relative
humidity was kept at 50 %. After additional 12 hours the samples were cleansed, weighted
and new electrolyte was added. This cycle was repeated 48 times. The corrosion resistance
of the materials were then evaluated in two regards – average pit depth and total mass loss.
Steel grade 1.4404 showed best corrosion resistance, while 1.4376 showed lowest mass
loss, but highest average pit depth. Results from this study indicates that molybdenum has
a strong positive effect on corrosion resistance in exhaust gas condensate. 1.4526 and 1.4376
are both alloyed with manganese, but manganese addition in 1.4376 is much higher, see table
4.1. This lower addition of manganese resulted in a higher mass loss, but lower average pit
depth. Low alloying with manganese proves to more beneficial than high alloying, since
lower average pit depth is regarded more beneficial than the increased mass loss. 1.4509 has
similar chromium content as 1.4526, while having no alloying with manganese and results
show that that these two grades have similar corrosion resistance. Therefore, the alloying
effect of manganese is shows to be limited.
4.5.2 Corrosion Study for Automotive Mufflers

Exhaust gas condensate cause severe corrosion inside automotive mufflers. In order to
simulate the corrosion effect of exhaust-gas condensate on mufflers Hirasawa et. al [23]
performed cycling immersion tests with synthetic exhaust gas condensate. Table 4.2, list
the grades and their composition of the stainless steels included in this study.
Table 4.2: Test materials and steel composition (wt%) [23].
Steels C Si Mn Cr Cu Mo Ti Nb
∗
Aluminised Steel 0.047 0.12 0.31 0.03 - - - -
Type409L 0.012 0.49 0.48 11.2 - - 0.24 -
Type439L 0.012 0.09 0.23 17.6 - - 0.26 -
Type430J1L 0.016 0.44 0.22 19.3 0.53 - - 0.45
Type436L 0.010 0.1 0.29 18.1 - 1.2 0.27 -
∗) Al-plating weight: 80g/m2
In order to simulate the effect of oxidation, caused by the hot exhaust gases, the test
samples were pre-oxidized at 400o C for 5 hours prior to immersion in the synthetic solution.
The synthetic exhaust-gas condensate consisted of Cl− (50 ppm), SO32 − (250 ppm), SO42 −
(1250 ppm), CO32 − (2000 ppm), N H4+ (2500 ppm) and 50 g/L active carbon. Solution and
test samples where kept in a beaker and test temperature was set to 80o C. The cover of
the beaker was adjusted so that the solution evaporates completely in 24 hours. After 24
hours the beaker was cleansed and refilled with new solution. Test specimens were also
soft brushed. Corrosions behavior was analysed by maximum corrosion depth and prior to
measurement the corrosion products were removed by nitric acid or diammonium hydrogen
citrate.
Results obtained after 10 cycles show that the maximum corrosion depth for 430J1L
and 436L were lowest, while the corrosion resistance of the aluminised steel was very poor.
From measured pit depths a pitting index could be derived, accordingly: [%Cr + 3x%M o +
1.5x%Cu]. Thus, molybdenum was three times as effective as chromium, and copper one
and a half times as effective as chromium.
A long time field test was conducted for comparison reasons. Special mufflers were de-
signed so that small test specimens could be built in. Results from the field study showed
good agreement with the relative ranking retrieved from the accelerated tests, but the over-
all corrosion rate in the field study was lower than in the accelerated tests. Thus, simulation
conditions were too severe. The following two reasons were suggested by the authors: 1)
chloride content in the test solution was too high. 2) test specimens in the field study did
not become fully immersed in condensate, because of their placement.
Results from this study show that molybdenum greatly increase corrosion resistance,
alloying with copper also shows to be beneficial.
4.6 Engineers Diagram

It is common that steel manufactures provide corrosion data on their different steel grades.
These data are usually based on laboratory experiments in clean acid solutions. Results
from these tests are often presented as isocorrosion curves and corrosion tables. The isocor-
rosion curve shows how critical pitting temperature, CPT, varies with temperature and acid
concentration. Exhaust gas condensate contains both sulphuric and nitric acid. Therefore,
corrosion resistance towards these acids are of interest, see figure 4.6 and 4.7 for example on
isocorrosion curves for stainless steels in sulphuric acid. However, exhaust gas condensate
4.6. Engineers Diagram 25
contains several other corrosive elements which may have a synergistic effect. Isocorrosion
curves should therefore be used with caution.
Figure 4.6: Isocorrosion curves, 0.1 mm/year, Figure 4.7: Isocorrosion curves, 0.1 mm/year,
in sulphuric acid [24]. in sulphuric acid containing 2000 ppm chloride
ions [24].
Steel 4404 is the Outokumpu name for the frequently used 1.4404 austenitic steel (17Cr-
8Ni-2Mo) and 2304 and 2205 is the Outokumpu name for duplex steel 1.4362 (23Cr-4.8Ni-
0.3Mo) and 1.4462 (22Cr-5.7Ni-3.1Mo) respectively. Duplex 2304 show similar corrosion
resistance as 1.4404 in most sulphuric acid concentrations, but in low sulphuric acid concen-
trations duplex 2304 has superior corrosion resistance compared to 1.4404, see figure 4.6.
Duplex 2205 show significantly better corrosion resistance compared to 1.4404 in both these
solutions, because of its high alloying of chromium and molybdenum. Austenitic stainless
steels are known to be more sensitive to chlorides compared to duplex grades, and this can
be seen in figure 4.7. These two isocorrosion curves can be summarised to show that du-
plex grades are attractive alternatives to austenitic grades regarding corrosion resistance,
especially in chloride containing environments.
Chapter 5
Experimental
This chapter presents the materials and the synthetic exhaust gas condensates selected for
the electrochemical measurements, followed by a brief experimental description.
5.1 Materials
Eight different commercial stainless steel grades, as shown in table 5.1, were investigated in
this study. The chemical compositions, pitting resistance equivalent number (PREN) and
alloy adjustment factor (AAF) for these alloys are shown in the table. More details of the
ferritic and duplex grades can be found in appendix figure 14 to 18. For martensitic steel
grade 1.4112, nitriding treatments was in addition performed on the metal surface. The
experiments were then performed in order to compare the results obtained on martensitic
steel with and without surface treatments.
Table 5.1: Stainless steels grades selected for electrochemical evaluation.
Grade C (%) N (%) Cr (%) Ni (%) Mo (%) Other PREN AAF

(EUR)**
Austenitic
1.4404 0.02 - 17.2 10.2 2.1 - 24.1 2167
1.4301 0.04 - 18.1 8.3 - - 18.1 1414
Ferritic
1.4521 0.018 - 18.2 - 2.02 Ti 25 954
1.4509 0.02 - 18.02 - - Nb, Ti 18 642
Duplex
2205 0.014 0.187 22.4 5.76 3.18 1.5Mn 35 1714
LDX2404 0.022 0.271 23.94 3.78 1.59 3.03Mn 33.6 1482
2304 0.018 0.131 23.51 4.86 039 1.5Mn 25.6 1103
Martensitic
1.4112* 0.9 - 18 - 1.1 Si, 0.1V 22 -
*) Tested plasma and gas nitrided. **) Outokumpu Flat Products, June 2012 [26].
Furthermore, two different commercial aluminium grades EN AW-3003 and EN AC-

43000KF were investigated in this study. Table 5.2 shows the chemical compositions of
these aluminium alloys. For aluminium alloy 43000KF, anodising surface treatment was
27
28 Chapter 5. Experimental
performed according to ISO 7599 [25]. Experiments were performed in order to investigate
if the anodising treatments have any positive effects on corrosion resistance in exhaust
condensate environments.
Table 5.2: Aluminium grades selected for electrochemical evaluation.
Grade Si (%) Fe (%) Cu (%) Mn (%) Mg (%) Ni (%) Zn (%) Ti (%)

EN AW-3003 0.6 0.7 0.05-0.20 1.0-1.5 - - 0.10 -
EN AC-43000 KF* 9.0-11.0 0.55 0.05 0.45 0.20-0.45 0.05 0,10 0.15
*) Also tested anodised
Austenitic stainless steel 1.4404 is a common used material for EGR components. The
general corrosion resistance of 1.4404 is good even in environments with low pH value. The
pitting corrosion resistance at room temperature is improved due to molybdenum. However,
nickel and molybdenum are expensive alloying elements.
Even though with high nickel and some molybdenum contents, the pitting corrosion re-
sistance of steel 1.4404 will be greatly decreased at elevated temperature, with increased
chloride content and under low pH value environment. Sensitivity to chloride induced pit-
ting corrosion has been an Achilles heel for austenitic stainless steels.
Whilst, duplex stainless steels consisting of ferritic and austenitic microstructure have
usually better pitting resistance in addition to the good corrosion resistance in low pH value
environments. The chloride induced stress corrosion cracking resistance is usually higher
for duplex stainless steels than austenitic stainless steels. Further, duplex and ferritic stain-
less steels are also more cost effective, since they contain less nickel compared to austenitic
grades, see table 5.1.
Duplex stainless steels have about twice the tensile strength as regular austenitic stain-
less steels, but the toughness and ductility of duplex stainless steels are lower compared to
austenitic stainless steels.
If increased corrosion resistance of austenitic grades are sought, it will most definitely
mean higher alloy costs. There is a possibility to maintain corrosion resistance at low mate-
rial cost with the use of duplex and ferritic grades. Duplex grade 2304 should provide similar
corrosion properties as the austenitic 1.4404 grade, in addition with better resistance against
chlorides and stress corrosion cracking (SCC). This makes duplex grades such as 2304 to an
attractive alternative to the austenitic 1.4404 steel grade.
Use of EGR may cause increased wear on piston rings, resulting in reduced engine life
time. Wear resistance might be increased by nitriding treatment on the contact surface.
However, there are concerns that nitriding will results in decreased corrosion resistance.
1.4112 was therefore selected in order to investigate if the nitriding treatment will have a
negative effect on corrosion resistance. Two techniques were selected for this purpose: gas
and plasma nitriding.
The two aluminium grades selected are typical commercial grades used for constructing
charge-air-coolers. Aluminium has significantly higher thermal conductivity and lower den-
sity compared to stainless steel and is therefore a more preferable material to use for cooling
5.2. Synthetic Exhaust Gas Condensate 29
of exhaust gases. It is therefore of interest to compare corrosion resistance of aluminium

and stainless steel. Corrosion resistance of aluminium can be improved by anodising, which
is a technique used to increase the thickness of the passive oxide film.
5.2 Synthetic Exhaust Gas Condensate

Five different synthetic exhaust gas condensates, as shown in Table 5.3, were used for this
study. Condensate acidity varies between pH 2.5 and pH 1.5. Chloride concentrations varies
between 32 ppm, 200 ppm and 3300 ppm. Condensate 1 is the least corrosive condensate
with pH 2.5 and a relatively low chloride concentration. Condensate 2 has a significantly
increased nitric acid concentration and chloride concentration, and also small concentrations
of organic weak acids.
Table 5.3: Chemical composition of synthetic exhaust gas condensates (C) used for anodic
polarisation measurements.
mg/l C1 C2a C2b C3a C3b

H2 SO4 110 15 15 2900 2900
HN O3 200 2900 2900 15 15
F ormic Acid - 20 20 20 20
Acetic Acid - 20 20 20 20
Cl− 32 200 3300 200 3300
pH* 2.5 1.5 1.5 1.5 1.5
*) Control measured after each batch.
Three different alternatives to condensate 2a were additionally selected (2b, 3a, 3b). The
purpose was to investigate the parameters that have most influences on pitting corrosion
resistance. Test temperature was 60o C,since it is the temperature which can still be expected
to produce substantial amount of exhaust gas condensate in the EGR system. The test
matrix is shown in Table 5.4.
Table 5.4: Grades and condensates (C) experimentally tested in this study.
Grade C1 C2a C2b C3a C3b

1.4404 X X X X X
1.4301 X X X X X
2205 X - X - X
LDX2404 - - X - X
2304 X X X X X
1.4521 X X X X X
1.4509 X - - - -
1.4112 X - - - -
1.4112 Plasma N X - - - -
1.4112 Gas N X - - - -
AW-3003 X - - - -
AC-43000KF X - - - -
AC-43000KF Anod X - - - -
30 Chapter 5. Experimental
As seen, not all combinations could be tested, due to limited time.
5.3 Sample Preparation

Test samples were wet abraded by using #120 to #600 grit SiC paper and cleansed af-
terwards by deionised water and ethanol. Test samples were then dried with a hairdryer
and left exposed to room environment for 12-16 hours before electrochemical measurements.
Test samples with surface treatments (4112 and AC 43000KF) were tested as delivered.
A water bath was used to preheat the condensate to 65o C. This higher temperature
was chosen because of expected heat loss during test preparation. The measurements were
then performed in a climate chamber at 60o C. The reference electrode was preheated in the
climate chamber. A brief description of the electrochemical experiments is given in the next
section.
5.4 Potentiodynamic Polarisation Measurements

All electrochemical experiments were carried out in a three-electrode cell with Ag/AgCl (3M
KCl) as reference electrode and a platinum net as counter electrode. The working electrode
with an immersed area of 1cm2 was mounted into the cell before measurements.
The electrochemical instrument set-up, as shown in figure 5, consists of a Solartron 1287

potentiostat, using CorrWare software package. Temperature were kept constant at 60o C
with a B.I.A climate chamber. The scan rate was 0.1667 mV/s, which is the recommended
by ASTM, and scan was programmed to be reversed at a current density of 10 µA/cm2 or
2 volt above OCP (open circuit potential). For aluminium grades and stainless steel grade
1.4112 the polarisation scan were programmed to be reversed at 0.8V vs reference electrode.
Figure 5.1: Experimental Setup: 1) Solartron 1287 potentiostat 2) Ag/AgCl reference

electrode (3M KCl 3) Platinum plate counter electrode 4) Working Electrode 5) Corrosion
Cell 6) B.I.A climatic climate chamber 7) Corrware
R Software Package.
Chapter 6
Results
This chapter presents the results obtained from the electrochemical measurements.
6.1 Condensate 1
Condensate 1 consists of 110 ppm sulphuric acid, 200 ppm nitric acid and 32 ppm Cl− . The
pH value of condensate 1 was 2.5.
6.1.1 Condensate 1 – Austenitic, Ferritic and Duplex Grades

The anodic polarisation curve is a useful method to evaluate the pitting corrosion resistance.
Figure 6.1 shows the anodic polarisation curves of different stainless steels in condensate 1
at 60o C. A relatively stable corrosion potential (also called open circuit potential, OCP), as
indicated in figure 6.1, was obtained after 10 minutes exposure in the condensate. The anodic
polarisation was then started at the corrosion potential with a scan rate of 0.1667 mV/s. The
current density increased with increasing potential, indicating that more corrosion occurred.
When the potential was increased to about 0.1 V vs. reference electrode, the material went
into a passive range. This is because a thin passive film has formed on the metal surface
consisting of chromium oxide. Corrosion continues but with a very small current density;
the passive current, ip . However, with further increased anodic potential, the passive layer
will breakdown. This potential is called the breakdown potential. The breakdown potential,
passive range, passive current density and maximum current density before passivation are
parameters to be used to evaluate the pitting corrosion resistance of stainless steels. A
stainless steel with good pitting corrosion resistance will have a higher pitting potential,
bigger passive range, lower passive current and maximum current.
31
32 Chapter 6. Results
Figure 6.1: Anodic polarisation plot for stainless steel grades Duplex 2205, Duplex 2304,
1.4301, 1.4404, 1.4509, 1.4521.
Even though the pH value was only 2.5, the pitting corrosion resistance of all tested
austenitic, ferritic and duplex stainless steels was good. Most notable difference was on the
maximum current density before passivation. Plots show that the duplex grades require
lowest current before passivation. This indicates that duplex grades passivate more easily
than austenitic and ferritic grades. The relatively low alloyed 1.4509 ferritic grade required
highest current before passivation. For all tested austenitic, ferritic and duplex grades, the
anodic polarisation predicts low risk for pitting corrosion in condensate 1, since repassivation
potential was similar to the breakdown potential, as can be observed in table 6.1.
Table 6.1: Eb , Ep and Imax obtained from anodic polarisation of austenitic, ferritic and
duplex grades in condensate 1.
Grade Eb (V ) Ep (V ) Imax (µA/cm2 )

1.4404 ∼0.70 ∼Eb ∼10
1.4301 ∼0.70 ∼E b ∼10
2205 ∼0.70 ∼Eb ∼10
2304 ∼0.70 ∼Eb ∼10
1.4521 ∼0.70 ∼Eb ∼10
1.4509 ∼0.70 ∼Eb ∼10
6.1. Condensate 1 33
6.1.2 Condensate 1 - Grade 4112 & Effect of Nitriding
Figure 6.2 shows the anodic polarisation curves obtained on steel 1.4112, with and without
nitriding treatment, in condensate 1 at 60o C. Appendix figure 1 shows photos of tested
samples after polarisation scan.
Figure 6.2: Plot from anodic polarisation measurement of nitride coated and uncoated
1.4112 martensitic stainless steel in condensate 1.
It was observed that the tested materials was experiencing serious general corrosion and
pitting corrosion. No significant passivation can be seen. The current densities are in the
magnitude 1000 times higher compared to the austenitic, ferritic and duplex steels at 0.8V
vs reference electrode.
There is no significant differences between the nitrided samples and the steel without
any surface treatment.
Microscopic examination of nitrided samples were carried out to investigate if the two
nitriding films had been affected differently by the anodic polarisation. Appendix figure
5 to 10 show nitride film before and after the anodic polarisation. The nitride film for
both test samples is observed to have been damaged by both general and pitting corrosion.
Subsequently, no difference between nitriding techniques can be observed. However, it is
unclear if similar damage can be expected to occur naturally (un-polarised), since the anodic
polarisation, which these two samples were subjected to, forced the material to undergo large
corrosion currents.
6.1.3 Condensate 1 – Aluminium Grades & Effect of Anodising

Figure 6.3 shows anodic polarisation curves for aluminium samples in condensate 1 at 60o C.
It is observed that tested aluminium grades show similar polarisation behavior as that of
the martensitic stainless steel 1.4112 steel.
Figure 6.3: Anodic polarisation plot for aluminium grades AC43000KF and AW-3003 in
condensate 1. AW-3003 and AC-43000KF-anodised also tested after 12 hour exposure.
The general corrosion rate was high. No passivation can be seen, see appendix figure 2.
It can be observed that the area which is exposed to the condensate is severely damaged by
the anodic polarisation. These results confirm the general understanding of aluminium, i.e.
low pH cause dissolution of the formed aluminium oxide film.
Anodised aluminium show initially relatively high OCP and low Icorr compared to the
un-anodised test sample. However, when potential is increased in the anodic direction the
current density increases rapidly. When the potential approach 0.8 V vs ref, all specimens
have similar current densities. This shows that not even the anodised film is stable at anodic
polarisation at low pH. To further investigate this, anodic polarisation were preformed after
twelve hours of exposure to condensate at 60o C. OCP was measured during this twelve hour
period and result is displayed in figure 6.4.
Figure 6.4: Open circuit potential (OCP) for Al AC-43000KF in condensate 1.
During this entire measuring period, it is observed that OCP moves in the cathodic
direction. This indicates loss of passivity and dissolution of anodised aluminium oxide film.
After twelve hours of exposure OCP for the anodised test sample is observed to be close to
OCP for unanodised aluminium. The beneficial effect of anodising aluminium is therefore
very limited at low pH condensate environment. Grade AW-3003 was also tested after twelve
hours of exposure. The OCP moved slightly in the anodic direction, but did not notably
affect the anodic polarisation behavior.
6.2 Condensate 2
Condensate 2 is more corrosive than condensate 1, due to lower pH and higher chloride
concentrations. The aluminium alloys and the martensitic stainless steel did undergo serious
general corrosion in condensate 1. It is therefore not useful to test these materials in more
corrosive condensates, such as condensates 2 and 3. The ferritic steel 1.4509 was expected
to undergo general corrosion in such a low pH condensate and was therefore also excluded
from further testing, because of shortage of time.
6.2.1 Condensate 2a – Austenitic, Ferritic and Duplex Grades

Condensate 2a contains 15 ppm sulphuric acid, 2900 ppm nitric acid, 20 ppm formic acid,
20 ppm acetic acid and 200 ppm Cl− . The chloride concentration is about 6 times higher
than that for condensate 1.
Figure 6.5 shows the anodic polarisation curves for two austenitic, one duplex and one
ferritic stainless steel in condensate 2a at 60o C. The breakdown potential, Eb , respassivation
potential, Ep , and maximum breakdown current, Imax is shown in Table 6.2.
Figure 6.5: Anodic polarisation plot for stainless stainless steel 2304, 1.4301, 1.4404, 1.4521
in Condensate 2a.
Eb for all tested steels are about 0.75V vs. reference electrode, which is similar to the
results obtained in condensate 1. All hysteresis loops where completed at or above Eb. The
results indicate a low risk for pitting corrosion of these steels in condensate 2a. Critical
current density, Icc , is similar to what was observed in condensate 1, see table 6.2. However,
passive current density, Ip , is slightly higher for all tested grades compared to condensate 1.
This is expected since condensate 2a is more acidic than condensate 1. Duplex grades 2205
and LDX 2404 were excluded in this test series, since the low alloyed duplex grade 2304
showed low likeliness for pitting corrosion in this condensate.
Table 6.2: Eb , Ep and Imax obtained from anodic polarisation scan in condensate 2a.

1.4404 ∼0.75 ∼Eb ∼10
1.4301 ∼0.75 ∼E b ∼10
2304 ∼0.75 ∼Eb ∼10
1.4521 ∼0.75 ∼Eb ∼10
6.2.2 Condensate 2b – Austenitic, Ferritic and Duplex Grades

Condensate 2b contains 15 ppm sulphuric acid, 2900 ppm nitric acid, 20 ppm formic acid,
20 ppm acetic acid and 3300 ppm Cl-. Hence, chloride concentration is considerably higher
compared to condensate 2a (increased from 200 ppm to 3300 ppm). Three duplex stainless
steels (2205, LDX 2404 and 2304), two austenitic stainless steels (1.4404 and 1.4301) and
one ferritic stainless steel (1.4521) were tested in condensate 2b.
Figure 6.6 shows the anodic polarisation curves for all tested steels in condensate 2b.
Eb and Ep for austenitc grades 1.4301 and 1.4404 are about 0.73V and 0.3V lower, which
is lower compared to that in condensate 1 and 2a. Whilst Eb and Ep for duplex 2205,
LDX2404, 2304 and ferritic 1.4521 are similar to that observed in condensate 1 and 2a, see
table 6.3.
Figure 6.6: Anodic polarisation plot for stainless steel grades 2205, LDX2404, 2304, 1.4404,
1.4301, 1.4521 in condensate 2b.
Pits were visually observed for the austenitic grades after polarisation scan, see appendix
figure 3. More pits can be observed for the low alloyed 1.4301 grade. The reduced Eb and
Ep for 1.4301 and 1.4404 indicates that these two austenitic steels are susceptible to pitting
corrosion in this high chloride concentrated solution. It is unclear if 1.4301 was actually
passivated, since Eb and Ep is very close to the OCP. Steel 1.4404 shows higher resistance
against pitting corrosion compared to 1.4301, which is expected due to its alloying with
molybdenum.
Table 6.3: Eb , Ep and Imax obtained from anodic polarisation of austenitic, ferritic and
duplex grades in condensate 2b.

1.4404 0.45 0.05 125
1.4301 0.02 -0.15 957
2205 ∼0.75 ∼Eb ∼10
LDX2404 ∼0.75 ∼Eb ∼10
2304 ∼0.75 ∼Eb ∼10
1.4521 ∼0.75 ∼Eb ∼10
Tendency for pitting corrosion in condensate 2b at 60o C follow the sequence:
1.4301<1.4404<duplex 1.4521∼duplex 2304∼duplex LDX2404∼duplex 2205
Clearly, duplex stainless steels have better pitting corrosion resistance in low pH envi-
ronment when chloride concentration is increased.
6.3 Condensate 3
Concentrations of sulphuric and nitric acid were switched in condensate 3, relative to con-
densate 2 in order to investigate the effect of increased concentration of sulphuric acid on
pitting corrosion resistance. The pH was kept the same as that in condensate 2 (pH 1.5).
6.3.1 Condensate 3a – Austenitic, Ferritic and Duplex Grades

Condensate 3a contains 2900 ppm sulphuric acid, 15 ppm nitric acid, 20 ppm formic acid,
20 ppm acetic acid and 200 ppm Cl− .
Two austenitic stainless steels 1.4404 and 1.4301, one duplex stainless steel 2304 and
one ferritic stainless steel 1.4521 were tested in condensate 3a. Figure 6.7 shows the anodic
polarisation curves for these tested materials in condensate 3a at 60o C. The results were
similar to that obtained in condensates 1 and 2a, i.e. predicting low risk for pitting corrosion.
Figure 6.7: Anodic polarisation obtained for stainless steels 2304, 1.4404, 1.4301, 1.4521 in
condensate 3a.
Clearly, high concentration of sulphuric acid, low pH value but lower chloride content
(200 ppm) do not increase the risk for pitting corrosion for austenitic steels 1.4404 and
1.4301, duplex 2304 and ferritic 1.4521.
6.3.2 Condensate 3b – Austenitic, Ferritic and Duplex Grades

Condensate 3b contains 2900 ppm sulphuric acid, 15 ppm nitric acid, 20 ppm formic acid,
20 ppm acetic acid and 3300 ppm Cl− . Hence, condensate 3b has considerably higher con-
centrations of chlorides compared to condensate 3a.
Three duplex stainless steels (2205, LDX 2404 and 2304), two austenitic stainless steels
(1.4404, 1.4301) and one ferritic stainless steel (1.4521) were tested in condensate 3b. The
passivation behavior for 1.4301 was again very poor and similar to the behavior in conden-
sate 2b. As a result the programmed reverse current density (10µ Amps/cm2 ) was triggered
shortly after polarsation start. The measurements performed on the steel could therefore
not yield in a usable polarisation curve. This poor passivation behavior was only observed
for 1.4301 and is therefore regarded as inferior to all other tested stainless steels.
Figure 6.8 shows the anodic polarisation curves for the tested materials in condensate 3b
at 60o C. The breakdown potential, Eb , repassivation potential, Ep , and maximum break-
down current, Imax are shown in Table 6.4.
Figure 6.8: Anodic polarisation for stainless steel grades 2205, 2304, LDX2404, 1.4301,
1.4404, 1.4521 for condensate 3b.
As shown in figure 6.8, the Eb and Ep decreased drastically for all tested materials,
except for the duplex grade 2205.
Eb and Ep for austenitic 1.4404, duplex 2304 and LDX2404 are similar. 1.4404 has
slightly higher Eb , but slightly lower Ep compared to the two duplex steels. These three
grades show similar pitting corrosion resistance for initiation, but when pitting has been
initiated, the two duplex steels especially LDX2404 show better tendency for repassivation.
The ferritic grade 1.4521 had a relatively high Eb , but Ep was low and Imax high. The
stochastic nature of Eb makes Ep to a more conservative value. Ep should therefore be given
more weight. Pitting resistance for 1.4521 is therefore only regarded superior to 1.4301.
Table 6.4: Eb , Ep and Imax for tested austenitic, ferritic and duplex grades in condensate
3b.

1.4404 0.35 -0.06 378
1.4301 - - -
2205 0.75 ∼Eb ∼10
LDX2404 0.34 -0.03 88
2304 0.31 -0.04 245
1.4521 0.53 -0.11 620
Tendency for pitting corrosion in condensate 3b at 60o C follows the sequence:
1.4301<1.4521<1.4404∼duplex 2304∼duplex LDX2404<duplex 2205
Clearly again, duplex stainless steels have better pitting corrosion resistance in low pH
environment when chloride concentration is increased.
Chapter 7
Discussion and Summary
The focus of this Master’s thesis is to investigate the pitting corrosion resistance of materials
in EGR system where exhaust gas condensation may occur. Based on the knowledge from
the literature review and earlier studies performed at Scania, totally eight different stainless
steels and two aluminium alloys were included. Furthermore, nitriding surface treatments
on one martensitic stainless steel and anodising surface treatments on one aluminium alloy,
were also included in this study.
Totally five condensates with different concentrations of sulphuric acid, nitric acid and
chlorides were chosen to perform the electrochemical measurements. Two pH values, 2.5
and 1.5, were included. The testing temperature was 60o C since this is the temperature
when most exhaust condensation is expected to happen.
Materials in the EGR system are exposed to several changing environmental factors,
e.g. material stress, temperature; dry to wet phase and vice versa. During electrochemical
experiments these environmental factors need to be held constant, since this makes compar-
ison possible. Furthermore, composition of exhaust gas condensate is complex, it contains
several impurities. Active carbon was initially thought to be included in the synthetic ex-
haust gas condensate composition, to simulate the effect of soot particles which is present in
actual condensate. During initial tries it was observed that the active carbon distribution
in the corrosion cell becomes very inhomogeneous, due to its poor solubility in the solution.
This inhomogeneous distribution increased experimental uncertainty, decision was therefore
made not to include active carbon in the synthetic condensates.
How differences between experimental condition and operating condition factors will
affect the risk for pitting corrosion is difficult to predict. The term ”predict” is therefore
commonly used for anodic polarisation, since results from anodic polarisation have in some
cases been observed to differ from actual service performance.
7.1 Pitting Corrosion in the EGR System

The general corrosion resistance of stainless steels are good in exhaust gas condensate envi-
ronments with low pH value. This is why the main EGR components such as EGR-coolers
are made of stainless steels. However, the pitting corrosion resistance at elevated temper-
ature, with increased chloride content and under low pH value environment will be greatly
43
44 Chapter 7. Discussion and Summary
decreased. Pitting corrosion is localised and difficult to prevent by surface treatment tech-
nique. Pitting can result in leakage for EGR components. Pits can also initiate stress
corrosion cracking.
From the results obtained in this study, it can be clearly seen that the pitting corrosion
resistance of all tested austenitic stainless steels, duplex stainless steels and ferritic stainless
steels is good in condensates with low chloride ions (condensate 1, 2a and 3a), even though
the pH values were low and sulphuric acid contents high.
Condensate 2a and 3a is a solution containing 200 ppm Cl− at pH 1.5. This is a solution
which should be regarded as more corrosive than actual exhaust gas condensate, especially
if low sulfur diesel fuels are used.
Results from this study indicates that high chloride concentrations are necessary for
pitting corrosion to occur. Accumulation of chlorides in localised positions in the EGR
system is therefore thought to be the root cause if pitting corrosion happens. The question
is where the chlorides comes from? The chloride content in the air in offshore or near sea
environment may be somewhat high, but still very limited and studies regarding chlorides
in the EGR system have not been found.
7.2 Material Choice for EGR Components

Austenitic stainless steel 1.4404 is a commonly used material for EGR components. Even
though with high nickel and some molybdenum contents, the pitting corrosion resistance of
steel 1.4404 was greatly decreased at 60o C, with increased chloride content and under low
pH value environment. Sensitivity to chloride induced pitting corrosion is an Achilles heel
for austenitic stainless steels.
1.4404 is alloyed with about 2% of molybdenum and 10% nickel. This gives the steel
good corrosion resistance against most acids, but the molybdenum and nickel addition also
considerably increase alloying cost, see the AAF factor for grades 1.4301 and 1.4404 in
table 5.1. Engineers diagrams and corrosion tables from steel manufactures show that du-
plex steels are attractive alternatives for austenitic steels and results obtained in this study
clearly supports this.
Duplex steel grades are known to have higher pitting resistance against chlorides and
SCC. Duplex steel grades also have higher mechanical strength compared to austenitic steel
grades. This may enable reduction of component thickness, which will save cost and weight.
The mixed microstructure of duplex steel grades also results in limitations, most pronounced
are lower formability, weldability and limited operating temperatures (-40 to 250-325o C).
Price for stainless steel is compromised of two factors: AAF and Base Price. All selected
duplex steels grades in this study have lower AAF compared to 1.4404, see 5.1. However,
the base price for duplex steel grades are higher compared to austenitic grades, because of
increased manufacturing complexity and lower manufacturing yield.
The ferritic steel 1.4521 seems to have good resistance against initiation of pitting corro-
sion, but once pits have formed repassivation ability compared to the austenitic and duplex
7.2. Material Choice for EGR Components 45
steel grades shows to be inferior. The reason is that ferritic steels are not alloyed with nickel,
which is known to increase repassivity for stainless steel grades.
Repassivation ability have a major role on the applicability of stainless steel grades, since
initiation of pitting corrosion is a matter of probability and is expected to occur sooner or
later.
Aluminium grades AW-3003 and AC-43000KF are subjected to both general and pitting
corrosion in condensate 1 (pH 2.5) and high current densities were observed during anodic
polarisation. The use of aluminium as engineering material in EGR environment can there-
fore not be recommended. Anodising aluminium may increase the corrosion resistance, but
the anodised oxide film was observed to dissolve in low pH environment, and after 12 hours
of exposure at 60o C the protective effect of the anodised film was negligible. It is well known
that aluminium is not passive in solutions with low pH so these results are not surprising.
Nitriding treatments of the martensitic steel 1.4112 show no effect on corrosion resis-
tance in condensate environments. However, since the steel grade was subjected to general
corrosion anodic polarisation is not a suitable method for studying this issue. Further in-
vestigation is therefore recommended.
Field studies are needed before implementing a possible material change. Results ob-
tained in this study are what can be expected; lower alloyed grades austenitic and duplex
grades showed to be inferior to the higher alloyed grades of the same type. However, the low
risk for pitting corrosion in condensate 1, 2a, 3a are somewhat surprising, since this indi-
cates that the exhaust gas condensate should not cause pitting corrosion in the EGR system,
instead results indicates that elevated levels of corrosive elements need to be reached, before
pitting corrosion should become prominent.
46 Chapter 7. Discussion and Summary
Chapter 8
Conclusions
• Eight different stainless steels and two aluminium alloys were investigated in this study
regarding the pitting corrosion resistance in exhaust gas condensate environment. Fur-
thermore, nitriding surface treatments on one martensitic stainless steel and anodising
surface treatments on one aluminium alloy were also included in this study.
• Five condensates with different concentrations of sulphuric acid, nitric acid and chlorides
were chosen to perform the electrochemical measurements. Two pH values, 2.5 and
1.5, were included. The testing temperature was 60o C, since it is the temperature
which can still be expected to produce substantial amount of exhaust gas condensate
in the EGR system.
• The anodic polarisation curve is a useful method to evaluate the pitting corrosion re-
sistance. The results obtained in this study indicates that condensate formed during
normal EGR operation should not cause pitting corrosion, in an environment with low
chloride content and pH value 1.5 and 2.5. Accumulation of chlorides to critical levels
is therefore predicted to be required to initiate pitting corrosion in the EGR system.
• Decreasing pH value from 2.5 to 1.5 and increasing chloride content from 32 ppm to 200
ppm will not increase the risk for pitting corrosion in condensate environments at
60o C. This was observed at two austenitic stainless steels (1.4404 and 1.4301), three
duplex stainless steels (duplex 2205, 2304 and LDX2404) and one ferritic stainless
steel (1.4521).
• Chloride contents of 0.33 wt% in the tested condensates significantly increase the risk for
pitting corrosion, especially for austenitic steels. Duplex stainless steels show better
pitting resistance in addition to the good corrosion resistance in low pH value and
high chloride content environments.
• Aluminium alloys are subjected to general and pitting corrosion in condensate environ-
ment at pH 2.5. The anodised film on the surface was not stable in condensate
environments with low pH value. After twelve hours of exposure to condensate 1 at
60o C, the protective effect of the film became negligible.
47
48 Chapter 8. Conclusions
• Martensitic stainless steel 1.4112 was subjected to general and pitting corrosion in con-
densate environment at low pH value. There is no difference in corrosion resistance
between the nitride coated 1.4112 steel and the steel without coatings. No differences
can be seen between the plasma and gas nitrided samples. Further investigation in less
corrosive environment is needed since the anodic polarisation curves are not suitable
to study the general corrosion behavior.
• The pitting corrosion resistance in condensates with high chloride concentrations at 60o C
follows the sequence 1.4301<1.4521<1.4404<duplex 2304<duplex LDX2404<duplex
2205. Clearly, duplex stainless steels have better pitting corrosion resistance in low
pH environment when chloride concentration is increased. Considering the operating
conditions of the EGR components, the element prices, it is probably more beneficial
to consider the duplex stainless steels for use in the EGR system where temperatures
allows it (<250-325o C).
Chapter 9
Recommendations
The austenitic stainless steels 1.4404 and 1.4301; duplex stainless steels 2205, 2304 and LDX;
and ferritic stainless steels 1.4521 show similar pitting corrosion resistance in condensates
at low pH vales. Long term exposure testing in condensates with low chlorides are therefore
recommended in order to investigate if it is possible to choose the cheaper steels in EGR
condensates.
Duplex stainless steels have better pitting corrosion resistance in low pH environment
when chloride concentration is increased. It is therefore recommended to do more investiga-
tions on how manufacturing cost is affected if components are changed to duplex stainless
steel. If the cost savings are notable, or if increased robustness is significant, continuous
field studies are needed before changing of material.
It is recommended to further investigate the effect of nitriding surface treatment with

long term exposure testing.
Accumulation of corrosive species in soot deposits are important to make any decision of
material choice. Characterisation of soot deposits is therefore also recommended. It should
be investigated if cleaner exhaust gases can be introduced into the EGR system, since this
will reduce fouling and engine wear. LP-EGR systems are advantageous in this regard, since
exhaust gases for such systems passes through the after treatment system.
Very limited studies were found regarding corrosiveness of exhaust gas condensates de-
rived from use of biodiesel fuels. Future work is therefore recommended and investigation
should include collection and analysis of biodiesel condensate.
49
50 Chapter 9. Recommendations
Bibliography
[1] Euro 6, Emission for heavy duty vehicles: European commission. [Online] [Cited: 07
03 2012.],
http://ec.europa.eu/enterprise/sectors/automotive/environment/eurovi/
index_en.htm
[2] Magdi K. Khair. Hannu Jääskeläinen, Exhaust Gas Recirculation. [Online] [Cited: 08
03 2012.]
http://www.dieselnet.com/tech/engine_egr.php
[3] Guillaume Bourgoin, Eva Tomas Jose Lujan and Benjamin Pla, Acidic Condensation
in HP EGR Systems Cooled at Low Temperature Using Diesel and Biodiesel Fuels. SAE
TECHNICAL PAPPER SERIES, Michagan Technological University, 1999
[4] Admir M. Kreso, John H. Johnson, Linda D. Gratz, Susan T. Bagley and David G.
Leddy, A Study of the Vapor- and Particle-Phase Sulfur Species in the Heavy-Duty
Diesel Engine EGR Cooler. SAE TECHNICAL PAPPER SERIES, Michagan Techno-
logical University, 1998
[5] M. Mosburger, J. Fuschetto, D. Assanis & Z. Filipi, Impact of High Sulfur Military
JP-8 Fuel on Heavy Duty Diesel Engine EGR Cooler Condensate. SAE International,
USA, 2008
[6] Thomas L. Mckinley, Modeling sulphuric Acid Condensation in Diesel Engine EGR
Coolers. SAE TECHNICAL PAPPER SERIES, Warrendale, 1997
[7] Huijbregts Corrosion Consultancy, KEMA BV, LATEST ADVANCES IN THE

UNDERSTANDING OF ACID DEWPOINT CORROSION: CORROSION AND
STRESS CORROSION CRACKING IN COMBUSTION GAS CONDENSATES. Anti-
Corrosion Methods and Materials, Vol. 51, 3, 2004, pg 173-188.
[8] M. D. Kass, J. F. Thomas, D. Wilson, S. A. Lewis, Sr. Assessment of Corrosivity Associ-

ated with Exhaust Gas Recirculation in a Heavy-Duty Diesel Engine. SAE TECHNICAL
PAPPER SERIES, Oak ridge National Laboratory, 2005
[9] S. Moroz, G. Bourgoin, J. M. Luján, B. and B. Pla, Acidic condensation in Low Pressure
EGR systems using Diesel and Biodiesel Fuels. SAE int. J. Fuels Lubr. 2(2):305-312,
2009
[10] International Fuel Quality Center, Maximum On-Road Diesel Sulfur Limits. [Online]
[Cited: 27 03 2012.] http://www.ifqc.org/NM_Top5.aspx
51
52 BIBLIOGRAPHY
[11] James W. Girard, Lida D. Grataz, John H. Johnson, Susan T. Bagley and David G.
Leddy, A Study of the Character and Deposistion Rates of Sulfur Species in the EGR
Cooling System of a Heavy-Duty Diesel Engine. SAE TECHNICAL PAPPER SERIES,
Michagan Technological University, 1999
[12] L L Shreir, R A Jarman & G TBurstein, Corrosion. Butterworth-Heinemann, Oxford,
Volym 1, 3nd Edition, 1994.
[13] Leffler Béla, Stainless – Stainless steels and their properties. [Online] [Cited: 24 02
2012.]
http://www.outokumpu.com/files/Group/HR/Documents/STAINLESS20.pdf
[14] ASM Handbook, Corrosion. ASM International, USA, Volume 13, 1987.
[15] Einar Mattson & Kucera Vladimir, Elektrokemi och Korrosionslara. SWERA Kimab,
Stockholm 2009.
[16] ASM Handbook, Corrosion Fundamentals, Testing, and Protection. ASM International,
Ohio, Volume 13A, 2003.
[17] Maria Ohman, Literature Review of Stainless steels for Automotive and Bus Applica-
tions. Swerea-KIMAB, Stockholm, 2012.
[18] L L Shreir, R A Jarman & G TBurstein, Corrosion. Butterworth-Heinemann, Oxford,
Volym 2, 3nd Edition, 1994.
[19] Outukumpu produktblad Materials for winning Ideas – Outokumpu Duplex Stainless
Steels. [Online] [Cited: 28 02 2012.]
http://www.outokumpu.com/applications/upload/pubs_10583455.pdf
[20] Outokumpu, Corrosions Hanbook. Outokumpu Oyj, 10nd Edition, 2009.
[21] University of South Carolina BASICS OF CORROSION MEASUREMENTS. [Online]
[Cited: 11 06 2012.]
http://www.che.sc.edu/faculty/popov/drbnp/ECHE789b/Corrosion\
%20Measurements.pdf
[22] C. Hoffman & P. Gümpel, Pitting corrosions in the wet section of the automotive ex-
haust system. Journal of Achievements in Materials and Manufacturing Engineering,
Volume 34, Issue 2 1987. 115-121
[23] J. Hirasawa, T. Ujiro, S. Satoh & O. Furukimi, Development of High Corrosion Resis-
tant Stainless Steels for Automotive Mufflers Based on Condensate Corrosion Test and
Fielt Investigation. SAE TECHNICAL PAPER SERIES, USA, 2001.
[24] Outukumpu produktblad Duplex Stainless Steels. [Online] [Cited: 04 03 2012.]
http://www.outokumpu.com/SiteCollectionDocuments/Duplex_Stainless_
Grade_Datasheet.pdf
[25] ISO 7599:2010, Anodizing of aluminium and its alloys - General specifications for anodic
oxidation coatings on aluminium. International Organization for Standardizationl, 2010
[26] Outukumpu Pricing/AAF Flat Products AAF Jun 2012. [Online] [Cited: 24 05 2012.]
http://www.outokumpu.com/SiteCollectionDocuments/
Flat-Products-AAF-June-2012.pdf
Appendix
Figure 1: Photo of 4112 grade test samples: 1) Before polarisation. 2) After polarisation.
3) Gas nitrided after polarisation. 4) Plasma nitrided after polarisation.
Figure 2: Photo of aluminium grade test samples: 1) AC-43000KF before polarisation 2)

AC-43000KF after polarisation 3) AC-43000KF anodised before polarisation 4) AC-43000KF
anodised after polarisation 5) AW-3003 before polarisation 6) AW-3003 after polarisation
53
54 APPENDIX
Figure 3: Photo of test samples after polarisation in condensate 2b. Pitts visible by visual
inspection have been marked.
Figure 4: Photo of test samples after polarisation in condensate 3b. Pitts visible by visual
inspection have been marked.
APPENDIX 55
Figure 5: Microscopic examinations of plasma nitride film before anodic polarisation.

56 APPENDIX
Figure 6: Microscopic examination of gas nitride film before anodic polarisation.

APPENDIX 57
Figure 7: Microscopic examinations of plasma nitride film after anodic polarisation in con-
densate 1. Marks indicate exposed area.
58 APPENDIX
Figure 8: Microscopic examinations of plasma nitride film after anodic polarisation in con-
densate 1.
APPENDIX 59
Figure 9: Microscopic examination of gas nitride film after anodic polarisation in condensate
1.
60 APPENDIX
Figure 10: Microscopic examination of gas nitride film after anodic polarisation in conden-
sate 1.
APPENDIX 61
Figure 11: Test coupon documentation for duplex steel 2205.

62 APPENDIX
Figure 12: Test coupon documentation for duplex steel LDX2404.

APPENDIX 63
Figure 13: Test coupon documentation for duplex steel 2304.

64 APPENDIX
Figure 14: Test coupon documentation for steel grade 1.4521.

APPENDIX 65
Figure 15: Test coupon documentation for steel grade 1.4509.

Investigation of Materials For Use in Exhaust Gas Condensate Environment With Focus On EGR Systems PDF

Uploaded by

Copyright:

Available Formats

Investigation of Materials For Use in Exhaust Gas Condensate Environment With Focus On EGR Systems PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Investigation of Materials For Use in Exhaust Gas Condensate Environment With Focus On EGR Systems PDF

Uploaded by

Copyright:

Available Formats

Investigation of Materials for

Use in Exhaust Gas Condensate

The pitting corrosion resistance in condensates with high chloride concentrations at

environment when chloride concentration is increased. Considering the operating conditions

Kartläggning av material för användning i

Gropfrätningsmotståndet för de testade rostfria stålen i kondensat med höga kloridhalter

2 Diesel Exhaust Gas Condensation 5

3 Stainless Steels and Aluminium 9

4 Corrosion & Electrochemistry 15

7 Discussion and Summary 43

1.1 N Ox and PM legislation for Euro 3, Euro 4, Euro 5 and Euro 6. . . . . . . . 1

2.1 Formation mechanisms of exhaust gas condensate summarised from literature

3.1 Pourbaix diagram for a Al/H2 O system at 25o C. . . . . . . . . . . . . . . . . 13

4.1 Schematic of the autocatalytic process of pitting corrosion. . . . . . . . . . . 16

5.1 Experimental Setup: 1) Solartron 1287 potentiostat 2) Ag/AgCl reference

1 Photo of 4112 grade test samples: 1) Before polarisation. 2) After polarisa-

3.1 Composition ranges for different stainless steels [13]. . . . . . . . . . . . . . . 9

4.1 Test materials and steel composition (wt%) [22]. . . . . . . . . . . . . . . . . 23

5.1 Stainless steels grades selected for electrochemical evaluation. . . . . . . . . . 27

– Why is then HP-EGR more common than LP-EGR?

2) Charge-air-coolers (CAC) have very narrow cooler-passages, which may be problem-

3) Most commercial charge-air-coolers are made of aluminium, because of the materials

Diesel Exhaust Gas

2.1 Condensation of Sulphuric Acid

S(g) + O2 (g) → SO2 (g) (2.1.1)

SO2 (g) +1 /2 O2 (g) → SO3 (g) (2.1.2)

SO3 (g) + H2 O(l/g) → H2 SO4 (l/g) (2.1.3)

In a study conducted by M. Mosburger et al. [5] no condensation of sulphuric acid could

In a study by A. M. Kreso et al. [4] small condensation of H2 SO4 could be measured

2.2 Condensation of Nitric Acid

N2 (g) + 02 (g) → 2 N O(g) (2.2.1)

2.3 Condensation of Water

If condensed water is present, formation of acidic elements will significantly increase,

2.4 Effect of Fuel-sulfur Quality

2.5 Exhaust Gas Condensate from Biodiesel

All biodiesels showed a significantly lower opacity compared to diesel, between 64 %

Stainless Steels and Aluminium

3.1 Stainless Steels

Table 3.1: Composition ranges for different stainless steels [13].

Crystal structure Composition (wt%)

3.1.1 Austenitic Stainless Steels

Table 3.2: Austenitic stainless steels and composition. Source: Outokumpu.

Steel: EN (ASTM/UNS) C N Cr Ni Mo Other

3.1.2 Ferritic Stainless Steels

Table 3.3: Ferritic stainless steels and composition. Source: Outokumpu.

Steel: EN (ASTM/UNS) C N Cr Ni Mo Other

3.1.3 Duplex Stainless Steels

Table 3.4: Duplex stainless steels and composition. Source: Outokumpu.

Steel: EN (ASTM/UNS) C N Cr Ni Mo Other

3.1.4 Martensitic Stainless Steels

Carbon contributes to the formation of chromium carbides, which cause inhomoge-

Steel: EN (ASTM/UNS) C N Cr Ni Mo Other

Figure 3.1: Pourbaix diagram for a Al/H2 O system at 25o C.

Table 3.6: Classification and description for aluminium alloys.

Series Classification Description [12, 16, 18]

Corrosion & Electrochemistry

4.1 Corrosion of Stainless Steels and Aluminium Alloys

4.1.1 General Corrosion

Uniform corrosion rate may be summarised to accelerate with increase in temperature,