i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9

Available online at www.sciencedirect.com

ScienceDirect

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

Experimental study and numerical simulation on

the SSCC in FV520B stainless steel exposed to
H2SþCl¡ Environment

Renchao Wei a,b,*, Xuedong Chen a,b, Zhibin Ai b, Youhai Jin a

a
College of Chemical Engineering, China University of Petroleum (Huadong), Qingdao 266580, China
b
Hefei General Machinery Research Institute, Hefei 230031, China

article info abstract

Article history: Constant displacement loading tests using wedge opening loading specimens were carried
Received 9 November 2017 out in aqueous hydrogen sulfide solution containing sodium chloride to investigate the
Received in revised form susceptibility of stress corrosion cracking (SCC) of FV520B precipitation hardening
4 March 2018 martensitic stainless steel. Results of the SCC tests indicated that the stress corrosion
Accepted 18 March 2018 critical stress intensity factor (KISCC) dramatically decreased in the corrosion medium
Available online xxx investigated and decreased with the increasing of H2S concentration. Microstructures of
fracture surfaces were analyzed using a scanning electron microscope (SEM) with an en-
Keywords: ergy dispersive X-ray spectroscopy (EDS). The fracture surface was typical of sulfide stress
FV520B steel corrosion fracture. In addition, large amount of intermittent arc-crack on the side surfaces
Stress corrosion cracking around the tip of main crack formed even no main crack propagated.
Hydrogen sulfide corrosion A sequentially coupling finite element analysis (FEA) program was utilized to simulate
Hydrogen diffusion the stress field and calculate the diffused hydrogen concentration distribution of specimen
Finite element analysis exposed to the corrosion medium investigated. The FEA results indicated that corrosion pit
affected the stress and diffusion hydrogen distribution around the corrosion pit where
large stress gradients formed. Side surface cracks initiated from those corrosion pits and
propagated under the synergy of stress and hydrogen. The effect of the corrosion pit on
hydrostatic stress distribution was limited in superficial zone near the side surface, thus
side surface cracks propagated along the hoop direction rather than along the direction of
specimen thickness. Based on the morphology observation and FEA results, it can be
concluded that the SCC mechanism of FV520B steel was hydrogen embrittlement mainly
and combination of anodic dissolution. Simultaneously, corrosion pitting was the
precondition of side surface crack formation while the stress induced hydrogen diffusion
was the dominant factor.
© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

absorb atomic hydrogen which is produced via the H2S

Introduction reduction reaction on the surface. Atomic hydrogen can
penetrate into material and accumulate in the matrix-
Aqueous H2S environments can easily form in the oil and inclusion interfaces, react with metal to form hydride, which
chemical industries when equipment is in operation, overhaul will cause the embrittlement of materials, and therefore lead
or shutdown. Materials exposed to aqueous H2S environments

* Corresponding author. College of Chemical Engineering, China University of Petroleum (Huadong), Qingdao 266580, China.
E-mail address: lajiao6687@126.com (R. Wei).
https://doi.org/10.1016/j.ijhydene.2018.03.119
0360-3199/© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
2 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9

to the premature damage of equipment, which significantly The material tested in this study was treated by the following
affects their long period safety operation. The sulfide stress steps:
corrosion cracking (SSCC) of steels have attracted great
concern [1e6]. (1) Solution treatment: temperature was risen to 1050  C
With the development of computer technology, the finite and maintained for 2 h, then quenched in oil until the
element method (FEM) has proved to be a powerful tool to temperature dropped to 200  C and then cooled in air.
calculate the distribution of diffusion hydrogen concentration (2) Intermediate treatment: temperature was risen to
and analysis influence of hydrogen on the material properties 780  C and maintained for 2 h, then quenched in oil.
[7e14]. Vigdis et al. [7] used the FEM to simulate the onset of (3) Aging: temperature was risen to 620  C and maintained
hydrogen-induced fracture of API X70 pipeline steel through a 3 h, then cooled in air.
finite element cohesive element approach. The results shown
that the base metal of X70 steel revealed a low susceptibility to The chemical composition (wt.%) of the testing material
hydrogen embrittlement while the weld metal of X70 steel was determined using a SPECTRO MAXx analyzer and was
showed marginal susceptibility to hydrogen. Jiang et al. [10] listed in Table 1. It met the requirements of EN10088-1. In
developed a 3D sequential coupling calculating method to order to observe the surface microstructure of FV520B steel,
calculate the residual stresses and distribution of diffused the metallographic sample was first mechanically wet ground
hydrogen. The effects of welding residual stress and micro- to 2000 grit silicon carbide paper, then polished with 1 mm
structure on the hydrogen diffusion were taken into account diamond paste, at last it was etched in a solution consisted of
[11]. Vergani et al. [13] used a numerical model to simulate 5 ml hydrochloric acid, 1 g picric acid and 100 ml ethanol.
crack propagation of compact tensile specimens for evaluating Subsequently, the metallographic sample was observed with a
the effect of hydrogen embrittlement on low-alloy steels. ZEISS SUPRA40 field emission scanning electron microscope.
Abderrazak et al. [14] provided a comprehensive finite element Fig. 1 displayed the typical SEM image of FV520B steel, it can be
model for the numerical simulation of hydrogen induced seen that the microstructure of the steel mainly consists of
cracking in steel pipelines exposed to sulfurous compounds, martensite. The mechanical properties of FV520B steel was
such as hydrogen sulphide (H2S). The model is able to mimic measured by tensile test under the loading rate of 1 min/min.
the pressure build-up mechanism related to the recombination The sample was a standard round bar with a diameter of
of atomic hydrogen into hydrogen gas within the crack cavity. 8 mm. The stress-strain curve of this material was shown in
FV520B is one of the most widely-used materials for Fig. 2. The Vickers hardness (HV) of FV520B was measured
rotating equipment in the oil and chemical industries due to its with a load of 10 kgf for 15 s on a DVK-1s hardness tester. The
high strength, excellent plasticity and corrosion resistance. mechanical properties of FV520B steel used in this study were
Many investigations on the FV520B steel have been done so far listed in Table 2.
[15e22]. There were failure cases that FV520B steel used in The test was conducted in 5% sodium chloride solution
compressor impeller premature damaged because of SSCC, with different concentration of aqueous hydrogen sulfide in
which was accelerated by hydrogen induced cracking [20e22]. ambient temperature. The wedge opening loading (WOL)
Based on the actual service condition of FV520B steel used as sample was machined according to the national standard GB/
compressor impeller, KISCC of FV520B steel exposed to aqueous T 15970.6e2007 [23]. In order to reveal the state of crack
hydrogen sulfide solution with the presence of sodium chlo-
ride were measured through constant displacement loading
tests using wedge opening loading specimens. Scanning elec-
tron microscopy with energy dispersive X-ray spectroscopy
was also utilized to analyze the morphology of fracture sur-
face. In addition, a sequential coupling stress-hydrogen
diffusion analysis program was established by FE software
ABAQUS according to the intermittent arc-crack generated on
the side surface. Stress field and diffused hydrogen distribu-
tion around the tip of main crack were calculated to investigate
the mechanism of side surface cracking.

SCC experiment for FV520B steel

The testing material used in this study was a hot-rolled plate

Fig. 1 e Metallographic microstructure of the FV520B steel.
of FV520B martensitic precipitated hardening stainless steel.

Table 1 e Chemical composition of FV520B steel (wt%).

C Mn P S Si Ni Cr Cu Mo Nb Fe
FV520B 0.057 0.456 0.025 0.0032 0.314 5.43 13.49 1.30 1.45 0.319 Bal.
EN10088-1 0.07 1.00 0.040 0.015 0.70 5.0e6.0 13.0e15.0 1.2e2.0 1.2e2.0 0.15e0.60 Bal.

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9 3

which was the focus of this work, were observed using a

ZEISS 55 scanning electron microscope.

Finite element analysis

Based on the results of constant displacement SCC test, a

sequential coupling stress-hydrogen diffusion analysis pro-
gram was utilized to calculate the distribution of stress fields
and hydrogen around the tip of crack by FE software ABAQUS.
The stress analysis was first implemented to obtain the dis-
tribution of stress and strain fields around the tip of crack, and
then the hydrostatic stress distribution was set as a predefined
field to calculate the hydrogen diffusion concentration. For
Fig. 2 e Stress-strain curve of the FV520B steel. analyzing the formation mechanism of intermittent arc-crack
on the side surfaces around the tip of crack, a hemispherical pit
with a diameter of 0.05 mm was built to simulate the actual pit
and analyzed on the same loading condition.
Table 2 e Mechanical properties of FV520B.
Yield Tensile Elongation Shrinkage Hardness Geometrical model and meshing
strength strength d (%) j (%) HV
Rp0.2 (MPa) Rm (MPa) (kgf$mm2) For convenience of calculation and the fact of the symmetry of
745.76 961.08 22.28 70.55 302.50 the WOL sample, a 1/2 3D elastic-plastic model of WOL sample
was built for the FE analysis. The FE meshing of the area
around the crack tip was refined for ensuring the calculation
existed in the structure, prefabricated crack was carried out
accuracy in the vicinity of crack tip, as shown in Fig. 3. The FE
on the sample with a MTS810 hydraulic servo fatigue test
model contained 80953 nodes in total with 72386 associated
machine. The maximum fatigue load value Pmax was 11.87 kN
elements. The same nodes and units were used in the calcu-
with a frequency of 18 Hz, stress ratio was 0.1. The initial ratio
lation of the stress field and diffusion hydrogen distribution.
of pre-crack length to width of sample was about 0.5.
The element used in the stress model was C3D8R, an 8-node
Different preloadings were applied on the samples to
linear brick reduced integration element. The element used
measure the KISCC. Since the solution concentration may
in the hydrogen diffusion model was DC3D8, an 8-node linear
change during the long time test, time was recorded as the
heat transfer brick element.
specimen was exposed to the medium. The solution was
changed by an increasing period during the whole test. The
Boundary conditions and assumptions
surface crack length was measured and recorded with a
reading microscope until the crack propagation rate was
In the calculation of stress field, an elastoplastic constitutive
basically maintained at 107 mm/s. Samples finished tests
relation of FV520B steel was used. Boundary conditions were
were pulled off machinery and then the corrosion products
set as same as the actual open displacement in the constant
were removed from the sample. The crack lengths of a1, a2
displacement test. In the hydrogen diffusion concentration
and a3 at the three locations of 25%, 50% and 75% of the
calculation, temperature was set as ambient temperature
thickness of the specimen were measured by a reading mi-
293 K. Hydrogen permeating into material is a complicated
croscope to calculate the stress intensity factor KI as the KISCC
process. The model is formulated with the following as-
of FV520B steel in the relative corrosion medium. The fracture
sumptions for simple calculation and qualitative analysis:
surface and side surface around the tip of fatigue pre-crack,

Fig. 3 e FE mesh of WOL specimen.

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
4 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9

(1) The influence of defects, such as pores and slags, on the  

vf vs
diffusion coefficient was ignored. The diffusion coeffi- J ¼ sD$ þ ks (3)
vx vx
cient of hydrogen in the material was isotropous and its
value is 7.034  106 mm2 s1 [24]. The parameters of stress field mentioned above were the
(2) During the calculation of hydrogen diffusion, concen- primary inputs for the hydrogen diffusion analysis.
tration of internal hydrogen of FV520B steel was set as
1.1  106 [25].
(3) External hydrogen was mainly from the proton reduction Results and discussion
reaction on the surface of specimen. Concentration of
atomic hydrogen produced by the hydrogen sulfide Stress corrosion critical stress intensity factor KISCC
reduction reaction was dependent on the concentration
of hydrogen sulfide (partial pressure of hydrogen sulfide), During the propagation of crack, stress intensity factor KI
pH, and compactness of the corrosion products on the decreased with the increasing of length of crack. When there
surface and so on. For convenience of calculating, the was no more crack propagation or the crack propagation rate
concentration of atomic hydrogen was set as a constant. reached the critical rate, the final KI was clearly approaching
to the KISCC. Based on the real-time experimental data, da/dt-
KI curves of FV520B immersed in different concentration of
Basic theory of stress induced hydrogen diffusion H2S þ Cl can be used to indicate the crack propagating
behavior, as shown in Fig. 4. It can be seen that the KISCC of
Based on the mass transfer theory, the diffusion of hydrogen FV520B steel immersed in 5% NaCl containing saturated H2S,
in an inhomogeneous medium is given by constitutive equa- and 5% NaCl containing 500 ppm H2S mediums were
tion [10,11]. 43.8 MPa m1/2 and 60.8 MPa m1/2, respectively. These values
  are considerably lower than that of KIC measured in air [27].
vf v vs Furthermore, the KISCC of FV520B steel decreased with the
J ¼ sD$ þ kS ðln TÞ þ ks (1)
vx vx vx increasing of the H2S concentration.
where J is diffusion flux, s is the solubility of hydrogen in steel,
which is 0.631  107 according to the Hirth solubility relation, Fracture morphology analysis
D is the coefficient of hydrogen diffusion, f is the activity of
hydrogen, ks is the coefficient of “Soret effect” related to tem- Fracture pattern of FV520B steel immersed in aqueous H2S
perature gradient, T is temperature, ks is the equivalent stress solution containing chloride was shown in Fig. 5. It can be
gradient coefficient. s is the hydrostatic stress, f is a function seen that the area with stress corrosion crack propagation was
of hydrogen concentration C and hydrogen solubility s which relatively smooth, and the surface was covered with many
can be represented as f ¼ C=s. Equivalent stress gradient co- corrosion products. The sign of crack branching was also
efficient ks can be calculated through the following equation found on the initial zone for crack propagation. It was showed
that the microscopic fracture morphology was the trans-
VH $f granular cracking, which had the fracture morphology of river
ks ¼ (2)
RT shape and sector. It can be concluded that the microscopic
where VH is the partial mole volume of hydrogen in steel, fracture morphology has the typical characteristics of sulfide
which is 2.0  103 mm3 mol1 [26], R is gas constant. In the stress corrosion fracture.
constant displacement SCC test, the temperature was kept During the SCC test, corrosion pits formed on the side
constant, therefore vxv ðln TÞ ¼ 0, so equation (1) can be surface of the specimen, and the large amount of intermittent
simplified as following concentric arc-cracks on the side surface around the tip of

0.1
0.1

0.01
1E-3

1E-3
1E-5
da/dt(mm/h)
da/dt(mm/h)

5% NaCl+H2S sat. 1E-4

1E-7 5% NaCl + 500ppm H2S

1E-9 1E-5

1E-11 1E-6

1E-13 1E-7
54 55 56 57 58 59 60 60 61 62 63 64 65
1/2 1/2
KI(MPam ) KI(MPam )

(a) 5% NaCl + saturated H2S (b) 5% NaCl + H2S 500 ppm

Fig. 4 e da/dt-KI curve of FV520B immersed in different concentrations of H2S þ Cl¡.

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9 5

image in Fig. 6(b). The enlarged images of side surface crack

also revealed that the crack was deep.
The WOL specimen was opened along the plane perpen-
dicular to the side surface crack. Fig. 7(a) displayed the prop-
agation of crack on the side surface around the tip of fatigue
pre-crack. It can be seen that the side surface crack was
perpendicular to the side surface of the specimen. The side
surface crack propagated into matrix transgranularly along
with secondary cracks, as shown in Fig. 7(b). The inside
morphology of side surface crack was displayed in Fig. 8(a). It
was covered with many corrosion products, which was
analyzed by EDS, as shown in Fig. 8(b). The chemical compo-
sition (wt.%) of the corrosion product was listed in Table 3.
Comparing to the composition of the original sample, the
Fig. 5 e Fracture pattern of FV520B immersed in aqueous portion of S and O were much higher than the original sample.
H2S solution containing chloride for SCC test. From the perspective of experimental environment, it is
concluded that the main forms of corrosion products are
Fe2O3, Fe3S4, FeS2 and FeS, the formation of corrosion products
are mainly related to the concentration of H2S [21,28]. These
crack were generated eventually, as shown in Fig. 6(a). Fatigue side surface cracks had characteristics of the SSCC.
pre-cracks of two samples immersed in the 5%
NaCl þ 500 ppm H2S solution were not propagated because FE analysis results
their initial stress intensity factor KI0 were as low as
54.4 MPa m1/2 and 56.8 MPa m1/2. The similar intermittent According to the SCC experimental results, a 1/2 3D elastic-
concentric arc-cracks were also found around the tip of crack plastic model based on the SCC test in the aqueous sodium
on these samples, as shown in Fig. 6(b). With increasing of chloride containing H2S with initial load of KI0 as
immersing time, the concentric arc-crack expanded into the 59.54 MPa m1/2 was built for the FE analysis. In order to
specimen. The concentric arc-crack gradually stopped analyze the results, two reference paths named P1 and P2
spreading in the late phase of SCC test. Secondary crack was were defined and shown in Fig. 9. P1 was along the extension
observed on the side surface crack as shown in the enlarged

Fig. 6 e Cracking morphology on the side surface of FV520B exposed to aqueous H2S þ Cl¡ þ H2O environment.

Fig. 7 e Propagation of side surface crack on the side surface around the tip of fatigue precrack.

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
6 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9

Fig. 8 e SEM images and corresponding EDS analysis of crack on the side surface.

cord of fatigue pre-crack on the side surface of specimen. P2

Table 3 e Chemical composition of the crack on the side
was along the extension cord of fatigue pre-crack in the
surface (wt%).
middle plane of the specimen.
O S Cr Fe Ni Cu
Fig. 10 shows the hydrostatic stress distribution along the
Corrosion product 20.69 7.40 18.37 39.64 10.11 3.78 reference path 1 and path 2. It can be seen that the hydrostatic
stress distribution along path 1 was similar with that along
path 2 while the hydrostatic stress along path 1 was larger
than that along path 2. There was a maximum hydrostatic
stress on the tip of crack. The hydrostatic stress decreased
sharply when the distance was far away from the crack tip.
Similarly, the distribution of the diffused hydrogen along path
1 and path 2 were displayed in Fig. 11(a) and (b), respectively.
Due to the influence of stress gradient, diffused hydrogen
concentrated at the high hydrostatic stress zone which was
around the tip of crack. The maximum hydrogen concentra-
tion formed at the crack tip on the middle plane which was
about 215 ppm. The diffusion of hydrogen rapidly decreased
with the increased distance from the crack tip, which had a
similar varying trend to the hydrostatic stress.
When the WOL specimens were immersed in the aqueous
sodium chloride containing H2S, corrosion pits formed on the
side surfaces. According to the corrosion pits and crack
formed on the side surfaces of specimen in the SCC test, a
hemispherical corrosion pit with radium of 0.05 mm was set at
3 mm away to the extension cord of fatigue pre-crack on the
side surface of specimen. Under the same loading and
Fig. 9 e Location of reference paths. boundary condition, stress field and distribution of diffused
hydrogen in the pit model were calculated. The corresponding

Fig. 10 e Distribution of hydrostatic stresses along path 1 and path 2.

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9 7

Fig. 11 e Distribution of diffusion for hydrogen concentration along path 1 and path 2.

hydrostatic stress and diffused hydrogen concentration cal- concentration of absorbed hydrogen and hydrogen concen-
culations were displayed in Figs. 10 and 11, respectively. The tration gradient leaded to a high penetration kinetics which
results indicated that hemispherical corrosion pit with promoted the hydrogen permeating into material. On the
radium of 0.05 mm had a little influence on the overall hy- other hand, the FEA results demonstrated that hydrostatic
drostatic stress distribution. Hydrostatic stress around the stress around the corrosion pit caused by anodic dissolution
corrosion pit was obviously different from that with no was larger than that without corrosion pit which leaded to
corrosion pit, as shown in Fig. 10(b). Hydrostatic stress on the atomic hydrogen absorbed on the surface to concentrate in
bottom of corrosion pit was about 415 MPa which was almost the high stress areas. Simultaneously, there existed relatively
60% higher than that without corrosion pit (about 255 MPa). In high stress gradient which accelerated the absorbed atomic
addition, there existed a relative large pressure gradient hydrogen to diffuse into steel furtherly. Atomic hydrogen
around the corrosion pit. Similarly, diffused hydrogen gath- penetrated into steel can precipitate as molecular hydrogen
ered on the bottom of corrosion pit. The maximum diffused and concentrate in the matrix-inclusion interfaces. There was
hydrogen concentration reached 109 ppm on the bottom of a positive correlation between hydrogen pressure and
corrosion pit which was 16% higher than that with no corro- hydrogen concentration, the higher the hydrogen concentra-
sion pit (about 94 ppm). tion, the higher the hydrogen pressure [34]. Local plastic
deformation was accelerated under the coefficient action be-
tween hydrogen pressure and applied load which increased
Discussion the crack propagation power. Then the plasticity of the crack
tip will decrease [35,36] which decreased the crack propaga-
Combined with morphology observation of side surface crack tion resistance.
and FE analyzing about stress and hydrogen distribution, the Corrosion pits generated around the tip of fatigue pre-
side surface crack forming mechanism was analyzed below. crack on the side surface developed into intermittent arc-
During the SSCC test of FV520B steel, nonuniform stress field crack under the combined stress and hydrogen action and
formed on the side surface of specimen. Anions in the me- propagated along the circumferential directions centered on
dium like Cl, S2 and HS absorbed on the side surface leaded pre-crack tip. The FEA results illuminate that stress on the
to the deference of electrochemical behavior. Then local middle plane of sample was higher than that on the side
anodic dissolution occurred which accelerated the breaking of surface because of the plane stress state of side surface. When
surface passivation film [29e31] and resulted in the formation corrosion pits formed on the side surface, stress states around
of corrosion pits. Local electrochemical reaction was acceler- corrosion pits changed. Another reference path named P3 (as
ated by the large cathode and small anode couple which shown in Fig. 9) was chosen to analyze the effect of corrosion
promoted the dissolution of corrosion pits. Atomic hydrogen pit on stress distribution along the thickness direction. Fig. 12
produced on the surface of specimen by proton reduction re- displayed the hydrostatic stress distribution along P3. It can be
action was suppressed to precipitate as molecular hydrogen seen that hydrostatic stress on the bottom of corrosion pit was
by H2S in the medium [32,33]. much higher than that without corrosion pit. With distance
On one hand, the internal hydrogen concentration was far away from side surface, hydrostatic stress decreased
relatively low [25] as compared to those absorbed on the sur- rapidly to a minimum value and then increased again which
face. A large hydrogen concentration gradient formed along caused a very large stress gradient on the bottom of the
the thick of specimen which resulted in permeation of atomic corrosion pit. As shown in Fig. 12, the influence of the corro-
hydrogen from surface to inner of specimen according to sion pit on hydrostatic stress distribution was limited in su-
equation (1). Mass transfer process in the electrochemical perficial zone near the side surface. The thickness of the
reaction was fast due to the immediate contacting between affected zone calculated in this condition was about 0.3 mm.
the side surface of specimen and corrosion medium. Rapid Figs. 7 and 13 displayed the longitudinal section morphologies
mass transfer process resulted to a large amount of atomic of side surface cracks. The observation indicated that the
hydrogen absorbed on the side surface of the specimen. High depths of side surface cracks were range from 0 to 0.3 mm

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
8 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9

Corrosion pits resulted from anodic dissolution pro-

moted the formation of side surface crack.
(3) FE models with and without corrosion pit were estab-
lished to investigate the effect of corrosion pit on the
distribution of stress and diffused hydrogen. Corrosion
pit can lead to large stress gradients, and affect the
stress and diffused hydrogen distribution around the
corrosion pit. Diffused hydrogen gathered on skin layer
of side surface of the specimen. The depth of side sur-
face crack was limited in thickness in hydrogen diffu-
sion zone which was consistent with the experimental
results. In addition, side surface crack propagated along
the circumferential direction was more than thickness
direction.

Fig. 12 e Distribution of hydrostatic stresses along path 3.

Acknowledgement

This work has been supported by National Basic Research

Program of China (No. 2012CB026003). Thanks to Dr. LIU,
University of British Columbia, for linguistic advice.

references

[1] Tsay LW, Chen YC, Chan SLI. Sulfide stress corrosion
cracking and fatigue crack growth of welded TMCP API 5L
X65 pipe-line steel. Int J Fatig 2001;23(2):103e13.
[2] Beidokhti B, Dolati A, Koukabi AH. Effects of alloying
elements and microstructure on the susceptibility of the
welded HSLA steel to hydrogen-induced cracking and sulfide
stress cracking. Mater Sci Eng A 2009;507(1e2):167e73.
[3] Ren CQ, Liu DX, Bai ZQ, Li TH. Corrosion behavior of oil tube
Fig. 13 e SEM images of the crack on the side surface
steel in simulant solution with hydrogen sulfide and carbon
around the tip of fatigue pre-crack. dioxide. Mater Chem Phys 2005;93(2e3):305e9.
[4] Chen YY, Liou YM, Shih HC. Stress corrosion cracking of type
321 stainless steels in simulated petrochemical process
which was in good agreement with the thickness of affected environments containing hydrogen sulfide and chloride.
zone. Mater Sci Eng A 2005;407(1e2):114e26.
[5] Chen TC, Chen ST, Tsay LW. The role of induced alpha-
martensite on the hydrogen-assisted fatigue crack growth of
Conclusions austenitic stainless steels. Int J Hydrogen Energy
2014;39(19):10293e302.
[6] Eadie RL, Szklarz KE, Sutherby RL. Corrosion fatigue and
(1) KISCC of FV520B steel exposed to 5% NaCl containing near-neutral pH stress corrosion cracking of pipeline steel
saturated H2S and 500 ppm H2S were 43.8 MPa m1/2 and and the effect of hydrogen sulfide. Corrosion
60.8 MPa m1/2, respectively, which dramatically reduced 2005;2(61):167e73.
compared to the fracture toughness of FV520B steel in [7] Vigdis O, Antonio A, Odd MA. Hydrogen diffusion and
air. Fracture surface observation indicated that the hydrogen influenced critical stress intensity in an API X70
pipeline steel welded joint-experiments and FE simulations.
microscopic fracture morphology exhibited the typical
Int J Hydrogen Energy 2012;37(15):11474e86.
characteristics of sulfide stress corrosion fracture. The
[8] Antonio A, Vigdis O, Odd MA. 3D cohesive modelling of
SCC mechanism of FV520B steel was mainly hydrogen hydrogen embrittlement in the heat affected zone of an X70
embrittlement and anodic dissolution in acidic pipeline steel. Int J Hydrogen Energy 2013;38(18):7539e49.
hydrogen sulfide solutions. [9] Antonio A, Vigdis O, Odd MA. 3D cohesive modelling of
(2) During the SCC tests, large amount of intermittent arc- hydrogen embrittlement in the heat affected zone of an X70
crack formed on the side surfaces around the tip of fa- pipeline steel e Part II. Int J Hydrogen Energy
2014;39(7):3528e41.
tigue pre-crack resulted from the synergy effect of
[10] Wenchun Jiang, Jianming Gong, Jianqun Tang, Hu Chen.
anodic dissolution and stress induced hydrogen Numerical simulation of hydrogen diffusion under welding
cracking. Secondary crack was observed on the side residual stress. Trans China Weld Inst 2006;27(11):57e60. 64.
surface crack. The side surface cracking mechanism [11] Jianming Gong, Wenchun Jiang, Jianqun Tang, Hu Chen,
was hydrogen induced sulfide stress corrosion cracking. Shandong Tu. 3D finite element simulation of hydrogen

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e9 9

diffusion in 16MnR steel weldment. J Mech Eng China. GB/T 15970.6e2007 corrosion of metals and alloys-
2007;43(9):113e8. stress corrosion testing Part 6: preparation and use of pre-
[12] Ilin DN, Saintier N, Olive JM, Abgrall R, Aubert I. Simulation cracked specimens for tests under constant load or constant
of hydrogen diffusion affected by stress-strain heterogeneity displacement. Beijing: Standards Press of China; 2007.
in polycrystalline stainless steel. Int J Hydrogen Energy [24] Fujii T, Hazama T, Nakajima H, Horita R. A safety analysis on
2014;39(5):2418e22. overlay disbonding of pressure vessels for hydrogen service. J
[13] Vergani L, Gobbi G, Colombo C. A numerical model to Am Soc Metals 1982:361e8.
evaluate the effect of hydrogen embrittlement on low-alloy [25] Qianqing Zhou, Xingping Yong, Yuchun Zhai. Effect of aging
steels. CONVEGNO IGF XXII ROMA 2013:153e61. Italia, 1-3 treatment on hydrogen embrittlement of FV520B maraging
Luglio 2013. steel. Corros Sci Protect Technol 2008;20(6):416e9.
[14] Abderrazak T, Marco A, Gilles L, Sebastien D, [26] Serebrinsky S, Carter E, Ortiz M. A quantum-mechanically
Abdelmounam S. An effective finite element model for the informed continuum model of hydrogen embrittlement [J]. J
prediction of hydrogen induced cracking in steel pipelines. Mech Phys Solid 2004;52(10):2403e30.
Int J Hydrogen Energy 2012;37(21):16214e30. [27] Qiaoguo Wu, Xuedong Chen, Zhichao Fan, Defu Nie,
[15] Guangcun Wang, Jianfeng LI, Xiujie JIA, Fangyi Li, Ziwu Liu. Jianhua Pan. Engineering fracture assessment of FV520B
Study on erosion behavior and mechanism of impeller's steel impeller subjected to dynamic loading. Eng Fract Mech
material FV520B in centrifugal compressor. J Mech Eng 2015;146:210e23.
2014;50(19):182e90. [28] SUN J. Study on stress corrosion cracking behavior and
[16] Ziwu Liu, Jianfeng Li, Xiujie JIA, Guangcun Wang, mechanism of impeller in centrifugal compressor. Jinan:
Wenhan Xu. Establishment and analysis of erosion depth Shandong University; 2016.
model for impeller material FV520B. Int J Precis Eng Manuf [29] Huifang Pang. Research on the stress corrosion resistance of
Green Technol 2016;3(1):27e34. 17-4PH martensite stainless steel. Dalian: Dalian University
[17] Jinlong Wang, Yuanliang Zhang, Qingchao Sun, Shujie Liu, of Technology; 2014.
Bowen Shi, Huitian Lu. Giga-fatigue life prediction of FV520B- [30] Tsai WT, Mingshan Chen. Stress corrosion cracking behavior
I with surface roughness. Mater Des 2016;89:1028e34. of 2205 duplex stainless steel in contained NaCl solution.
[18] Jinlong Wang, Yuanliang Zhang, Shujie Liu, Qingchao Sun, Corros Sci 2000;42(3):545e59.
Huitian Lu. Competitive giga-fatigue life analysis owing to [31] Zhiyong Liu, Chaofang Dong, Xiaogang Li, Lixian Wang,
surface defect and internal inclusion for FV520B-I. Int J Fatig Qing Zhi, Ping Liang. Stress corrosion cracking behaviour of
2016;87:203e9. two stainless steels in hydrogen sulfide environment. J Univ
[19] Wu Qaioguo, Chen Xuedong, Fan Zhichao, Nie Defu, Sci Technol Beijing 2009;31(03):318e23.
Wei Renchao. Corrosion fatigue behavior of FV520B steel in [32] Wenkui Hao, Zhiyong Liu, Cuiwei Du, Xiaogang Li. Stress
water and salt-spray environments. Eng Fail Anal corrosion cracking behavior of 35CrMo steel in acidic
2017;79:422e30. hydrogen sulfide solutions. J Mech Eng 2014;50(4):39e46.
[20] Qiaoling Chu, Min Zhang, Jihong Li. Failure analysis of [33] Jiao Sun, Songying Chen, Jin Ding, Jianfeng Li. Stress
impeller made of FV520B martensitic precipitated hardening corrosion behavior of FV520B steel in H2S/CO2 environment. J
stainless steel. Eng Fail Anal 2013;34:501e10. Northeast Univ Nat Sci 2015;vol. 36(12):1790e4.
[21] Ming Qin, Jianfeng Li, Songying Chen, Yanpeng Qu. [34] Wenchun Jiang, Jianming Gong, Jianqun Tang, Hu Chen.
Experimental study on stress corrosion crack propagation Finite element simulation of hydrogen blistering of steel
rate of FV520B in carbon dioxide and hydrogen sulfide 16MnR serving in wet H2S environment. J Jilin Univ Eng
solution. Results Phys 2016;6:365e72. Technol Ed 2008;vol. 38(1):61e5.
[22] Defu Nie, Xuedong Chen, Zhichao Fan, Qiaoguo Wu. Failure [35] Wuyang Chu. Hydrogen damage and delayed fracture, vol.
analysis of a slot-welded impeller of recycle hydrogen 57. Beijing: Metallurgical Industry Press; 1988.
centrifugal compressor. Eng Fail Anal 2014;42:1e9. [36] Chatzidouros EV, Papazoglou VJ, Tsiourva TE, Pantelis DI.
[23] General Administration of Quality Supervision. Inspection Hydrogen effect on fracture toughness of pipeline steel
and Quarantine of the People's Republic of China, welds, with in situ hydrogen charging. Int J Hydrogen Energy
Standardization Administration of the People's Republic of 2011;36(9):12626e43.

Please cite this article in press as: Wei R, et al., Experimental study and numerical simulation on the SSCC in FV520B stainless steel
exposed to H2SþCl Environment, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/j.ijhydene.2018.03.119