Engineering Failure Analysis 93 (2018) 100–110
Contents lists available at ScienceDirect
Engineering Failure Analysis journal homepage: www.elsevier.com/locate/engfailanal
Hydrogen embrittlement behavior of high strength bainitic rail steel: Effect of tempering treatment
T
⁎
Yongjian Zhang , Zhibao Xu, Xiaoli Zhao, Guhui Gao, Weijun Hui, Yuqing Weng School of Mechanical, Electronic and Control Engineering, Beijing Jiaotong University, Beijing 100044, PR China
A R T IC LE I N F O
ABS TRA CT
Keywords: Hydrogen embrittlement Rail steel Bainitic steel Tempering Microstructure
The present study was attempted to evaluate the effect of tempering treatment on the hydrogen embrittlement (HE) behavior of a novel bainitic rail steel by using slow strain rate test (SSRT) with notched round bar specimens. The microstructure of the as-hot rolled rail steel mainly consists of granular bainite and ~9 vol% martensite. It was found that tempering treatment could increase the HE resistance and decrease the susceptibility to HE of the as-hot rolled bainitic rail steel except at temper embrittlement temperature of ~400 °C, and excellent HE properties could be obtained when tempered at 500 °C at less expense of strength and ductility. The enhanced HE properties are ascribed mainly to the gradual decomposition of blocky M/A constituents as well as the tempering of the martensite. It is thus suggested that suitable tempering after hot rolling could be applied to enhance the HE properties as well as to obtain maximum mechanical properties, and thus to guarantee the safety of bainitic steel rails in service.
1. Introduction High C-Mn steels with full pearlitic microstructure have been widely used as rail steels in railway systems because of their superior mechanical properties, which is the primary requirement for withstanding heavy traffic loads [1]. Recently, higher train speeds and increased axle loads to increase the efficiency of rail transport have given rise to larger wheel/rail contact forces [2]. This trend has been increasing the severity of the environment in which rails are used and thus promoting the research and development of next generation rail steels with enhanced properties and longer rail service life [3]. It seems that bainitic steels particular low-C ones have the highest potential to substitute the traditional pearlitic rail steels due to their excellent mechanical properties, such as high strength and toughness as well as excellent fatigue performance [1,4–8]. Bainitic rail steels with strength levels as high as 1200–1500 MPa have been used to manufacture rails and crossings in recent years [6–9]. However, hydrogen related brittle fracture occurred during the process of service of bainitic steel rails and crossings, which is one of the reasons for the limited use of this kind of steel in the past decades [9]. Since bainitic rail steels are of relatively high strength and exposed to conditions that favor hydrogen entry (through corrosion), increasing interests have been paid to their hydrogen embrittlement (HE) behavior [9–12]. Zhang et al. found that the content of hydrogen in the bainitic steel used for crossings played a key role in its failure mechanism; when the hydrogen content is higher than a critical value of ~0.7 ppm, hydrogen-induced brittle fracture was responsible for the failure of the crossings in a short time in service [9]. Zheng et al. reported that the susceptibility to HE significantly decreased with increased Al content in MneAl bainitic steels for crossings mainly due to the increased volume fraction of retained austenite (RA) [10]. Li et al. showed that pre-deformation treatment on a carbide-free bainitic steel for
⁎
Corresponding author. E-mail address:
[email protected] (Y. Zhang).
https://doi.org/10.1016/j.engfailanal.2018.07.005 Received 26 March 2018; Received in revised form 15 May 2018; Accepted 5 July 2018 Available online 05 July 2018 1350-6307/ © 2018 Published by Elsevier Ltd.
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
railway crossings notably increased its HE sensitivity [11]. Our previous research also revealed that the bainitic rail steel was more susceptible to HE than conventional pearlitic rail steel [12]. Therefore, it is necessary and important to further investigate the HE behavior of bainitic rail steel for the purpose of increasing the lifetime and reliability of railway rails and crossings. The microstructures of commercial bainitic steels which are produced by continuous cooling transformation such as environmentfriendly air cooling process after hot rolling/forging are often quite complex compared to those formed from austenite by isothermal transformation. Thus mixed microstructures including granular bainite, lath bainite or even martensite are often found in continuously cooled bainitic rail steels, which influence final mechanical properties through phase type, morphology and size [7,8,12]. It was also reported that bainite/martensite (B/M) multiphase microstructure of bainitic rail steel exhibited superiority to conventional pearlitic rail steel [3,7]. However, the existence of brittle untempered martensite is usually detrimental to the mechanical properties of bainitic steels as well as the non-equilibrium microstructure. In fact, it was found that tempering treatment had significant influence on the mechanical properties of bainitic and B/M multiphase steels [8,13,14]. Therefore, tempering treatment after hot rolling/forging began to be applied to obtain maximum performance as well as to relieve the residual stress and to stabilize phases of bainitic rails [8]. It has been confirmed that suitable tempering treatment is also beneficial to relieve the susceptibility to HE of high strength steels [15–20]. However, there are few published data available concerning the effect of tempering treatment on the susceptibility to HE of bainitic rail steels. Therefore, in the present study, the influence of tempering treatment on the HE behavior of a novel bainitic rail steel was studied by using slow strain rate test (SSRT), in an attempt to enhance the resistance to HE thereby ensuring the reliability of bainitic rail steels.
2. Material and methods 2.1. Materials and specimen preparation The samples used in the current investigation were obtained from hot rolled and air cooled steel rails produced in actual production line with the chemical composition as listed in Table 1. Specimens were cut from the head region of the hot rolled rails in the rolling direction. The as-hot rolled specimens (designated as B1) were divided into four groups and were then tempered at 280 °C, 350 °C, 400 °C and 500 °C for 2 h, respectively, which were henceforth designated as BT2, BT3, BT4 and BT5 specimens, respectively. Moreover, part of the as-hot rolled specimens were austenitized at 880 °C for 0.5 h, oil quenched and then tempered at 450 °C for 2 h to obtain quenched and tempered (Q&T) microstructure for comparison, these were designated QT4. Circumferentially notched round bar specimen with notch root radius of 0.15 mm (Kt = 3.2) [21] was used for the SSRT as shown in Fig. 1. The presence of a notch allows for obtaining a hydrostatic stress state to increase the embrittling influence of hydrogen. Smooth round specimens for tensile test are standard round bars with 5 mm diameter and 25 mm gauge length. Specimens with diameter of 5 mm and length of 25 mm were used to study the hydrogen desorption behavior. All the specimens were rinsed with deionised water and then degreased with ethanol before hydrogen-charging. Hydrogen was introduced into the SSRT and the thermal desorption spectrometry (TDS) specimens by electrochemical charging in a 0.1 mol/L NaOH aqueous solution at 8 mA/cm2 current density for 72 h at room temperature to ensure an equilibrium and constant hydrogen content throughout the specimens [20].
2.2. Measurement of HE susceptibility and hydrogen content SSRT was performed within 10 min after the termination of hydrogen-charging using a WDML-100 kN type uniaxial tensile machine at room temperature with a duration time of about 1.5 h to 2 h. The strain rate was 2.1 × 10−6 s−1 and the test results represent the mean value of at least four specimens. Meanwhile, corresponding specimens without hydrogen-charging were also tested as a reference. After SSRT, the notch tensile strengths (σN0 and σNH for the uncharged and hydrogen-charged specimens, respectively), which were defined as the nominal maximum tensile stresses, were obtained. HE was evaluated using the so-called HE index (HEI), which was determined by calculating the relative notch tensile strength loss according to the following equation:
σ HEI (%) = ⎛1 − NH ⎞ × 100% σN 0 ⎠ ⎝ ⎜
⎟
(1)
The TDS specimen for the analysis of hydrogen was heated from ambient temperature to 800 °C at a constant heating rate of 100 °C/h, and then the hydrogen effusing out of the specimen was analyzed by the quadrupolar mass spectrometer and the hydrogen content could be obtained through the integration of the hydrogen evolution curve. Table 1 Chemical composition of the tested bainitic rail steel (wt%). C
Si
Mn
P
S
Cr
Ni
Mo
0.20
0.80
2.00
0.021
0.010
0.80
0.49
0.31
101
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 1. Geometry and dimensions (in mm) of the notched specimen with notch root radius of 0.15 mm and Kt = 3.2.
2.3. Microstructural observation and mechanical evaluation Microstructures of specimens at different conditions were examined using an optical microscope (OP, Zeiss Scope A1), a scanning electron microscope (SEM, Zeiss EVO 18) and a transmission electron microscope (TEM, Hitachi H-800) attached with an energy dispersive X-ray spectrometer (EDS). Specimens for OP and SEM observation were prepared using standard grinding and polishing and were then etched in 3% nital solution. A solution of 5 vol% perchloric acid and 95 vol% alcohol at −20 °C was used for the preparation of the TEM thin foils in a twin-jet electropolishing apparatus. The operating voltage for TEM was 200 kV. A Rigaku D/ MAX 2500 X-rays diffraction (XRD), with a Cu-Kα radiation operated at 40 kV and 150 mA, was used to determine the volume fraction of RA. Tensile tests were performed on a SUNS 5305 type universal testing machine using a starting strain rate of 4.2 × 10−4 s−1 at room temperature. The fracture surfaces of the fractured specimens were observed using SEM operated at 20 kV. 3. Results 3.1. Microstructure characteristics Figs. 2–4 show the OP, SEM and TEM microstructures of the specimens at different conditions, respectively. The as-hot rolled specimen shows a mixed microstructure consisting of granular bainite (GB) and lath martensite. The volume fraction of martensite using image analysis is about 9.1 ± 1.7 vol%. The GB microstructure consists of martensite/austenite (M/A) constituent in a matrix of bainitic ferrite, and there are exist two types of M/A constituent based on its size and morphology, i.e., thin M/A strip and large blocky M/A island in the tested bainitic rail steel (Fig. 3(a)), as was also found in other bainitic rail steels [8,12]. Moreover, the GB contains twin martensite because of the significant enrichment of carbon in the austenite as shown in Fig. 4(a). Tempering treatment of the as-hot rolled specimen gave rise to the decomposition of M/A as well as the tempering of the martensite phase. It is obvious that the thin M/A changed relatively slightly with an increase in tempering temperature up to 400 °C compared with that of the blocky M/A which began to decompose (Figs. 3(d) and 4(d)), indicating much higher thermal stability of the former. Almost all the M/A decomposed into ferrite and cementite for the specimen tempered at 500 °C as shown in Figs. 3(e), 4(e) and 4(f). Further XRD analysis revealed that the volume fraction of RA continuously decreases with increasing tempering temperature till a slight increase at 400 °C and then a sharp decrease up to 500 °C (Fig. 5). Similar variations of RA with tempering temperature were also found in a bainitic rail steel except the peak temperature [8]. As expected, the Q&T treatment gave rise to fine and uniform tempered martensitic microstructure with negligible amount of RA for the QT4 specimen (Figs. 3(f) and 5). 3.2. Mechanical properties Fig. 6 shows the tempering temperature dependence of the tensile properties of the tested as-rolled bainitic rail steel, and the tensile properties of the QT4 specimen is also given. When the as-rolled bainitic rail steel was tempered at 280 °C, there are a notable
Fig. 2. OP micrographs of the specimens at different conditions: (a) B1; (b) BT4. M refers to martensite and GB refers to granular bainite. 102
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 3. SEM micrographs of the specimens at different conditions: (a,b) B1; (c) BT2; (d) BT4; (e) BT5; (f) QT4. TM refers to tempered martensite.
increase of both yield strength (YS) and reduction of area (Z) while ultimate tensile strength (UTS) and total elongation (A) remain almost unchanged. Further increasing tempering temperature, UTS, YS and Z decrease while A increases slightly. This kind of variations of tensile properties with tempering temperature is regarded to be mainly due to the decomposition of M/A constituents and the tempering effects in martensite [14,22]. Moreover, the QT4 specimen exhibits higher strength level as well as similar ductility to that of the bainitic rail steel at identical tempering temperature. 3.3. Hydrogen desorption behavior Fig. 7 shows the hydrogen desorption rate curves of the hydrogen-charged specimens. It is obvious that all the tested specimens exhibited remarkable low-temperature desorption peaks, which is similar to that obtained for bainitic rail steel with similar chemical composition [12] and other granular bainitic steel [23] and carbide-free nanobainitic steel [24]. The peak temperatures for the B1, BT2, BT5 and QT4 specimens are 195 °C, 170 °C, 175 °C and 145 °C, respectively, i.e., the low-temperature peak shifted to lower temperature with tempering temperature, which indicates that the activation energy of hydrogen trapping decreases. This lowtemperature peak is assumed to arise from the release of hydrogen from reversible hydrogen traps [25]. None of the tested specimens showed notable evidence of high-temperature peaks corresponding to irreversible hydrogen traps. Therefore, the total hydrogen content (CH) could be considered to be equal to the diffusible hydrogen content (Cr) for reversible traps and is also given in Fig. 7. 3.4. SSRT properties Fig. 8 shows the SSRT tensile stress versus displacement curves of both the uncharged and hydrogen-charged specimens. For each 103
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 4. TEM micrographs of the specimens at different conditions: (a,b) B1; (c) BT2; (d) BT4; (e,f) BT5.
Fig. 5. (a) X-ray diffraction spectra and (b) changes of the volume fraction of RA with tempering temperatures of the tested bainitic rail steel.
104
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 6. Variations of tensile properties with tempering temperature of the tested bainitic rail steel: (a) strength; (b) ductility.
Fig. 7. Hydrogen desorption rate curves (TDS curves) of hydrogen-charged specimens.
kind of specimen, the result of only one of the two (uncharged) or four (hydrogen-charged) tensile tests is given. It is obvious that hydrogen charging significantly deteriorates the notch tensile behavior, i.e., the notch ultimate tensile strength decreases notably after hydrogen charging. As can be seen from Fig. 9(a), both σN0 and σNH firstly increase when tempered at 280 °C, and then they slightly decrease with increasing tempering temperature up to 400 °C; this variation is somewhat like that of the UTS for smooth specimen as shown in Fig. 6(a). Interestingly, there is a sharp increase of σNH when tempered at 500 °C. As a result, HEI remains almost constant with increasing tempering temperature up to 400 °C, and then it sharply decreases when tempered at 500 °C as shown in Fig. 9(b). In general, the tempered specimens exhibit higher notch UTS and superior resistance to HE when compared with the ashot rolled one, and a rather lower susceptibility to HE could be obtained for the 500 °C tempered specimen. 3.5. SSRT fracture surface Fig. 10 shows the fracture surfaces in the crack initiation region of the SSRT notched specimens before and after hydrogencharging. As shown in Fig. 10(a), the fractographic analysis shows a mixed mode fracture of brittle quasi-cleavage and intergranular fracture for the uncharged as-hot rolled B1specimen, while the fracture surface of the hydrogen-charged B1 specimen is dominated by brittle intergranular failure with a few ductile tearing as shown in Fig. 10(b), which corresponds to the rather lower HE resistance. When the as-hot rolled steel was tempered at 280 °C (BT2), the fracture mode changed from quasi-cleavage for the uncharged to mainly brittle intergranular fracture with some ductile tearing and intergranular secondary cracks for the hydrogen-charged specimen as shown in Figs. 10(c) and 10(d) and also found in previous study [12]. The fracture characteristic of the 400 °C tempered specimen (BT4) is similar to that of the as-hot rolled specimen except more intergranular secondary cracks for the uncharged specimen as shown in Figs. 10(e) and 10(f). As can be seen from Figs. 10(g) and 10(h), both the uncharged and hydrogen-charged specimen of the 500 °C tempered specimens (BT5) exhibit primarily quasi-cleavage fracture. It is obvious that the change of the 105
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 8. SSRT curves of notched specimens of the tested bainitic rail steel before and after hydrogen-charging: (a) B1; (b) BT2; (c) BT3; (d) BT4; (e) BT5; (f) QT4.
facture mode with increasing tempering temperature is consistent with those of both notch UTS and HEI as shown in Fig. 9. Similar fracture characteristic to that of the BT2 specimen was found for the QT4 specimen as shown in Fig. 11. It is obvious that the prior austenite grain size of the QT4 specimen is finer than those of the as-hot rolled and tempered specimens, indicating a refining effect of cycle quenching treatment. Post-tensile specimens were further sectioned in order to highlight the microstructural constituents involved in the crack initiation and propagation. Fig. 12 presents SEM micrographs showing the formation and propagation of micro-cracks near M/A constituents near the fracture surface of fractured SSRT specimen. It is obvious that numerous microvoids could be found to be 106
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 9. Variations of (a) notch UTS of the uncharged and hydrogen-charged specimens and (b) HEI with tempering temperature.
initiated at blocky M/A constituents, and these voids tend to propagate along the interface of M/A and BF for the as-hot rolled specimen (Fig. 12(a)), as also found by other researchers [23,26,27]. It should also be mentioned that large cracks are almost all associated with big blocky M/A, indicating that microvoids were formed at those constituents at an early stage of deformation. Microvoids could also be found to be associated with decomposed M/A for the specimen tempered at 500 °C (Fig. 12(b)), but these voids seem to stay isolated from each other not to form cracks, and the number of voids is much lower than that of the as-hot rolled specimen. 4. Discussion There have been many investigations concerning the influence of tempering treatment on the HE of high strength steels. It was found that improvements in the resistance of martensitic microstructure to HE could be obtained through increasing tempering temperature [15–20]. This progression of decreased susceptibility to HE with increasing tempering temperature was primarily correlated with the microstructural evolution, including the reduction in the density of dislocations, and the formation of carbides as well as the morphology change from film-like cementites at prior austenite grain boundaries to rod- or spherical-like particles with increasing tempering temperature [17–20]. It should be noted that increasing tempering temperature usually causes a corresponding loss of strength for conventional low alloy martensitic steels such as AISI 4140 and AISI 4340, since it has been demonstrated to strongly influence the susceptibility to HE in high strength steels [17–20,28]. However, unlike that of martensitic steels, rather different result was obtained for the present bainitic rail steel as shown in Fig. 9, i.e., both σNH and HEI change slightly with increasing tempering temperature up to 400 °C, and then σNH increases while HEI decreases significantly when it was tempered at 500 °C. As there is only a slight decrease of UTS (less than 5%) during the tested temperature range for the tested bainitic rail steel (Fig. 6), the UTS influence could be eliminated as a cause of the difference in the HE behavior, and thus it is reasonable to regard that other factors rather than strength in governing the susceptibility to HE of the tested bainitic rail steel. As mentioned above, one of the most significant microstructural evolutions during tempering is the decomposing of the M/A constituents as shown in Figs. 3 and 4. Studies have revealed that the existence of large blocky M/A constituents is detrimental to the resistance to HE mainly due to the lower mechanical stability of blocky RA which could transform to martensite at the onset of deformation [26,27,29]. Recent investigation using secondary ion mass spectrometry (SIMS) analysis of granular bainitic steel revealed that the M/A constituent is a strong hydrogen trapping site due to the high dislocation density of martensite [23]. Moreover, it is regarded that the hydrogen concentration in austenite is much higher than in other phases mainly due to its much higher hydrogen solubility and extremely lower hydrogen diffusivity [26]. Thus, it is reasonable to suppose that a great part of the charged hydrogen was trapped in M/A constituents. Compared to that of thin M/A constituents, large M/A constituents could give rise to greater local stress concentration and plasticity at the interface between M/A and bainitic ferrite under tensile deformation, which thus causes accelerated hydrogen diffusion and accumulation there [27]. Therefore, micro-crack would initiate at the interface of M/A and bainitic ferrite when hydrogen content reaches the critical value as shown in Fig. 12, and thus strongly affects the HE fracture process of the tested bainitic rail steel. For the as-hot rolled bainitic rail steel, the presence of the untempered brittle martensite as well as the blocky M/A constituent makes it have higher susceptibility to HE and lower HE resistance [12] as shown in Fig. 9. When the as-hot rolled bainitic rail steel was tempered at 280 °C, there is a notable increase of both σN0 and σNH, an increase of 15.5% and 13.7%, respectively, which could be attributed primarily to the tempering of brittle martensite and the decomposing of M/A constituents (Figs. 3(c) and 4(c)) as well as the relaxation of internal stress, as indirectly indicated by a slight decrease of CH. Further increasing tempering temperature up to 400 °C, both σN0 and σNH notably decrease and are comparable to those of the as-hot rolled specimen. This behavior is supposed to be mainly due to the temper embrittlement at ~400 °C. The studies of one bainitic rail steel with similar microstructure revealed significant impact toughness decrease when tempered at ~400 °C [8]. Similar results were also reported in bainitic forging steels with 107
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 10. SEM fractographs of (a,c,e,g) uncharged and (b,d,f,h) hydrogen-charged specimens of the as-hot rolled and tempered bainitic rail steel in crack initiation region: (a,b) B1; (c,d) BT2; (e,f) BT4; (g,h) BT5.
granular bainite and martensite [13,14], although the tempering temperature corresponding to the valley of impact toughness is a little different. This temper embrittlement is suggested to be mainly resulted from the decomposition of blocky M/A along prior austenite grain boundaries as well as the precipitation of plate-shaped cementites [14], while the brittle fracture of quasi-cleavage and intergranular fracture for the uncharged BT4 specimen (Fig. 10(e)) confirmed this. The decomposed cementites are located mainly near grain boundaries, which caused more hydrogen trapped there and thus promoted brittle intergranular fracture for the hydrogen-charged BT4 specimen (Fig. 10(f)). 108
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
Fig. 11. SEM fractographs of (a) uncharged and (b) hydrogen-charged specimens (QT4) in crack initiation region.
Fig. 12. SEM micrographs showing the longitudinal section of fractured SSRT specimens in the region beneath the fracture surface. (a) B1; (b) BT5.
When the tempering temperature was increased to 500 °C, σNH increases significantly though σN0 decreases slightly. This result can be explained as follows. Firstly, as can be seen from Figs. 3–5, most (~70%) of the original RA (the RA in the as-hot rolled condition) decomposed into ferrite and cementite, and it is reasonable to believe that all the blocky M/A constituents decomposed into ferrite and cementite when tempered at 500 °C. Therefore, the microstructure of the BT5 specimen became more homogeneous compared to other specimens except the one of QT4. Moreover, both the decomposed and precipitated cementites are strong hydrogen traps. Thus, it is suggested that the distribution of hydrogen trapping sites is more uniformly distributed in the BT5 specimen, and thus the amount of hydrogen per trapping sites should be lower although the total hydrogen content is the highest among the tested specimens. This situation lowered the local plasticity caused by internal hydrogen, and thus enhanced the HE resistance [23]. Secondly, there is still a small amount of RA (~2.4 vol%) which is supposed to be primarily in the form of filmy morphology, and it was revealed that filmy RA is less susceptible to HE than the blocky RA [26]. Finally, the highest tempering temperature is also beneficial to the improvement of HE resistance as mentioned above. Interestingly, tempering at 500 °C did not change much both UTS for the smooth specimen (decrease by ~44 MPa) and σN0 for the notched specimen (increase by ~55 MPa) in the uncharged condition, but significantly enhanced the HE resistance (increase by 715 MPa, an increase of ~93.5% compared to the as-hot rolled specimen), which is even much higher than that of the QT4 specimen at almost identical strength level. These results strongly suggest that tempering at 500 °C is a promising strategy to notably enhance the HE resistance of the tested bainitic rail steel. The variation trend of both σN0 and σNH is almost the same with increasing tempering temperature up to 400 °C, and thus HEI remains almost unchanged; when the steel was tempered at 500 °C, σNH increases significantly while σN0 decreases slightly, and thus HEI decreases significantly. It should be noted that the BT5 specimen has the highest hydrogen content at the same hydrogen charging current density (i.e., the identical amount of hydrogen content on the surface of specimen [30]), indicating that the specimen possessed a considerable number of hydrogen trapping sites. This result suggests that both the extrinsic (industrial, i.e., hydrogen charged at the same current density) and the intrinsic (compare at similar hydrogen content) HE resistance [23] of the BT5 specimen are superior to those of other specimens. 5. Conclusions (1) The as-hot rolled specimen exhibits a mixed microstructure consisting of granular bainite and ~9 vol% lath martensite. The applying of tempering treatment gives rise to the gradual decomposition of M/A constituents as well as the tempering of the martensite, and almost all the blocky M/A decomposed into ferrite and cementite when tempered at 500 °C, causing a sharp decrease of the RA fraction. (2) Both notch tensile strength of uncharged (σN0) and hydrogen-charged (σNH) specimens firstly increase when tempered at 280 °C, and then they slightly decrease with increasing tempering temperature up to 400 °C. There is a sharp increase of σNH while a 109
Engineering Failure Analysis 93 (2018) 100–110
Y. Zhang et al.
continuous decrease of σN0 when tempered at 500 °C. As a result, the susceptibility index to HE (HEI) still maintains at high level with increasing tempering temperature up to 400 °C, and then it decreases significantly when tempered at 500 °C. (3) The hydrogen-charged SSRT fracture surface is dominated by brittle intergranular failure with a few ductile tearing for the as-hot rolled specimen and specimens tempered at temperatures no more than 400 °C, while it is primarily quasi-cleavage fracture for the 500 °C tempered specimen. Massive hydrogen-induced cracks are found to be initiated at blocky M/A constituents of the ashot rolled specimen, while only isolate voids are found in the 500 °C tempered specimen. (4) Tempering treatment could enhance the HE resistance and decrease the susceptibility to HE of the as-hot rolled bainitic rail steel, and extremely higher HE resistance and rather lower susceptibility to HE could be obtained when tempered at 500 °C at less expense of strength and ductility. Acknowledgements This work was supported by the Fundamental Research Funds for the Central Universities (No.2014JBM108 and No.2017RC024). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]
K.M. Lee, A.A. Polycarpou, Wear of conventional pearlitic and improved bainitic rail steels, Wear 259 (2005) 391–399. W. Zhong, J.J. Hu, Z.B. Li, Q.Y. Liu, Z.R. Zhou, A study of rolling contact fatigue crack growth in U75V and U71Mn rails, Wear 271 (2011) 388–392. H.A. Aglan, Z.Y. Liu, M.F. Hassan, M. Fateh, Mechanical and fracture behavior of bainitic rail steel, J. Mater. Process. Technol. 151 (2004) 268–274. U.P. Singh, B. Roy, S. Jha, S.K. Bhattacharyya, Microstructure and mechanical properties of as rolled high strength bainitic rail steels, Mater. Sci. Technol. 17 (2001) 33–38. K. Sawley, J. Kristan, Development of bainitic rail steels with potential resistance to rolling contact fatigue, Fatigue Fract. Eng. Mater. Struct. 26 (2003) 1019–1029. P. Pointner, High strength rail steels-the importance of material properties in contact mechanics problems, Wear 265 (2008) 1373–1379. Z.L. Tan, B.F. An, G.H. Gao, X.L. Gui, B.Z. Bai, Analysis of softening zone on the surface of 20Mn2SiCrMo bainitic railway switch, Eng. Fail. Anal. 47 ( (2015) 111–116. K.K. Wang, Z.L. Tan, G.H. Gao, X.L. Gui, R.D.K. Misra, B.Z. Bai, Ultrahigh strength-toughness combination in bainitic rail steel: the determining role of austenite stability during tempering, Mater. Sci. Eng. A 662 (2016) 162–168. F.C. Zhang, C.L. Zheng, B. Lv, T.S. Wang, M. Li, M. Zhang, Effects of hydrogen on the properties of bainitic steel crossing, Eng. Fail. Anal. 16 (2009) 1461–1467. C.L. Zheng, B. Lv, C. Chen, Z.G. Yan, F.C. Zhang, L.H. Qian, Hydrogen embrittlement of a manganese-aluminum high-strength bainitic steel for railway crossings, ISIJ Int. 51 (2011) 1749–1753. Y.G. Li, F.C. Zhang, C. Chen, B. Lv, Z.N. Yang, C.L. Zheng, Effects of deformation on the microstructures and mechanical properties of carbide-free bainitic steel for railway crossing and its hydrogen embrittlement characteristics, Mater. Sci. Eng. A 651 (2016) 945–950. W.J. Hui, Z.B. Xu, Y.J. Zhang, X.L. Zhao, C.W. Shao, Y.Q. Weng, Hydrogen embrittlement behavior of high strength rail steels: a comparison between pearlitic and bainitic microstructures, Mater. Sci. Eng. A 704 (2017) 199–206. Y. Luo, J.M. Peng, H.B. Wang, X.C. Wu, Effect of tempering on microstructure and mechanical properties of a non-quenched bainitic steel, Mater. Sci. Eng. A 527 (2010) 3433–3437. Z.B. Xu, W.J. Hui, Z.H. Wang, Y.J. Zhang, X.L. Zhao, X.M. Zhao, Mechanical properties of a microalloyed bainitic steel after hot forging and tempering, J. Iron Steel Res. Int. 24 (2017) 1085–1094. Y. Nakatani, T. Higashi, K. Yamada, Effect of tempering treatment on hydrogen-induced cracking in high-strength steel, Fatigue Fract. Eng. Mater. Struct. 22 (1999) 393–398. F. Sarıoglu, The effect of tempering on susceptibility to stress corrosion cracking of AISI 4140 steel in 33% sodium hydroxide at 80 °C, Mater. Sci. Eng. A 315 (2001) 98–102. S.J. Lee, J.A. Ronevich, G. Krauss, D.K. Matlock, Hydrogen embrittlement of hardened low-carbon sheet steel, ISIJ Inter. 50 (2010) 294–301. Y.J. Zhang, C. Zhou, W.J. Hui, H. Dong, Effect of C content on hydrogen induced delayed fracture behavior of Mn-B type steels, J. Iron Steel Res. 26 (5) (2014) 49–55. S.Z. Han, W.J. Hui, R.P. Liu, C.W. Shao, Effect of tempering temperature on hydrogen-induced delayed fracture behavior of 30CrMnSiA steel, Trans. Mater. Heat Treatment 35 (7) (2014) 114–119. X.F. Li, J. Zhang, S.C. Shen, Y.F. Wang, X.L. Song, Effect of tempering temperature and inclusions on hydrogen-assisted fracture behaviors of a low alloy steel, Mater. Sci. Eng. A 682 (2017) 359–369. W.J. Hui, Y.J. Zhang, X.L. Zhao, C.W. Shao, K.Z. Wang, W. Sun, T.R. Yu, Influence of cold deformation and annealing on hydrogen embrittlement of cold hardening bainitic steel for high strength bolts, Mater. Sci. Eng. A 662 (2016) 528–536. Y.L. Zhou, T. Jia, X.J. Zhang, Z.Y. Liu, R.D.K. Misra, Investigation on tempering of granular bainite in an offshore platform steel, Mater. Sci. Eng. A 626 (2015) 352–361. D.H. Shim, T. Lee, J. Lee, H.J. Lee, J.Y. Yoo, C.S. Lee, Increased resistance to hydrogen embrittlement in high-strength steels composed of granular bainite, Mater. Sci. Eng. A 700 (2017) 473–480. B.A. Szost, R.H. Vegter, P.E.J. Rivera-Díaz-Del-Castillo, Hydrogen-trapping mechanisms in nanostructured steels, Metall. Mater. Trans. A 44 (2013) 4542–4550. F.G. Wei, T. Hara, T. Tsuchida, K. Tsuzaki, Hydrogen trapping in quenched and tempered 0.42C-0.30Ti steel containing bimodally dispersed TiC particles, ISIJ Int. 43 (2003) 539–547. X. Zhu, K. Zhang, W. Li, X.J. Jin, Effect of retained austenite stability and morphology on the hydrogen embrittlement susceptibility on quenching and portioning treated steels, Mater. Sci. Eng. A 658 (2016) 400–408. X.B. Shi, W. Yan, W. Wang, L.Y. Zhao, Y.Y. Shan, K. Yang, HIC and SCC behavior of high-strength pipeline steels, Acta Metall Sin (Engl Lett) 28 (2015) 799–808. N. Nanninga, J. Grochowsi, L. Heeldt, K. Rundman, Role of microstructure, composition and hardness in resisting hydrogen embrittlement of fastener grade steels, Corros. Sci. 52 (2010) 1237–1246. X.C. Xiong, B. Chen, M.X. Huang, J.F. Wang, L, the effect of morphology on the stability of retained austenite in a quenched and partitioned steel, Scr. Mater. 68 (2013) 321–324. M. Koyama, E. Akiyama, K. Tsuzaki, Hydrogen embrittlement in Al-added twinning-induced plasticity steels evaluated by tensile tests during hydrogen charging, ISIJ Int. 52 (2012) 2283–2287.
110
