Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments

Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments

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 7 ) 1 e1 4 Available online at www.sciencedirect.com ScienceDi...

3MB Sizes 0 Downloads 11 Views

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 7 ) 1 e1 4

Available online at www.sciencedirect.com

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

Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments Lijun Gan a, Feng Huang a,*, Xiaoyu Zhao a, J. Liu a,**, Y. Frank Cheng a,b a

State Key Laboratory of Refractories and Metallurgy, Wuhan University of Science and Technology, Wuhan, Hubei 430081, China b Department of Mechanical Engineering, University of Calgary, Calgary, Alberta T2N 1N4, Canada

article info

abstract

Article history:

Hydrogen induced cracking (HIC) susceptibility of the welded X100 pipeline steel was

Received 23 August 2017

evaluated in NACE “A” solution at room temperature according to the NACE TM0284-2011

Received in revised form

standard. Both the kinetic parameters of the permeability ðJ∞ LÞ, the apparent diffusivity

28 October 2017

(Dapp) and the concentration of reversible and irreversible hydrogen in the base metal and

Accepted 27 November 2017

welded joint of X100 pipeline steel were quantitatively investigated by hydrogen perme-

Available online xxx

ation test. The results showed that the welded joint with an inhomogeneous microstructure had a higher trap density and more susceptible to HIC due to two orders of magnitude

Keywords:

larger in the concentration of irreversible hydrogen than that of base metal, though all

X100 pipeline steel

presenting poor HIC resistance for both base metal and the welded joint. The HIC cracks

The welded joint

initiated from the inclusions enriching in Al, Ca, Si, Mn. The cracks are primarily trans-

Hydrogen induced cracking

granular, accompanying with limited intergranular ones.

Hydrogen trapping

© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Hydrogen concentration

Introduction Pipelines used to transport oil and natural gas are susceptible to hydrogen induced cracking (HIC) and sulfide stress corrosion cracking (SSCC) in wet H2S environments [1e3]. The HIC is a common type of hydrogen damage phenomenon [4], which refers to the initiation and propagation of microcracks in pipeline steels, especially those used in H2S environment, even in the absence of external stress, leading to cracking of the pipeline [5]. When pipeline steels are exposed to wet H2S environments, hydrogen atoms generated by the cathodic hydrogen evolution reaction during corrosion of the steels can penetrate into the steel matrix [6] due to the

“poisonous” effect of H2S and HS. The hydrogen atoms are usually trapped at various metallurgical defects, such as nonmetallic inclusions, dislocations, grain boundaries, etc. When the pressure buildup caused by the concentration of hydrogen between the inclusions and metal matrix reaches a critical value, the HIC cracks initiate [7e9]. The HIC susceptibility of pipeline steels with varied strengths from X52 to X100 grade has been investigated extensively in terms of the effect of metallurgical microstructure on the cracking processes. For example, Amin et al. [10] and Beidokhti et al. [11] found that the acicular ferrite (AF) in welded X65 and X70 pipeline steels possessed a higher resistance to HIC because the AF could act as effective reversible H trapping sites to reduce the susceptibility of the

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (F. Huang), [email protected] (J. Liu). https://doi.org/10.1016/j.ijhydene.2017.11.155 0360-3199/© 2017 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

steels to HIC [12]. Arafin et al. [13] thought that the bainite (B) in pipeline steels was highly susceptible to HIC. However, the homogeneously quenched and tempered B microstructure with little martensite (M) had the best performance against HIC, as confirmed by Carneiro et al. [14]. The authors also found that the granular B and martensite and austenite (M/A) constituents were more susceptible to HIC for X120 pipeline steel [15]. In addition, the role of crystallographic texture on HIC of steels cannot be ignored. Mohammad et al. [3,16] found that the high fraction of grains oriented in {001}//ND textures caused a high HIC susceptibility of API 5L X70 pipeline steel. Venegas et al. [17] suggested that strong {111}//ND fiber textures greatly increased the HIC resistance of pipeline steels, whereas {001}//ND and close-to-random textures made steels HIC-prone. The {100}//ND oriented grains of hot-rolling X60 pipeline steels were not the only paths that HIC cracks tended to propagate [18]. Obviously, the effect of structure of pipeline steels on their HIC susceptibility was complex and would be affected by different experimental conditions, environments and thermal-mechanical processing. In addition to the metallurgical features mentioned above, various non-metallic inclusions and dislocations contained in the steels act as either reversible or irreversible H traps [19e24]. Particularly, dislocations are basically categorized as reversible H traps [24], where cold rolling and annealing treatments produced a higher dislocation density in deformed and recovered regions. The increase of dislocation density caused the increased hydrogen trap density, which, combined with specific texture structure, mitigated against any possible benefits of other microstructural parameters such as coincidence site lattice boundaries and grain size. It has been accepted [4e7] that both the reversibly diffusible H and irreversibly trapped H can affect the HIC susceptibility of steels. Electrochemical cyclic H permeation tests conducted by Fallahmohammadi et al. [23] investigated the roles of reversible and irreversible H in HIC of the steel. The results were also confirmed by others' results [25,26]. Xue and Cheng [27] found that the irreversible H would increase the HIC susceptibility of X80 pipeline steel. Furthermore, a uniform distribution of irreversible H traps could disperse the internal pressure and avoid locally elevated H pressures, greatly improving the resistance of the steels to HIC [28]. However, Dadfarnia et al. [29] thought that the effect of irreversible trapping sites on hydrogen embrittlement was not ignorable. Obviously, there have been different viewpoints about the role of irreversible H on HIC of pipeline steels. In terms of the reversible H, they can result in a high HIC susceptibility of steels [30]. However, Amin et al. [10] believed that the reversible H reduced the HIC of welded X65 steel. The authors' previous work [15] also showed that the irreversible H affected remarkably the HIC susceptibility of X120 pipeline steel in saturated H2S environments. From the brief review of published work, it is seen that, although a large number of work have been performed to study the effect of reversible and irreversible H on HIC susceptibility of pipeline steels, a commonly accepted consensus has still been lacking. The X100 pipeline steel has been proposed to construct energy west-east pipeline project in China due to its high strength and good ductility [31,32]. Pipelines are welded infrastructure, and thermal cycling during welding can

generate unique microstructure at the welded joint from the steel matrix [33,34]. To date, relevant investigations on HIC of welded steels have been focused on the effect of local microstructure and non-metallic inclusions on initiation and propagation of hydrogen-induced cracks [2,6,11,35]. The essential role of reversible and irreversible H on pipeline HIC, especially X100 pipeline steel, has remained unknown. In this work, the HIC susceptibility of both base X100 pipeline steel and its welded joint was investigated in NACE A solution [36]. Surface characterization techniques, i.e., optical microscopy (OM), scanning electron microscopy (SEM) and energy-dispersive x-ray spectrum (EDS), were used to observe the microstructure of the steel and the fracture feature. The reversible and irreversible H concentrations were measured by electrochemical cyclic H permeation tests.

Experimental Materials Specimens used in this work were cut from a sheet of X100 pipeline steel, which was supplied by Wuhan Iron and Steel Corporation (WISCO). The yielding and tensile strengths of the steel were 695 MPa and 830 MPa, respectively. The welding was conducted using the submerged-arc welding (SAW) process. Table 1 shows the chemical compositions of the steel and the weld metal, as analyzed by the inductive coupled plasma (ICP) technique.

Microstructural characterization, inclusion observation and harness measurement For microstructural characterization, the welded specimen ground up to 2000 grit silicon carbide papers sequentially, polished with diamond pastes, and then etched in a Nital solution (i.e., a mixture of 4% nitric acid and ethano). The metallographic microstructure of the specimens was examined by optical microscopy (OM, Axioplan2 imaging) and field emission SEM (FE-SEM, Czechoslovakia Nova nano 400 coupled with an EDS). The morphology and size distribution of inclusions contained in the X100 pipeline base metal and its weld metal were respectively characterized by OM, and the composition of inclusions was analyzed by further EDS spectra analysis in the FE-SEM. The Vickers with a 300 g external stress was applied to measure the distribution of hardness of the welded steel specimen. Taking the center of the weld bead as start point, the indenter was moved either left or right successively with a distance interval of 1 mm to measure the hardness distribution.

Measurements of HIC susceptibility The HIC susceptibility of X100 pipeline steel base metal and the welded joint was measured according to NACE TM02842011 standard [37]. The specimens were machined into the dimension of 100 mm in length and 20 mm in width. The weld bead is at the center of the specimen. The NACE A solution, which consisted of 5.0 wt. % NaCl, 0.50 wt. % CH3COOH and

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

3

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 7 ) 1 e1 4

Table 1 e Chemical composition of X100 pipeline steel and the weld metal (wt. %). Specimens

C

Mn

Si

P

S

Cr

Cu

Ni

Mo

Fe

X100 steel Weld metal

0.05 0.07

2.00 1.58

0.28 0.17

0.019 0.019

0.010 0.017

0.28 0.001

0.56 0.05

0.55 2.17

0.35 0.31

95.42 95.56

94.5 wt. % distilled water, was used for testing at ambient temperature and pressure. The initial pH of the solution was 2.7±0.1. After 1 h of deaerating by nitrogen (N2, 99.9%), hydrogen sulfide (H2S) gas was bubbled into the solution at a flow rate of 50 cc/min for 30 min prior to the test. The total time for HIC test was 96 h, and the H2S gas was purged into the solution to maintain a saturated concentration. After testing, the specimens were taken out of the solution and inspected for internal cracks. Three parameters were used for HIC evaluation by Ref. [37]: P Crack length ratio: CLR ¼

a  100% W P

Crack thickness ratio: CTR ¼

b

T

 100%

(1)

(2)

P Crack sensitivity ratio: CSR ¼

ða  bÞ  100% WT

(3)

where a is the crack length (mm), b is the crack thickness (mm), W is the specimen width (mm), and T is the specimen thickness (mm). The hydrogen-induced cracks, once initiated, were observed using OM and the FE-SEM.

Hydrogen permeation testing The H permeation testing was conducted on the X100 base metal and its welded joint in the H2S saturated solution by a modified Devanathan-Stachurski double cells [38e40], with a 1.00 cm2 area of the specimen exposed to electrolytes in both cells. The dimension of the specimen used was 20 mm  30 mm  2.00 mm, and both sides were pre-ground and polished. The deaerated 300 mL 0.1 M NaOH solution and NACE A solution were added in the H detection cell and the H charging cell, respectively. It was noted that the corrosion of X100 pipeline steel in NACE A solution purged with H2S did not affect the steady-state H permeation current density (I∞) and H permeation curve, as demonstrated by multiple parallel tests. The specimen at the H detection side was maintained at a potential of 300 mV more positive than the open-circuit potential (OCP) to fully oxidize the H diffusing through the steel membrane from the H charging side. In the H charging cell [15], H2S gas was continuously purged into the NACE A solution. Prior to H2S purging, the background current density of the H detected side was lower than 2  107 A/cm2 [23]. Once the H permeation current density was in a steady state for 1 h, the test was stopped. The H permeation parameters, including permeability (J∞L), apparent H diffusivity (Dapp), and apparent H solubility (Capp), were calculated by Refs. [15,36]: J∞ L ¼

I∞ L FA

(4)

Dapp ¼

L2 6tL

(5)

Capp ¼

J∞ L Dapp

(6)

where I∞ was the steady-state H permeation current density, L was the thickness of the specimens, and tL was the relaxation time, corresponding to the time when the H permeation current density was equal to 0.63 times of I∞ [36]. All tests were performed at room temperature. In order to ensure the repeatability of the experimental data, each test was repeated at least three times. In some cases, five measurements were made to convince the reliability of the results.

Electrochemical cyclic hydrogen permeation testing The electrochemical cyclic H permeation testing was conducted to obtain H diffusion coefficient and the reversible/ irreversible H concentrations. The experimental setup was identical to that in H permeation testing in the saturated H2S solution. The only difference was that 300 mL of solution of 0.5 M H2SO4 and 3.1  103 M Na4P2O7 was added in the H charging cell, as shown in Fig. 1, where the Na4P2O7 was used a poison since it was non-toxic, and had a good effect on prevention of the combination of H atoms to form molecular H on the specimen surface [41]. A cathodic current density of 10 mA/cm2 was applied by the constant potential current meter at the H entry side for hydrogen evolution. The anodic current density vs. time, i.e., i-t, curve measured at the H detection side contained two phases, i.e., the H permeation phase and the H permeation current decay phase. By fitting the normalized H permeation current curves for each phase, the lattice H diffusion coefficient (DL) and effective H diffusion coefficient (Deff) can be determined according to Eqs. (7) and (8) [23,26,42,43]: it  i0 The H permeation phase: i∞  i0 # " ∞ ð2n þ 1Þ2 L2 2L X ¼ pffiffiffiffiffiffiffiffiffi exp  4Dt pDt n¼0 it  i∞ The H current decay phase: i0  i∞ " # ∞ 2L X ð2n þ 1Þ2 L2 ¼ 1  pffiffiffiffiffiffiffiffiffi exp  4Dt pDt n¼0

(7)

(8)

where D was the H diffusion coefficient, it was the measured H permeation current density at time of t, i0 was the initial H permeation current density, i∞ was the steady-state H permeation current density for each phase, and the number n

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

Fig. 1 e Schematic diagram of the experimental device for electrochemical hydrogen permeation testing, where WE, RE and CE refer to work electrode, reference electrode, and counter electrode, respectively.

of 0, 1, 2, 3 could be generally taken for an accurate curve fitting [23,26]. For the first-cycle of electrochemical H permeation test, the i0 was the steady-state background current density, which was close to 0. It is noted that Eq. (7) represents the rising part of the permeating current transient, including hydrogen in irreversible and reversible traps and in lattice, while Eq. (8) represents the current decay part, including hydrogen in reversible traps and in lattice. H entered into a metal matrix diffuses manly in the lattice with the tetrahedral interstitial sites at ambient temperature. In the presence of trapping sites, H atoms exist in three forms, i.e., H in irreversible trapping sites (Cir), H in reversible trapping sites (Cr) and H in the lattice interstitial sites (Cl). In this work, the reversible H concentration (CR) included both Cr and Cl, and the irreversible H concentration (CIR) is equal to Cir. The irreversible and reversible H concentrations can be obtained by fitting of the hydrogen permeation curves. First, the effective diffusion coefficient (Deff) is obtained by fitting the first-cycle rising part of the current transient. The Deff is then used in Eq. (8) to obtain the effective H decay curve. Finally, the total H concentration (CT) in the specimen can be calculated by integrating the area beneath the effective H decay curve. The reversible H concentration (CR) is the area under the first-cycle decay of the current transient curve, and the irreversible H concentration (CIR) is the difference between CT and CR. Generally, the H trapping sites were assumed uniformly in the steel. The diffusion of H atoms is identical in each direction. Compared to the H permeating phase, the H permeation decay curve reflected a changing concentration of H permeating into the specimen. Thus, the average concentration of H (CH ) in each form can be obtained from the electrochemical H decay curves. The reversible and irreversible H concentrations can be calculated by integrating the area beneath the decay curve according to Eq. (9) [44,45]:

0 1B C¼ @ L

Ztf 0

1

0

1 B C JL ðtÞdtA ¼ @ FL

Ztf

1 C IL ðtÞdtA

(9)

0

where C (mol/m3) was the average H concentration in the steel, and JL ðtÞ and IL ðtÞ were the H diffusion flux and the H permeation current density at the detection side of the specimen at time t, respectively. The electrochemical cyclic H permeation tests were performed successively for three cycles at room temperature. In order to ensure the reliability of the testing results, each test was repeated at least three to five times.

Results and discussion Microstructure and hardness of the welded X100 pipeline steel The macroscopic view of the welded X100 pipeline steel specimen is shown in Fig. 2, where weld metal (WM), fusion line, heat affected zone (HAZ) and base metal (BM) are marked. The welded joint specimen used for HIC susceptibility and electrochemical H permeation tests including the regions of BM, HAZ and WM is marked with a red rectangle. The metallographic microstructure of different zones of the welded joint is shown in Fig. 3, where the HAZ is divided into intercritical heat affected zone (ICHAZ), fine grain heat affected zone (FGHAZ) and coarse grain heat affected zone (CGHAZ). The fusion zone (FZ) is between the HAZ and WM. The BM (Fig. 3a) mainly consists of quasi-polygonal ferrite (F) with a few lath bainite (B) distributing inside the F grain. Fig. 3b shows an apparent boundary between the BM and ICHAZ, and the ICHAZ is composed of granular F with varied sizes and

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

5

Fig. 2 e Photo and schematic diagram of the welded joint of X100 pipeline steel. some lath B. The black dots distributed along grain boundaries are pearlite (P). The FGHAZ (Fig. 3c) contains fine-grained F. Fig. 3d shows the boundary between the CGHAZ and the fusion zone (FZ) due to a large number of P distributing along the grain boundaries in the FZ. The CGHAZ is composed of equiaxial coarse-grained F and a few lath B. The FZ contains acicular F, and the grain boundaries are not evident. Fig. 3e shows the obvious fusion line between FZ and the WM. For the WM (Fig. 3f), fine acicular F grains with an equiaxial dispersion are observed. The distribution of Vickers hardness of various zones on the welded specimen is shown in Fig. 4. Three measurements were conducted at each zone, and the standard deviations of the data point were marked in the figure. It is seen that the Vickers hardness generally increases from the weld metal to base metal. There is the lowest hardness of about 160 HV at the weld metal, and the highest value of 260 HV in the base metal. The softening of HAZ in the welded joint of highstrength pipeline steel is a widely observed phenomenon [46]. This is associated with the presence of hardening phase of lath bainite in base metal, while the HAZ contains soft ferrite, as seen in Fig. 3. In general, the recommended hardness of pipeline steels is less than 250 HV in order to resist sulfide stress cracking, which is a type of HIC in nature [47]. However, the average Vickers hardness of the tested base metal exceeds 250 HV. It implies that the X100 pipeline steel would be susceptible to HIC, especially in H2S environments.

Inclusions contained in the welded specimen Fig. 5 shows the OM morphology of inclusions contained in the steel specimen, and the distribution of their sizes and

composition. It is noted that sixty fields are selected randomly on the specimen surface to characterize about 8000 inclusions to obtain the statistic results. From Fig. 5b, it is seen that over 50% of inclusions have a size between 1 and 2 microns, and about 90% of them are smaller than 5 microns. There is a similar size distribution of the inclusions located in the base metal and the weld metal. In addition, with further EDS analysis (Fig. 5c), the inclusions were mainly the AleCaeSieO compound ones, and some Al2O3, CaO and SiO2 inclusions either for X100 base metal or the weld metal. The results are in agreement with the findings of Cheng et al. [22,36]. It is noted that MnS inclusions, which are regarded as the most detrimental inclusions to HIC, are not detected due to the strictly controlled amount of element sulfur in the steel. Combining the results of size statistics and component analysis, it can be found that the size, type and component of the inclusions are similar in both X100 pipeline steel base metal and the weld metal. It can be implied that the effects of inclusions on HIC susceptibility should be the same for the base metal and the welded joint of X100 pipeline steel.

Testing for HIC susceptibility of the X100 pipeline steel Hydrogen bubbles are observed on the surface of the welded joint of X100 pipeline steel, as shown in Fig. 6. It shows that H atoms could enter the specimen when immersed in NACE “A” solution saturated with H2S. Table 2 shows the parameters of CLR, CTR and CSR for X100 pipeline steel base metal and welded joint specimens determined by NACE TM 0284-2011 criterion [48] for evaluation of HIC susceptibility. It is seen that, while both the base metal and welded joint do not pass the criterion for HIC resistance,

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

Fig. 3 e Metallographic microstructure of the welded joint of X100 pipeline steel (a) base metal, (b) The base metal/heat affected zone, (c) fine grain heat affected zone, (d) coarse grain heat affected zone/fusion zone, (e) fusion zone/weld metal, (f) the weld metal. F refers to ferrite, LB is lath bainite, P is pearlite and AF is acicular ferrite.

the HIC susceptibility of the welded joint is much bigger than the base metal. It is believed that the microstructural heterogeneity at the welded joint is responsible for the high susceptibility to HIC. For the X100 base metal, the presence of dual-phase microstructure of B and F is usually associated with to a high HIC susceptibility, as confirmed by Arafin et al. [13].

Initiation and propagation of hydrogen-induced cracks in base metal and the weld of X100 pipeline steel

Fig. 4 e Distribution of Vickers hardness of the welded joint of X100 pipeline steel.

Fig. 7 shows the propagation paths of hydrogen-induced cracks in X100 base metal and the weld. It is seen that the cracks is in the typical stepwise pattern in both the base metal and the weld metal, and the cracks propagates mainly in transgranular form accompanied with some intergranular cracks. For base metal, the cracks pass through the adjacent F grains, propagate along the FB microstructure, and then

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

7

Fig. 5 e The OM Morphology (a), the size distribution (b) and composition EDS analysis (c) of inclusions contained in both the base metal and the weld metal of X100 pipeline steel.

split at the interfaces between the lath B and polygonal F. For the weld metal, the cracks cross the adjacent F grains and propagate intergranularly along the AF microstructure. Fig. 8 shows the FE-SEM back scattered electron (BSE) morphology imaging of the inclusion-associated cracks and the EDS spectra analysis of the inclusions in both the X100 base metal and the weld metal, respectively. It is seen that the hydrogen-induced cracks pass through the AleCaeSieO

inclusions for both the base metal and the weld metal. The interface between the inclusions and steel matrix are usually the preferential sites to accumulate hydrogen, and initiate cracks, as confirmed previously [22,49]. It is noted that, in addition to inclusions, center segregation zones can also be an important phenomenon in steels to affect HIC behavior. The relevant work to investigate the effect of center segregation zones on HIC initiation and propagation is being conducted in authors' lab.

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

Fig. 6 e Digital photo of hydrogen bubbles on the surface of the welded joint of X100 pipeline steel immersed in standard NACE “A” solution.

Electrochemical hydrogen permeation parameters of X100 base metal and welded joint Fig. 9 shows the H permeation current curves for X100 base metal and the welded joint in NACE A solution saturated with H2S. The calculated H permeation parameters of Dapp, J∞L, and Capp are listed in Table 3. It is seen that the H permeation currents of the base metal and the welded joint achieve gradually steady-state values of 7  106 A and 8.3  106 A, respectively. There is a smaller Dapp and a larger Capp at the welded joint than the base metal. Compared with the HIC susceptibilities in Table 2, the smaller Dapp and the larger Capp of the specimen are associated with a higher HIC susceptibility, which is consistent with Park's [12] and the authors' previous work [5,15]. Furthermore, it was found [3,8e16] that, in addition to the H concentration, the H diffusion and trapping in steels are also influenced by the microstructures, such as inclusions. In this

Table 2 e HIC susceptibility parameters of base metal and the welded joint of X100 pipeline steel in NACE A solution. Specimens

X100 steel Welded joint

Average value of three parallels CSR (%)

CLR (%)

CTR (%)

1.36 2.47

11.28 48.84

5.83 10.12

work, the inclusions contained in the base steel and the welded joints are almost identical. Thus, the different HIC susceptibility of the base steel and the welded joint can be attributed to the difference of their microstructures. The X100 steel contains polygonal F and lath B, and the lath B usually severs as the sites to trap H and cause HIC [12,13]. The inhomogeneous microstructure of the welded joint, primarily at the HAZ, as shown in Fig. 3, has a large number of interfaces, i.e. grain boundaries and phase boundaries. According to literatures [50e53], the interfaces can be traps for the permeated H atoms, making H easier to be concentrated and trapped here to directly affect the diffusion of the permeated H atoms, as shown in this work. Thus, it could be inferred that the Deff of the welded joint is lower and the HIC susceptibility is higher for the welded joint.

Effects of reversible and irreversible H on HIC susceptibility of base metal and the welded joint of X100 pipeline steel Fig. 10 shows the electrochemical cyclic H permeation current curves for X100 base metal and the welded joint in NACE A solution. The present results are based on three times of parallel testing under the identical conditions. It is seen that, for the welded joint (Fig. 10b), the steady-state current (I∞) in the first-cycle of the H permeation testing process is much higher than those in the second and third cycles, which is due to the effect of irreversible trapping sites on H permeation. The difference between the second and third cycles is slight, which is attributed to the H trapping at irreversible trapping

Fig. 7 e FE-SEM images of the propagation path of HIC cracks in base metal and the weld metal of X100 pipeline steel, where F is ferrite, LB is lath bainite, and AF is acicular ferrite. Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

9

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 7 ) 1 e1 4

Fig. 8 e FE-SEM morphology of the inclusion-associated cracks and the EDS spectra of the inclusions in the (a) X100 base metal and (b) the weld metal of X100 pipeline steel.

Fig. 9 e Electrochemical hydrogen permeation curves of base metal and the welded joint of X100 pipeline steel in NACE “A” solution. sites occurring mainly in the first cycle of the test [23e26]. In other words, irreversible H trapping occurs not only in the 1st run, but also in the second and third cycles, with a reduced amount compared to that of the first cycle. For the base metal (Fig. 10a), the current increases in the first polarization cycle, but at a slow speed of 1  109 A/cm2 h. For the second and third cycles, there is no effect on the calculation of diffusion coefficient. Fig. 11 shows the normalized I-t hydrogen permeation current curves and the obtained H diffusion coefficients for X100 steel base metal and the welded joint. It is assumed that

the trapping sites filled with H atoms would not affect the H diffusivity at the very beginning part of current decay transient [35]. Thus, the obtained H diffusion coefficient by fitting 1-0.8 part of the first-cycle normalized experimental data in the current decay transient is regarded as DL, as shown in Fig. 11b and d. The DL is a constant, and depends on the material properties only. Prior to the H permeation, the specimen is under anodic oxidation to obtain a background current density less than 2  107 A/cm2, which is regarded as the current that the H in the trapping sites of the specimen itself are all almost driven out of the trapping sites. It is thus approximately assumed that the content of H in the metal matrix is negligible [23,26]. Moreover, the detected hydrogen at the detection side is a part of hydrogen permeating into the steel. The other part becomes trapped in the irreversible traps.

Table 3 e The electrochemical hydrogen permeation parameters of base metal and the welded joint of X100 pipeline steel in NACE “A” solution. Parameters

Specimens Base metal

I∞ (A) tL (s) J∞L (mol/cm s) Dapp (m2/s) Capp (mol/m3)

6

(7.0 ± 0.2)  10 1183 ± 13 (1.5 ± 0.1)  1011 (5.6 ± 0.4)  1010 2.6 ± 0.3

Welded joint (8.3 ± 0.3)  106 2150 ± 25 (1.7 ± 0.2)  1011 (3.1 ± 0.6)  1010 5.6 ± 0.4

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

10

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 7 ) 1 e1 4

Fig. 10 e Electrochemical cyclic H permeation current curves of base metal and the weld joint of X100 pipeline steel.

Fig. 11 e Normalized I-t hydrogen permeation curves and the fitted H diffusion coefficients of base metal (a, b) and the welded joint (c, d) of X100 pipeline steel.

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

Table 4 e H diffusion coefficients of base metal and the welded joint of X100 pipeline steel. Specimens X100 base metal Welded joint

Deff (m2/s)

DL (m2/s)

(1.401 ± 0.185)  1010 (2.393 ± 0.177)  1011

(2.117 ± 0.201)  1010 (2.209 ± 0.189)  1010

Thus, it is assumed that the first H transient could include the effect of irreversible traps. Therefore, the diffusion coefficient obtained by fitting the rising current part of the first cycle of the permeating curve by Eq. (7) is Deff, considering the reversible and irreversible trapping effect on permeated H atoms, as shown in Fig. 11a and c. The fitted DL and Deff values for the X100 pipeline steel base metal and its welded joint are shown in Table 4. It is seen that the Deff of the base metal is one order of magnitude larger than that of the welded joint, whereas the DL is almost identical for both the base metal and the welded joint. The results indicate that the H diffusion is faster in the base metal than in the welded joint, which is mainly due to the effect of H trapping sites.

11

Put DL and Deff into Eq. (8), the relations between the H permeation decay current and time were obtained. Thus, the corresponding lattice and effective H decay curves of base metal and the welded joint can be obtained, as shown in Fig. 12. The average Cl is calculated by integrating the area beneath the lattice H decay curve via Eq. (9), with the assumption that H atoms in lattice sites are completely released. By measuring the area beneath the effective decay current curve, the total released H concentration (CT ) in the specimen is calculated by Eq. (9). The irreversible and the reversible H stay in the specimen as long as it is not heated up between transients. It is impossible for H atoms to diffuse from irreversible trapping sites to lattice or reversible trapping sites at ambient temperature due to the high escape energy [43,54]. It is noted that the irreversible and the reversible H concentration can't be directly obtained. However, the concentrations can be calculated by fitting method. The total H concentration in the lattice and reversible/irreversible trapping sites takes an equivalent value. The CR in the specimen can be calculated by integrating the area beneath the first-cycle H decay curve according to

Fig. 12 e The fitted lattice H decay curves (a, b) and effective H decay curves (c, d) in base metal and the welded joint of X100 steel with electrochemical cyclic H permeation test. Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

12

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 7 ) 1 e1 4

Fig. 13 e Determination of the reversible and irreversible H concentrations in base metal (a) and the welded joint (b) of X100 pipeline steel.

Eq. (9). Thus, the average CIR in the specimen is the difference between CT and CR during the first-cycle H decay process, which is described by: CIR ¼ CT  CR

(10)

According to Eqs. (9) and (10), the areas highlighted in green and red are used to describe the CR and CIR, respectively, in Fig. 13. Hence, the average Cr in the steel for the first-cycle H decay process is the difference between CR and Cl : Cr ¼ CR  Cl

(11)

The calculated H concentrations in the base metal and the welded joint are listed in Table 5. The results show that the CIR in the welded joint is much larger than that in the base metal, and the CR is almost the same for the base metal and its welded joint. Thus, the concentration of H atoms in irreversible trapping sites is the essential reason to cause the different susceptibilities to HIC between the base metal and the welded joint. When the local pressure caused by H atoms in irreversible trapping sites, i.e. the interfaces between inclusion and metal matrix, exceeds the critical value, the micro-cracks can be initiated to result in HIC [1e5]. Apparently, the larger the concentration of H atoms in irreversible trapping sites, the greater the local pressure will be. Therefore, the HIC susceptibility of the welded X100 pipeline steel is greater than that of the base metal. It is noted that this work focuses on investigations of the effect of hydrogen trapping on the HIC susceptibility. Actually, the HIC susceptibility of steel depends on a synergistic effect of multiple factors. For example, hardness plays a crucial role

Table 5 e The calculated H concentrations in base metal and welded joint of X100 pipeline steel. CH (mol/m3)

Steel

Base metal Welded joint

in HIC of metals. The conclusions drawn herein are based primarily on the results obtained from this work, while the effect of other factors is not included.

Cir

Cr

Cl

CIR

CR

0.144 19.797

0.279 0.321

1.162 3.020

0.144 19.797

1.441 3.341

Conclusions Both the X100 pipeline steel base metal and the welded joint do not meet the NACE criterion of resistance to HIC. The welded joint has a higher hydrogen trapping efficiency and HIC susceptibility due to its inhomogeneous microstructure. The hydrogen-induced cracks in both base metal and the weld usually initiate from the AleCaeSieO inclusions, and propagate in a typical stepwise pattern with the transgranular mode. The low Dapp and large Capp indicate more hydrogen entrapped and a larger HIC susceptibility for X100 base metal and its welded joint in saturated H2S environments. The irreversible hydrogen concentration in the welded joint, which is almost 100 times higher than that of X100 steel base metal, is the primary factor to lead to a higher HIC susceptibility.

Acknowledgments This work was supported by the National Natural Science Foundation of China (No: 51571154 and 51201119).

references

[1] Chatzidouros EV, Papazoglou VJ, Pantelis DI. Hydrogen effect on a low carbon ferritic-bainitic pipeline steel. Int J Hydrogen Energy 2014;39:18498e505. [2] Huang F, Cheng P, Zhao XY, Liu J, Hu Q, Cheng YF. Effect of sulfide films formed on X65 steel surface on hydrogen permeation in H2S environments. Int J Hydrogen Energy 2017;42:4561e70.

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

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 7 ) 1 e1 4

 res M, Ladino DH, Abreu HFG. Role of [3] Masoumi M, Silva CC, Be crystallographic texture on the improvement of hydrogeninduced crack resistance in API 5L X70 pipeline steel. Int J Hydrogen Energy 2017;42:1318e26. [4] Garcia DCS, Carvalho RN, Lins VFC, Rezende DM, Dos Santos DS. Influence of microstructure in the hydrogen permeation in martensitic-ferritic stainless steel. Int J Hydrogen Energy 2015;40:17102e9. [5] Huang F, Liu S, Liu J, Zhang KG, Xi TH. Sulfide stress cracking resistance of the welded WDL690D HSLA steel in H2S environment. Mater Sci Eng A 2014;591:159e66. [6] Saleh AA, Hejazi D, Gazder AA, Dunne DP, Pereloma EV. Investigation of the effect of electrolytic hydrogen charging of X70 steel: II. Microstructural and crystallographic analyses of the formation of hydrogen induced cracks and blisters. Int J Hydrogen Energy 2016;41:12424e35. [7] Dong CF, Liu ZY, Li XG, Cheng YF. Effects of hydrogencharging on the susceptibility of X100 pipeline steel to hydrogen-induced cracking. Int J Hydrogen Energy 2009;34:9879e84. [8] Mohtadi-Bonab MA, Eskandari M. A focus on different factors affecting hydrogen induced cracking in oil and natural gas pipeline steel. Eng Fail Anal 2017;79:351e60. [9] Mohtadi-Bonab MA, Eskandari M, Karimdadashi R, Szpunar JA. Effect of different microstructural parameters on hydrogen induced cracking in an API X70 pipeline steel. Met Mater Int 2017;23:726e35. [10] Amin LV, Miresmaeili R, Abdollah-Zadeh A. The mutual effects of hydrogen and microstructure on hardness and impact energy of SMA welds in X65 steel. Mater Sci Eng A 2017;679:87e94. [11] 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;507:167e73. [12] Park GT, Koh SU, Jung HG. Effect of microstructure on the hydrogen trapping efficiency and hydrogen induced cracking of linepipe steel. Corros Sci 2008;50:1865e71. [13] Arafin MA, Szpunar JA. Effect of bainitic microstructure on the susceptibility of pipeline steels to hydrogen induced cracking. Mater Sci Eng A 2011;528:4927e40. [14] Carneiro RA, Ratnapuli RC, Lins VFC. The influence of chemical composition and microstructure of API linepipe steels on hydrogen induced cracking and sulfide stress corrosion cracking. Mater Sci Eng A 2003;357:104e10. [15] Huang F, Liu J, Deng ZJ, Cheng JH, Lu ZH, Li XG. Effect of microstructure and inclusions on hydrogen induced cracking susceptibility and hydrogen trapping efficiency of X120 pipeline steel. Mater Sci Eng A 2010;527:6997e7001. [16] Mohammad M, Cleiton CS, Hamilton FGA. Effect of crystallographic orientations on the hydrogen-induced cracking resistance improvement of API 5L X70 pipeline steel under various thermomechanical processing. Corros Sci 2016;111:121e31.  ndez JH, [17] Venegas V, Caleyo F, Baudin T, Espina-Herna Hallen JM. On the role of crystallographic texture in mitigating hydrogen induced cracking in pipeline steels. Corros Sci 2011;53:4204e12. [18] Mohtadi-Bonab MA, Eskandari M, Szpunar JA. Texture, local misorientation, grain boundary and recrystallization fraction in pipeline steels related to hydrogen induced cracking. Mater Sci Eng A 2015;620:97e106. [19] Hejazi D, Haq AJ, Yazdipour N, Dunne DP, Pereloma EV. Effect of manganese content and microstructure on the susceptibility of X70 pipeline steel to hydrogen cracking. Mater Sci Eng A 2012;551:140e9.

13

[20] Beidokhti B, Koukabi AH, Dolati A. Influences of titanium and manganese on high strength low alloy SAW weld metal properties. Mater Charact 2009;60:225e33. [21] Zhao MC, Yang K, Shan Y. The effects of thermo-mechanical control process on microstructures and mechanical properties of a commercial pipeline steel. Mater Sci Eng A 2002;335:14e20. [22] Jin TY, Liu ZY, Cheng YF. Effect of non-metallic inclusions on hydrogen-induced cracking of API 5L X100 steel. Int J Hydrogen Energy 2010;35:8014e21. [23] Fallahmohammadi E, Bolzoni F, Lazzari L. Measurement of lattice and apparent diffusion coefficient of hydrogen in X65 and F22 pipeline steels. Int J Hydrogen Energy 2013;38:2531e43. [24] Mohtadi-Bonab MA, Eskandari M, Szpunar JA. Effect of arisen dislocation density and texture components during cold rolling and annealing treatments on hydrogen induced cracking susceptibility in pipeline steel. J Mater Res 2016;31:3390e400. [25] Chen YX, Chang QG. Effect of traps on diffusivity of hydrogen in 20g clean steel. Acta Metall Sin 2011;47:548e52. [26] Liu Q, Atrens A. Reversible hydrogen trapping in a 3.5 NiCrMoV medium strength steel. Corros Sci 2015;96:112e20. [27] Xue HB, Cheng YF. Characterization of inclusions of X80 pipeline steel and its correlation with hydrogen-induced cracking. Corros Sci 2011;53:1201e8. [28] Pressouyre GM, Bernstein IM. A kinetic trapping model for hydrogen-induced cracking. Acta Metall 1979;27:89e100. [29] Dadfarnia M, Sofronis P, Neeraj T. Hydrogen interaction with multiple traps: can it be used to mitigate embrittlement? Int J Hydrogen Energy 2011;36:10141e8. [30] Kim WK, Koh SU, Yang BY, Kim KY. Effect of environmental and metallurgical factors on hydrogen induced cracking of HSLA steels. Corros Sci 2008;50:3336e42. [31] Zhao WG, Chen M, Chen SH, Qu JB. Static strain aging behavior of an X100 pipeline steel. Mater Sci Eng A 2012;550:418e22. [32] Al-Mansour M, Alfantazi AM, El-boujdaini M. Sulfide stress cracking resistance of API- X100 high strength low alloy steel. Mater Des 2009;30:4088e94. [33] Sun XJ, Yuan SF, Xie ZJ, Dong LL, Shang CJ, Misra RDK. Microstructure-property relationship in a high strength-high toughness combination ultra-heavy gauge offshore plate steel: the significance of multiphase microstructure. Mater Sci Eng A 2017;689:212e9. [34] Eliyan FF, Alfantazi A. Corrosion of the Heat-Affected Zones (HAZs) of API-X100 pipeline steel in dilute bicarbonate solutions at 90 C e an electrochemical evaluation. Corros Sci 2013;74:297e307. [35] Lan LY, Kong XW, Qiu CL, Zhao DW. Influence of microstructural aspects on impact toughness of multi-pass submerged arc welded HSLA steel joints. Mater Des 2016;90:488e98. [36] Huang F, Li XG, Liu J, Qu YM, Jia J, Du CW. Hydrogen-induced cracking susceptibility and hydrogen trapping efficiency of different microstructure X80 pipeline steel. J Mater Sci 2011;46:715e22. [37] NACE TM 0284-2011. Evaluation of pipeline and pressure vessel steels for resistance to hydrogen induced cracking. 2011. [38] Devanathan MAV, Stachurski Z. The adsorption and diffusion of electrolytic hydrogen in Palladium. Proc Roy Soc A 1962;270:90. [39] Zakroczymski T. Electrochemical determination of hydrogen in metals. J Electroanal Chem 1999;475:82e8. [40] Zakroczymski T. Adaptation of the electrochemical permeation technique for studying entry, transport and

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155

14

[41]

[42]

[43] [44]

[45]

[46]

[47]

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 7 ) 1 e1 4

trapping of hydrogen in metals. Electrochim Acta 2006;51:2261e6. Yuan W, Huang F, Liu J, Dong YY. Study on hydrogen induced cracking of X120 pipeline steel by electrochemical methods. Chin J Appl Chem 2012;29:1065e9. ISO 17081:2004(E). Method of measurement of hydrogen permeation and determination of hydrogen uptake and transport in metals by an electrochemical technique. Switzerland: ISO; 2004. McNabb A, Foster PK. A new analysis of the diffusion of hydrogen in iron. J Trans Metall Soc AIME 1963;227:618e27. Olden V, Alvaro A, Akselsen OM. Hydrogen diffusion and hydrogen influenced critical stress intensity in an API X70 pipeline steel welded joint e experiments and FE simulations. Int J Hydrogen Energy 2012;37:11474e86. Zhang BY, Diao FQ, Tang XY. Effect of critical hydrogen concentration and stress on hydrogen-induced cracking in plain steel 20. Acta Metall Sin 1990;26:68e74. Zhang C, Cheng YF. Corrosion of welded X100 pipeline steel in a near-neutral pH solution. J Mater Eng Perf 2010;19:834e40. European Federation of Corrosion. Guidelines on material requirement for carbon and low alloy steels for H2S containing environments in oil and gas production. 2009.

[48] Jiang F, Rao W. Pressure vessel steel and new development of corrosion protection in acid environment. Nat Gas Ind 1999;19:94e7. [49] Huang F, Li XG, Liu J, Qu YM, Du CW. Effects of alloying elements, microstructure, and inclusions on hydrogen induced cracking of X120 pipeline steel in wet H2S sour environment. Mater Corros 2012;63:59e66. [50] Wang SH, Luu WC, Ho KF, Wu JK. Hydrogen permeation in a submerged arc weldment of TMCP steel. Mater Chem Phys 2002;77:447e54. [51] Mohtadi-Bonab MA, Szpunar JA, Razavi-Tousi SS. A comparative study of hydrogen induced cracking behavior in API 5L X60 and X70 pipeline steels. Eng Fail Anal 2013;33:163e75. [52] Mohtadi-Bonab MA, Eskandari M, Rahman KMM, Ouellet R, Szpunar JA. An extensive study of hydrogen-induced cracking susceptibility in an API X60 sour service pipeline steel. Int J Hydrogen Energy 2016;41:4185e97. [53] Silverstein R, Sobol O, Boellinghaus TH, Unger W, Eliezer D. Hydrogen behavior in SAF 2205 duplex stainless steel. J Alloy Compd 2017;695:2689e95. [54] Turnbull A. Perspectives on hydrogen uptake, diffusion and trapping. Int J Hydrogen Energy 2015;40:16961e70.

Please cite this article in press as: Gan L, et al., Hydrogen trapping and hydrogen induced cracking of welded X100 pipeline steel in H2S environments, International Journal of Hydrogen Energy (2017), https://doi.org/10.1016/j.ijhydene.2017.11.155