Engineering Failure Analysis 119 (2021) 105011

Contents lists available at ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal

A corrosion failure analysis of heat exchanger tubes operating in

petrochemical refinery
Milad Rezaei *, Zeynab Mahidashti, Sajad Eftekhari, Ehsan Abdi
Department of Materials and Metallurgical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Hafez Ave., P.O. Box 15875-
4413, Tehran, Iran

A R T I C L E I N F O A B S T R A C T

Keywords: The aim of this study was to analyze corrosion failure of heat exchanger tubes used in petro­
Erosion-corrosion chemical refinery. Thermodynamic and kinetics analysis showed that the major chemical sub­
Heat exchanger stance playing role in electrochemical corrosion of the tubes is solute CO2 present in crude
Crude methanol
methanol passing by the tubes. However, polarization tests revealed that corrosion caused by CO2
Hard chromium coating
is not so significant to produce cavities observed on the outer surface of the tube. By further
investigation of the morphology and previous processes in producing crude methanol, it was
revealed that crude methanol contains significant amount of suspended solid particles with
relatively high hardness, which were detached from the catalysts of the previous phases.
Therefore, besides the electrochemical corrosion, the main mechanism of tubes corrosion was
attributed to the erosion-corrosion. According to the corrosion cause, appropriate methods were
proposed for corrosion prevention or corrosion reduction of the tubes.

1. Introduction

Corrosion is considered as one of the most serious problems, annual losses of which is in the hundreds of billions of dollars [1].
Replacement and repair of the equipment and facilities of most industries are among frequently losses caused by corrosion. Heat
exchangers are kind of structures prone to corrosion and subsequent failure [2]. The use of heat exchangers in various industries is
quite common; for example, they can be found in a variety of boilers and condensers in oil, petrochemical and gas industries [3,4]. In
principle, heat exchanger systems are used to heat or cool process fluids, and the operation of a process unit depends on the proper
functioning of this equipment. Design of many heat exchangers is such a way that embedment of slots, connection of dissimilar metals,
sedimentation, thermal gradients at the surface of metal and fluid, high speed of fluid in specific regions, or presence of stagnant fluid
in some other regions is inevitable [5]. All of these, along with the corrosive fluid in the heat exchanger, leads to the increment of
corrosion, acceleration of destruction, failure of equipment and eventually phases the line out of the production cycle.
Corrosion of heat exchangers in petrochemical refineries is one of the most common reasons behind the system shut down [6]. In
this case, corrosion of the heat exchanger tube can be very problematic for petrochemical plants, because the perforation of the tubes
leads to the coolant fluid leakage from tubes into the process fluid. Pollution of the fluid process causes significant damage and loss of
products. There are various reports regarding heat exchangers failure due to main corrosion mechanisms such as uniform [7], pitting
[8], crevice [9], stress cracking [10], erosion [11], intergranular [12], galvanic [13], selective [14] and microbiologically induced

* Corresponding author.
E-mail address: miladrezaei@aut.ac.ir (M. Rezaei).

https://doi.org/10.1016/j.engfailanal.2020.105011
Received 7 August 2019; Received in revised form 8 October 2020; Accepted 12 October 2020
Available online 15 October 2020
1350-6307/© 2020 Elsevier Ltd. All rights reserved.
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

corrosion [4,15]. For example, Deen et al. reported the failure of a 316L plate heat exchanger due to pitting corrosion. The authors
concluded that high chloride contents in the cooling water caused a local breakdown of the passive oxide film [16]. Prithiraj et al.
showed pitting and intergranular corrosion of carbon steel and stainless steel in the presence of a mixed industrial culture of bacteria
found in industrial heat exchangers [15]. Consequently, corrosion control in heat exchangers plays an important role in the operation
of the system.
Due to the non-corrosiveness of many processes, carbon steel is widely used as a construction material in petrochemical complexes.
During methanol production process, however, possible substances such as CO2 gas is produced additionally [17]. This gas, along with
water present in the process, can lead to carbonic acid formation and pH reduction of the fluid. As a result of acidity increment of the
process fluid, corrosion of pipes and equipment made of carbon steel increases and equipment failure may happen [18]. There are
various studies conducted on CO2 induced corrosion of mild steel in different pH ranges [19,20]. Gray et al. studied the influence of
temperature and pH on the mechanism of mild steel corrosion in aqueous CO2 solutions and modelled Tafel constants for iron
dissolution [21]. Tanupabrungsun studied CO2 corrosion of carbon steel in a temperature range of 80–300 ◦ C in 1 wt% NaCl solution
and analysis of corrosion product showed that FeCO3 layers were formed on all the test samples [20]. Moreover, presence of some
erosive side products in the fluid such as solid particles generated as a result of reaction with chloride ions can also accelerate corrosion
through erosion-corrosion [22].
Failure of a heat exchanger is investigated in this work, which is in contact with crude methanol containing solute CO2 gas, water
and side products. The construction material is carbon steel according to ASTM A179. Based on the refinery reports, the tube is the part
experiencing corrosion. All the processing units were analyzed and possible reasons for corrosion and failure of the heat exchangers
were examined, such as inappropriate processing conditions, unsuitable inspection and maintenance programs or improper material
selection. For this purpose, in addition to sampling the tubes and corrosion products from the desired heat exchanger and analyzing
them, appropriate laboratory tests were designed and performed to determine the mechanism and causes of corrosion. Finally, suitable
solutions for dealing with this problem in the framework of the research is proposed.

2. Experimental

2.1. Material

The test samples in this research were gathered from the tubes of failed carbon-steel heat exchangers taken from an oil refinery in
Iran. The composition of the carbon steel is given in Table 1.
The failed tube bundle and positions of the corroded areas on it are depicted in Fig. 1. Corroded samples along with deposits of
corrosion products on the tubes and shell section of the heat exchanger were gathered from two sites of the field, site I and site II. The
sample taken from site II has experienced longer corrosion times.
The heat exchanger transports crude methanol with input and output temperature of 38.3 ◦ C and 67.7 ◦ C, respectively with a pH of
6.2. Composition of crude methanol is reported in Table 2.

2.2. Characterization

2.2.1. Surface characterization techniques

Surface morphology of the samples taken from site I and II and composition of corrosion products were obtained using scanning
electron microscope model VEGA\\TESCAN-LMU equipped with an energy dispersive X-ray spectrometer (EDS). Elemental charac­
terization was performed by X-ray fluorescence (XRF) model ARL PERFOMIR’X, while phase characterization was carried out using X-
ray diffraction (XRD) model XRD-EQuinox 3000 with CuKα radiation. PANalytical X’pert highscore plus (PXHP) software was utilized
for fitting the XRD pattern. For better examination of elemental composition, ionic chromatography (IC) and inductively coupled
plasma optical emission spectrometry (ICP-OES) model Optima 8000 were used.

2.2.2. Electrochemical measurements

The electrochemical experiments were carried out on samples immersed in 300 mL of simulated crude methanol. The cell was
composed of steel samples with a chemical composition similar to that of the tubes as the working electrode, Ag/AgCl electrode as a
reference electrode, and platinum plate as a counter electrode. In cyclic potentiodynamic polarization tests, the sample was polarized
in forward scan from − 0.2 V to around +1 V versus OCP (open circuit potential) and then scan became reversed to − 0.2 V versus OCP
at the same scan rate. The data were recoded using Autolab PGStat30 workstation and corrosion parameters were extracted by fitting
the curves with Nova software (ver 1.6).

Table 1
Chemical composition of carbon steel according to ASTM A179.
Element C Si Mn P S

Content (wt%) 0.06–0.08 – 0.27–0.63 <0.035 <0.035

2
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

Fig. 1. Macroscopic images of (a) failed tube bundle and (b) position of corroded areas on the tubes.

Table 2
Composition (molar percentage) of the crude methanol passing through the heat exchanger.
Compound CH3OH H2O CO2 CH4 H2 CO N2

Molar percentage 80.61 18.49 0.73 0.05 0.02 0.01 0.01

3. Results and discussion

3.1. Corrosion investigation

Fig. 2a and b shows cross sectional SEM images of corroded tube samples. As can be seen, outer surface of the tubes looks like a gear
wheel with a relatively regular pattern and thickness reduction can clearly be observed in specific areas. Relatively curved cavities
formed on the outer surface (shown by arrow in Fig. 2b) can be attributed to erosion-corrosion on the outer surface. Furthermore,
corrosion products deposited on the outer surface in some regions are obvious, which can lead to acceleration of local or under-deposit
corrosion. Such deposits can also result in erosion-corrosion in the presence of a moving fluid as will be discussed in the next sections.
Fig. 2c shows SEM image of deposits in the shell. Spherical morphology of the deposits is clear. These deposits belong either to
detached corrosion products from tube surface or to other solid pollutants within crude methanol, which have experienced erosion
during their transport to the sell. In either case, absence of sharp edges can be a sign of erosion in system [23].
EDS analysis of corrosion deposits on the tube surface (Fig. 2a and b) shows that major fraction of deposits consists of Fe, C and O.
The amount of carbon on the tube surface has increased from 0.06 wt% in the original sample to more than 6 wt% in corroded samples.
Besides O which is absent in the original sample is obviously detected in the corroded samples. In fact, the corrosion of carbon steel is
the percentage loss of Fe, and the addition of other elements such as C and O [24]. Therefore, it can be concluded that possible
corrosion product is a combination of iron oxide and carbonate. Furthermore, there is higher amount of O in the corrosion product
taken from the heat exchanger that has experienced longer corrosion (site II), compared to the tube taken from site I (30 wt% vs. 21.5
wt%). In the other words, the ratio of iron carbonate to iron oxide is higher in more corroded sample. It can be concluded that corrosion
product at the beginning is iron oxide, which converts to iron carbonate due to contact with crude methanol containing CO2. This
formation of iron carbonate in the presence of CO2 is well established in the literature [13,25] and its possibility can also be shown by
thermodynamic calculations. In presence of water with pH of 6.2, iron can be oxidized and Fe2+ can be produced through reactions (1)
and (2) (ΔE > 0) [20]:

Fe → Fe2+ + 2e (1)

2H2O + 2e → H2 + 2OH− (2)

On the other hand, carbonic acid is generated due to CO2 presence in crude methanol (0.73 mol.%) through following reaction
[26]:

CO2 + H2O → H2CO3 Keq = 2.434 × 10-2 to 1.1 × 10-2 (38.3 ◦ C to 67.7 ◦ C) (3)

3
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

Fig. 2. SEM-EDS results of (a) tube from site I, (b) tube from site II, and (c) deposits in the shell.

This acid is reduced and generates carbonate alongside with bicarbonate anions (Eq. (4) and (5)) [25], which react with iron cations
in the next step and iron carbonate is formed (Eq. (6)) [27]:

2H2CO3 + 2e → 2HCO−3 + H2 (4)

2HCO−3 + 2e → 2CO2−
3 + H2 (5)

Fe2+ + CO2−
3 → FeCO3 (6)

From EDS analysis of shell deposits (Fig. 2c), detection of elements such as Cu, Zn, Al, Si and K confirms presence of abrasive
particles in crude methanol. Moreover, high content of C, O and Fe implies that possible brittle corrosion product of the system, i.e.
FeCO3, has been detached from the outer surface of the tubes and deposited in the shell.
To better understand the distribution of elements throughout the tube, EDS line scan analysis of site I and site II tubes is shown in
Fig. 3. From Fig. 3a, it is clear that oxygen and carbon content have increased in the outer surface of the tube (right side of the curve).
The length of this region is about 5 µm, which is approximately equal to the thickness of the corrosion products present on the tube
surface. Due to this increase in carbon and oxygen concentration, it can be concluded that corrosion products on the surface of the tube
are iron carbonates. Similar result is obtained for tube of site II, as shown in Fig. 3b. As this tube has experienced lower corrosion,

4
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

Fig. 3. SEM-EDS line analysis of the sample taken from (a) site I tube and (b) site II tube.

Fig. 4. XRD results of corrosion products found in the shell of heat exchanger.

5
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

oxygen content is lower in corrosion products of the tube surface.

Fig. 4 shows result of X-ray diffraction analysis for deposits in the shell side of the heat exchanger. In this spectrum, various phases
including CuO, Cu2O, Cu, ZnCO3, SiO2, Al2O3 and Fe0.98O can be observed. Alumina, silica, and carbonate are among abrasive par­
ticles. Based on the type of fluid motion and intensity of the flow, these particles can result in a sharp increase in mechanical
degradation and are one of the factors affecting erosion-corrosion [28]. Furthermore, presence of copper ion in crude methanol can
increase electrochemical corrosion rate of steel tubes according to the following spontaneous reaction:

Fe + Cu2+ = Cu + Fe2+ (7)

According to EDS analysis, however, no Cu was detected on the tube surface. For further evaluation, ICP and XRF tests were
performed on the corrosion products and 0.27% Cu was detected in corrosion products of the tube side (Table 3), which is far less than
the amount found in the corrosion product of the shell side (app. 44%).
In this case, one can make two conclusions. First, it can be said that Cu reduces on the tube surface according to Eq. (7) and
subsequently detaches due to the relative motion of crude methanol, which leads to its deposition in the shell side. Second, it can be
hypothesized that Cu enters heat exchanger as an external source within crude methanol and trace amount of it traps in corrosion
products of the tube surface. To evaluate these possibilities, immersion test was performed on three representative samples of the tube
side at 53 ◦ C for 48 h in three different media as shown in Fig. 5: i) pure methanol +0.8 ppm Cu2+, ii) pure methanol +18.49% water,
iii) pure methanol +0.73% CO2. As can be seen from the results reported in Table 4, the sample immersed in solution containing CO2
gas has experienced the highest corrosion rate of 0.8 mpy, while corrosion rate of the one immersed in solution containing Cu2+ is less
than 0.05 mpy. This result confirms that Cu detected in corrosion products of the shell side cannot be produced as a result of corrosion
reaction since the corrosion rate in presence of Cu2+ is significantly low.
Table 5 represents ionic chromatography results of four common corrosive ions usually found in corrosion systems. Chloride and
sulphate ions are detected in the deposits, which may have entered crude methanol in previous processing phases. These ions can result
in local corrosion and deposition of corrosion products on the surface. The distance between deposits and metal surface, can itself act as
a crevice and accelerate corrosion rate. Furthermore, theses brittle deposits can be detached from the surface as a result of turbulent
stream and act as abrasive particles.
In order to evaluate corrosion rate of the tubes before formation of oxide layers, potentiodynamic polarization tests was performed
in different rotation speeds (0–2000 rpm) at room temperature and different temperatures (32–69 ◦ C) at 300 rpm, on steel samples
with a chemical composition similar to that of the tubes. The electrolyte used is a simulated crude methanol solution. Polarization
curves and parameters obtained are reported in Fig. 6 and Table 6, respectively.
It is clear that corrosion rate increases by increasing temperature, reaching to 28.1 µA.cm− 2 at 69 ◦ C (three times increase).
Corrosion potential is also shifted to more negative values, implying that temperature increase affects kinetics of corrosion reactions,
and thermodynamically, tubes become more active. As can be seen, rotation speed have no significant effect on the corrosion current
and potential. However, it should be mentioned that it can increase the rate of detachment of passive layers or corrosion products from
substrate in the presence of abrasive particles in the electrolyte.
A cyclic polarization test was performed in order to investigate the possibility of pitting corrosion (Fig. 7). As shown in Fig. 7,
backward anodic branch have not intersected forward one in a wide range of potential (from − 0.5 to 0.8 V vs. Ag/AgCl). Therefore,
pitting in this condition is not possible [29].

3.2. Summery of corrosion mechanism

Overall, macroscopic and microscopic evidence of tube surface confirmed that the cavities are result of incident of solid particles
such as Al2O3, SiO2, ZnCO3, FeO, etc., which were detected in corrosion precipitates. These particles are of relatively high hardness
(2600 HV for alumina, 2200 HV for silica and 420 HV for carbonate compared to 129 HV for carbon steel of the tube) and are capable
of producing cavities on the substrate as a result of collision. Furthermore, iron carbonate layer, presence of which was confirmed by
EDS results, has a low adhesion to the substrate in temperatures lower than 70 ◦ C according to the literature [30,31]. Considering that
the working temperature of the heat exchanger is about 53 ◦ C, carbonate layer cannot act as a protective layer and can be eroded easily
[26]. These results confirm that the main mechanism behind the exchanger failure is erosion-corrosion. Other possibilities of corrosion
in the system such as pitting, cavitation, galvanic and under-deposit corrosion can be rejected with evident. In the case of pitting

Table 3
ICP and XRF analysis performed on corrosion deposits of shell and tube sides.
Compound XRF Results (wt%) Compound ICP Results (ppm) Compound ICP Results (ppm)

Shell side Cu 44.054 Cu >100,000 Tube Side Cu 2751

Zn 10.14 Zn >100,000 Zn 590

Al2O3 15.538 Al 66,500 Al 1000
SiO2 15.352 Fe 22,600 Fe 933,200
Na2O 3.424 Na 2000 Na 300
Fe2O3 3.162 K 10,900 K 0
MgO 1.006 Mg 4600 Mg 300
K2O 0.839 Cr 885 Cr 1173

6
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

a b c

Fig. 5. Immersion test method conducted on (a) pure methanol +0.8 ppm Cu2+, (b) pure methanol +18.49% water and (c)) pure methanol
+0.73% CO2.

Table 4
Corrosion rate (mpy) of the tube samples immersed in three different
mediums.
Medium Corrosion rate (mpy)

Methanol + 0.79% CO2 0.8

Methanol + 0.8 ppm Cu2+ < 0.05
Methanol + 18.49% H2O 0.2

Table 5
Ionic chromatography results performed on the corrosion products of tube samples.
Ion type Deposits in the shell side (ppm) Deposits on the tubes (ppm)

F- 1 0
Cl- 290 120
NO3- 1 3
SO2-4 460 490

Fig. 6. Potentiodynamic plots obtained at (a) different rotation rates and temperature of 25 ◦ C and (b) different temperatures and rotation speed of
300 rpm (scan rate: 100 mV.s− 1).

7
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

Table 6
Corrosion current and potential obtained at different rotation speeds and temperatures.
Corrosion parameters Temperature (◦ C)

25 32 36 44 49 54 60 65 69

Ecorr (V) − 0.14 − 0.21 − 0.31 − 0.23 − 0.31 − 0.30 − 0.25 − 0.29 − 0.3
icorr (µA.cm− 2) 1.46 9.92 12.5 10.6 13 11.5 15.5 14.4 28.1

Rotation speed (rpm)

0 1000 2000

Ecorr (V) − 0.07 − 0.06 − 0.08

icorr (µA.cm− 2) 1.04 0.078 1.1

Fig. 7. Cyclic polarization test carried out on tube sample in simulated crude methanol (scan rate: 100 mV.s− 1).

corrosion, it should be mentioned that cyclic polarization test showed that pitting is not probable in the system in absence of solid
particles [29]. In presence of solid particles, SEM results showed that cavities formed are not too deep and their depth is approximately
equal or smaller than the mouth, which rejects the possibility of pitting corrosion [32]. Cavitation is another possibility, which due to
the low vapor pressure of CO2 in average temperature of the crude methanol, i.e. 53 ◦ C, is not able to produce such holes observed on
the tube [33]. Galvanic coupling of Fe and Cu is the third possibility. According to XRD analysis, Cu exists in the system in the form of
oxide and carbonate not the metal form that is able to make electrical contact (although there is a little amount of elemental copper at
51◦ , but the presence of high amount of other impurities hinders proper electrical connection between tube and metal copper). Besides,
galvanic coupling should result in Cu production in corrosion deposits. However, no Cu was detected by ICP analysis of corrosion
products on tube surface. Moreover, galvanic coupling results in thickness reduction and brightness increment of active metal, here
steel tube. However, visual inspection confirmed a rough and matte tube surface. Furthermore, ratio of cathode to anode area, which is
an influential factor in galvanic coupling, is significantly low in this case [34]. All these together reject possibility of galvanic coupling
in the system. Finally, under-deposit corrosion results in pits with a depth larger than mouth. Besides, due to the relatively high speed
of crude methanol (66.5 Kg/s), shielded zone can barely be generated under the deposits, and hence this corrosion cannot be the main
mechanism of corrosion [35]. Summary of the corrosion failure possibilities is shown in Fig. 8.

3.3. Corrosion prevention procedure

Considering the main corrosion mechanism of the system, i.e. erosion-corrosion, following prevention methods were proposed: i)
Using carbon steels with higher carbon content and hence higher yield strength and hardness such as ASTM A566, which is a carbon

Different Possibiities for Corrosion Failure

Erosion-Corrosion Pitting Cavitation Galvanic Corrosion Under-deposit Corrosion

Rejected by Rejected by Rejected by High

The main Low Vapor Rejected due
Cyclic Speed of Crude
Cause of Pressure of to the Absence
Polarization and Methanol and
Failure CO2 of Matellic Cu
Cavities Shape Cavities Shape

Fig. 8. Summary of corrosion failure possibilities and reasons of their rejection.

8
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

steel with 0.3% carbon and its two commercial B2 and C2 grades in the form of seamless tubes are used in heat exchangers. ii) Using
alloy steels such as ASTM A213, which contains chromium (app. 1%) and molybdenum (app. 0.5%) and provides higher hardness and
abrasion resistance compared to ASTM A179. These steels (suggested in (i) and (ii)) have a thermal conductivity similar to that of
ASTM A179. Therefore, performance of the exchanger is not noticeably affected. iii) Utilizing hard materials may have some limi­
tations during metal forming such as bending. However, by applying surface hardening methods such as carburizing, nitriding, and
flame hardening, one can use relatively mild ASTM A179 carbon steel for metal forming and meanwhile increase material hardness. iv)
Installing proper filters at the entrance of heat exchanger or in preceding units in order to filter crude methanol and remove large
abrasive particles detached from catalysts before entering heat exchanger. In this case, erosion takes place only as a result of relative
motion between fluid and tube surface. Considering low velocity of the fluid, no intensive shear stress is produced at the surface. v)
Coating the tubes with abrasion resistant materials, which may be the simplest and lowest cost method in this case since there is no
need to tube replacement as in (i) and (ii) methods. In this work combination of method (iv) and (v), i.e. filtering and hard chromium
electroplating of the tubes was proposed. Filtering per se cannot remove abrasive particles of all sizes and small particles can enter the
crude methanol. Hard chromium coating is a relief in this case. Chromium electroplating is an easy to perform procedure, which can be
done during retubing of heat exchangers, after bending and before inserting the tubes inside the heat exchangers. This kind of coating
is of strong adhesion to the substrate, acceptable corrosion resistance and high hardness and abrasion resistance [36]. Besides, due to
its low thickness (about 20 µm), it will not affect thermal conductivity of the steel substrate according to Fourier’s law [37].

4. Conclusion

This study was performed to find the reason behind corrosion failure of tube heat exchangers containing crude methanol. In­
vestigations revealed that crude methanol contained significant amount of suspended solid particles with relatively high hardness,
which resulted in erosion-corrosion of the tube. Accordingly, appropriate methods were proposed for corrosion prevention, including:
i) using carbon steel tube with higher carbon content, such as A556; ii) utilizing alloy steels, such as A213; iii) surface hardening of the
tube surface, such as carburizing or flame hardening; iv) crude methanol filtering before entering heat exchanger to remove suspended
hard particles; v) applying hard chromium coating on the surface of the heat exchanger tube. Among these methods, methanol filtering
and using hard chromium coating with a thickness of 30 µm were recommended.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

References

[1] R. Bhaskaran, N. Palaniswamy, N. Rengaswamy, M. Jayachandran, Global cost of corrosion—A historical review, Corrosion: Mater. 13 (2005) 621–628.
[2] M. Ali, A. Ul-Hamid, L.M. Alhems, A. Saeed, Review of common failures in heat exchangers – Part I: Mechanical and elevated temperature failures, Eng. Fail.
Anal. 109 (2020) 104396, https://doi.org/10.1016/j.engfailanal.2020.104396.
[3] X.u. Xiu-qing, B. Zhen-quan, F. Yao-rong, M.a. Qiu-rong, Z. Wen-zhen, The influence of temperature on the corrosion resistance of 10# carbon steel for refinery
heat exchanger tubes, Appl. Surf. Sci. 280 (2013) 641–645.
[4] T.Y. Rizk, K.M. Al-Nabulsi, M.H. Cho, Microbially induced rupture of a heat exchanger shell, Eng. Fail. Anal. 76 (2017) 1–9.
[5] T.K. Nguyen, M. Sheikholeslami, M. Jafaryar, A. Shafee, Z. Li, K.V.V. Chandra Mouli, I. Tlili, Design of heat exchanger with combined turbulator, J. Therm. Anal.
Calorim. 139 (1) (2020) 649–659.
[6] W. Faes, S. Lecompte, Z. Yunus Ahmed, J. Van Bael, R. Salenbien, K. Verbeken, M. De Paepe, Corrosion and corrosion prevention in heat exchangers, Corrosion
Rev. 37 (2) (2019) 131–155.
[7] J. Xie, K. Yazdanfar, G. Anning, T. Rockwell, A. Ahmed, Corrosion of a vertical shell and tube heat exchanger, NACE Int. (2015).
[8] Z. Qiankun, S. Yafei, R. Sixian, L. Huifeng, Z. Xingjiang, Corrosion failure analysis on heat exchanger pipes, J. Failure Anal. Prevention 17 (2) (2017) 349–353.
[9] Y. Gong, Z.G. Yang, J.Z. Yuan, Failure analysis of leakage on titanium tubes within heat exchangers in a nuclear power plant. Part II: Mechanical degradation,
Mater. Corrosion 63 (1) (2012) 18–28.
[10] C.R. Corleto, G.R. Argade, Failure analysis of dissimilar weld in heat exchanger, Case Stud. Eng. Fail. Anal. 9 (2017) 27–34.
[11] U. Klein, A. Zunkel, A. Eberle, Breakdown of heat exchangers due to erosion corrosion and fretting caused by inappropriate operating conditions, Eng. Fail. Anal.
43 (2014) 271–280.
[12] D. Ifezue, F.H. Tobins, Corrosion failure of aluminum heat exchanger tubes, J. Fail. Anal. Preven. 15 (4) (2015) 541–547.
[13] K. Thulukkanam. Heat exchanger design handbook, CRC press, 2013.
[14] Y.D. Li, N. Xu, X.F. Wu, W.M. Guo, J.B. Shi, Q.S. Zang, Failure analysis of the condenser brass tube in 150 MW thermal power units, Eng. Failure Anal. 33 (2013)
75–82.
[15] A. Prithiraj, I.O. Otunniyi, P. Osifo, J. van der Merwe, Corrosion behaviour of stainless and carbon steels exposed to sulphate – reducing bacteria from industrial
heat exchangers, Eng. Failure Anal. 104 (2019) 977–986.
[16] K.M. Deen, M.A. Virk, C.I. Haque, R. Ahmad, I.H. Khan, Failure investigation of heat exchanger plates due to pitting corrosion, Eng. Fail. Anal. 17 (4) (2010)
886–893.
[17] R. Sharma, H. Poelman, G.B. Marin, V.V. Galvita, Approaches for selective oxidation of methane to methanol, Catalysts 10 (2) (2020) 194.
[18] R. Julian, K. Lichti, E. Mroczek, B. Mountain, Heavy metal scaling and corrosion in a geothermal heat exchanger, Proceedings World Geothermal Congress
(2015).
[19] F.M. Song, D.W. Kirk, J.W. Graydon, D.E. Cormack, Predicting carbon dioxide corrosion of bare steel under an aqueous boundary layer, Corrosion 60 (8) (2004)
736–748.
[20] T. Tanupabrungsun, B. Brown, S. Nesic, Effect of pH on CO2 corrosion of mild steel at elevated temperatures, Proc. Corros. (2013) 2348.
[21] L.G. Gray, B.G. Anderson, M.J. Danysh, P.R. Tremaine, Effect of pH and temperature on the mechanism of carbon steel corrosion by aqueous carbon dioxide,
Corrosion 40 (1990).
[22] B. Kuźnicka, Erosion–corrosion of heat exchanger tubes, Eng. Fail. Anal. 16 (7) (2009) 2382–2387.

9
M. Rezaei et al. Engineering Failure Analysis 119 (2021) 105011

[23] L.L. Li, Z.B. Wang, Y.G. Zheng, Interaction between pitting corrosion and critical flow velocity for erosion-corrosion of 304 stainless steel under jet slurry
impingement, Corros. Sci. 108084 (2019).
[24] X. Xu, S. Liu, Y. Liu, K. Smith, Y. Cui, Corrosion of stainless steel valves in a reverse osmosis system: Analysis of corrosion products and metal loss, Eng. Fail.
Anal. 105 (2019) 40–51.
[25] S. Nešić, Key issues related to modelling of internal corrosion of oil and gas pipelines – A review, Corros. Sci. 49 (12) (2007) 4308–4338.
[26] N. Mundhenk, S. Carrero, K. Knauss, R. Wonneberger, A. Undisz, Y. Wu, Kinetic and thermodynamic analysis of high-temperature CO2 corrosion of carbon steel
in simulated geothermal NaCl fluids, Corros. Sci. 108597 (2020).
[27] B. Craig, Corrosion product analysis—a road map to corrosion in oil and gas production, Mater. Perform. 41 (8) (2002) 56–58.
[28] S. Skal, Y. Kerroum, A. Guenbour, A. Bellaouchou, R. Boulif, H. Idrissi, J. García-Antón, A. Zarrouk, Erosion-corrosion effect on the alloy 316L in polluted
phosphoric acid, J. Bio- and Tribo-Corrosion 5 (3) (2019) 78.
[29] M. Javidi, S.M.S. Haghshenas, M.H. Shariat, CO2 corrosion behavior of sensitized 304 and 316 austenitic stainless steels in 3.5 wt.% NaCl solution and presence
of H2S, Corros. Sci. 163 (2020), 108230.
[30] Y. Yang, B. Brown, S. Nešic, M.E. Gennaro, B. Molinas, Mechanical strength and removal of a protective iron carbonate layer formed on mild steel in CO2
corrosion, NACE Int. (2010).
[31] A. Korchef, M. Touaibi, Effect of pH and temperature on calcium carbonate precipitation by CO2 removal from iron-rich water, Water Environ. J. (2019) 1–11.
[32] F. Pessu, R. Barker, A. Neville, CO2-corrosion of carbon steel: the synergy of chloride ion concentration and temperature on metal penetration, Corrosion (2020).
[33] T. Magne, R. Paridaens, F. Ravelet, S. Khelladi, F. Bakir, P. Tomov, L. Pora, Effect of gas content on the cavitating and non-cavitating performance of an axial
three-bladed inducer, Multiphase Sci. Technol. 32 (1) (2020).
[34] J.X. Jia, G. Song, A. Atrens, Influence of geometry on galvanic corrosion of AZ91D coupled to steel, Corros. Sci. 48 (8) (2006) 2133–2153.
[35] Y. Hou, C. Aldrich, K. Lepkova, B. Kinsella, Detection of under deposit corrosion in a CO2 environment by using electrochemical noise and recurrence
quantification analysis, Electrochim. Acta 274 (2018) 160–169.
[36] E.H. Williamson, M. Gee, D. Robertson, J.F. Watts, M.J. Whiting, J.A. Yeomans, Wear performance and characterisation of coatings for nuclear applications:
WC-(W, Cr)2C-Ni and hard chromium plate, Wear 430–431 (2019) 169–182.
[37] P.P. Adsul, L. Dineshkumar, On code verification of 2D transient heat conduction in composite wall, IOP Conf. Ser. Mater. Sci. Eng. (2018), 012128.

10