ArticlePDF Available

Sulfidic corrosion in refineries - A review

De Gruyter
Corrosion Reviews
Authors:

Abstract and Figures

Sulfidic corrosion of steels in refineries is a prevalent phenomenon that occurs in oil containing sulfur species between 230°C and 425°C. There are several internal and external variables controlling the occurrence of sulfidic corrosion. The most important external factors are temperature, concentration and type of sulfur species, and presence of naphthenic acid. The most important internal or metallurgical factor to control sulfidic corrosion is the amount of chromium in the steel. The refinery industry relies today in a vast industrial experience on the variables affecting sulfidic corrosion but very little is known on the basic mechanism of attack. There is ample room for research and the basic understanding of this phenomenon.
Content may be subject to copyright.
Corros Rev 29 (2011): 123–133 © 2011 by Walter de Gruy ter • Berlin • Boston. DOI 10.1515/CORRREV.2011.021
Sulfi dic corrosion in refi neries a review
Raul B. Rebak
GE Global Research , Schenectady, NY , USA,
e-mail: rebak@ge.com
Abstract
Sulfi dic corrosion of steels in refi neries is a prevalent
phenomenon that occurs in oil containing sulfur species
between 230 ° C and 425 ° C. There are several internal and
external variables controlling the occurrence of sulfi dic cor-
rosion. The most important external factors are temperature,
concentration and type of sulfur species, and presence of
naphthenic acid. The most important internal or metallurgical
factor to control sulfi dic corrosion is the amount of chromium
in the steel. The refi nery industry relies today in a vast indus-
trial experience on the variables affecting sulfi dic corrosion
but very little is known on the basic mechanism of attack.
There is ample room for research and the basic understanding
of this phenomenon.
Keywords: corrosion; refi nery; sulfi dic; temperature.
1. Introduction
Iron (Fe) or steel reacts readily with hydrogen sulfi de (H
2 S)
to form iron sulfi de (FeS or Fe
x S y ). Since H
2 S is a ubiqui-
tous compound in the oil and gas industry, sulfi dation of steel
occurs under upstream (exploration and production) and
downstream (refi nery) conditions. Reaction of Fe with sul-
fur (S) also occurs in power generation during the burning of
fossil fuels. Table 1
shows three areas of sulfi dation corrosion
of steels.
In the temperature range from ambient to approximately
200 ° C, the sulfi dation of carbon steel is a common occurrence
under upstream aqueous conditions (Kane , 2006 ). Several
types of FeS were identifi ed; including amorphous, pyr-
rhotite, mackinawite, troilite, cubic, greigite, marcasite and
pyrite (Vedage , Ramanarayanan, Mumford, & Smith, 1993 ;
Harmandas & Koutsoukos , 1996 ; Sun & Nesic , 2007 ; Smith ,
Brown, & Sun, 2011 ). The establishment of the various stoi-
chiometric and non-stoichiometric forms of Fe
x S y may be
infl uenced by factors such as the partial pressure of H
2 S, the
pH, and the temperature (Smith et al. , 2011 ). For example, at
ambient temperature and at pH 4 mostly mackinawite forms
at all concentrations of H
2 S; however when the pH is 7, the
formation of pyrrhotite is favored for the higher concentra-
tions of H
2 S (Smith et al. , 2011 ).
At temperatures higher than 538 ° C the sulfi dation mode of
attack of steel by sulfur compounds may change from mostly
an external uniform corrosion reaction of a component to an
internal localized attack, when sulfur diffuses inside the bulk
metal (Table 1 ). The higher temperature sulfi dation attack
generally occurs in gases, for example, in the gasifi cation
of coal, in the refi ning industry and in gases that result from
burning fossil fuels (gas, liquid and coal) (Lai , 2007 ). In other
cases steel could be attacked by the combined presence of S
and O
2 , such as in the phenomenon known as hot corrosion
(Type I and Type II) (Rapp , 2002 ). The presence of molten
alkali metal sulfate, sulfur trioxide as well as vanadium pen-
toxide may destroy a protective oxide fi lm on the metal sur-
face accelerating corrosion locally. The hot corrosion process
may also happen in refi neries in the fi re side of heater tubes
(Wen & Mucek , 2011 ).
The higher (above 500 ° C gaseous) and lower (upstream
wet) temperature sulfi dation issues are not part of this review.
Only the sulfi dation corrosion in refi neries, often called sul-
dic corrosion, is reviewed here. Sulfi dic corrosion is the
corrosion process of engineering alloys (mainly steels) in
presence of hydrocarbons containing sulfur species, mainly
H
2 S. The temperature range in which the phenomenon of
sulfi dic corrosion occurs is 232 427 ° C (Table 1 ). Several
subgroups of sulfi dic corrosion in refi neries may be listed,
including presence of H
2 S, presence of H
2 S plus S and/
or S-containing compounds (mercaptans), presence of H
2 S
plus H
2 , and presence of H
2 S plus naphthenic acids, and the
combination of all of above.
Typical sulfi dic corrosion in refi neries can occur in a large
number of components such as fi ttings and pressure ves-
sels but it seems more prominent in piping. Sulfi dic corro-
sion manifests itself as more or less uniform thinning of the
wall of the component, but in horizontal pipes the 12 o clock
position may be preferentially attacked (Niccolls , Gallon, &
Yamamoto, 2008 ). The mechanism of sulfi dic attack is by
direct reaction of the sulfur with the metal.
Several review documents on sulfi dic corrosion have
been published (NACE , 2004 ; API , 2008 ) to capture the
current understanding and consensus on the sulfi dic phe-
nomenon in refi neries. However, very little is found in the
literature regarding fundamental research of sulfi dic cor-
rosion including kinetics and thermodynamics (Farrell &
Roberts , 2010 ).
The main objective of the current review is to analyze the
most recent (newer than 2006) published data on sulfi dic
corrosion and to highlight incomplete and contradictory data.
1.1. Crude oil in a refi nery
A crude oil refi nery contains more than a thousand com-
ponents housing nearly a hundred internal environments
(Jenkins , 1998 ). In such a refi nery crude oil or petroleum is
converted into end products such as gasoline, kerosene, die-
sel oil, etc. Crude oil is a complex mixture of several 100
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
124 R.B. Rebak: Sul dic corrosion in re neries a review
hydrocarbons containing approximately 84 % carbon, 14 %
H and 1 3 % S and < 1 % each of metals, salts and oxygen
(Ruschau & Al -Anezi, 2001 ; Guedes Soares , Garbatov,
Zayed, & Wang, 2008 ). There are more than 150 varieties of
crudes produced worldwide (Bacon & Tordo , 2005 ). Crude
oils may be classifi ed in different ways, for example: (1) as
paraffi nic, naphthenic or aromatic based on the predominant
type of hydrocarbon molecule (Ruschau & Al -Anezi, 2001 ;
Guedes Soares et al. , 2008 ); (2) as sour or sweet depending on
the amount and type of reactive sulfur species, such as H
2 S.
Sour oils contain more H
2 S; (3) as light or heavy depending
on the API gravity number (oils with a high API number and
high proportion of H are considered light) and (4) as acidic or
not acidic depending of the total acid number (TAN). A TAN
value higher than 1 can be considered acidic (Bacon & Tordo ,
2005 ). The acidity in the oil is given by the amount and type
of carboxylic organic acids. A crude with a TAN number
lower than 0.5 will be less corrosive to a refi nery plant (Kane
& Cayard , 2002 ).
Refi neries may use blends of different varieties of crude
and at the same time there is variation in the oil that is being
processed (distilled) in one refi nery from point to point of
the fl ow. Similarly, the contaminants that the oil carries are
different from different blends that come into the refi nery
and for a typical blend, the amount and type of species also
change from sector to sector in the refi nery. During the refi n-
ing process, other substances may be added to the oil, such
as hot water (hydro treating) or emulsifi ers to remove salts
and metals (desalting). In other treatments, the oils may be
sweetened by neutralizing H
2 S using amines or other caustics
(Jayaraman & Saxena , 1995 ).
2. The sulfi dation corrosion process
The actual sulfi dic corrosion mechanism of steel is not known
(Niccolls , 2005 ). It is generally accepted that the corrosion
progresses in the steels via a fi lm on the surface. This fi lm is
formed by metal sulfi des and it is pseudo-passive, i.e., it
is semi-protective and not tenacious (Jayaraman & Saxena ,
1995 ).
The following steps could lead to the formation of an FeS
lm
Fe Fe 2 + + 2 e - (1)
H 2 S HS - + H +
(2)
as iron oxidizes, the hydrogen cation gets reduced and then it
may get dissolved in the metal or it may evolve as molecular
hydrogen gas in the stream
H +
+ e - H o
(3)
H
o
Ho
diss (4)
H o
+ H o
H 2 (5)
If Equation 4 is dominant over Equation 5, the dissolved
hydrogen in the steel may cause hydrogen embrittlement,
especially in welds. FeS forms on the surface by the reactions
in Equations 6 and 7
HS - + Fe 2 + FeS + H +
(6)
or it may form non-stoichiometric sulfi de products such as
yHS - + xFe 2 + Fe x
S y
+ yH +
(7)
The stability of the fi lm will depend among other vari-
ables on the presence of fl ow or turbulence. As turbulence
increases, the pseudo-passive fi lm becomes minimal. The
transport of sulfur through the pseudo-passive metal sulfi de
layer is confi rmed by a parabolic decrease in the fi lm thick-
ness as a function of time suggesting a diffusion controlled
mechanism (Qu , Zheng, Jing, Yao, & Ke, 2006 ).
If the temperature is high enough ( 500 ° C and higher) it
is possible that besides forming a metal sulfi de scale on the
surface, sulfur may diffuse inside the metal (mainly though
grain boundaries) and react with the metal forming internal
sulfi des. The presence of sulfur inside the metal may interfere
with the dissolved and carbide precipitated carbon. The metal
carbides disappear where the sulfi des form and the liber-
ated carbon diffuses deeper into the metal (Hucinska , 2006 ).
This process of carbide decomposition is accelerated if H
2 is
present in the environment (Hucinska , 2006 ).
3. Previous reviews on sulfi dic corrosion
Sulfi dic corrosion has occurred since modern refi neries were
in operation and it is still pervasive today. However, refi n-
eries are familiar with sulfi dation and know how to control
it by alloy selection and corrosion and materials monitoring
and management. Niccolls et al. (2008) state that refi ning is a
Table 1 Sulfi dation corrosion of steels.
Application Temperature range Corrosion characteristics
Upstream oil a nd gas exploration
and production, geothermal wells
Ambient to 230 ° C Uniform type of corrosion in presence of water. Different type of iron
sulfi de scales may form depending on the environmental cha racteristics
including oxidizing vs. reducing environments, pH, bacterial activity, etc.
Downstream oil refi neries 232 – 427 ° C Sulfi dic corrosion could be in the presence of liquid or vapor phase oil
containing sulfur species. The attack is generally uniform.
Fossil fuel power production 538 – 1100 ° C Sulfi dation attack could be localized. It is generally a gaseous phase
corrosion. There are several types of corrosion environments, including
gasifi ers (syngas), oxidizing conditions (SO
2 ), reducing conditions (H
2 S),
hot corrosion, coal ash corrosion, etc. (Lai , 2007 )
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
R.B. Rebak: Sulfi dic corrosion in refi neries a review 125
mature industry and therefore almost all of the types of degra-
dation modes are well known to seasoned materials engineers
and inspectors. Shut downs or failures due to sulfi dic corro-
sion are currently rare. The previous knowledge on sulfi dic
corrosion has been summarized in a series of comprehensive
reports that were prepared by the experts in the fi eld. These
reports include the NACE 34103 Overview and the API
939-C Guidelines.
3.1. NACE 34103 overview of sulfi dic corrosion
in petroleum refi ning
The NACE International Specifi c Technology Group (STG)
34 on Petroleum Refi ning and Gas Processing formed the task
group TG176 called Prediction Tools for Sulfi dic Corrosion.
This task group issued a consensus document in February
2004. This document states that little fundamental research
has been published studying the mechanism, thermodynamics
and kinetics of sulfi dic corrosion in the temperature range of
interest for the refi ning industry. Most of the current under-
standing on the corrosion behavior of materials is based on
plant experience.
Some basic facts from the NACE 34103 document:
Carbon steels form an iron defi cient sulfi de scale (Fe 1. 1-x S)
on the surface that, as it grows, slows down the corrosion
process as a function of time.
Steels that contain chromium form a two-layer scale the 2.
inner layer is a sulfo-spinel (FeCr
2 S 4 ) and the outer layer
is the Fe
1-x S. It is generally accepted that the sulfo-spinel
layer is more protective than the Fe
1-x S.
Several steps may be involved in the corrosion process in 3.
presence of sulfur compounds chromium (Cr) may poi-
son a critical decomposition step of the sulfur compounds
previous to the incorporation of sulfur to the scale and
H
2 may promote the decomposition of these sulfur com-
pounds (therefore H
2 may counteract the benefi cial effect
of Cr).
Carbon steels can be used until 260 ° C. The corrosion 4.
resistance of carbon steel is marginal at 316 ° C.
Lower than 5 % Cr steels are not currently used in refi neries, 5.
5Cr steels are used up to 343 ° C and 9Cr steels are used up
to 400 ° C. Austenitic type 18/8 stainless steels have excel-
lent resistance to sulfi dic corrosion even under long-term
use at high temperature like in furnace tubes and furnace
transfer lines. Sensitization of the stainless steel does not
decrease its resistance to sulfi dic general corrosion. If H
2
is present in the system the austenitic 18/8 stainless steels
may need to be used above 260 ° C.
The mechanism of increased aggressiveness due to the 6.
presence of H
2 has not been established yet. Some pos-
tulated that H
2 does not allow for benefi cial formation of
coke, others argue that H
2 promotes the decomposition of
other sulfur bearing compounds into H
2 S, thus increasing
the aggressiveness of the environment.
The FeS scale that forms on the surface generally contains 7.
cracks, fi ssures and spalls that may provide avenues for
sulfur ingress and promote more corrosion. Some argue
that the presence of coke may seal these paths for sulfur
ingress. That is, when coking starts occurring, the corro-
sion attack by sulfi dation generally decreases.
3.2. API RP 939-C guidelines for avoiding sulfi dic
corrosion
In January 2008 the API Subcommittee on Corrosion and
Materials issued the Version 5.0 of the RP 939-C Guideline
for Avoiding Sulfi dation Corrosion (API , 2008 ). RP stands
for recommended practice to provide a practical guidance to
corrosion engineers and other personnel, such as inspectors,
project and maintenance engineers on how to address sulfi dic
corrosion. The Subcommittee on Corrosion and Materials is
part of the API Committee on Refi nery Equipment (CRE).
The mission of the CRE is to promulgate safe and proven
engineering practices for the design, fabrication, installation,
inspection, and use of materials and equipment in refi neries and
related processing facilities. The Subcommittee on Corrosion
& Materials deals with issues, such as: (1) Fabrication require-
ments; (2) Corrosion mechanisms; (3) Equipment reliability;
(4) Refractory systems; and (5) Reducing capital and mainte-
nance costs. The RP 939-C document is applicable to hydro-
carbon process streams containing sulfur compounds, with
and without the presence of H
2 , which operate at temperatures
above approximately 230 ° C up to about 540 ° C. A threshold
limit for sulfur content is not provided because within the past
decade signifi cant corrosion has occurred in the reboiler/frac-
tionator sections of some hydroprocessing units at sulfur or
H
2 S levels as low as 1 ppm.
The API RP 939-C document captures the state of the art
consensus in the industry of the state of knowledge dealing
with sulfi dation corrosion. That is, any signifi cant contribu-
tions on the understanding of variables that affect sulfi dation
corrosion are captured in this document. The information in
RP 939-C has been mined from published technical papers,
information exchanges at the API and NACE levels and also
from refi nery owners and operators. This does not mean that
some fi ndings previous to 2008 that may also be relevant may
not be captured in this document since they may be proprie-
tary or because consensus has not been reached in the matter.
The following basic fi ndings can be summarized from the
API RP 939-C document:
The sulfi dic corrosion rate increases with the temperature 1.
from 230 ° C until 425 ° C and then it decreases as the tem-
perature increases until 540 ° C. The rate of increase in the
corrosion rate between 230 ° C and 425 ° C is faster as the
temperature is increased.
Chromium is a benefi cial alloying element protecting 2.
against sulfi dation corrosion, i.e., the corrosion rate of steel
with 5 % Cr is lower than the corrosion rate of carbon steel.
The corrosion rate increases with the amount of sulfur in 3.
the stream. The infl uence of sulfur on the corrosion rate is
less important than the effect of the temperature.
The presence of H 4. 2 tends to accelerate the corrosion rate.
The benefi cial effect of Cr in the steel may be less effec-
tive when H
2 is present in the oil mixture.
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
126 R.B. Rebak: Sul dic corrosion in re neries a review
4. Variables affecting sulfi dic corrosion
As stated in the NACE 34103 document Overview of
Sulfi dic Corrosion in Petroleum Refi ning very little basic
research has been dedicated to the understanding of the ther-
modynamics and kinetics of sulfi dic corrosion. The little
interest in research may have resulted from the 40 years plant
experience that led to the preparation of guidelines to deal
with materials selection and plant performance in the fi eld
(Chambers & Kane , 2008 ). Materials engineers may have felt
comfortable predicting the lifetime of several types of steels
according to the temperature and sulfur content of the process
and replacing the part at scheduled shut downs rather than
trying to fully understand the sulfi dic corrosion mechanisms.
Table 2
shows a few of the internal and external variables that
infl uence sulfi dic corrosion in refi nery applications. Internal
variables pertain to the metallic component (pipe), such as
composition and phase distribution, heat treatment, etc.
External variables include temperature, type of crude, sulfur
content in the crude, fl ow velocity, etc.
4.1. Effect of the temperature
It is accepted that the most important variable affecting cor-
rosion of a component in a refi nery is the temperature. The
temperatures of interest for sulfi dic corrosion are in the range
230 – 425 ° C. The temperature generally defi nes what material
is used in that application. For example, depending on the
particular application in the refi nery and crude blend, carbon
steel may be used up to 260 288 ° C. Similarly, the upper limit
for the 5Cr steel could be 329 343 ° C and for temperatures
up to 400 ° C 9Cr steel may be used. In general, for all materi-
als, the corrosion rate increases as the temperature increases;
however, the relationship between corrosion rate and tempera-
ture may not be the same in the entire range of temperature.
As the temperature increases the activation energy value
seems to increase (API , 2008 ). For temperatures higher than
427 ° C, sulfi dic corrosion may actually decrease due to addi-
tional protection from coking (Gutzeit , 1986 ).
Carbon steels containing < 0.1 % silicon (Si) have a faster 5.
rate of sulfi dic corrosion than carbon steels with higher
contents of Si.
Other variables that may affect the corrosion rate include 6.
ow velocity, coking, and presence of steam (stripping).
3.3. McConomy and Couper-Gorman curves
McConomy and Couper-Gorman curves are prediction
tools based on survey data from refi nery experience that
are used to estimate the corrosion rate of different steels
in crude oil as a function of the temperature (API , 2008 ).
The McConomy curves developed by API in 1963 are
used in the absence of dissolved hydrogen and the Couper-
Gorman curves developed in 1971 are used when H
2
and H
2 S are present (Qu et al. , 2006 ; API , 2008 ; Farrell
& Roberts , 2010 ). Neither the McConomy nor the Couper-
Gorman curves take into consideration the effect of fl ow
velocity. The McConomy curves were later modifi ed in
1986 since the original ones were too conservative, i.e.,
the original curves predicted higher sulfi dic corrosion rates
than the actually observed in refi neries (API , 2008 ; Farrell
& Roberts , 2010 ). The modifi ed McConomy curves were
developed from empirical data at a total sulfur concentra-
tion of 0.6 wt % in the crude oil. A correction factor may
be applied for higher sulfur levels (API , 2008 ; Farrell &
Roberts , 2010 ). The McConomy curves (Figure 1 ) show that
the corrosion rate of the different steels increases monoton-
ically as the temperature increases. The highest corrosion
rates are for carbon steel and the lowest are for type 18/8
stainless steels. The curves do not cross each other in the
entire range of temperature reported from 260 ° C to 400 ° C.
The predictions from both types of curves (McConomy and
Couper-Gorman) are still conservative compared to indus-
try experience (Riley , 2005 ). Materials behave differently
in H
2 S environment and in H
2 S + H 2 environments, since the
benefi cial effect of chromium for the H
2 S environment is
less effective in the H
2 S + H 2 environment (Setterlund , 1991 ;
Niccolls , 2005 ).
10.000
1.000
Carbon steel
1-3Cr
4-6Cr
7Cr
9Cr
12Cr
18Cr/8
0.100
0.010
0.001
250 270 290 310 330 350 370
0.4
10
1
0.1
0.01
Sultur content WT%
0.8 1.2 1.6 2
Corrosion rate multiplier
390 410
Temperature (°C)
Corrosion rate (mm/year)
Figure 1 Modifi ed McConomy curves to predict sulfi dic corrosion.
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
R.B. Rebak: Sulfi dic corrosion in refi neries a review 127
Activation energy values for sulfi dic corrosion are
scarce. No data were found for pure sulfi dic corrosion in
contact with crude oil in the temperature range 230 425 ° C.
The activation energy was measured for the sulfi dation
of Fe exposed to 10 mm Hg sulfur vapor (gas phase) for
up to 10 h in the temperature range from 250 ° C to 500 ° C
(Foroulis , 1978 ). It was reported that the formation of the
sulfi de scale on the Fe surface followed an Arrhenius rela-
tionship between the parabolic rate constant and the inverse
of the absolute temperature. However, two ranges were
found; the activation energy was 59 kJ/mol in the higher
temperature range from 370 ° C to 500 ° C and 113 kJ/mol in
the lower temperature range (250 370 ° C) (Foroulis , 1978 ).
For the lower temperature range the scale consisted in an
inner layer of pyrrhotite and a thin external layer of pyrite;
however, for the higher temperature range the scale was
only pyrrhotite (Foroulis , 1978 ). The activation energy val-
ues reported by Foroulis (1978) may not be fully relevant
to oil refi neries since it was in presence of sulfur vapor and
for pure Fe instead of steel.
For naphthenic acid corrosion Gutzeit reported that for
both carbon steel and 410SS the Arrhenius activation energy
was approximately 69 kJ/mol at temperatures higher than
288 ° C (Gutzeit , 1977 ). More recently, an Arrhenius relation-
ship was also reported for the sulfi dic corrosion rate of carbon
steel between 210 ° C and 300 ° C in oil containing two types
of naphthenic acid. Values of activation energy from 23.8 to
31.8 kJ/mol were reported (Slavcheva , Shone, & Turnbull,
1998 ).
Two steels (carbon steel and 5Cr0.5Mo) were tested for
their response to sulfi dic corrosion in the temperature range
between 230 ° C and 270 ° C (Qu et al. , 2006 ). The carbon steel
was mostly a ferrite phase and the 5Cr0.5Mo steel was mostly
a pearlite phase. The tests were performed in a static auto-
clave for up to 65 h of testing time. As expected, for both
steels, the corrosion rate increased linearly with the tempera-
ture; however, the corrosion rate was higher for the carbon
steel than for the 5Cr0.5Mo steel. For the carbon steel the
corrosion rate increased from approximately 0.4 mm/year
at 230 ° C to approximately 2 mm/year at 270 ° C (Qu et al. ,
2006 ). However Qu et al. did not report the value of activa-
tion energy for these experiments (Qu et al. , 2006 ).
It seems a little surprising that activation energy for the
“ pure ” or single sulfi dic corrosion mechanism were not found
in the literature for carbon or alloy steels. A few activation
energy values are published for the combined mechanism of
sulfi dic corrosion and naphthenic acid corrosion. In general
the activation energy values available for steels in presence
of sulfur species and naphthenic acid seem to suggest that the
activation energy above 300 ° C could be approximately two-
fold higher than the activation energy below 300 ° C.
4.2. Sulfur species
The second materials selection criteria are the presence in the
stream of H
2 S and other sulfur containing species. As the con-
centration of sulfur species increase the corrosion rate of refi n-
ery plants components increase. For example, it was reported
that the corrosion rate of carbon steel exposed to a 329 ° C oil
stream in a hydro-processing distillation column increased
practically 10-fold from 5 mpy to 45 mpy when the H
2 S con-
centration increased approximately three-fold from 800 ppm
to 2500 ppm (Niccolls , 2005 ). Sulfur is present in crude oil as
H
2 S, as thiols, mercaptans, sulfi des, benzothiophenes, polysul-
des, or as elemental sulfur (de Jong et al. , 2007 ; API , 2008 ;
Guedes Soares et al. , 2008 ). Sulfur becomes aggressive to steel
if its proportion in the crude oil is 0.2 % or higher (Ruschau &
Al -Anezi, 2001 ). At temperatures higher than 230 ° C, Fe reacts
with S to form FeS. In general, in the refi neries crudes are clas-
sifi ed as sweet ( < 1 % S) and as sour (more than 0.5 % S) (Bota ,
Qu, Nesic, & Wolf, 2010 ). Sour crudes are blamed for sulfi dic
corrosion in refi neries. During refi ning some crude oils may
be treated with a caustic wash to remove sulfur in a so-called
sweetening process. Since the reactivity or corrosivity of these
different forms of sulfur varies from molecule to molecule, the
corrosion rate of steel generally is not proportional to the total
sulfur content in the oil stream (Bota et al. , 2010 ). Smaller
molecules of sulfur compounds tend to be more corrosive than
the large ones (Setterlund , 1991 ). It was also reported that
sulfur species that have the ability to decompose into H
2 S at
the exposure temperature would have an impact on the corro-
sion rate of the steel (Kane & Cayard , 2002 ). Sometimes it is
reported that mercaptans are even more aggressive than H
2 S in
affecting sulfi dic corrosion (Niccolls , 2005 ).
A study was carried out to determine the corrosion behav-
ior of 1018 carbon steel in presence of mercaptans in crude
oil in the liquid phase in the temperature range 200 300 ° C
(de Jong et al. , 2007 ). Coupons of carbon steel were exposed
to crude containing four different mercaptans at different con-
centrations ranging from 100 ppm to 3000 ppm (de Jong et
al. , 2007 ). It was reported that at 275 ° C the corrosion rate of
steel increased signifi cantly with the concentration of mer-
captans up to 1000 ppm and more slowly for the higher mer-
captan concentrations. The highest corrosion rate at 275 ° C
corresponded to the higher molecular weight mercaptan but
an explanation was not given for this behavior (de Jong et
al. , 2007 ). It was also reported that the corrosion rate did
not increase monotonically as the temperature increased but
peaked at approximately 280 290 ° C and then decreased for
the higher temperatures (de Jong et al. , 2007 ).
In general the information in the published literature regard-
ing the effect of sulfur species on sulfi dic corrosion is highly
contradictory, some argue that thiols and mercaptans are more
corrosive than H
2 S and some argue just the opposite.
Table 2 Variables infl uencing sulfi dic corrosion.
Internal variables Exter nal variables
Type of steel Temperature
Heat treatment amount and
distribution of pearlite colonies
Type and concentration of
sulfur species in the oil stream
Cr content Naphthenic (carboxylic) acids
Si content Fluid velocity
Mo content Hydrogen in oil stream
Steam
Amines and other chemicals
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
128 R.B. Rebak: Sul dic corrosion in re neries a review
4.3. Effect of naphthenic acid
The third important criteria for materials selection in refi ner-
ies is the total acid number (TAN), which is an indication of
the relative amount of carboxylic acids or naphthenic acid in
the oil. Naphthenic acid corrosion (NAC) is most prevalent
in the temperature range 200 400 ° C and it seems to peak at
approximately 370 ° C (Jayaraman & Saxena , 1995 ). The cor-
rosion caused by naphthenic acid is more important for TAN
higher than 0.5 (Laredo , Lopez, Alvarez, & Cano, 2004 ).
There is also a complex relationship between sulfi dic corro-
sion and naphthenic acid corrosion. For example, it has been
claimed that the naphthenic acid and sulfi dic mechanisms
act synergistically, i.e., the presence of sulfur may acceler-
ate naphthenic acid corrosion, but the opposite has also been
defended, i.e., that the presence of sulfur may inhibit naph-
thenic corrosion (Slavcheva et al., 1998 ; Tebbal , 1999 ; Kane
& Cayard , 2002 ; Kanukuntla , 2008 ; O ’ Kane, Rudd, Cooke,
Dean, & Powell, 2010 ). It is claimed sometimes that there is
a continuum between naphthenic acid corrosion and sulfi dic
corrosion (Kane & Cayard , 2002 ). Other researchers state
that the relationship between the two modes of corrosion may
change depending on the levels of sulfur and naphthenic acid
in the system (Messer , Tarleton, Beaton, & Phillips, 2004 ;
Chambers & Kane , 2008 ). Since FeS is insoluble in oil, this
may protect the steel from attack by naphthenic acid (Piehl ,
1988 ; Turnbull , Slavcheva, & Shone, 1998 ; Kanukuntla,
2008; Bota et al. , 2010 ). However, since iron naphthenate is
soluble in oil, the presence of naphthenic acid may weaken
the protectiveness of FeS on the surface promoting scale de-
bonding and favoring more corrosion. In a corroded refi nery
component, the evidence of naphthenic acid corrosion may be
supported by the absence of a surface scale on the corroded
component (Gutzeit , 1977 ).
The following reactions have been proposed to explain the
interaction between naphthenic acid (RCOOH) and Fe
x S y
Fe + 2RCOOH Fe(RCOO) 2 + H 2 (8)
Fe(RCOO) 2 + H 2 S FeS + 2RCOOH (9)
The corrosion of iron by naphthenic acid is given by Equation
8. Iron naphthenate is soluble in oil, but if enough sulfur is
present in the oil, it may react with iron naphthenate to reform
the FeS layer on the surface of the component (Equation 9).
However, this reaction regenerates naphthenic acid in the sys-
tem. Therefore, it is claimed that crude with high naphthenic
acid and low sulfur may be more corrosive than crude with
a similar naphthenic acid content with a higher sulfur level
(Turnbull et al. , 1998 ; Laredo et al. , 2004 ).
Two type of steels (carbon steel and 5Cr0.5Mo) were
tested for 24 h in presence of two naphthenic acids at a con-
centration of 0.25 mol/L, and in presence of the same naph-
thenic acid plus 0.1 % H
2 S (in argon) at 275 ° C in two types
of oil (heavy vacuum gas oil and mineral oil) (Slavcheva et
al. , 1998 ). Figure 2 shows that the presence of H
2 S reduced
the corrosion rate of the steels promoted by naphthenic acids.
The inhibitive effect also depended on the type of naphthenic
acid and the type of steel. Figure 2 also shows that in the
presence of the naphthenic acid mixture the corrosion rate of
the 5Cr0.5Mo steel was higher than the corrosion rate of the
carbon steel. The latter result is one of the surprising fi ndings
in which a 5Cr steel is found more prone to corrosion than
plain carbon steel. The authors claim that the inhibitive effect
of sulfur also depended of the type of carrier oil (not shown in
Figure 2 ) (Slavcheva et al. , 1998 ).
Two steels (5Cr and 9Cr) were tested for resistance to
impingement corrosion in Tuffl o 1200 oil at 343 ° C using
ow velocities of 16–97 m/s. The tests were performed in
oil containing only naphthenic acid (up to TAN 3.5) and in
oil containing naphthenic acid plus two different levels of
H
2 S (0.2 and 0.45 psia) (Kane & Cayard , 2002 ). For the pure
naphthenic acid environment, the impingement corrosion
was evident for TAN values higher than 1.5. Both the 5Cr
and the 9Cr steel had similar degradation rate. As the TAN
value increased, the velocity to onset impingement corrosion
decreased. When the 0.2 psia level of H
2 S was added it caused
inhibition of impingement corrosion in both steels. When the
H
2 S was increased to the higher level (0.45 psia) the 5Cr steel
started to corrode but the 9Cr still maintained the inhibition
(Kane & Cayard , 2002 ).
The interaction between naphthenic acid and sulfi dic cor-
rosion was also investigated at 270 ° C using 5Cr0.5Mo steel
(Qu et al. , 2006 ). For a solution containing a TAN between 6
and 16, the corrosion rate was inhibited by adding 1 % S (Qu
et al. , 2006 ). However, then the TAN was 32, the addition of
1 % S was detrimental (increased the corrosion rate promoted
by naphthenic acid). For the carbon steel, the addition of 1 % S
always increased the corrosion rate by naphthenic acid. Sulfur
inhibited corrosion caused by naphthenic acid only in the
5Cr0.5Mo steel (Qu et al. , 2006 ). It was more recently claimed
10
Inhibitive effect of H2S on
naphthenic acid corrosion
mineral oil, 24 h, 275°C
Cyclohexane carboxilic acid
Naphthenic acid mixture+H2S
Naphthenic acid mixture
Cyclohexane carboxilic acid+H2S
8
6
4
2
0
Carbon steel 1018 5Cr 0.5Mo steel
Corrosion rate (mm/year)
Figure 2 Inhibiting effect of hydrogen sulfi de on the corro-
sion caused by naphthenic acid. Plot prepared from table 9 data in
Slavcheva et al. (1998) .
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
R.B. Rebak: Sulfi dic corrosion in refi neries a review 129
that no fundamental studies were carried out to investigate the
mechanism of interaction between the sulfi de layer and the
naphthenic acid in solution (El Kamel et al. , 2010b ). Coupons
of four different alloys (carbon steel, 2.25Cr1Mo, 5Cr0.5Mo
and 304L) were pre-sulfi dized in 150 mbar of pure H
2 S at
300 ° C. The FeS that formed on the surface was pyrrhotite.
Then the coupons were exposed to white oil containing naph-
thenic acid to a total TAN = 4 at 260 ° C for different amount of
times and the effect of the naphthenic acid on the pre-existing
H
2 S scale was monitored (El Kamel et al. , 2010b ). It was
reported that the naphthenic acid attacked locally the sulfi de
scale on the coupons causing eventual detachment of the scale
from the surface of the coupons. Figure 3 shows a representa-
tion of the required testing time to produce detachment of the
FeS scale from the coupons (El Kamel et al. , 2010b ). Scale
detachment from the 304L steel (18Cr) did not occur even
after 9 h of exposure to the oil. Figure 3 also shows an expo-
nential fi t of the data for the fi rst three points (carbon steel,
2.25Cr1Mo and 5Cr0.5Mo). An extrapolation of these results
show that detachment may have occurred for the 304L steel
only after more than 100 h exposure (Figure 3 ).
Recent laboratory results show the intricacy of the relation-
ships between steel compositions, temperature, sulfur species,
naphthenic acid, etc. Current laboratory data do not contradict
ndings from the plants but at the same time do not provide
too much insight on the corrosion mechanism or mechanisms.
Interestingly, the interaction between sulfi dic corrosion and
naphthenic acid seems to be one of the most widely tested
phenomena in the laboratory.
4.4. Effect of composition of the steel
The selection of materials in a refi nery is not only based on
the ability to resist corrosion but also on price, availability and
ability to weld. Table 3
shows some regular materials used in
the petroleum refi ning industry. For carbon steels it is known
that steels containing < 0.1 % Si corrode faster than steels with
higher Si content (API , 2008 ). The presence of Si in the steel
may help to form a more adherent and stable sulfi de scale
on the surface (API , 2008 ). The modifi ed McConomy curves
(Figure 1 ) do not differentiate between low and high Si steel.
These curves show that at each temperature the corrosion rate
is practically reduced about half its value for series carbon
steel > 2.25Cr steel > 5Cr steel > 9Cr steel > 12Cr > 18/8 steel. It
has also been reported that the corrosion rate can be reduced
10-fold when ferritic 9Cr steel is used instead of carbon steel
(Hucinska , 2006 ). In general the resistance of the steels to
sulfi dic corrosion increases according to the following order
(Farraro & Stellina , 1996 ; Qu et al. , 2006 ): Carbon steel,
Carbon steel + 0.5Mo, 5Cr + 0.5Mo, 9Cr + 1Mo, 12Cr (410),
17Cr (430), 304SS, 316SS, and 317SS.
Stainless steels are used to resist high temperature sulfi dic
corrosion (Farraro & Stellina , 1996 ). Stainless steels contain-
ing molybdenum are used to combat corrosion mainly by
naphthenic acid (Farraro & Stellina , 1996 ). The effect of Cr
to protect against sulfi dic corrosion may be more important
under fl ow conditions (Qu et al. , 2006 ). The benefi cial effect
of Cr may originate of its ability to poison the decomposition
of sulfur compounds (Farrell & Roberts , 2010 ).
Two steels (carbon steel and 5Cr0.5Mo) were tested for
their response to naphthenic acid corrosion and sulfi dic
corrosion (Qu et al. , 2006 ). The carbon steel was mostly a
ferrite phase and the 5Cr0.5Mo steel was mostly a pearlite
phase. The tests were performed in a static autoclave for up
to 65 h of testing time. Coupons were exposed to the liquid
and vapor phase inside the vessel. Testing temperature was
from 220 ° C to 320 ° C (at 20 ° C intervals). The base fl uid (car-
rier) was transformer oil to which (1) naphthenic acid to a
TAN = 2 to 14.51 and (2) dimethyl disulfi de with [S] = 1 % were
added separately (Qu et al. , 2006 ). It was reported that in
the naphthenic acid environment with TAN = 2 at 270 ° C, the
carbon steel was found more resistant to corrosion than the
5Cr0.5Mo steel. However, in the dimethyl disulfi de solution
([S] = 1 % ) the 5Cr0.5Mo steel was found more resistant to
corrosion. Cross sections of the scale formed on the testing
coupons showed that on the 5Cr0.5Mo steel the scale con-
sisted of two layers while in the carbon steel the scale con-
sisted of only one layer (Qu et al. , 2006 ). Laboratory tests and
eld results may suggest that the presence of Cr in the steel
may not be benefi cial to protect against naphthenic acid cor-
rosion (Qu et al. , 2006 ).
In the case when hydrogen is present, the Couper-Gorman
curves do not predict a large decline of the corrosion rates
between carbon steel and 5Cr0.5Mo steel. Only when the 9Cr
steel was used there is modest decrease in the corrosion rate,
and a distinctive improvement was noticed when the 18 % Cr
austenitic steel was used (Hucinska , 2006 ).
Coupons of three steels (carbon steel, P5 and 304L) were
exposed to fl owing high sulfur crude oil at 300 ° C and 35 bar
pressure for times as long as 98 h (El Kamel et al. , 2010a ).
Mass losses after cleaning the sulfi de layer were transformed
to pyrrhotite layer thickness. For carbon steel and P5 steel
(5Cr0.5Mo) the same thicknesses were found, indicating
Exponential fit, R2=0.97
304L
5Cr
2.25Cr
Carbon steel Sulfided coupons
white oil, TAN=4, 260°C
04
1000
100
10
Sulfide scale detachment time (min)
8121620
Weight % Cr
Figure 3 Detachment time for a sulfi de scale when exposed to
white oil containing naphthenic acid TAN = 4 at 260 ° C (plotted from
data by El Kamel et al. , 2010b ).
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
130 R.B. Rebak: Sul dic corrosion in re neries a review
little benefi cial effect of the presence of 5 % Cr in the steel.
However, there was a strong benefi t with the 18 % Cr since
they could not fi nd weight change for the 304L coupons (El
Kamel et al. , 2010a ).
Current laboratory fi ndings and some plant operation
experience may not support the use of 5Cr steel in refi ner-
ies since in many applications its corrosion behavior cannot
be fully differentiated from the behavior of carbon steels.
Moreover, some current commercial 5Cr steel pipes now
contain < 4.5 % Cr. In plant applications when there are limi-
tations to the performance of carbon steel, some plant engi-
neers may fi nd it appropriate to upgrade the pipe material
directly to 9Cr steel, skipping the recommendation of 5Cr
steel.
4.5. Effects of velocity, hydrogen and steam
Fluid velocity and turbulence are important factors affect-
ing sulfi dic attack (Kane & Cayard , 2002 ). It is accepted
that the lowest corrosion is found when the surface of the
metal is completely wetted with the hydrocarbon under low
ow (Gutzeit , 1986 ). Fluid velocity up to 60 m/s may ham-
per the ability of the steel to form a semi protective sulfi de
lm on the surface and therefore high velocity may acceler-
ate corrosion (API , 2008 ). An adherent sulfi de scale may
control further attack of the steel either by more sulfi da-
tion or by naphthenic acid attack. However, if the sulfi de
scale is removed by the shear stress resulting from fl uid
ow, the attack of the underlying steel may be accelerated
(Kane & Cayard , 2002 ; Qu , Liu, Jiang, Lan, & Shan, 2011 ).
This is especially true in presence of naphthenic acids (Bota
et al. , 2010 ; Qu et al. , 2011 ). It has also been argued that
too little fl ow may also be detrimental since more H
2 S may
be allowed to evolve and concentrate in certain pipe areas
(API , 2008 ).
Materials behave differently in H
2 S environment than in
H
2 S + H 2 environments, since the benefi cial effect of chro-
mium for the H
2 S environment may seem less effective
in the H
2 S + H 2 environments (Setterlund , 1991 ; Niccolls ,
2005 ). It has been argued that the presence of H
2 may be
detrimental for the corrosion resistance of the steels since
it inhibits the formation of semi-protective coke on the sur-
face (NACE , 2004 ). Another explanation for the hydrogen
effect is that it reacts with less corrosive sulfur containing
species and forming more H
2 S and therefore increasing the
corrosiveness of the system (Gutzeit , 1986 ; NACE , 2004 ).
In general, if H
2 is present in the stream above 260 ° C it is
recommended to use 18Cr steel (e.g., type 316SS) and avoid
carbon steel and lower Cr steels altogether (Gutzeit , 1986 ;
NACE , 2004 ).
The presence of vaporization and steam may yield higher
corrosion rates in steels. The most detrimental corrosion may
happen within a spray fl ow with vapor loads higher than
60 % containing liquid droplets that may destroy the sulfi de
scale via impingement (Gutzeit , 1986 ). It was reported that
in sulfi dic environments the corrosion rate is approximately
six-fold higher when the metal surface is exposed to vapor
vs. liquid (McLaughlin , 2005 ). Carbon steel and 5Cr0.5Mo
steels were tested for resistance to sulfi dic corrosion at
230 ° C, 250 ° C and 270 ° C (Qu et al. , 2006 ). In the sulfi dic
solution ([S] = 1 % ), the corrosion rate of both steels increased
monotonically as the temperature increased both for the
liquid and vapor phases. For both steels the corrosion rate
was the same in the liquid and vapor phases (Qu et al. , 2006 ).
However, the corrosion rate was higher for the carbon steel
than for the 5Cr0.5Mo steel (Qu et al. , 2006 ). The effect
of the sulfur concentration on the corrosion rate was also
studied at 270 ° C. When the sulfur concentration increased
from 0.5 to 1.25, the corrosion rate increased up to [S] = 1
and then decreased from 1 to 1.25. The corrosion rate for
both steels was higher in the vapor phase than in the liquid
phase and the corrosion rate in both phases of the carbon
steel was higher than the corrosion rate of the 5Cr0.5Mo.
When mixtures of naphthenic acid with sulfur compounds
were tested, a complex relationship of the variables involved
was reported (Qu et al. , 2006 ). Nevertheless, the corrosion
rate of the 5Cr0.5Mo steel was lower than the corrosion rate
of the carbon steel (Qu et al. , 2006 ).
Table 3 Typical materials in sulfi dic corrosion applications.
Material Designation Typical composition, weight %
Single gures are maximum
A53 Grade B ASTM A53 Fe, 0.3C, 0.4Cr, 1.2Mn, 0.045S, 0.05P, 0.4Cu, 0.4Ni, 0.15Mo,
0.08V, (Cu + Ni + Cr + Mo + V = max 1.0 % )
A106 Grade B ASTM A106
(K03006)
Fe, 0.3C, 0.4Cr, 0.29 1.06Mn, 0.1Si, 0.035S, 0.035P, 0.4Ni,
0.4Cu, 0.08V
5Cr 0.5Mo (pipe) ASTM A335 P5 Fe, 0.15C, 4 6Cr, 0.45 0.65Mo, 0.3 0.6Mn, 0.5Si,
0.025S + 0.025P
9Cr 1Mo (pipe) ASTM A335 P91 Fe, 0.08 – 0.12C, 8 – 9.5Cr, 0.85 – 1.05Mo, 0.3 – 0.6Mn, 0.2 – 0.5Si,
0.18 – 0.25V, 0.01S, 0.02P
12Cr 410SS UNS S41000 Fe, 0.15C, 11.5 13Cr, 1.00Mn, 0.040P, 0.030S, 1.00Si
316LSS (pipe) ASTM A312 (S31603) Fe, 0.03C, 16 18Cr, 10 14Ni, 2 3Mo, 2Mn, 1Si, 0.03S, 0.045P
317SS (pipe) ASTM A312 (S31700) Fe, 0.08C, 18 20Cr, 11 15Ni, 3 4Mo, 2Mn, 1Si, 0.03S, 0.045P
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
R.B. Rebak: Sulfi dic corrosion in refi neries a review 131
4.6. Effect of other chemical species, such as amines
and ammonium disulfi de
Crude oil may contain many other chemicals, such as amines
added to control wax deposition, enhance fl ow characteris-
tics, aid in water separation, etc. (Kapusta , van den Berg,
Daane, & Place, 2003 ). For example, amines such as mono-
ethanolamine, di-ethanolamine, aminoethoxyethanol, methyl
di-ethanolamine and di-isopropanolamine are added at dif-
ferent points in the refi nery to remove H
2 S, mercaptans and
carbon dioxide from the process stream (Shahid & Faisal ,
2009 ; Lagad , Cayard, & Srinivasan, 2010 ). The presence of
amines and other chemicals add another degree of complexity
to the sulfi dic corrosion mechanism. Little or no information
is available in this area.
Some crude oils may also contain species such as cya-
nide ions (CN
- ), which may destroy the protective layers of
Fe
x S y on the surface and promote hydrogen ingress into the
steel what eventually may cause hydrogen induced cracking
(Groysman , Feldman, Kaufman, & Balali, 2011 ).
4.7. Effect of testing time
The sulfi de scale that forms on the test metallic coupons
could be semi-protective depending on the testing condi-
tions. That is, as the time increases the corrosion rate may
decrease. When coupons of carbon steel and 5Cr0.5Mo steel
were tested in the liquid and vapor phase of transformer oil
containing 1 % [S] as dimethyl disulfi de at 270 ° C, the cor-
rosion rate initially increased (up to 8 h) and then decreased
as the time increased up to 65 h (Qu et al. , 2006 ). These
results suggest that diffusion of sulfur through the sulfi de
scale is the rate limiting step in the corrosion rate (Qu et al. ,
2006 ). Since the parabolic law rate constant for the carbon
steel was higher than for the 5Cr0.5Mo steel, it was con-
cluded that under the tested conditions the 5Cr0.5Mo steel
was more resistant to sulfi dic corrosion than the carbon steel
(Qu et al. , 2006 ).
4.8. Plant experience
Corrosion issues in crude refi neries are generally solved
by alloy selection based on experience accumulated by the
industry in many decades. Trade documents cited before
(API RP 939-C, NACE 34103) provide a guide for alloy
selection under the different conditions found in the plants.
Some refi neries may use additional measures, such as inhibi-
tion or neutralization (Tuttle , 2005 ). Continuous monitoring
is essential in many plants to measure the degradation rate
of some components and to determine if these components
need replacement in the next scheduled shut down. Most of
the recent failures due to sulfi dic corrosion were traced to the
use of carbon steel pipes with insuffi cient amount of silicon
(API , 2008 ).
An unusual high temperature sulfi dic corrosion has been
reported in the catalytic refi ning unit in a plant in Indiana
(Wilks , 2000 ). The failure occurred in a hot dip aluminized
steel pipe elbow that was exposed to a turbulent two-phase
ow. Failure occurred after the aluminized layer was cor-
roded or eroded (Wilks , 2000 ). This pipe operated above
316 ° C. Some of the corrosion attack progressed under the
aluminized layer lifting it away into the fl owing stream. The
leaking failure was ductile overload due to the thinning of the
pipe wall. As a consequence of the reported failure, all pip-
ing in the area operating above 260 ° C has been upgraded to
5Cr0.5Mo (Wilks , 2000 ).
A corrosion study was conducted for 20 days at a refi nery
during a sour operation when a blend of high sulfur crudes
were processed (Farrell & Roberts , 2010 ). The total amount
of sulfur in the blend was 1.8 ± 0.2 wt % and the TAN num-
ber was relatively low (0.24). Weight loss coupons of fi ve
different steels were exposed to the oil stream in piping at
two points in the plant, (1) in the heavy atmospheric gas
oil (HAGO) with a fl ow of 1.5 m/s at 354 ° C and (2) in the
light vacuum gas oil (LVGO) with a higher fl ow of 2.4 m/s
but at a lower temperature of 204 ° C. The tested materials
were: (a) carbon steel, (b) 5Cr, (c) 9Cr, (d) 410SS 12Cr and
(e) cast CA6NM 12Cr4Ni0.5Mo (Farrell & Roberts , 2010 ).
For the coupons exposed to the LVGO (204 ° C), the carbon
steel, the 5Cr and the 9Cr steel performed well with corro-
sion rates below the detection limit of 0.1 mpy ( < 0.0025
mm/year). This agrees well with the common knowledge
that sulfi dic corrosion is becoming an important issue at
temperatures above 232 ° C, which is above the temperature
in the LVGO. Surprisingly it was reported that in the LVGO
both the 410SS and the CA6NM coupons showed a measur-
able corrosion rate of 1.8 mpy (0.05 mm/year) for the 410SS
and 0.2 mpy (0.005 mm/year) for the CA6NM steel (Farrell
& Roberts , 2010 ). Figure 4 is a graphic representation of
results reported by Farrell and Roberts in their Table 2 for
the HAGO system at 354 ° C. Figure 4 shows that the corro-
sion rates of carbon steel, 5Cr and 9Cr materials were higher
than 10 mpy ( > 0.254 mm/year). The highest corrosion rate
was for the 5Cr steel at 29.1 mpy (0.74 mm/year) (Farrell
& Roberts , 2010 ). It may seem unanticipated that under the
tested conditions the corrosion rate of the 5Cr steel was
higher than the corrosion rate of the carbon steel. Figure 4
also shows a sharp decline in the corrosion rates between
the 9Cr and the 12Cr steels, since for the latter materials
the corrosion rate was < 1 mpy ( < 0.025 mm/year). Other
results from Farrell and Roberts show a signifi cant effect
of the temperature on the sulfi dic rate of the steels between
204 ° C and 354 ° C, since the corrosion rate increased more
than two orders of magnitudes between these temperatures.
Under the tested conditions, the effect of fl uid fl ow was not
signifi cant since for the higher velocity of 2.4 m/s at 204 ° C
the corrosion rate was approximately two orders of magni-
tude lower than for the lower velocity of 1.5 m/s at 354 ° C
(Farrell & Roberts , 2010 ). These results seem to indicate that
the temperature is a more important factor controlling
corrosion than the fl uid velocity. It was also noted by Farrell
and Roberts that, after the in-situ plant tests, the carbon steel
coupons yielded corrosion rates lower than the values pre-
dicted from the McConomy curves, while the 5Cr and 9Cr
steel yielded higher corrosion rates than the McConomy
curves predicted values (Farrell & Roberts , 2010 ).
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
132 R.B. Rebak: Sul dic corrosion in re neries a review
5. Summary and conclusions
Sulfi dic corrosion is a complex mechanism of steel degrada-
tion in crude refi neries occurring between 232 ° C and 427 ° C.
Internal and external factors affect the sulfi dic corrosion
degradation rate including alloy composition, temperature,
total sulfur content in the oil, and presence of naphthenic
acid. Two review consensus documents have been issued by
NACE International in 2004 and API in 2008 capturing the
state of knowledge in plant experience of how the environ-
mental variables affect the sulfi dic corrosion performance of
the engineering steels. Most of the process of material selec-
tion to replace degraded parts or for new refi neries is based on
almost a century of data from plant experience.
The main aim of the current review was to bring together
the most recent research results from laboratory testing and
evaluate their fi ndings in perspective of the plant experience.
Based on the reviewed literature it seems apparent that the
basic mechanism of sulfi dic corrosion of carbon steel and
alloy steels is not fully understood. Little or no systematic
research has been carried out in laboratory or in plant to
measure, for example, kinetics of sulfi de scale growth, the
mechanical properties and adherence of the scales, the activa-
tion energy for scale formation and its dependence of tem-
perature ranges, alloy composition, sulfur species, etc. Little
or no information exists to determine why the presence of
Cr in the steel is benefi cial for sulfi dic corrosion resistance.
There is ample room for systematic laboratory testing on the
sulfi dic corrosion of engineering alloys in simulated and plant
refi nery environments, mainly on the effect of temperature,
sulfur species, alloy composition, and microstructure.
References
API. RP 939-C Guidelines for avoiding sulfi dation (sulfi dic) corro-
sion railures in oil refi neries. API Subcommittee on Corrosion &
Coupons in plant piping
sour crude ~2% [S], 20 days, 354°C
100
10
1
0.1
Carbon
steel 5Cr 9Cr 410SS CA6NM
0.01
0.1
1
Corrosion rate (mpy)
mm/year
Figure 4 Corrosion behavior of engineering alloys coupons
exposed to a plant stream containing sulfur (plotted from tabular data
by Farrell & Roberts , 2010 ).
Materials, Version 5.0, February 2008. New Orleans, LA: API,
2008.
Bacon R, Tordo S. Crude oil price differentials and differences in
oil qualities: a statistical analysis. ESMAP Technical paper 081,
October 2005. Washington, DC: ESMAP, 2005.
Bota GM, Qu D, Nesic S, Wolf HA. Naphthenic acid corrosion of
mild steel in the presence of sulfi de scales formed in crude oil
fractions at high temperature. Corrosion/2010, Paper 10353.
Houston, TX: NACE International, 2010.
Chambers BD, Kane RD. Refi ning high acid crudes. PTQ Q4 2008;
137 – 149.
De Jong J-P, Dowling N, Sargent M, Etheridge A, Saunders-Tack A,
Fort W. Effect of mercaptans and other organic sulfur species on
high temperature corrosion in crude oil and condensate distilla-
tion units. Corrosion/2007, Paper 07565. Houston, TX: NACE
International, 2007.
El Kamel M, Galtayries A, Vermaut P, Albinet B, Foulonneau G,
Roumeau X, Roncin B, Marcus P. Sulfi dation kinetics of indus-
trial steels in a refi nery crude oil at 300 ° C: reactivity at the nano-
meter scale. Surface Interface Anal 2010a; 42: 605 609.
El Kamel M, Galtayries A, Foulonneau G, Roumeau X, Morel G,
Marcus P. High temperature corrosion of sulfi ded steels in naph-
thenic acid environments. Eurocorr 13 17 September 2010,
Moscow. Paper 9377. 2010b.
Farraro T, Stellina RM Jr. Materials of construction for refi nery appli-
cations. Corrosion/1996, Paper 614, Conference and Exposition.
Houston, TX: NACE International, 1996.
Farrell D, Roberts L. A study of high temperature sulfi dation
under actual process conditions. Corrosion/2010, Paper 10358.
Houston, TX: NACE International, 2010.
Foroulis ZA. Kinetics and mechanism of the reaction of iron with sul-
fur vapor in the temperature range of 250 to 500 ° C. Werkstoffe
Korros 1978; 29: 385 393.
Groysman A, Feldman B, Kaufman A, Balali R. Hydrogen damage
and prevention in the oil refi nery. Corrosion/2011, Paper 11295.
Houston, TX: NACE International, 2011.
Guedes Soares C, Garbatov Y, Zayed A, Wang G. Corrosion wastage
model for ship crude oil tanks. Corros Sci 2008; 50: 3095 3106.
Gutzeit J. Naphthenic acid corrosion in oil refi neries. Mater Perform
1977; 16: 24.
Gutzeit J. Refi nery corrosion overview. In: Process industry corrosion
the theory and practice. Houston, TX: NACE International,
1986. pp. 171 – 189.
Harmandas NG, Koutsoukos PG. The formation of iron(II) sulfi des
in aqueous solutions. J Crystal Growth 1996; 167: 719 724.
Hucinska J. Infl uence of sulphur on high temperature degradation of
steel structures in the refi nery industry. In: Advances in materials
science. Gda n´ sk, Poland: Versita Open, 2006.
Jayaraman A, Saxena RC. Corrosion and its control in petro-
leum refi neries a review. Corros Prevent Control 1995; 42:
123 – 131.
Jenkins W. A refi nery process equipment database. Corrosion/1998,
Paper 401. Houston, TX: NACE International, 1998.
Kane RD. Corrosion in petroleum production operations. In: ASM
metals handbook, Volume 13C Corrosion: environments and
industries. Metals Park, OH: ASM International, 2006. pp.
922 – 966.
Kane RD, Cayard MS. A comprehensive study on naphthenic acid
corrosion. Corrosion/2002, Paper 02555. Houston, TX: NACE
International, 2002.
Kanukuntla V. Formation of sulfi de scales and their role on the naph-
thenic acid corrosion of steels. Master of Science Thesis. Athens,
OH: Ohio University, 2008.
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
R.B. Rebak: Sulfi dic corrosion in refi neries a review 133
Kapusta S, van den Berg F, Daane R, Place MC. The impact of oil
eld chemicals on refi nery corrosion problems. Corrosion/2003,
Paper 03649. Houston, TX: NACE International, 2003.
Lagad VV, Cayard MS, Srinivasan S. Prediction and assessment
of rich amine corrosion under simulated refi nery conditions.
Corrosion/2010, Paper 10183. Houston, TX: NACE International,
2010.
Lai GY. (2007). Sulfi dation. In: High-temperature corrosion and
materials applications. Metals Park, OH: ASM International,
2007. pp. 201 – 234.
Laredo GC, Lopez CR, Alvarez RE, Cano JL. Naphthenic acids,
total acid number and sulfur content profi le characterization in
Isthmus and Maya crude oils. Fuel 2004; 83: 1689 1695.
McLaughlin J. Sulfi dation corrosion due to H
2 S/H 2 , effect of H
2
partial pressure, effect of fl uid fl ow. API Spring Conference, 19
April 2005. New Orleans, LA: API, 2005.
Messer B, Tarleton B, Beaton M, Phillips T. New theory for naphthenic
acid corrosivity of Athabasca oilsands crudes. Corrosion/2004,
Paper 04634, Conference and Exposition. Houston, TX: NACE
International, 2004.
NACE International. Overview of sulfi dic corrosion in petroleum
refi ning. Publication 34103. Task Group 176 of Prediction Tools
for Sulfi dic Corrosion. Houston, TX: NACE International,
2004.
Niccolls EH, Gallon AE, Yamamoto K. Systematic integration of
advanced NDE and corrosion monitoring for improved refi n-
ery reliability. Corrosion/2008, Paper 08280, Conference and
Exposition. Houston, TX: NACE International, 2008.
Niccolls N. Sulfi dation: thoughts on theory and practice. Energy
Technology Company, ChevronTexaco. Augusta, GA: Doe
Hydrogen Pipeline Working Group, 2005.
O Kane JM, Rudd TF, Cooke D, Dean FWH, Powell SW. Detection
and monitoring of naphthenic acid corrosion in a visbreaker
unit using hydrogen fl ux measurements. Corrosion/2010, Paper
10351. Houston, TX: NACE International, 2010.
Piehl RL. Naphthenic acid corrosion in crude distillation units. Mater
Perform 1988; 27: 37 43.
Qu DR, Zheng YG, Jing HM, Yao ZM, Ke W. High temperature
naphthenic acid corrosion and sulphidic corrosion of Q235 and
5Cr0.5Mo steels in synthetic refi ning media. Corros Sci 2006;
48: 1960 – 1985.
Qu D, Liu X, Jiang X, Lan Z, Shan G. Setting critical operational
TAN and sulfur level for crude distillation units. Corrosion/
2011, Paper 11362. Houston, TX: NACE International,
2011.
Rapp RA. Hot corrosion of materials: a fl uxing mechanism ? Corros
Sci 2002; 44: 209 221.
Riley J. Roundtable on Sulfi dation. API Spring Conference, 19 April
2005. New Orleans, LA: API, 2005.
Ruschau GR, Al-Anezi MA. Petroleum refi ning: corrosion control
and prevention. In Corrosion costs and preventive strategies in
the United States, Appendix U. Washington, DC: US Department
of Transportation, 2001.
Setterlund RB. Selecting process piping materials. Hydrocarb
Process 1991; 70: 93 100.
Shahid M, Faisal M. Effect of hydrogen sulfi de gas concentration
on the corrosion behavior of ASTM A-106 Grade A carbon
steel in 14 % diethanol amine solution. Arab J Sci Eng 2009; 34:
179–186.
Slavcheva E, Shone B, Turnbull A. Factors controlling naphthenic
acid corrosion. Corrosion/1998, Paper 579. Houston, TX: NACE
International, 1998.
Smith SN, Brown B, Sun W. Corrosion at higher H
2 S concentra-
tions and moderate temperatures. Corrosion/2011, Paper 11081.
Houston, TX: NACE International, 2011.
Sun W, Nesic S. A mechanistic model of H
2 S corrosion of mild steel.
Corrosion/2007, Paper 07655. Houston, TX: NACE International,
2007.
Tebbal S. Critical review of naphthenic acid corrosion. Paper 380.
NACE Corrosion Conference, 1999. Houston, TX: NACE
International, 1999.
Turnbull A, Slavcheva E, Shone B. Factors controlling naphthenic
acid corrosion. Corrosion 1998; 54: 922 930.
Tuttle RN. Petroleum production and refi ning. In: Corrosion tests
and standards: application and interpretation, 2nd ed., Chapter
76. West Conshohocken, PA: ASTM International, 2005. pp.
812 – 821.
Vedage H, Ramanarayanan TA, Mumford JD, Smith SN.
Electrochemical growth of iron sulfi de fi lms in H
2 S-saturated
chloride media. Corrosion 1993; 49: 114 121.
Wen S, Mucek M. Low-temperature hot corrosion in the refi ning
industry. Corrosion/2011, Paper 11365. Houston, TX: NACE
International, 2011.
Wilks GW. Unusual aspects of corrosion failures in refi nery
hydrotreater units. Corrosion/2000, Paper 00702, Conference
and Exposition.
AUTHOR’S COPY | AUTORENEXEMPLAR
AUTHOR’S COPY | AUTORENEXEMPLAR
... Un report del 2019 dell'American Petroleum Institute [2] riporta almeno 45 incidenti gravi in 30 anni causati da sulfidazione dovuta agli alti contenuti di zolfo nelle cariche alle raffinerie. Le conoscenze scientifiche disponibili sull'argomento sono alla base del metodo pratico adottato per controllare il fenomeno, ma molti aspetti non sono ancora ben compresi e richiederebbero approfondimenti di ricerca [3]. La corrosione da sulfidazione si manifesta a temperature superiori a 220°C e provoca un assottigliamento generale e in qualche caso localizzato. ...
... • Il tenore di zolfo è determinato dalla presenza di vari composti alcuni dei quali sono meno attivi ai fini corrosione, stesso discorso vale per l'acidità [7]; • L'esposizione prolungata determina un forte rallentamento della corrosione, per passivazione del materiale, favorita dalla presenza di ossigeno nel fluido [8]; • Le temperature sono misurate al centro del flusso, mentre i valori sulle pareti possono essere più elevati [9], la presenza di particelle solide in sospensione, bolle o vortici possono avere effetti erosivi, sinergici [10]. La corrosione è anche influenzata dalla presenza di altre sostanze, inclusi metalli pesanti e vapore acqueo [3]. • Un fattore moltiplicativo per velocità di flusso superiori a 30,5 m/s tiene conto della fluidodinamica [11], tuttavia flussi superiori alla soglia possono staccare materiale e quelli troppo lenti o stagnanti possono accelerare la corrosione [2]. ...
Article
Full-text available
In refineries, high-sulfur crude oils reduce procurement costs, but require attention to the risks involved, particularly those related to sulfidifation corrosion, which can lead to breakdowns, leaks and accidents. Depending on temperature and sulphur content, under operating conditions, corrosion rates vary by two orders of magnitude for the same material. In current practice, operators refer to McConomy curves, which associate, for the most commonly used materials, the rate of corrosion with temperature, sulfur content and acidity. This approach introduces large approximations and overlooks important details. To manage and overcome the uncertainties inherent in current practice, a Bayesian network has been developed, where the data of the McConomy curves represent only the starting point, which is updated according to experience. The network, fed by operating data, makes it possible to estimate the useful life consumed and safely manage the life cycle of exchangers subject to sulfifiation. Similarly, it is also possible to control the risk of random line breakage. The model has been validated over a multi-year period, with different operating conditions. The proposed method represents a dynamic alternative to current practice and is of interest to control the risks of breakage and accident in refineries using the most difficult oils.
... Different specific degradation methods can occur in the refinery and power industry [1][2][3][4]. Among the degradation mechanisms, corrosion processes can lead to significant damages, and thus to the risk of safe operation [5][6][7][8]. ...
Article
Full-text available
The crude distillation unit is the most critical elements in the refining process. Moreover, most of the equipment in the distillation unit are made of general carbon steels. Data analysis models, machine learning techniques can predict corrosion degradation rates. We used Pearson’s correlation coefficient and multiple linear regression, to predict the impact of process parameters. Altogether, we have analysed 84 channels of technological parameters, and 22 different types of crude oils. Among the corrosion agents, the chloride content strongly affected the weight loss of coupons, where the highest coefficient was 0.68. The most influential parameter is found to be the pH value. Thus, an estimation method of the pH value is set up to predict the corrosion degradation rate. The regression correlation for estimating the pH value is 0.53 if the corrosion agents are not used, which can be improved to 0.76 if the corrosion agents are also used in the regression analysis.
... Generally, H2S accidents in oil refineries are associated, directly and indirectly, with corrosion of industrial equipment. H2S-associated corrosion can be defined as the total, partial, superficial or structural deterioration of materials, causing the breakdown of industrial equipment, resulting in leaks [25,26]. ...
Chapter
Full-text available
Hydrogen sulfide (H2S) is a colorless gas with a characteristic unpleasant odor (like rotten eggs) that is extremely toxic and corrosive. Due to its toxicity, H2S irritates the eyes and/or affects the nervous and respiratory systems. Concentrations on the order of 700 to 1500 ppm can kill a human in a matter of minutes. The media has presented several examples of H2S leaks that jeopardize human lives, threaten the integrity of industrial assets, and endanger the environment. This work aims to show the importance of knowledge of the origins and physicochemical properties of hydrogen sulfide in the direct and indirect relations with man, the environment and industrial equipment.
... Ferritic steels such as Fe-9Cr-1Mo (P91) are commonly used in the metallic infrastructure of diverse equipment due to their good mechanical properties and acceptable corrosion resistance at high temperatures (Alviz et al., 2018). P91 steel Corrosion rates have been deeply studied in water vapor and H2S-containing environments (Ehlers et al., 2006;Rebak, 2011) whereas under the dual effect of the dual environment of furnaces from oil refineries, where crude oil flows in the inner pipes' face and flue gases in the outer face, it hasn't been assessed. To sum up, the gap that covers this work is the lack of experimental and theoretical studies to determine the role of H2 in the corrosion rate of P91 alloy in the dual environment of a furnace from the oil refining sector. ...
Article
Full-text available
Hydrogen transport and storage represent a challenge due to its high diffusivity capacity, which causes the pitting of pipes. This study focused on addressing this problem by simulating the effect caused by the hydrogen generated during the heating pre-processing step of crude oil before addressing it to the distillation unit. For this purpose, the software COMSOL Multiphysics® was employed to determine how the hydrogen diffusion across ferritic steel containers increases the corrosion rate in a furnace. Then, a dual corrosion process was assessed at 650 °C, simulating the flow of water vapor at the outer part of the pipe whereas hydrogen sulfides-as contamination from crude oil-at the inner part. The ferritic steel selected for this research was the 9Cr-1Mo steel due to its usual implementation in processes developed at high temperatures in industry. Besides, we also studied the corrosion rate of both pipe's faces independently, omitting the effect of the other fluid on the other side of the pipe-singular environments-. The influence of hydrogen sulfide on the corrosion rate on the water vapor side was found negligible. Meanwhile, water vapor boosted the corrosion rate on the hydrogen sulfide side, making it 1.2 times greater compared to its singular estimation. The corrosion rate induced by the water vapor is higher due to the higher number of corrosion products-current density that it delivers when reacting with the steel elements at high temperatures. These approaches aimed to quantify the hydrogen diffusion contribution to the corrosion rate of the ferritic steels used in oil refining furnaces.
Book
Full-text available
Functional Materials for the Oil and Gas Industry: Characterization and Applications discusses the latest techniques in characterization and applications of functional materials in the oil and gas industry. It provides an expert review of recent developments in a variety of materials, such as ceramics, composites, and alloys, and covers all major aspects relevant to the industry, including asset management (corrosion), operation (pipeline engineering), energy management, and applications in extreme environments. This book: Discusses modern characterization techniques, such as in situ TEM, SAXS, SANS, X-ray, and neutron tomography Covers conventional and advanced nondestructive techniques (NDTs), such as ultrasonic testing and radiography for asset integrity checking in oil and gas sectors Describes advanced properties of a variety of functional materials and their applications to the oil and gas field Explains self-cleaning coating technologies and their applications and materials for renewable energy sources Details advances in synthesis methods for functional materials Features industrial aspects of afunctional materials application in each chapter Written for an interdisciplinary audience of industrial practitioners, academics, and researchers in petroleum, materials, chemical, and related disciplines of engineering, this work offers significant insight into the state-of-the-art in the development and characterization of advanced functional materials.
Article
Full-text available
It is well known that delayed Coker unit is one of the most important refinery units in the downstream industry, coke drums are exposed to number of cycles of thermal loads which initiate mechanical deformation in form of bulges and subsequently cracks which occur because of low-cycle fatigue. To increase the coke productivity, some operators have introduced short operational cycles to increase the production rate at some point to increase the revenue, since by shortening the operational cycle, the coke drum will be able to make more cycles in short period and subsequently produce high volume of the final product. But the fact that increasing the revenue now is reducing the fatigue-life time of the coke drum and subsequently the coke drum retires before its normal time, added to that production loss, human safety, and environmental impacts due to failures because of working beyond design parameters. Therefore, this paper discusses the consequences on the coke drum after introducing short operational cycles especially on the cladding materials in the bottom section of the coke drum, and this paper explores how the sulfidation corrosion occurred and what is the impact on the reliability of coke drum itself.
Article
Full-text available
Sulfidation and naphthenic acid corrosion have been the main causes of damage in oil refinery units for decades, with serious safety and reliability issues. Improving the understanding of the involved reactions and mechanisms is a key factor to better control the corrosion in refinery processes. We present here an investigation based on the surface reaction of model sulfide layers with naphthenic acid diluted in white oil, at 260°C. Iron sulfide layers were formed in pure gaseous H2S at 300°C on carbon steel (XC18), Fe-2.25Cr-1Mo (P22) steel, Fe-5Cr-0.5Mo steel (P5) and Fe-18Cr-8Ni stainless steel (304L). The sulfide layers were characterized by grazing X-Ray Diffraction (XRD), X-ray Photoelectron Spectroscopy (XPS) and Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS). The diffraction patterns showed that the pyrrhotite (Fe1-xS) phase was formed, thus confirming that the iron sulfide layers were satisfactorily modelling the sulfidation occurring in industrial conditions (in refinery sulfur crude oil at 300°C). The ToF-SIMS profiles allowed us to evidence a duplex layer, for all sulfided steels: an external pyrrhotite layer, an internal oxide layer, the composition of which changing with the type of steels. The thicknesses of the pyrrhotite layers were in the range of 60-100 nanometers, depending on the samples. The experiments with naphthenic acid were performed in a dedicated laboratory reactor, designed for controlled short reaction times: from tens of seconds to several minutes. The surface was analyzed by ToF-SIMS. The ToF-SIMS data (spectra, images and profiles) confirm the surface modifications after interaction with naphthenic acid. Except 304L which did not react in these conditions (after 9 hours of interaction), the naphthenic acid interaction resulted in a localized deep reaction with the metal. In the regions where no reaction was observed, the sulfide film was chemically intact and (pseudo)-resistant (no decrease of the initial thickness). The total detachment of the sulfide layer occured after a reaction time of 16 min with naphthenic acid, in the case of XC18. On P22 and P5 steels, the time required before observation of the localized attack (15 min and 30 min, respectively) and the time of total detachment of the sulfide layer (40 min and 80 min, respectively) were longer than for XC18 steel. The corrosion mechanisms explaining these observations most probably imply (i) the steel chromium content which has a beneficial effect on the corrosion behavior through the presence of chromium oxide at the metal/sulfide interface, but also (ii) metallurgical heterogeneities such as iron carbides, also localized at the metal/sulfide interface.
Article
Corrosion has always been one of the major problems in oil refineries. Although crude oils consist predominantly of non-corrosive hydrocarbons, they contain varying amounts of potentially-corrosive impurities. In addition, refinery processes use chemicals known to be corrosive. In this article, corrosion problems in a refinery environment are discussed under two categories: low-temperature corrosion and high-temperature corrosion. The causes of low-temperature corrosion in a refinery are the presence of hydrochloric acid, hydrogen sulphide, oxygen, oxides of sulphur, amines, or cyanides. High-temperature corrosion in refineries is caused by sulphur compounds, naphthenic acids, or hydrogen gas. The control measures for these corrosion problems are discussed in the light of recent literatures. Corrosion monitoring of refinery equipment and optimizing control measures are discussed separately.
Article
Refiners are continuously challenged with processing crudes that have increased concentrations of sulfur and naphthenic acid. Very often the original equipment metallurgy was not designed for such corrosive conditions, and therefore substantial upgrades are required in the high temperature areas (greater than 428°F (220°C)) where this type of corrosion will show up. This paper will present new concepts for the evaluation of critical TAN (Total Acid Number) and sulfur level based on the comprehensive evaluation of high temperature corrosion risk. A model consisting of public and proprietary data has been developed for this assessment. The primary factors evaluated include TAN / sulfur content, materials of construction, and temperature. Utilization of this model enhances the predictability of corrosion due to TAN and sulfur. Historical data of corrosion monitoring is used to validate the accuracy of the model. Minimum wall thickness required for pressure vessels and piping is also calculated. Then, high risk parts of distillation units are identified. Critical operational TAN and sulfur level of a crude distillation unit is determined by the identified maximum allowable TAN and sulfur concentration of any one circuit. This advanced knowledge will allow for increased reliability of crude distillation units where this is implemented.
Article
The aim of this work is to analyse hydrogen damage, monitoring, and possible prevention at the oil refineries' units. Hydrogen gas occupies an essential place in the processes in the oil refining industry: hydrodesulfurizers, hydrocrackers, and catalytic reformers. In addition to these processes, there are some sources of hydrogen gas arising from corrosion of carbon steel equipment in contact with aqueous solutions of acids, such as H 2S (sour water), HC1, HCN, H 2SO 4, and HF. High temperature chemical attack of carbon steel equipment by naphthenic and other organic acids can be related to hydrogen damages owing to appearing of hydrogen gas on metal surface as a result of such attack. Historically many names of hydrogen damages appeared: hydrogen degradation, hydrogen embrittlement, hydrogen blistering, decarburization, hydrogen stress cracking (HSC), hydrogen attack, high temperature hydrogen attack (HTHA), hydrogen-induced cracking (HIC), also known as stepwise cracking (SWC), stress-oriented hydrogen-induced cracking (SOHIC), sulfide stress cracking (SSC). This profusion of names does not always correctly elucidate and explain them, and can even confuse and complicate their detection, understanding, monitoring, and control. From analyzing the literature on this subject and based on our own experience, we differentiate all hydrogen damage failures into two main groups based on two mechanisms: electrochemical processes (mainly at low temperature, up to ~100°C) arising from acid corrosion and cathodic protection, and high temperature (between 200 and 900°C) arising from the presence of hydrogen gas at high pressures. Examples of hydrogen damage are given for the various units of oil refineries. Different monitoring methods were developed for detection of possible hydrogen damages. Hydrogen can be detected either in intrusive or non-intrusive devices called hydrogen probes. Hydrogen that penetrates through a metallic wall can be detected with manometric (hydrogen pressure,) or vacuum method, electrolytically (hydrogen ionization from H atoms into H + ions), heat conduction (gas chromatography), vacuum extraction at 400°C, or hydrogen effusion. Monitoring methods are critically reviewed. Preventive measures of hydrogen failures are differentiated into two groups according to low (electrochemical) and high temperature (dissociation of hydrogen molecules) mechanisms. The first group includes protective measures from hydrogen blistering and sulfide stress cracking (SSC): metallurgical measures, change of environmental conditions (removing aggressive species such as sulfides, cyanides, and arsenic compounds, neutralization, injection of inhibitors of hydrogen penetration), use of organic, inorganic, and metallic coatings, heat treatment and proper welding. The second group includes metallurgical measures (use of steels containing chromium and molybdenum, and decrease of carbon content in steel), heat treatment, and proper welding. Analysis of preventive measures of hydrogen failures was carried out and recommendations were given. Examples of hydrogen damages are given for the isomerization and hydrodesulfurizer units of oil refinery.
Article
Hot corrosion is an accelerated degradation (oxidation and/or sulfidation) of metals and alloys in the presence of liquid melts of salts and/or oxides. It's generally divided into two forms: Type I and Type II. Type I hot corrosion usually occurs at 850-950 C (1550-1750 F), and it is also known as high-temperature hot corrosion. Type II hot corrosion occurs at 600-800 C (1112-1472 F). Type II hot corrosion is also called low-temperature hot corrosion. It may take the form of fuel-ash or oil-ash corrosion in refinery heater fire side tube surfaces as the result of fuel impurities. Another potential location for Type II hot corrosion is FCCU regenerators, where the operating temperatures are normally in the range of 650 to 788 C (1200 to 1450 F). The mechanisms and characteristics of Type II hot corrosion will be presented. The analytical methods and mitigation strategies for Type II hot corrosion will also be discussed.
Article
A discussion on refining high acid crudes covers a revised procedure that includes analysis, testing, and a corrosion model accounting for critical parameters for the evaluation of operational risk, unit upgrading, and maintenance planning; refinery experience; naphthenic acid corrosion; sulfidic corrosion; impingement corrosion on 5Cr steel; API RP 581 base resource document; specific crude oil properties influencing corrosivity; and naphthenic acid structures from crude oil and reagent naphthenic acid.
Article
This paper reviews sulphidation behaviour of steel structures in the refinery industry. The data presented are related to two aspects of the corrosion. One aspect refers to external degradation of the structures due to formation of sulphide scales on the steel surface and metal losses; another is connected with internal degradation, i.e. changes of steel microstructure and formation of internal corrosion products. The so-called modified McConomy curves, and the Couper-Gorman curves used to predict corrosion rates of steels in hydrogen-free refinery streams, and hydrogen-containing environments are shown. Decreased stability of carbides in steel due to the presence of sulphur on the steel surface is considered, and a role of hydrogen in this process is outlined.