2018 Volume 58 Issue 1 Pages 173-178
The hydrogen embrittlement behavior of an ultra-high strength (1180 MPa grade) dual phase steel sheet composed of ferrite and tempered martensite, as compared with that of a single phase steel sheet composed of tempered martensite, has been investigated by a sustained tensile-loading test. No fracture of the dual phase steel occurs under the low hydrogen-charging current density of 5 A/m2 except under high applied stress substantially larger than the yield stress. With the high current density of 50 A/m2, the time to fracture of the dual phase steel varies widely, but is almost the same as that of the single phase steel. The critical applied stress for fracture of the dual phase steel is higher than that of the single phase steel. Under the high applied stress, however, the time to fracture of the dual phase steel is shorter than that of the single phase steel, and a unique intergranular-like morphology is observed at the crack initiation area on the fracture surface. Upon plastic deformation before the sustained tensile-loading test under the high applied stress, the time to fracture of the dual phase steel increases and the initiation area on the fracture surface exhibits typical quasi-cleavage features. The results of the present study indicate that the hydrogen embrittlement of the dual phase steel displays some anomalous behavior.
Dual phase steel sheets composed of ferrite and martensite phases, which are produced by cold-rolling and continuous annealing, have been widely used as automotive parts because of their excellent elongation. Recently, the tensile strength of steel sheets for automotive structural parts has risen to 1180 MPa or more to satisfy governmental and corporate targets for weight reduction and crashworthiness. However, high strengthening of steel sheets leads to increasing susceptibility to hydrogen embrittlement, including delayed fracture.
The hydrogen embrittlement behavior of ultra-high strength steel (UHSS) with tensile strength exceeding 1180 MPa is particularly well known in bolts composed of a single phase of tempered martensite.1) Various methods for evaluation of the hydrogen embrittlement of martensitic steel have been reported, such as a bend test, a slow strain rate technique and a sustained tensile-loading test.2,3,4) In contrast, the hydrogen embrittlement of dual phase UHSS has been often evaluated by the U-bending method.5,6,7,8,9) This method is used because dual phase UHSS is generally applied to door impact beams and bumper reinforcements, which are mainly formed by bending. Nevertheless, the results of the sustained tensile-loading test appear to be able to simulate automobile parts that may be subjected to delayed fracture under applied residual stress after press forming. As an additional advantage of this method, the results of tests for dual phase steel can easily be compared with those for martensitic steel reported by numerous researchers. For instance, under the sustained tensile-loading test for martensitic steel, the scattering of the time to fracture is usually small and the fracture surface shows a quasi-cleavage or intergranular morphology.10,11,12,13) Therefore, the hydrogen embrittlement behavior of dual phase UHSS should be evaluated by the sustained tensile-loading test and will thereby provide versatile data.
Susceptibility to hydrogen embrittlement generally increases with increasing tensile strength of steels.14) The primary cause of embrittlement is considered to be explained by the brittleness of martensite. Moreover, the cracks propagate along the prior austenite grain boundary in martensite single phase steel; thus, most studies15,16,17) have been reported from the viewpoint of grain boundary strengthening. In dual phase UHSS, a significant strength difference between the ferrite and martensite phases probably results in the preferential deformation of ferrite. It is difficult to predict the effects of ferrite on hydrogen embrittlement behavior under a high applied stress (higher than 1000 MPa). Additionally, dual phase steel for automobile parts is usually used after cold-press forming, i.e., plastic deformation, and in this case, the ferrite phase is considered to be preferentially subjected to plastic deformation.18) If such deformation has marked effects on the hydrogen embrittlement behavior of the dual phase steel, it would be the first observation of this phenomenon and would also be an important finding. Therefore, an evaluation of the effects of pre-deformation on hydrogen embrittlement behavior is essential to understanding the optimum approach to using dual phase steel.
The purpose of the present study is to investigate the hydrogen embrittlement behavior of a dual phase UHSS composed of ferrite and tempered martensite with tensile strength of 1180 MPa grade by the sustained tensile loading test. Furthermore, effects of the pre-deformation in the ferrite phase on the hydrogen embrittlement behavior of this UHSS are examined.
The chemical composition of the UHSS sheet used in the present study is shown in Table 1. Cold-rolled steel sheets with the thickness of 1.2 mm were prepared after hot-rolling in the manufacturing process. Two types of specimens, one composed of a tempered martensite single phase and the other composed of ferrite and tempered martensite phases, were prepared by the heat treatments shown in Fig. 1, and are referred to as the single phase steel and the dual phase steel, respectively, unless otherwise stated. The annealing temperatures to recrystallization of the single phase and dual phase steels were 1153 K (austenite phase region) and 1093 K (ferrite + austenite phase region) for 10 min, respectively. After water quenching, tempering was carried out at 673 K (single phase steel) and 423 K (dual phase steel) to adjust the tensile strength of the steels. Scanning electron microscope (SEM) images of the microstructure of the single phase and dual phase steels etched with a 3% nital solution are shown in Figs. 2(a) and 2(b), respectively. The single phase steel exhibited tempered martensite with carbides, and the grain size of the prior austenite was approximately 10 μm. The dual phase steel had a martensite volume fraction of approximately 60%, and the size of the carbides was smaller than that in the single phase steel.
| C | Si | Mn | P | S | Al | N | Fe |
|---|---|---|---|---|---|---|---|
| 0.13 | 1.4 | 2.2 | 0.01 | 0.002 | 0.03 | 0.003 | Bal. |
Heat treatment diagrams of (a) single phase steel and (b) dual phase steel.
SEM microstructures of (a) single phase steel and (b) dual phase steel.
For the tensile test, Japanese Industrial Standards (JIS) No. 5 type tensile specimens were machined parallel to the rolling direction, and the tensile tests were carried out at strain rate of 2.78 × 10−3 s−1 at 298 K. The tensile properties of the specimens are summarized in Table 2. The tensile strengths of both steels showed 1180 MPa grade. The yield stress of the dual phase steel (711 MPa) was lower than that of the single phase steel (1146 MPa) owing to the large volume fraction of ferrite.
| Steel | Yield stress (MPa) | Tensile strength (MPa) | Elongation (%) |
|---|---|---|---|
| Single phase steel | 1146 | 1253 | 9 |
| Dual phase steel | 711 | 1217 | 13 |
A sustained tensile-loading test with hydrogen charging was carried out under various applied stress. The hydrogen charging was started after applying the stress by cathodic electrolysis using a 3% NaCl + 0.3% NH4SCN aqueous solution at room temperature (298 ± 2 K) with a current density of 5 A/m2 or 50 A/m2. A schematic illustration of the specimen size used in the sustained tensile-loading test is shown in Fig. 3. The hydrogen charged length of the specimens immersed in the solution was 15 mm. The applied stress was calculated as the ratio of the applied load to the initial cross-sectional area and was varied to determine the fracture life characteristics. The time to fracture of the specimens was measured, and the test was terminated when no fracture occurred after 100 h. The fracture surface was examined by SEM.
Schematic diagram of test specimen for sustained tensile-loading test; dimensions in mm.
To investigate the effects of pre-deformation on hydrogen embrittlement behavior, the specimens were subjected to the applied stress of 1100 MPa in air for 1 h immediately before the sustained tensile-loading test. The characteristics of plastic deformation by pre-deformation were examined with a transmission electron microscope (TEM). The thin foils for TEM observation were prepared by twin-jet electropolishing in an 8% HClO4 + 82% C2H5OH aqueous solution.
The sustained tensile-loading test results for the single and dual phase steels are plotted in Fig. 4 in terms of the time to fracture as a function of the applied stress at the current density of 5 A/m2. The yield stress of each steels is shown by a dotted line in the figure. The arrows in the figure denote the result for a non-fractured specimen at the indicated elapsed time. Under the applied stress of 1000 MPa, the single phase steel did not fracture within 100 h, whereas the dual phase steel fractured after approximately 4 h. Under the other applied stresses, no fracture occurred. One of the reasons for the fracture of the dual phase steel is that the applied stress is substantially higher than the yield stress of the material. Under the high applied stress levels, in particular, in the plastic deformation region the susceptibility to hydrogen embrittlement of the dual phase steel at this current density is presumably higher than that of the tempered martensite single phase steel.
Time to fracture versus applied stress in sustained tensile-loading test at current density of 5 A/m2.
Similarly, the results of the sustained tensile-loading test at the current density of 50 A/m2 are plotted in Figs. 5(a) and 5(b) for the single and dual phase steels, respectively. For the single phase steel, the time to fracture decreased with increasing applied stress and the critical applied stress for fracture was 300 MPa. For the dual phase steel, the time to fracture varied widely, although it tended to decrease with increasing applied stress. Under the high applied stress (1000 MPa), the time to fracture was shorter than 1 h. The critical applied stress for fracture of the dual phase steel, i.e., 400 MPa, was higher than that of the single phase steel. The scattering of the time to fracture and the high critical applied stress for fracture appear to be characteristics of the hydrogen embrittlement behavior of the dual phase steel under the sustained tensile-loading test. In the applied stress regions from 500 to 900 MPa, the time to fracture of the dual phase steel was almost the same as that of the single phase steel.
Time to fracture versus applied stress in sustained tensile-loading test at current density of 50 A/m2: (a) single phase steel and (b) dual phase steel.
The time to fracture of each steels decreased with increasing current density because of the increase of the amount of absorbed hydrogen. Upon increasing the current density from 5 to 50 A/m2 under the applied stress of 1000 MPa, the single phase steel fractured after approximately 10 h, whereas the time to fracture of the dual phase steel decreased from a few hours to within 1 h. This result suggests that the resistance to hydrogen embrittlement of the dual phase steel under an applied stress higher than the yield stress depends considerably on the applied stress as well as the amount of absorbed hydrogen.
Figure 6(a) shows general view of the fracture surface for the single phase steel subjected to the sustained tensile-loading test under the applied stress of 600 MPa with the current density of 50 A/m2. The reduction in area was scarcely observed. Magnified views of the fracture initiation and the crack propagation areas on the fracture surface are shown in Figs. 6(b) and 6(c), respectively. The fracture initiation area was located at approximately 150 μm from the side surface of the specimen. Quasi-cleavage was observed in the vicinity of the crack initiation area, whereas the crack propagation area displayed, a mixture of quasi-cleavage and intergranular fracture at the prior austenite grain boundary. The morphologies of the fracture surface changed in the order of quasi-cleavage, intergranular and dimple from the crack initiation point to the propagation area. The result of the present study is consistent with that of the previous study,19) although the strength levels and chemical compositions are different.
SEM images of fracture surface of single phase steel after sustained tensile-loading test under applied stress of 600 MPa at current density of 50 A/m2 in (a) general view, and magnified views of (b) fracture initiation area and (c) crack propagation area.
The fracture surfaces of the dual phase steel after the sustained tensile-loading test under the applied stress of 600 MPa with the current density of 50 A/m2 are shown in Fig. 7. The arrows in the micrographs indicate the direction of crack propagation. In the fracture initiation area (Fig. 7(a)), the crack propagated radially. The quasi-cleavage morphology was observed from the crack initiation to the propagation areas. Figure 8 shows a magnified view of the fracture surface in the vicinity of the crack initiation area with the applied stress of 1000 MPa and the current density of 50 A/m2. A unique morphology, i.e., mixed quasi-cleavage and intergranular-like fracture, was observed. As shown in Fig. 5(b), the time to fracture decreased extremely under this applied stress. Such a drastic change in the fracture surface morphology and the time to fracture suggests that the fracture mechanism and/or process changed under the high applied stress. Under the high applied stress, it appears that the ferrite in the dual phase steel is subjected to high stress which is not normally applied to ferritic single phase steel. Therefore, the probable reason for the short time to fracture is related to the preferential plastic deformation of the ferrite under the high applied stress. In a U-bend bolting test of another dual phase steel,2) micro voids were generated at the interface between the ferrite and martensite. Although the susceptibility to hydrogen embrittlement of ferrite itself is not necessarily higher than that of martensite, it is likely that the inhomogeneous strain partitioning at the interface between the ferrite and martensite enhances hydrogen embrittlement. Hence, the susceptibility to hydrogen embrittlement of the dual phase UHSS appears to increase with increasing applied stress in ferrite accompanying the decreasing volume fraction of martensite.
SEM images of fracture surface of dual phase steel after sustained tensile-loading test under applied stress of 600 MPa at current density of 50 A/m2 in (a) fracture initiation area and (b) crack propagation area.
SEM images of crack initiation area on fracture surface of dual phase steel after sustained tensile-loading test under applied stress of 1000 MPa at current density of 50 A/m2: (a) low magnification and (b) high magnification.
With regard to the fact that no fracture occurred under applied stress of lower than 400 MPa, hydrogen embrittlement is presumably inhibited by the effects of ferrite. The previous study20) demonstrated that the property of delayed fracture in dual phase high strength steel is improved by precipitating ferrite along austenite grain boundaries. Consequently, the reason why the critical applied stress of the dual phase steel is higher than that of the single phase steel is that the existence of ferrite appears to suppress delayed fracture. On the other hand, at the current density of 5 A/m2, the critical applied stress of the dual phase steel is lower than that of the single phase steel, as shown in Fig. 4. However, the critical applied stress is significantly higher than the yield stress of the dual phase steel. It is likely that the ferrite phase in the dual phase steel deforms greatly in the work hardened region, thereby considerably enhancing hydrogen embrittlement. The enhancement in hydrogen embrittlement is also observed at current density of 50 A/m2, as shown in Fig. 5(b). Role of plasticity of the ferrite phase in the hydrogen embrittlement behavior of dual phase steels is important. The dual phase steel appears to possess excellent resistance to delayed fracture, except under the high applied stress substantially larger than the yield stress. The critical applied stress for delayed fracture is an important property from the practical perspective.
3.2. Effects of Pre-deformation on Hydrogen Embrittlement Behavior of Dual Phase SteelFigures 9(a) and 9(b) show TEM images of the ferrite phase region in the dual phase steel before and after tensile deformed under the applied stress of 1100 MPa, respectively. The dislocation density tended to increase qualitatively after tensile deformation. In addition, each dislocation length was shortened, as dislocation intersection and tangled dislocations were observed. These results indicate that the ferrite phase of the dual phase steel was subjected to plastic deformation. Figures 10(a) and 10(b) show the results of the sustained tensile-loading test at the current density of 50 A/m2 after pre-deformation for the single and dual phase steels, respectively. The data of without pre-deformation are reproduced from Fig. 5. Upon plastic deformation before the sustained tensile-loading test under the applied stress of 1000 MPa, the time to fracture of the dual phase steel increased to approximately 10 h and was almost the same as that of the single phase steel. In the case of the single phase steel, the effects of the applied stress of 1100 MPa on the mechanical properties and microstructure were negligible because the applied stress is in the elastic deformation region of this steel. Moreover, the time to fracture of the dual phase steel changes only slightly under the applied stress of 800 MPa. Thus, these results indicate that pre-deformation markedly improves the resistance to hydrogen embrittlement of the dual phase steel under applied stress substantially larger than the yield stress. These results are probably consistent with those of the previous study8) that the critical hydrogen amount to fracture in U-bend and drawn cup tests increases with the increase in plastic strain. Davies21) also reported that the time to delayed fracture of a dual phase steel with tensile strength of 590 MPa grade with pre-strain is longer than that with no strain under the same applied stress. Although the volume fraction of ferrite in the present study is less than that in the previous report,21) resistance to hydrogen embrittlement is improved by pre-deformation. Strain concentration in dual phase steels as a result of uniaxial deformation has been observed in ferrite neighboring martensite.18,22,23,24) In dual phase steels having a high volume fraction of martensite, the martensite is also deformed when large strain is applied,25) and this deformation of the martensite may enhance hydrogen embrittlement. However, since the applied strain in this study was small, the strain is probably concentrated in the ferrite neighboring the martensite, as shown in Fig. 9. Pre-deformation leads to work-hardening of ferrite in the dual phase steel; hence, the yield stress after pre-deformation is larger than that before pre-deformation. As a result, because further plastic deformation of the ferrite in the dual phase steel is presumably inhibited during the sustained tensile-loading test, resistance to hydrogen embrittlement increases. Furthermore, the crack initiation area on the fracture surface of the dual phase steel changed from the unique intergranular-like morphology (Fig. 8) to the typical quasi-cleavage type (Fig. 11). These results suggest that the fracture mechanism and/or process are changed by pre-deformation, and the intergranular-like fracture is related to the conditions at the interface between the ferrite and martensite in the dual phase steel. Further study is needed to clarify these points. Nonetheless, it should be emphasized that the conditions at the interface between the ferrite and martensite phases are essential to the hydrogen embrittlement behavior of dual phase steels. Specifically, the conditions at the interface between the ferrite and martensite phases lead to a difference between yield stress and tensile strength, and when the difference between yield stress and tensile strength increases, the resistance to hydrogen embrittlement of the dual phase UHSS under the sustained tensile-loading test will decrease, but not necessarily vice versa.
TEM images of ferrite phase region in dual phase steel (a) before and (b) after tensile deformation.
Time to fracture versus applied stress in sustained tensile-loading test at current density of 50 A/m2 after pre-deformation: (a) single phase steel and (b) dual phase steel. The data of without pre-deformation are reproduced from Fig. 5.
SEM images of crack initiation area on fracture surface of dual phase steel subjected to sustained tensile-loading test under applied stress of 1000 MPa at current density of 50 A/m2 after pre-deformation.
We have demonstrated that an ultra-high strength dual phase steel sheet composed of ferrite and tempered martensite exhibits anomalous hydrogen embrittlement behavior under the sustained tensile loading test. By ultra-high strengthening, when the applied stress is substantially larger than the yield stress, fracture readily occurs in a short time, and the crack initiation area on the fracture surface exhibits a unique intergranular like morphology. Nevertheless, such behavior are improved by the pre-deformation in the ferrite phase. The anomalous behavior of hydrogen embrittlement observed in the dual phase steel is probably associated with the relationship between low yield stress and high tensile strength, which is caused by the characteristics of the interface between the ferrite and martensite phases. Although further study is necessary, we conclude that the ultra-high strength dual phase steel is intrinsically excellent in resistance to hydrogen embrittlement in the sustained tensile-loading test.
