Effects of metastable austenite and martensite on the susceptibility to hydrogen embrittlement of TBF steels, The

Abstract

A series of four experimental TRIP-aided bainitic ferrite (TBF) sheet steels was used to investigate the influence of microstructure (in particular secondary phase austenite) on hydrogen absorption and hydrogen embrittlement susceptibility. The TBF steels were designed to have similar carbon equivalent values (≈ 0.6) and similar tensile strengths (950-1150 MPa), allowing the effects of microstructure to be isolated. Quasi-static tensile tests were performed on the TBF steels after various durations of electrochemical hydrogen charging in order to characterize susceptibility to hydrogen embrittlement. Melt extraction was used to quantify the hydrogen absorbed by each steel during hydrogen infusion. The microstructure of each steel was characterized using electron back-scatter diffraction (EBSD) and x-ray diffraction (XRD). In particular, microstructural attributes believed to potentially contribute to hydrogen retention, such as austenite and martensite/austenite (MA) constituent volume fractions, grain and phase boundary areas, and austenite aspect ratio, were quantified. Variations in these microstructural components were compared to observed differences in hydrogen embrittlement susceptibility and hydrogen absorption behavior to identify the influence of key microstructural characteristics. Increased austenite volume fraction and/or increased austenite phase boundary area were found to have the dominant influence on increased hydrogen absorption and hydrogen embrittlement susceptibility. Increased austenite aspect ratio was found to have a minor influence. One of the TBF steels and an experimental dual phase (DP) steel were used to investigate the effects of secondary phases on hydrogen retention and microcrack initiation and propagation. The DP steel was designed to have a microstructure that was very similar to the TBF steel, but with untempered martensite as the secondary phase instead of austenite. In this way, the influence of the secondary phases could be isolated. Specimens of the TBF and DP steels were electrochemically hydrogen charged and aged. Melt extraction was then used to determine the remaining concentration of hydrogen after various durations of aging. The TBF steel was found to retain hydrogen for longer periods of time than the DP steel, but not indefinitely. This again demonstrated that austenite and/or its phase boundary was the primary microstructural feature that stored hydrogen, and that austenite was able to retain hydrogen significantly longer than untempered martensite. Tensile tests of hydrogen infused specimens were conducted on the TBF and DP steels over a range of strain rates ("ε" ̇ = 10 2, 10 3, 10 4, and 10 5 s-1). Increasing strain rate limits the mobility of hydrogen over the duration of the test. The wide range of strain rates thus allowed the effects of hydrogen stored within the secondary phase to be observed both with and without hydrogen redistribution during deformation. Microcrack populations were quantified after failure using FESEM. Microcracks were found to initiate within untempered martensite (in DP steel) and strain induced martensite (in TBF steel) even at the highest strain rate. This implies both untempered martensite and austenite store enough hydrogen to induce microcracking without additional accumulation during deformation. Additionally, decreasing strain rate was found to enhance microcrack initiation and propagation in the TBF steel more so than in the DP steel. This is evidence that hydrogen is released during the austenite to martensite transformation, increasing the diffusible hydrogen within the system and assisting in the nucleation and propagation of microcracks.