Introduction

Pipeline steels are used to carry natural gas and oil over long distances. For example, there are more than 96,000 kilometers of transmission pipeline in Canada and about 480,000 kilometers in the United States. No doubt, HIC in pipeline steels is the most important mode of damage in sour environment. Hydrogen atoms which are the product of surface corrosion of pipeline, welding, and cathodic protection accumulate at different structural defects, such as inclusions, precipitates, dislocations, and carbides. When the amount of hydrogen reaches a critical value, the crack may initiate and propagate through the pipe (Ref 1-3). Among different forms of damage in pipelines, HIC has been recognized as an important technological challenge to steel manufacturers and the petroleum industry. HIC typically produces catastrophic and unexpected pipeline failure. Therefore, HIC is a threat to the safety of the oilfield production and the security of the national energy supply. Since the pipeline steels are often built to serve in sour environment for several decades, such risk is not unexpected. National Energy Board (NEB) reported that 37% of pipeline failures, between 1991 and 2004, were related to the cracking (Ref 4). Moreover, based on Dot/OPS report, around 50% of all types of failures in pipeline steels are related to stress corrosion cracking (SCC), internal, and external corrosion including HIC (Ref 5). The mechanism of failure by HIC is not fully understood; however, there are several studies related to HIC failure (Ref 6-9). HIC in all manufactured pipeline steels depends on microstructural parameters, composition of steel, type, and morphology of non-metallic inclusions as well as mechanical and environmental parameters such as applied stress, type of liquid and gas being transported, ground water chemistry, and many others. There are also hydrogen traps, affecting HIC. Based on the binding energy of traps with hydrogen atoms, hydrogen traps are categorized as reversible and irreversible traps (Ref 10). The irreversible traps, such as non-metallic inclusions and carbides, retain hydrogen while reversible traps may release hydrogen continuously at ambient temperature (Ref 11). The irreversible traps play an important role in crack nucleation process; however, the reversible traps facilitate crack propagation. Mixed oxide and manganese sulfide inclusions with high binding energy are categorized as irreversible traps. Our previous work demonstrated that, in the X70 pipeline steel, some HIC cracks initiated from the manganese sulfide inclusion and propagated at the center of cross section area (Ref 12). Hydrogen permeation test provides useful information about hydrogen trapping behavior in steels. First buildup transient makes it possible to evaluate the total density of hydrogen traps including both reversible and irreversible traps while second polarization provides information on reversible traps.

Control of texture and grain boundary engineering is considered as a new approach to improve the HIC resistance in pipeline steel. There are already several studies about the role of texture in HIC claiming that {111} fiber texture decreases the possibility of crack coalescence (Ref 13, 14). Our previous research also illustrated that the control of texture improves resistance to SCC (Ref 15). Also, it has been documented that the {111} texture can improve HIC resistance by decreasing the number of paths for crack propagation (Ref 16). In this latter work, the authors claimed that several texture components, such as {001}, {112}, and {113}, increase HIC susceptibility while {332} texture creates resistance to HIC crack growth (Ref 17).

The main objective of the present research is to study the HIC susceptibility in X60 and X60SS pipeline steels. Experimental techniques such as hydrogen charging, discharging, hydrogen permeation, and crystallographic texture measurements were used to characterize steels. First, OM and SEM observations were used to investigate the microstructure of X60 and X60SS steels. Second, the steel specimens were electrochemically charged by hydrogen for different durations of time in order to find the crack nucleation sites and the path of propagation. Then, JIS test method was used to measure the content of discharged hydrogen after charging for different times. Hydrogen permeation test was carried out at different layers of X60 and X60SS steels in order to study the hydrogen trapping behavior in both steels. Finally, texture and EBSD measurements were used to image HIC cracking in investigated steels.

Experimental Procedure

Tested Materials

All experiments were carried out on API X60 and X60SS pipeline steels. The chemical composition of both steels is listed in Table 1. The rolling, transverse, and normal directions were named RD, TD, and ND, respectively. The surface of both steels (RD-TD plane) was polished with 1 µm diamond paste at the final stage and etched with 2% nital solution. A Nikon Eclipse MA100 optical microscope and SU 6600 Hitachi field emission scanning electron microscope (SEM) were used to observe the microstructure of both steels.

Table 1 Chemical composition of X60 and X60SS pipeline steels
Full size table

Electrochemical Hydrogen Charging and Discharging Measurements

Both X60 and X60SS steels were electrochemically charged with hydrogen by a mixed solution of 0.2 M H2SO4 and 3 g/l ammonium thiocyanate (NH4SCN). This method of charging was preferred to gaseous charging method since it creates more severe environment. Ammonium thiocyanate acts as a recombination poison that increases hydrogen diffusion rate by preventing hydrogen bubble formation on the surface of steel. Five specimens from both X60 and X60SS steels with dimensions of 130 mm (TD) × 25 mm (RD) × 6 mm (ND) and 130 mm (TD) × 25 mm (RD) × 9 (ND) mm were cut from pipeline plates, and all surfaces were polished with 600 grit SiC emery paper at the final stage. All specimens were washed with distilled water and ultrasonically degreased with acetone for half an hour. Each specimen was separately placed in a glass test vessel and the vessel was filled with two liters of charging solution. The specimens were electrochemically charged for 1, 3, 8, 15, and 24 h with a constant current density of 20 mA/cm2 using an Instek power supply. Since the duration of the test varied from 1 to 24 h, the charging vessel was firmly covered with a Para film to prevent the evaporation of charging solution. This test makes it possible to investigate the crack nucleation and even observe the propagation path. Moreover, it was possible to measure the content of discharged hydrogen from steels after electrochemical charging using Japanese Industrial Standard (JIS) test method (Ref 18). The JIS test setup which was used for hydrogen discharging is shown in Ref 34. After charging, each specimen was washed with distilled water and acetone and immersed in a scaled funnel-shaped glass tube. The glass tube was filled with ethylene glycol and was placed in a hot bath at 45 °C. The test continued for two days; however, no discharging was observed after 30 h. Hydrogen from the steel specimens was released at 45 °C, stored on top of the scaled tube, and its amount was registered in hour intervals.

Hydrogen Permeation Test

Hydrogen permeation test was used to study the hydrogen diffusion behavior in both X60 and X60SS steels. This test also provides information about hydrogen trapping in steel specimens. The test was carried out for the specimens representing the surface and center layers of the investigated steels. Figure 1 shows two layers from X60 and X60SS specimens that were used for hydrogen permeation test based on ISO 17081:2004 E standard test method (Ref 19). After machining, both surfaces of steels were polished with 1 µm diamond paste, and then the detection side of samples was coated with palladium. Palladium layer increases oxidation rate on the detection side of steel membrane. In this work, 100 ml of a mixed solution (ammonia 28% + PdCl2 5g/l) was used for palladium coating using a G750 Gamry potentiostat with a constant current of 2 mA/cm2 for 90 s (Ref 20). Refer to Ref 26, 34 to see how a steel specimen is mounted between two cells in Devanathan-Stachurski setup. Some modifications were made to increase the reliability of the test. For instance, the charging solution was poured into the cathodic cell after background current was stabilized to a value close to zero. It is worth-mentioning that background stabilization may take up to several hours depending on the type of steel and pH of the solution. Therefore, the charging cell should be emptied during this time. Moreover, the charging solution is highly corrosive and may corrode the surface of steel, and this may decrease the hydrogen diffusion rate through the steel membrane. During the time required for stabilization of the oxidation current, the anodic solution was deaerated with argon gas, and this process continued during the testing. The charging solution was also deaerated with the same gas outside of permeation setup. When the test started, the cathodic side of specimen was electrochemically charged with a mixed solution of 0.2 M sulfuric acid and 3 g/l ammonium thiocyanate using an Instek DC power supply with a constant current of 5 mA. The hydrogen atoms diffused through the steel membrane and oxidized in the detection side in reaction with a 0.1 N sodium hydroxide, and the oxidation current was registered. When the oxidation current reached a steady state level, the cathodic current was interrupted. Therefore, the first buildup transient was finished and the first decay transient was started. Equations 1-3 were used to calculate permeability (J ∞ L), effective diffusivity (D eff), and apparent solubility (C app). Equation 4 proposed by Oriani et al. in 1970 was used to estimate the density of hydrogen traps (N t) (Ref 21).

Fig. 1
figure 1

Surface and center layers of X60 and X60SS pipeline steels

Full size image
$$J_{\infty } L = \frac{{I{}_{\infty }L}}{FA},$$
(1)
$$D_{\text{eff}} = \frac{{L^{ 2} }}{{ 6t_{\text{l}} }},$$
(2)
$$C_{\text{app}} = \frac{{J_{\infty } L}}{{D_{\text{eff}} }},$$
(3)
$$N_{t} = \frac{{C_{\text{app}} }}{ 3}\left( {\frac{{D_{\text{l}} }}{{D_{\text{eff}} }} - 1} \right).$$
(4)

In the above equations, I ∞ (µA), L (cm), A (cm2), F (C/mol), t l (s), and D l (cm2 s−1) correspond to the steady state current, thickness of steel membrane, area of specimen subjected to charging and oxidation cells, Faraday constant, time lag and lattice diffusion coefficient, respectively. The time lag method which provides highly reliable experimental results was used to measure the diffusivity coefficient (Ref 22). Time lag represents the elapsed charging time when the proportion of J(t)/I ∞ is equal to 0.63. In this paper, the values for D l=1.28 × 10−4 cm2 s−1 and F = 96,500 C/mol were constant (Ref 23, 24).

This test also gives another opportunity to measure the density of reversible traps. For such measurements, the second buildup transient started after the first decay transient finished to estimate the density of reversible traps. The second modification was applied to the test in order to get a more reliable data for the second permeation parameters. Both anodic and cathodic sides of steel specimens was lightly polished with 1 µm diamond paste to remove the oxidation and corroded layers that formed during the first buildup and decay transients. These layers may decrease the hydrogen diffusion rate and overestimate the density of reversible traps. All hydrogen traps were filled with hydrogen during the first buildup transient while only reversible traps released their hydrogen during the first decay transient. Therefore, the second buildup transient is dominated by hydrogen accumulated in the reversible traps (Ref 25).

SEM Observations and Crystallographic Texture Measurements

Since inclusions and precipitates increase HIC susceptibility in pipeline steels, their type and morphology were studied. For this purpose, specimens with dimensions of 20 mm (RD) × 20 mm (TD) × 6 mm (ND) and 20 mm (RD) × 20 mm (TD) × 9 mm (ND) were cut from as-received X60 and X60SS steels, and the cross section area of specimens (RD-ND planes) was polished with 1 µm diamond paste at the final stage and etched with 2% nital solution. The inclusions that have been recognized as one of the main regions for HIC crack initiations (Ref 26-29) were studied at the cross sections of both steels using SEM and EDS analyses. The same sample preparation was carried out in hydrogen-charged specimens to observe the HIC cracks. Texture measurements were also done with the aim to evaluate what are the main texture components that make steel highly susceptible or resistant to HIC. For texture measurements, the same process of surface preparation was used on the cross section of both X60 and X60SS steels. Texture measurements were done at the surface and the center layers of both steels. It is worth-mentioning that the thickness of X60 and X60SS pipeline steel was 6 and 9 mm, respectively. A Bruker XRD apparatus was used to measure the texture in an area of 3 × 3 mm2. After the measurements, Multex 3, Tools and Resmat-TexTools softwares were used to calculate pole figures and ODFs. A SU 6600 Hitachi field emission scanning electron microscope equipped with an Oxford Instruments Nordlys nano EBSD detector was used to do EBSD measurements at the center layer of cross sections (RD-ND planes) in both as-received steels. After measurements, Tango and Mambo softwares were used to analyze the measured data.

Results and Discussion

Microstructure of X60 and X60SS Pipeline Steels

Figure 2(a) and (b) show the OM images of microstructure of X60 and X60SS steels in the surface of both steels (RD-TD plane). The grains with white color are of ferrite phase and the pearlite grains are shown in black color. The main phase in both steels is ferrite. Ferrite with a hardness value around 200 Vickers is recognized as the softest phase in steel and has a very high resistance to HIC (Ref 30). However, it is important to consider that the yield and tensile strength of ferrite phase is lower than other phases such as pearlite, bainite, and martensite. The pearlite microstructure is composed of ferrite and cementite. This microstructure can increase HIC susceptibility by trapping hydrogen at interfaces between ferrite and cementite (Ref 31). It should be considered that the volume fraction of pearlite in both steels is very low (less than 1%) compared with that of ferrite phase. Also, the average grain size for the X60SS steel is 9.1 µm in length, while it is 6.5 µm in length for the X60 one. The average grain size was the average size of 20 different grains. Yazdipour et al. has implied that the highest HIC resistance happens for an optimum grain size (Ref 32). Figure 2(c) and (d) shows the SEM images of microstructure of X60 and X60SS. It is worth-mentioning that the microstructure of center of cross section in the X60 steel is a little bit different with outer surfaces. At the center of cross section, several hard phases, such as bainite and small particles of martensite, are seen.

Fig. 2
figure 2

OM images of microstructure of API (a) X60 and (b) X60SS steels and SEM images of microstructure of API (c) X60 and (d) X60SS steels

Full size image

Inclusions, such as elongated manganese sulfide or mixed oxide with high binding energy, are recognized as irreversible hydrogen traps (Ref 33). These types of inclusions are mostly incoherent with the metal matrix and may increase HIC susceptibility. Hydrogen atoms are accumulated at the interfaces between these inclusions and metal matrix and when they combine to form hydrogen molecules, cracks initiate. Figure 3(a) and (c) shows two types of inclusions that were observed in a region close to the center of cross section of X60 steel. Based on the EDS scans as shown in Fig. 3(b) and (d), these inclusions have rather complex structure (Al-Mg-Ca oxide). Figure 3(e) and (g) shows other types of precipitates observed at the center of cross section of X60SS specimen. As shown in Fig. 3(f) and (h), these precipitates are Mn-Al-Ca oxide inclusions and complex nitride precipitates. These types of inclusions can be also considered as crack nucleation sites because they may increase the hardness and HIC susceptibility.

Fig. 3
figure 3

(a) Al-Mg-Ca oxide inclusion in X60 steel, (b) EDS point scan on the inclusion, (c) Al, Mg, Ca oxide inclusion in X60 steel, and (d) EDS point scan on the inclusion, (e) Al-Ca oxide inclusion in X60SS steel, (f) EDS point scan on the inclusion, (g) complex nitride precipitate in X60SS steel, and (h) EDS point scan on the inclusion. These inclusions and nitride precipitate were found in a region close to the center line of the cross section area in the X60 and X60SS steels

Full size image

Hydrogen Charging and Discharging Measurements

Hydrogen charging test was carried out in order to study the HIC crack nucleation and propagation sites in both steels. After charging the X60 steel for 3 h, several small cracks appeared at the center of the cross section of the plate and propagated in the rolling direction. Therefore, when charging time increased, the length of cracks also increased. Several large cracks were observed after 24 h charging at the center of the cross section in X60 steel. Surprisingly, no cracks were observed at the cross section of X60SS specimen after charging. The main conclusion is that this steel is not susceptible to HIC. Since there are a lot of non-coherent inclusions at the cross section of both steels, it was rather strange to discover that there were no cracks after hydrogen charging in the X60SS steel. Figure 4(a) showed two large cracks at the center of cross section in X60 steel. However, as shown clearly in Fig. 4(b), there were no cracks after hydrogen charging in the X60SS steel. It is worth-mentioning that there were a lot of blisters observed on the surface of both steels and the size and the volume fraction of blisters increased with charging time. However, at the same condition of charging, the density of blisters in the X60 steel was higher than that in the X60SS steel. In 24-h-charged samples, the size of most of blisters was more than 1 mm in length. To better understand HIC phenomenon and also whether or not hydrogen atoms can diffuse easily in both steels, hydrogen discharge measurements were done for both X60 and X60SS steels using JIS standard test method. Figure 5(a) and (b) shows the hydrogen discharged content versus discharging time. We see that the hydrogen discharge increased with the time and reached a steady state level. The main point is that the amount of discharged hydrogen after the same condition of charging was higher for X60 steel than for X60SS specimen. For instance, the highest amount of discharged hydrogen in X60 steel was 7 ppm while in the case of X60SS steel it was 4.9 ppm. Moreover, our previous work showed that as-received X70 pipeline steel can keep hydrogen up to 5.8 ppm (Ref 34). This work also showed that this amount of hydrogen is able to create several HIC cracks at the center of cross section in X70 steel. Despite the fact that the X60SS steel has very high resistance to HIC, it can trap a considerable amount of hydrogen. The other point is that hydrogen discharged content in 1- and 3-h-charged specimens reached a steady state level after a short period of time (2-3 h), as shown in Fig. 5(a) and (b), while in 24-h-charged specimen, hydrogen discharged content reached a steady state level after a long time (around 25 h). It is more probable that the higher amount of hydrogen is accumulated at the center of cross section in long-hour-charged steels, and it took more time to release hydrogen from that region. The main conclusion, here, is that the amount of hydrogen inside pipeline steel cannot be a reliable measure to evaluate HIC susceptibility in pipeline steel because the X60SS with a considerable amount of hydrogen inside its traps is very resistant to HIC.

Fig. 4
figure 4

SEM images of (a) crack propagation in the X60 steel and (b) no crack in the X60SS. Both images were taken from the center of cross section of steels after hydrogen charging

Full size image
Fig. 5
figure 5

Hydrogen discharged content histogram in (a) the as-received X60 and (b) the as-received X60SS steels at different discharging times

Full size image

Hydrogen Trapping Behavior in the X60 and X60SS Steels

Figure 6(a)-(d) shows the first buildup and decay transients at the surface and center layers of the X60 and X60SS steels. Hydrogen permeation test is used to study the hydrogen diffusion through steel specimens. Four parameters including permeability coefficient (J ∞ L), diffusivity coefficient (D eff), solubility coefficient (C app), and density of hydrogen traps (N t) are calculated from the first and the second buildup transients. Permeability is measured on the oxidation side of the specimen. The effective diffusivity shows the rate of hydrogen diffusion through the steel membrane. The solubility coefficient represents the lattice concentration of hydrogen in steel specimens. Finally, the density of traps is defined as the number of hydrogen traps per centimeter cube. These parameters, based on Eq 1-4, were calculated and are shown in Table 2. Based on Eq 4, a decrease in diffusivity and an increase in solubility result in higher density of hydrogen traps in steel. The diffusivity coefficient decreased from the surface to the center layer, while solubility coefficient increased. These changes resulted in a high density of total hydrogen traps at the center layer of X60 steel, and it is well-accepted that the higher density of hydrogen traps, the more susceptible is the steel to HIC. Therefore, based on evaluation of trap density, the center layer of X60 steel should be more susceptible to HIC than the surface layer. In X60SS steel, the diffusivity decreased from the surface layer to the center layer, and solubility increased. Considering the density of traps at the surface and center, we see that the density of hydrogen traps at the center layer is considerably higher, and it can be considered a clear sign of higher susceptibility of this layer to HIC. The density of reversible traps at the center layer of X60 specimen is higher than that in the X60SS steel. Since the reversible traps provides easy paths for hydrogen mobility, hydrogen atoms can move more easily at the center layer of the X60SS steel than the X60 one. Since the irreversible traps keep hydrogen permanently or for a long period of time, they play a major role in crack nucleation. In other word, based on the internal pressure theory (Ref 35, 36), when the stress between the interface of precipitates and metal matrix reaches a critical value, cracks nucleate. Figure 6(e)-(h) shows the second buildup and decay transients at the surface and the center layers of the X60 and X60SS steels. The hydrogen permeation parameters for the second buildup transient are also shown in Table 2. If we look at the second buildup transient parameters at the surface and center layers of X60 steel, we will see that both diffusivity and solubility coefficients increased toward the center layer. However, the density of reversible traps at the center layer was higher than that at the surface layer. The conclusion is that the crack propagation occurs easier at the center layer. Finally, the diffusivity coefficient notably increased from the surface layer to the center layer, while the solubility coefficient decreased in the X60SS steel. Therefore, the density of reversible traps at the center layer of X60SS was considerably lower than that at the surface layer. It is notable that the permeability coefficient was not discussed in this section because the effect of permeability, based on the Eq 3, was already accounted in the solubility, and second, the density of traps depends on the diffusivity and solubility coefficients, see Eq 4.

Fig. 6
figure 6

First buildup and decay transients at (a) the surface layer of X60, (b) the center layer of X60, (c) the surface layer of X60SS, and (d) the center layer of X60SS and second buildup and decay transients at (e) the surface layer of X60, (f) the center layer of X60, (g) the surface layer of X60SS, and (h) the center layer of X60SS

Full size image
Table 2 Hydrogen permeation test parameters at the surface and center layers of X60 and X60SS pipeline steels
Full size table

Beside the density of traps in both specimens, the center segregation zone was very important in HIC crack nucleation and propagation phenomenon. During casting, the outer surfaces of steel solidify quickly and elements with low melting points are rejected to the center part of casting plate. The concentration of these elements at the center of cross section zone of steel forms the center segregation zone. Table 1 shows the chemical composition of both steels. If we compare the carbon content in both steels, we will clearly see that the amount of carbon in the X60 steel (0.052 wt.%) is almost two times higher than in the X60SS steel (0.027 wt.%). Moreover, the manganese content in the X60 steel (1.5 wt.%) is much more higher than in the X60SS steel (1.26 wt.%). It is worth-mentioning that the amount of titanium, niobium, and sulfur in the X60 steel is higher than in the X60SS steel. Micro-hardness measurements of both X60 and X60SS specimens at the center of cross section showed that the average value of hardness for X60 and X60SS steels was 190 HV and 160 HV, respectively. It has also been implied that the hardness of center segregation zone is higher than the other areas in pipeline steel (Ref 39). Therefore, the center segregation zone in the X60 steel is more pronounced than in the X60SS steel. This area makes the center of cross section in the X60 steel very hard and prone to HIC cracking. The center of cross section in the X60 steel has higher hardness value than in the X60SS, and it was a good confirmation for the importance of segregation zone in the X60 steel.

Texture and EBSD Measurements

Figure 7(a)-(d) shows {111} pole figures at the surface and center layers of X60 and X60SS steels. Since {111} texture plays an important role in HIC resistance of steel, the {111} pole figures were used. Figure 7 shows {110} texture at the surface layer of X60 steel, while the center layer of X60 and both surface and center layers of X60SS steel show {111} texture. It is important to notice that the intensities of these pole figures vary from 1.5 to 2.3 which mean that textures are weak. However, the {111} texture is recognized as the crack-resistant texture that prevents crack propagation (Ref 13, 37). Since ODF at φ2 = 45° section shows the major texture components and α, ε, and γ fibers appear at this section, the ODFs at the surface and center layers for X60 and X60SS steel are shown at φ2 = 45° in Fig. 8(a)-(d). The intensities for these ODFs fibers varied from 2.1 to 3.4. The ideal orientations in BCC steel are schematically shown in Ref 38. For ferritic steels such as pipeline steels, the important fibers are the α-fiber, ε-fiber, and γ-fiber. In α-fiber, the fiber axis <110> is parallel to the rolling direction that includes {001}<110>, {112}<110>, and {111}<110> components. In ε-fiber, the fiber axis <011> is parallel to the transverse direction including {001}<110>, {112}<111>, {111}<112>, and {011}<100> components. Finally, in γ-fiber, the fiber axis <111> is parallel to the normal direction that includes {111}<110> and {111}<112> components. As shown in Fig. 9(a), the orientation density of the α-fiber is mainly centered on the {112}<110> orientation. The maximum intensity of this orientation, at the level 3.3, is related to the center layer of X60 steel. There are also some other and different results concerning the role of texture components in the literature. For instance, Verdeja et al. has recently showed that the presence of the {112} texture is expected to increase HIC susceptibility (Ref 12), while our recent study showed the opposite result (Ref 37). In other words, we found that the {112} dominant texture make steel resistance to HIC. It should be considered that the second highest intensity of {112}<110> component is related to the center layer of X60SS steel. Also, as clearly shown in Fig. 9(a), the maximum intensity of {001}<110> component is observed at the center layer of X60 steel. The {001} texture provides easy path for HIC crack propagation and increases the susceptibility of the steel to HIC (Ref 13, 14). The lowest intensity of {001}<110> component is observed on the surface layer of X60 and X60SS specimens.

Fig. 7
figure 7

{111} pole figures at the (a) surface layer of X60, (b) center layer of X60, (c) surface layer of X60SS, and (d) center layer of X60SS. All of measurements were done at the RD-TD planes and the thickness of as-received X60 and X60SS steels were 6 and 9 mm, respectively

Full size image
Fig. 8
figure 8

ODFs at the (a) surface layer of X60, (b) center layer of X60, (c) surface layer of X60SS, and (d) center layer of X60SS. The ODF results were shown for φ2 = 45°

Full size image
Fig. 9
figure 9

(a) α-fibers, (b) β-fibers, and (c) ε-fibers developed during hot rolling process in both X60 and X60SS steels

Full size image

If we look at Fig. 9(b), we will see that the orientation density of the γ-fiber is mainly centered on the {111}<112> component. The maximum intensity of this component with the amount of 3.1 is seen at the center layer of the X60 steel. As determined by the hydrogen charging test, the center layer of the X60 steel is the most susceptible region to the HIC so that all of HIC cracks are seen in this region. The second highest intensity of {111}<112> component is observed at the center layer of X60SS and the lowest intensity is related to the surface layer of X60 where there was no HIC cracks observed. In Fig. 9(b), the maximum intensity of {112}<111> component is seen at the surface of X60 and other layers have relatively low and almost the same intensities. Moreover, based on Fig. 9(a) and (b), the intensities of {011}<110> and {110}<001> components in both layers are low with an amount of less than 1 and only the intensity of {110}<001> component at the surface layer of X60 was higher than 1. Finally, as shown clearly in Fig. 9(c), the intensity of {111}<110> component vary from 1 to 1.4 at the center layers of both X60 and X60SS steels and the surface layer of X60SS, while this component has an intensity of less than 1 at the surface layer of X60 steel. We should consider that the {111} and {332} fiber textures increase HIC resistance while grains with <100>||ND orientation provide easy crack path for crack propagation (Ref 13, 14, 17). It is important to stress that it is not easy to make a conclusion about HIC susceptibility based on low intensity of these texture components. There are other factors affecting HIC, such as grain size, type of grain boundaries (low or high angle grain boundaries and CSL boundaries), local residual stresses and strains, etc.

To have a better understanding of a role of crystallographic texture on HIC, EBSD measurements were done at the center layer of cross section of both as-received X60 and X60SS steels. Figure 10(a) and (b) shows the EBSD map and inverse pole figure (IPF) at the center layer of X60 steel. As shown in Fig. 10(a), the orientation of high number of grains is <111>∥ND or <100>∥ND. The grains with <100>∥ND orientation increase the HIC susceptibility by providing easy path for crack propagation. Our previous work showed that the HIC cracks propagated along small grains (less than 3.5 µm in length) at the center of cross section in the X70 steel and the dominant texture for these grains was {100} (Ref 12). There are also small grains with <100>||ND orientation or an orientation close to the <100>||ND orientation at the center of the cross section in the X60 steel that can be potentially considered as HIC crack propagation path. Also, based on Fig. 10(a), the size of grains in X60 steel varies from very small to large grains. This means that, the length of grain boundaries is high and the boundaries can be considered as reversible hydrogen traps. As shown in Fig. 10(b), the intensity of IPF is 2.39 and this documents that the texture is weak. Figure 10(c) shows EBSD map at the center layer of cross section of X60SS steel. There is a high accumulation of grains with <111>||ND orientation, and based on IPF map as shown in Fig. 10(d), the dominant texture is {111} with the intensity of 4.35. The grains with <111>∥ND orientation increase the HIC resistance by providing resistant path for crack propagation. These factors make the X60SS steel highly resistance to HIC.

Fig. 10
figure 10

(a) and (b) EBSD orientation map and inverse pole figure at the center of cross section (RD-ND) plane of X60 steel, (c) and (d) EBSD orientation map and inverse pole figure at the center of cross section (RD-ND) plane of X60SS steel. The thickness of as-received X60 and X60SS steels were 6 and 9 mm, respectively

Full size image

Conclusions

The following results were obtained based on the hydrogen charging, discharging, and crystallographic texture studies:

  1. (1)

    Hydrogen charging test documented that X60SS steel was not susceptible to HIC while X60 steel showed a higher susceptibility against HIC. Large cracks were observed after hydrogen charging at the center of cross section in X60 steel. A lot of blisters appeared at the surface of both steels after hydrogen charging.

  2. (2)

    Hydrogen discharging results illustrated that both steels have ability to store hydrogen in hydrogen traps. However, the content of hydrogen in the X60 traps was higher than that in X60SS steel.

  3. (3)

    Hydrogen permeation test allowed to analyze both reversible and irreversible traps at the surface and center layers of X60 and X60SS steels. The density of total hydrogen traps at the center of cross section area was higher than at the surface area of steels. Interestingly, the density of reversible traps was very low at the center part of X60SS steel and this is considered as a possible reason for high resistance of this steel against HIC crack propagation.

  4. (4)

    The center segregation zone with higher hardness value in the X60 steel was more pronounced than in the X60SS steel. This zone made the X60 steel very susceptible to HIC cracking.

  5. (5)

    Crystallographic texture studies proved the presence of weak texture in both steels. However, based on EBSD measurement results, the accumulation of many equiaxed grains with <111>||ND orientation at the center of cross section of X60SS can be considered as one of the main reasons of high resistance of this steel to HIC. It is notable that most of the grains in X60SS steel became equiaxed at the time of dynamic recrystallization during hot rolling process. Such equiaxed grains with low stacking fault energy (SFE) have a low amount of stored energy of deformation and resist HIC crack propagation better than the other grains.