PSI - Issue 42

Simon Vander Vennet et al. / Procedia Structural Integrity 42 (2022) 813–820 S. Vander Vennet / Structural Integrity Procedia 00 (2019) 000–000

814

2

carbon content is achieved by allowing carbon di ff usion into the austenite ( γ ) during the partitioning step and is necessary to stabilise the austenite during the final quenching step down to room temperature, at which point it is often termed retained austenite (RA). The RA is metastable at room temperature and provides increased ductility by the TRIP-e ff ect. However, as is the case with most high strength steels (Laureys et al. (2016); Depover et al. (2018, 2014); Wang et al. (2007); Koyama et al. (2014)) Q&P steel is particularly susceptible to hydrogen embrittlement (Lovicu et al. (2013); Zhu et al. (2014)). Retained austenite plays a crucial role, since it has both advantages and disadvantages with respect to hydrogen. Initially, it was assumed that RA could be advantageous against hydrogen embrittlement due to its high solubility and low di ff usivity (Caskey and Sisson (1981); Sun et al. (1989)), thereby limiting the presence of hydrogen in the susceptible body centered cubic (BCC) matrix. However, when Q&P steel undergoes mechanical deformation and the metastable RA transforms into martensite, the assumption that austenite remains present in the material is no longer valid. At this point, the high hydrogen solubility of RA turns into a disadvantage since the hydrogen that was dissolved in the austenite phase is inherited by the newly formed fresh martensite, which has a considerably lower hydrogen solubility (Ronevich et al. (2012)) and an increased susceptibility to hydrogen embrittlement due to its high dislocation density and residual stresses after transformation (Luppo and Ovejero-Garcia (1991)). Previous studies on the hydrogen embrittlement of Q&P steel already showed that the interface between RA and ferrite / martensite is a preferential initiation site for transgranular cracks (Zhu et al. (2014); Lovicu et al. (2013)). Therefore, the e ff ect of stress in Q&P steel, and steels containing RA in general, should not be neglected since it essentially destabilises the metastable RA. Another factor to consider is the interplay between stress and hydrogen. The e ff ect of stress on hydrogen di ff usivity was illustrated by multiple studies using electrochemical hydrogen permeation (Van den Eeckhout et al. (2020); Zhao et al. (2016); Kim and Kim (2012)). A study by Van den Eeckhout et al. (2020) demonstrated the e ff ect of stress by conducting permeation experiments on dual phase (DP) steel stressed at a constant applied load. They found that di ff usivity increased with increasing elastic stress, similar to the results by Zhao et al. (2016). The authors proposed that the increase of the volume of the unit cell due to the applied elastic stress, which also causes an expansion of the interstitial sites, facilitates hydrogen di ff usion. In the plastic regime, i.e. when the constant load applied on the specimen exceeded 100% of the yield stress, hydrogen di ff usivity decreased again due to the formation of lattice defects, such as dislocations, which act as hydrogen traps and slow down di ff usion. In this study, a similar approach will be used to characterise the e ff ect of a constant applied load in steels containing retained austenite in di ff erent BCC matrices. Samples of Q&P and TRIP assisted steel will be subjected to permeation experiments at a constant applied load and the results will be compared with those on dual phase steel obtained by Van den Eeckhout et al. (2020).

2. Materials and methods

2.1. Material characterisation

Three industrial multiphase materials were compared: dual phase steel (DP) (Van den Eeckhout et al. (2020)), transformation induced plasticity steel (TRIP) and quenching and partitioning steel (Q&P). These steel grades will henceforth be referred to by their respective abbreviations given between brackets. The compositions and microstruc tural phases are given in Table 1. In order to determine the RA fraction of TRIP and Q&P, X-ray di ff raction (XRD) was used. These experiments were performed on a Siemens Kristalloflex D5000 di ff ractometer with a Mo-K α source operated at 40 kV and 40 mA, measuring the di ff raction spectrum over a range of 2 θ from 25 ◦ to 40 ◦ . The used range of di ff raction angles contains BCC peaks and two face-centered cubic (FCC) peaks corresponding to the α { 200 } , α { 211 } and γ { 220 } , γ { 311 } planes respectively. Using the integrated intensities of these peaks, the volume fraction of RA was determined using the formula by van Dijk et al. (2005), based on powder di ff raction standards. Prior to phase quantification, the raw data was post-processed by subtracting the background and the K α 2 doublet peaks.

Made with FlippingBook - Online catalogs