Hydrogen-Induced Cracking in CGHAZ of Welded X80 Steel under Tension Load

: X80 steel is extensively used in hydrogen environments and is susceptible to hydrogen embrittlement (HE). This paper studied the hydrogen-induced cracking (HIC) behavior in the coarse-grained heat-affected zone (CGHAZ) of X80 steel welds, through applying in situ hydrogen-charging tensile experiments, hydrogen permeation experiments, and various surface analysis techniques. It is shown that a few hydrogen atoms can signiﬁcantly decrease a material’s elongation and reduction of area. When the heat input (HI) was 29.2 kJ/cm, the material had minor sensitivity to hydrogen embrittlement. The tensile fractures were ductile without hydrogen. However, the fracture surface exhibited brittle fracture with hydrogen. With increased HI, the HE fracture showed a transition of intergranular fracture → intergranular and transgranular mixed fracture → transgranular fracture. In the presence of hydrogen, the grain boundaries of elongated strips were prone to the formation of intergranular cracks under a tension load, and the hydrogen embrittlement resistance of the bulk lath bainite (LB) was weak. The hydrogen embrittlement susceptibility of pure granular bainite (GB) was lower. Fine LB and GB composite structures could remarkably inhibit intergranular cracks, giving the steel a superior resistance to hydrogen embrittlement.


Introduction
The development of the hydrogen energy industry is a vital engine in optimizing the energy structure and promoting the transformation and upgrading of traditional industries [1][2][3][4].Hydrogen storage and transportation is the bridge between hydrogen production and hydrogen energy applications [5,6].China plans to build the first westto-east hydrogen pipeline, with a total length of more than 400 km.In order to save transportation costs, X80 steel, suitable for large-diameter and high-pressure pipelines, will be widely used in hydrogen environments.In addition, the X80 steel pipelines currently in service are affected by cathodic overprotection and high-voltage transmission line interference, which causes hydrogen generation and penetration into the pipeline [7][8][9][10].All these factors expose X80 steel to the risk of hydrogen embrittlement [11][12][13].
Whether exposed to high-pressure hydrogen or cathodic hydrogen evolution, once the hydrogen atoms are adsorbed on the metal surface and enter the pipeline [14], they accumulate in the hydrogen trap [15,16], which will cause hydrogen embrittlement (HE), especially in the weld area of a pipeline [17,18].The multiplicity of grain orientations, grain boundaries, and dislocations caused by welding changes the mechanical properties of the material [19][20][21][22].Pipe fracture failure due to insufficient toughness of the girth weld is one of the main failure modes of high-strength steel pipes.The welded joint's heat-affected zone (HAZ) [23][24][25] is a unique area between the weld and the base material (BM).Due to a non-uniformity of organization [26,27], it is easy to produce local defects and stress concentrations.This is the most vulnerable failure zone in the weld [28][29][30].
To date, the susceptibility of different metallographic structures of the X80 stee coarse-grained heat-affected zone (CGHAZ) to HE under tension load has not been fully characterized and understood.This knowledge is critical in helping to prevent hydrogeninduced cracking (HIC) in X80 steel hydrogen pipelines.This work investigated the HIC behavior in the CGHAZ of X80 steel welds using in situ hydrogen-charging tensile tests hydrogen permeation experiments, and various surface analysis techniques.These results are expected to be used as the reference for the design and fabrication process of hydrogen pipelines.

Materials and Specimens
Test specimens used in this work were taken from a sheet of X80 steel pipeline.The chemical composition of the steel contains (wt.%):C 0.07, Cu 0.221, Cr 0.266, Si 0.216, Mn 1.8, Ni 0.168, Mo 1.82, Nb 0.105, P 0.0137, Ti 0.013, Al 0.026, N 0.003, S 0.0009 and Fe balance.
The process of specimen machining is shown in Figure 1.First, slices of 16 × 180 × 2.2 mm 3 and 34 × 150 × 2.2 mm 3 were axially cut from X80 pipeline steel using wire cutting Subsequently, the Gleeble 3500 thermal simulator (DSI, New York, NY, USA) was employed to conduct the heat treatment of the sheet [31].The red part was the uniform temperature zone in the simulated welding heat treatment.Then, the heat-treated specimens were processed into tensile and hydrogen-permeable specimens with wire cutting.The surface of the specimen was ground with silicon carbide sandpaper to a grain size of 600 800, and 1200, then polished with W5 diamond slurry.The specimen was then degreased with acetone and dehydrated with ethanol.Through a spiral submerged arc welding experiment of X80 steel, heat input curves were obtained from thermocouples pre-buried at different positions from the weld [29] The cooling time from 800 °C to 500 °C (t8/5) was obtained according to the heat input curve of CGHAZ [32,33].The actual t8/5 was 15.5s, and the corresponding welding heat input was 18.1 kJ/cm.We changed the cooling time t8/5 and calculated the heat input correspond ing to different cooling times, according to the theoretical and empirical formula of welding.The calculated thermal simulation parameters are shown in Table 1.Through a spiral submerged arc welding experiment of X80 steel, heat input curves were obtained from thermocouples pre-buried at different positions from the weld [29].The cooling time from 800 • C to 500 • C (t 8/5 ) was obtained according to the heat input curve of CGHAZ [32,33].The actual t 8/5 was 15.5s, and the corresponding welding heat input was 18.1 kJ/cm.We changed the cooling time t 8/5 and calculated the heat input corresponding to different cooling times, according to the theoretical and empirical formula of welding.The calculated thermal simulation parameters are shown in Table 1.

Label
Heating Rate (

Metallographic Observation
The specimens were etched in a mixed solution of 4% nitric acid and ethanol and then observed with a ZEISS Axio Imager A2m (Carl Zeiss AG, Oberkochen, Germany) metallurgical microscope.The microstructures of CGHAZ under different heat inputs are shown in Figure 2. HI08.5 and HI18.1 were mainly composed of bulk lath bainite (LB), and LB was more finely interwoven in HI08.5.In HI29.2, LB was divided by granular bainite (GB), the composite structure was mainly composed of fine LB and GB.As the heat input continued to increase, the HI38.2 and HI48.5 were mainly composed of GB.With the increase in the heat input, the content of LB gradually decreased, while that of GB increased gradually [35].

Metallographic Observation
The specimens were etched in a mixed solution of 4% nitric acid and ethanol and then observed with a ZEISS Axio Imager A2m (Carl Zeiss AG, Oberkochen, Germany) metallurgical microscope.The microstructures of CGHAZ under different heat inputs are shown in Figure 2. HI08.5 and HI18.1 were mainly composed of bulk lath bainite (LB) and LB was more finely interwoven in HI08.5.In HI29.2, LB was divided by granular bain ite (GB), the composite structure was mainly composed of fine LB and GB.As the heat input continued to increase, the HI38.2 and HI48.5 were mainly composed of GB.With the increase in the heat input, the content of LB gradually decreased, while that of GB increased gradually [35].

Tensile and Hydrogen Permeation Testing
The in situ hydrogen charging tensile device [34] is shown in Figure 3a.The type of tensile machine used in the tensile experiment in this paper was a slow strain stress testing machine WDML-50, which was manufactured by LETRY in 2020, and the in situ electrochemical hydrogen charging method was used to charge the specimen with hydrogen during the stretching process [36].The electrochemical charging solution contained 0.5 mol/L H2SO4 + 0.2 g/L CH4N2S [37].The hydrogen charging current densities were 5 mA/cm 2 , 10 mA/cm 2 , 20 mA/cm 2 , 30 mA/cm 2 , and 40 mA/cm 2 ; the hydrogen charging

Tensile and Hydrogen Permeation Testing
The in situ hydrogen charging tensile device [34] is shown in Figure 3a.The type of tensile machine used in the tensile experiment in this paper was a slow strain stress testing machine WDML-50, which was manufactured by LETRY in 2020, and the in situ electrochemical hydrogen charging method was used to charge the specimen with hydrogen during the stretching process [36].The electrochemical charging solution contained Metals 2023, 13, 1325 4 of 16 0.5 mol/L H 2 SO 4 + 0.2 g/L CH 4 N 2 S [37].The hydrogen charging current densities were 5 mA/cm 2 , 10 mA/cm 2 , 20 mA/cm 2 , 30 mA/cm 2 , and 40 mA/cm 2 ; the hydrogen charging temperature was 20 • C; and the pre-charging time was 1.5 h.Stretching was performed at 20 • C, with a strain rate of 10 −6 s −1 .After the specimen was broken, the fracture was ultrasonically cleaned for 10 min in acetone and absolute ethanol solutions and dried and sealed for SEM observation.temperature was 20 °C; and the pre-charging time was 1.5 h.Stretching was performed at 20 °C, with a strain rate of 10 −6 s −1 .After the specimen was broken, the fracture was ultrasonically cleaned for 10 min in acetone and absolute ethanol solutions and dried and sealed for SEM observation.
Hydrogen permeation experiments [38,39] were carried out using the D-S double electrolytic cell shown in Figure 3b.The working electrode and the auxiliary electrode on the cathode side were connected to a DC stable power supply.The working electrode on the anode side, the auxiliary electrode, and the saturated calomel reference electrode constituted a three-electrode system connected to an electrochemical workstation.The electrochemical workstation recorded the hydrogen permeation current density change with time.The hydrogen permeation test temperature was 20 °C.

Characterization of the Fracture Surface
The fractured section was cut from the specimen using wire cutting, ultrasonically cleaned with acetone and anhydrous ethanol, and dried with cold air.The fracture surface was observed using a Hitachi S4800 cold field emission SEM (Hitachi, Tokyo, Japan).

Effect of Heat Input on Tensile Properties
To analyze the effect of heat input on the tensile properties of CGHAZ, engineering stress-strain curves of BM and CGHAZ without hydrogen were measured, as shown in Figure 4.The mechanical performance parameters were measured, and the results are shown in Table 2.The tensile curve of BM had an obvious yield point, while CGHAZ had no obvious yield point.As the heat input increased, the material's ultimate tensile strength (UTS) gradually decreased, but there was a slight increase for HI29.2.Simultaneously, the elongation (EL) of the material decreased with the increase in heat input, indicating that the higher the tensile strength, the worse the tensile toughness.When the heat input was ≤29.2 kJ/cm, the UTS of CGHAZ was significantly higher than BM.When the heat input was increased to 38.3 kJ/cm, the mechanical properties of CGHAZ were similar to BM.When the heat input continued to increase to 48.5 kJ/cm, the strength of CGHAZ was lower than that of BM, but the EL was higher than that of BM, showing a more substantial tensile toughness.The reduction in area (RA) in the CGHAZ was smaller than that of the BM.Hydrogen permeation experiments [38,39] were carried out using the D-S double electrolytic cell shown in Figure 3b.The working electrode and the auxiliary electrode on the cathode side were connected to a DC stable power supply.The working electrode on the anode side, the auxiliary electrode, and the saturated calomel reference electrode constituted a three-electrode system connected to an electrochemical workstation.The electrochemical workstation recorded the hydrogen permeation current density change with time.The hydrogen permeation test temperature was 20 • C.

Characterization of the Fracture Surface
The fractured section was cut from the specimen using wire cutting, ultrasonically cleaned with acetone and anhydrous ethanol, and dried with cold air.The fracture surface was observed using a Hitachi S4800 cold field emission SEM (Hitachi, Tokyo, Japan).

Effect of Heat Input on Tensile Properties
To analyze the effect of heat input on the tensile properties of CGHAZ, engineering stress-strain curves of BM and CGHAZ without hydrogen were measured, as shown in Figure 4.The mechanical performance parameters were measured, and the results are shown in Table 2.The tensile curve of BM had an obvious yield point, while CGHAZ had no obvious yield point.As the heat input increased, the material's ultimate tensile strength (UTS) gradually decreased, but there was a slight increase for HI29.2.Simultaneously, the elongation (EL) of the material decreased with the increase in heat input, indicating that the higher the tensile strength, the worse the tensile toughness.When the heat input was ≤29.2 kJ/cm, the UTS of CGHAZ was significantly higher than BM.When the heat input was increased to 38.3 kJ/cm, the mechanical properties of CGHAZ were similar to BM.When the heat input continued to increase to 48.5 kJ/cm, the strength of CGHAZ was lower than that of BM, but the EL was higher than that of BM, showing a more substantial tensile toughness.The reduction in area (RA) in the CGHAZ was smaller than that of the BM.

Tensile Properties of CGHAZ with Different Heat Inputs at Different Hydrogen Concentrations
Slow tensile tests were performed on BM and CGHAZ under different heat inputs and hydrogen concentrations (including no hydrogen).The engineering stress-strain curves are shown in Figure 5.For BM, in the elastic stretching stage, the slope of the stressstrain curve increased obviously with the increase in hydrogen charging current density.Furthermore, the Young's modulus of the material gradually increased, and the stiffness of the material gradually increased.Nevertheless, for CGHAZ, in the elastic stretching stage, the Young's modulus of the material was less affected by hydrogen.In the presence of hydrogen atoms, whether BM or CGHAZ, the tensile specimens broke rapidly in the elastic stage.Hydrogen reduced the tensile properties of the materials, but there were apparent differences in the hydrogen embrittlement susceptibility among the different materials.

Tensile Properties of CGHAZ with Different Heat Inputs at Different Hydrogen Concentrations
Slow tensile tests were performed on BM and CGHAZ under different heat inputs and hydrogen concentrations (including no hydrogen).The engineering stress-strain curves are shown in Figure 5.For BM, in the elastic stretching stage, the slope of the stress-strain curve increased obviously with the increase in hydrogen charging current density.Furthermore, the Young's modulus of the material gradually increased, and the stiffness of the material gradually increased.Nevertheless, for CGHAZ, in the elastic stretching stage, the Young's modulus of the material was less affected by hydrogen.In the presence of hydrogen atoms, whether BM or CGHAZ, the tensile specimens broke rapidly in the elastic stage.Hydrogen reduced the tensile properties of the materials, but there were apparent differences in the hydrogen embrittlement susceptibility among the different materials.
To clarify the effect of hydrogen on the tensile properties of the materials, we should pay attention to the concentration of hydrogen atoms rather than the hydrogen charging current density.Therefore, we used electrochemical hydrogen permeation experiments to determine the concentration of hydrogen atoms in the different materials at different hydrogen-charging current densities [40][41][42][43].The hydrogen permeation curves of the materials with different heat inputs at different hydrogen charging current densities are shown in Figure 6.To clarify the effect of hydrogen on the tensile properties of the materials, we should pay attention to the concentration of hydrogen atoms rather than the hydrogen chargin current density.Therefore, we used electrochemical hydrogen permeation experiments t determine the concentration of hydrogen atoms in the different materials at different hy   The constant concentration (CC) model was chosen as the hydrogen permeation model [44][45][46].For the CC model, the theoretical seepage transient current follows the dimensionless Formula (1): where I t is the transient current density at time t; I ∞ is the steady-state current density; τ = Dt/L 2 is a dimensionless parameter; D is the hydrogen diffusivity; and L is the thickness of the specimen.The hydrogen diffusion coefficient D in this experiment was calculated using the Laplace method [47][48][49].Moreover, the hydrogen concentration C 0 was calculated using the following formula: where F is the Faraday constant, and its value is 96,485 C/mol.The calculation results are shown in Table 3.The UTS and EL were obtained according to the stress-strain curve in Figure 5. RA was obtained from fracture dimension measurements.Measurements were taken at three locations of the fracture surface, to obtain an average value.According to the hydrogen concentrations of the different heat input specimens at different hydrogen charging current densities in Table 3, the hydrogen charging current density was replaced by the hydrogen concentration.The UTS, EL, and RA under different hydrogen charging current densities, hydrogen concentrations, and HI were obtained, as shown in Figures 7-9.As depicted in Figure 7c, it can be observed that a minor welding heat input led to an improvement in the UTS, with HI08.5 exhibiting the highest UTS value under a tension load.However, when hydrogen atoms were present, as shown in Figure 7b, the UTS of the CGHAZ gradually decreased with the increase in hydrogen concentration, and the smaller the heat input, the more significant the decrease in UTS.For this scenario, the UTS of HI08.5 and HI18.1 was the worst.
In the absence of hydrogen atoms, the CGHAZ achieved desirable elongation characteristics, with EL gradually increasing with the increase in welding heat input.Notably, at HI48.5, the elongation rate reached 14.48%, surpassing BM. Figure 8c shows that the elongation rate dropped significantly once the hydrogen atoms entered the material.From Figure 8d,b, with the increase in the hydrogen charging current density, the elongation of the material decreased.The UTS of HI08.5 and HI18.1 was the worst.EL refers to the ability of a material to deform when stretched.The greater the elongation, the better the toughness of the material.Hydrogen atoms can cause a material to lose its toughness rapidly, especially when the material microstructure has a substantial content of LB.In the absence of hydrogen atoms, the CGHAZ achieved desirable elongation characteristics, with EL gradually increasing with the increase in welding heat input.Notably, at HI48.5, the elongation rate reached 14.48%, surpassing BM. Figure 8c shows that the elongation rate dropped significantly once the hydrogen atoms entered the material.From Figure 8d,b, with the increase in the hydrogen charging current density, the elongation of the material decreased.The UTS of HI08.5 and HI18.1 was the worst.EL refers to the ability of a material to deform when stretched.The greater the elongation, the better the toughness of the material.Hydrogen atoms can cause a material to lose its toughness rapidly, especially when the material microstructure has a substantial content of LB.In the absence of hydrogen atoms, the CGHAZ achieved desirable elongation characteristics, with EL gradually increasing with the increase in welding heat input.Notably, at HI48.5, the elongation rate reached 14.48%, surpassing BM. Figure 8c shows that the elongation rate dropped significantly once the hydrogen atoms entered the material.From Figure 8d,b, with the increase in the hydrogen charging current density, the elongation of the material decreased.The UTS of HI08.5 and HI18.1 was the worst.EL refers to the ability of a material to deform when stretched.The greater the elongation, the better the toughness of the material.Hydrogen atoms can cause a material to lose its toughness rapidly, especially when the material microstructure has a substantial content of LB.As shown in Figure 9a, in the absence of hydrogen atoms, the RA of CGHAZ was lower than that of the BM.For CGHAZ, with the increase in HI, RA gradually increased, indicating that the plasticity of the material was enhanced.As shown in Figure 9c, the RA dropped significantly once the material came into contact with hydrogen atoms.From Figure 9d,b, with the increase in hydrogen concentration, the RA of the material decreased gradually, indicating a loss of plasticity of the material.The EL and RA of HI18.1 showed a continuous and sharp drop around the hydrogen concentrations of 1.5 × 10 −5 mol•cm −3 and 2.2 × 10 −5 mol•cm −3 .This proves that the plasticity of the material dropped sharply.As shown in Figure 9a, in the absence of hydrogen atoms, the RA of CGHAZ was lower than that of the BM.For CGHAZ, with the increase in HI, RA gradually increased, indicating that the plasticity of the material was enhanced.As shown in Figure 9c, the RA dropped significantly once the material came into contact with hydrogen atoms.From Figure 9d,b, with the increase in hydrogen concentration, the RA of the material decreased gradually, indicating a loss of plasticity of the material.The EL and RA of HI18.1 showed a continuous and sharp drop around the hydrogen concentrations of 1.5 × 10 −5 mol•cm −3 and 2.2 × 10 −5 mol•cm −3 .This proves that the plasticity of the material dropped sharply.
The relative elongation ΔEL was used to characterize the hydrogen embrittlement susceptibility of the material, and its expression is where ELw/oH is the elongation without hydrogen, and ELwithH is the elongation with hydrogen.The ΔEL of CGHAZ with different heat inputs at different hydrogen concentrations was calculated, respectively.The calculation results are shown in Figure 10.The relative elongation ∆EL was used to characterize the hydrogen embrittlement susceptibility of the material, and its expression is where EL w/oH is the elongation without hydrogen, and EL withH is the elongation with hydrogen.The ∆EL of CGHAZ with different heat inputs at different hydrogen concentrations was calculated, respectively.The calculation results are shown in Figure 10.When the heat input was set to 29.2 kJ/cm, CGHAZ had the lowest hydrogen embrittlement sensitivity under tensile load, even lower than the BM.When HI > 29.2 kJ/cm, the susceptibility to hydrogen embrittlement of CGHAZ was less affected by the hydrogen concentration gradient.At this stage, the susceptibility to hydrogen embrittlement was slightly higher than that of BM, steadily increasing with an escalating hydrogen concentration.When HI <29.2 kJ/cm, the hydrogen embrittlement susceptibility of HI18.1 was the highest, and the susceptibility to hydrogen embrittlement increased rapidly with the increase in the hydrogen concentration.The welding heat input associated with HI18.1 precisely matched the welding parameter used in the current operational pipeline.

Fracture Surface Analysis of CGHAZ with Different Heat Inputs
The tensile fracture surfaces of the CGHAZ specimens under different heat inputs without hydrogen charging and at a hydrogen charging current density of 20 mA/cm 2 are shown in Figure 11.When the heat input was set to 29.2 kJ/cm, CGHAZ had the lowest hydrogen embrittlement sensitivity under tensile load, even lower than the BM.When HI >29.2 kJ/cm, the susceptibility to hydrogen embrittlement of CGHAZ was less affected by the hydrogen concentration gradient.At this stage, the susceptibility to hydrogen embrittlement was slightly higher than that of BM, steadily increasing with an escalating hydrogen concentration.When HI <29.2 kJ/cm, the hydrogen embrittlement susceptibility of HI18.1 was the highest, and the susceptibility to hydrogen embrittlement increased rapidly with the increase in the hydrogen concentration.The welding heat input associated with HI18.1 precisely matched the welding parameter used in the current operational pipeline.

Fracture Surface Analysis of CGHAZ with Different Heat Inputs
The tensile fracture surfaces of the CGHAZ specimens under different heat inputs without hydrogen charging and at a hydrogen charging current density of 20 mA/cm 2 are shown in Figure 11.Without hydrogen, the tensile fracture surfaces under different heat inputs were characterized by ductile fracture.Dimple structures dominated the fracture surface, and dense small dimples were distributed around large ones.Furthermore, there was an apparent necking phenomenon.With the increase in heat input, the area and content of large dimples in the fracture surface increased, indicating that the material's toughness had increased.When it reached HI29.2, small cracks and inclusion holes appeared on the fracture surface.In the presence of hydrogen, the fracture surface showed brittle fracture characteristics.HI08.5 exhibited intergranular fractures, and deep cracks were observed on the fracture surface.The hydrogen accumulated in the specimen reduced the critical stress of crack initiation and had a specific auxiliary effect on crack propagation [50][51][52].Therefore, the large cracks propagated in different directions, leading to the rapid fracturing of the specimen under hydrogen charging.At HI18.1, intergranular and transgranular fractures existed in the fracture surface, and a clear boundary between them could be observed.At HI29.2, the fracture surface mainly showed a transgranular fracture, but there were apparent tearing edges on the fracture surface.The surface of the transgranular fracture was truncated at multiple points, and there were apparent step structures between the river-like patterns.The fracture surfaces resulting from HI38.3 and HI48.5 exhibited transgranular brittle fractures, with the surfaces becoming progressively flatter.The step area of the river-like pattern also gradually increased, and the secondary cracks gradually became more extensive and deeper.

Mechanism of the Difference in Sensitivity of LB and GB to HIC under a Tensile Load
HIC analysis was performed of the metallographic structure of CGHAZ with different heat inputs and of its tensile properties under different hydrogen concentrations.In the absence of hydrogen, the bulk LB enhanced the tensile strength of CGHAZ and improved the material performance [53,54].Nevertheless, when hydrogen was present, the bulk LB exhibited a high sensitivity to hydrogen embrittlement, and the materials predominantly composed of bulk LB demonstrated a weak resistance to hydrogen embrittlement.In the presence of hydrogen, the composite structure of fine LB and GB had the best resistance to hydrogen embrittlement.The susceptibility to hydrogen embrittlement of the pure GB structure was less affected by changes in the hydrogen concentration.The mechanism of the influence of LB and GB on HIC under tensile load was analyzed, as shown in Figure 12.Hydrogen atoms tended to accumulate in the grain boundaries [55].When the atoms gathered in the bulk LB grain boundaries, the grain boundaries of long strips tended to initiate intergranular cracks under a tensile load, as shown in Figure 12a.When the microstructure became a composite microstructure of fine LB and GB, as shown in Figure 12b, the extended grain boundaries in the original bulk LB were divided by GB.Segmentation made the intergranular cracks more likely to be blocked, so the fracture surface presented mixed intergranular and transgranular fractures.The fracture was transgranular when the metallographic structure became pure GB, as shown in Figure 12c.Hydrogen atoms only acted on cracks passing through the grain boundaries, so their susceptibility to hydrogen embrittlement was less affected by changes in the hydrogen concentration.Therefore, under a tension load, the fine LB and GB composite structure in Figure 12b shows a better resistance to hydrogen embrittlement.This is a more desirable structure in metal metallurgy and welding processes.Without hydrogen, the tensile fracture surfaces under different heat inputs were characterized by ductile fracture.Dimple structures dominated the fracture surface, and dense small dimples were distributed around large ones.Furthermore, there was an apparent necking phenomenon.With the increase in heat input, the area and content of large dimples in the fracture surface increased, indicating that the material's toughness had increased.When it reached HI29.2, small cracks and inclusion holes appeared on the frac- ture surface presented mixed intergranular and transgranular fractures.The fracture was transgranular when the metallographic structure became pure GB, as shown in Figure 12c.Hydrogen atoms only acted on cracks passing through the grain boundaries, so their susceptibility to hydrogen embrittlement was less affected by changes in the hydrogen concentration.Therefore, under a tension load, the fine LB and GB composite structure in Figure 12b shows a better resistance to hydrogen embrittlement.This is a more desirable structure in metal metallurgy and welding processes.According to the microstructure content and the arresting effect of the different structures on HIC, a comparison of HIC resistance of the different microstructures is summarized in Table 4.This verifies that 11% LB steel had a better HE resistance.Moreover, these LBs should be scattered in GB.
An increment in hydrogen concentration resulted in an increased Young's modulus of BM during the elastic stretching stage.However, this effect was not prominent in CGHAZ.The UTS of CGHAZ gradually decreased with the increase in hydrogen concentration.Once the material was exposed to hydrogen, the material plasticity dropped sharply at first and then decreased slowly with the increased hydrogen concentration.

2.
The HI 29.2 specimen had the lowest sensitivity to HE under tensile load.The HE sensitivity of HI38.3 and HI48.5 became less affected by the hydrogen concentration gradient.The HE susceptibility of HI18.1 was the highest, and its HE susceptibility increased rapidly with the hydrogen concentration.

Figure 1 .
Figure 1.Schematic diagram illustrating the processes for specimen manufacturing and the speci men dimensions, adapted with permission from [32] (unit: mm).

Figure 1 .
Figure 1.Schematic diagram illustrating the processes for specimen manufacturing and the specimen dimensions, adapted with permission from [32] (unit: mm).

Figure 3 .
Figure 3. Schematic diagram of the device: (a) in situ hydrogen charging tensile, (b) D-S double electrolytic cell for hydrogen permeation.

Figure 3 .
Figure 3. Schematic diagram of the device: (a) in situ hydrogen charging tensile, (b) D-S double electrolytic cell for hydrogen permeation.

Figure 4 .
Figure 4. Stress-strain curve of BM and CGHAZ under different heat inputs.

Figure 4 .
Figure 4. Stress-strain curve of BM and CGHAZ under different heat inputs.

Figure 7 .
Figure 7. UTS at different hydrogen charging current densities, hydrogen concentrations, and HI (a) under different I, (b) under different C0, and (c) under different HI.

Figure 8 .
Figure 8. EL under different HI, hydrogen charging current density, and hydrogen concentration (a) under different HI, (b) under different C0, (c) under different I, and (d) local amplification under different I.

Figure 7 . 17 Figure 7 .
Figure 7. UTS at different hydrogen charging current densities, hydrogen concentrations, and HI (a) under different I, (b) under different C 0 , and (c) under different HI.

Figure 8 .
Figure 8. EL under different HI, hydrogen charging current density, and hydrogen concentration (a) under different HI, (b) under different C0, (c) under different I, and (d) local amplification under different I.

Figure 8 .
Figure 8. EL under different HI, hydrogen charging current density, and hydrogen concentration (a) under different HI, (b) under different C 0 , (c) under different I, and (d) local amplification under different I.

Figure 9 .
Figure 9. RA under different HI, hydrogen charging current densities, and hydrogen concentrations: (a) under different HI, (b) under different C0, (c) under different I, and (d) local amplification under different I.

Figure 9 .
Figure 9. RA under different HI, hydrogen charging current densities, and hydrogen concentrations: (a) under different HI, (b) under different C 0 , (c) under different I, and (d) local amplification under different I.

Figure 10 .
Figure 10.ΔEL under different hydrogen charging current densities and hydrogen concentrations: (a) under different I, (b) under different C0.

Figure 10 .
Figure 10.∆EL under different hydrogen charging current densities and hydrogen concentrations: (a) under different I, (b) under different C 0 .

Figure 12 .
Figure 12.HIC of LB and GB under tension load: (a) LB, (b) LB and GB, (c) GB.Figure 12. HIC of LB and GB under tension load: (a) LB, (b) LB and GB, (c) GB.

Figure 12 .
Figure 12.HIC of LB and GB under tension load: (a) LB, (b) LB and GB, (c) GB.Figure 12. HIC of LB and GB under tension load: (a) LB, (b) LB and GB, (c) GB.

Table 2 .
Tensile parameters of BM and CGHAZ under different heat inputs.

Table 2 .
Tensile parameters of BM and CGHAZ under different heat inputs.

Table 3 .
Calculation results of hydrogen concentration for different heat input specimens at different hydrogen charging current densities.

Table 4 .
Comparison of the microstructure and HIC resistance of CGHAZ.