PSI - Issue 68

Renata Latypova et al. / Procedia Structural Integrity 68 (2025) 1115–1120 R. Latypova et al./ Structural Integrity Procedia 00 (2025) 000–000

1116

2

Keywords: steel; martensite; low-temperature tempering; hydrogen diffusion; hydrogen concentration; hydrogen embrittlement

1. Introduction The rising interest in hydrogen (H) as a green energy carrier has accelerated research into materials that can provide a safe infrastructure for its transport and storage. High-performance ultrahigh-strength steels (UHSS) are wanted candidates for this, but their utilisation requires further microstructural optimisation to minimize the risk of hydrogen embrittlement (HE). Low-temperature tempering (LTT) is a common heat treatment for enhancing ductility and toughness properties of UHSS, and it is often applied to cold-rolled steel sheets for automotive components through the paint baking process, where sheets are heated to 120–180 °C (Toji et al., 2012; Venezuela et al., 2020a). LTT also shows promise in improving resistance to HE, but the precise underlying mechanisms are not fully understood. During LTT, carbon (C) diffuses to dislocations and grain boundaries (25 – 100 °C), transition carbides such as eta (η) and epsilon (ɛ) carbides are formed (100 – 200 °C) and retained austenite (RA) transforms to ferrite/cementite (200 – 350 °C) (Porter et al., 2014). Because of C diffusion, carbon-saturated martensite will experience loss in tetragonality and partial relief of residual stresses (Bhadeshia and Honeycombe, 2006; Krauss, 2014). However, these phenomena are highly dependent on the amount of available C. For example, in low-C steels that do not have apparent tetragonality, there might not be sufficient driving force for transition carbide formation. These temperature-dependent processes can also have varying impacts on H diffusion, trapping, and susceptibility to HE. In 1976, an electrochemical hydrogen permeation (EP) test was used to study H diffusion and solubility in as quenched martensitic steels with C contents ranging from 0.13 to 1.1 wt.%. These steels were tempered at various temperatures between 100 and 650 °C. The results showed that as diffusivity increased in the tempered samples, solubility decreased. The as-quenched martensite exhibited the slowest H diffusion and the highest solubility, likely due to the greater presence of lattice imperfections and microstrain (Sakamoto and Mantani, 1976). Similarly, Venezuela et al. observed faster H diffusion in 180 °C tempered MS1500 steel (0.19 C), which also showed enhanced resistance to HE. This faster diffusion is thought to result from fewer H traps, as C atoms occupy dislocations, forming Cottrell atmospheres, along with a reduction in tetragonality and microstrain. However, in MS1700 (0.24 C) steel, H diffusion slowed down after the same heat treatment as for MS1500 and HE susceptibility increased. With higher C content in MS1700, the formation of transition carbides slows down H diffusion since they can increase H-trapping (Venezuela et al., 2020b). LTT is typically applied to reheated and quenched (RQ) steels to optimize mechanical properties following the final quenching process. However, UHSS can also be produced through a novel direct-quenching (DQ) process, a thermomechanical rolling method integrated with quenching (Kömi et al., 2016). During direct-quenching, complete suppression of C diffusion is not achievable, and diffusional phase transformations, such as fine cementite precipitation, can occur when the martensite start (M s ) temperature is sufficiently high. This phenomenon, known as auto-tempering, improves the toughness of the as-quenched steel, making DQ steels more cost-effective as LTT is often unnecessary. Despite this, the influence of additional LTT on HE susceptibility in DQ steels remains unclear. This study focuses on additional LTT treatment of DQ steel and how it affects H diffusion, trapping, and HE susceptibility. 2. Test materials and methods Four materials were studied: DQ steel (0.25C-0.1Si-0.25Mn wt.%) with auto-tempered lath-martensitic microstructure, and the same tempered at 50 °C (aging, T50), 150 °C (T150), and 250 °C (T250), all with a 1-hour holding time. The selected tempering temperatures were used to quantify the effects of C segregation (50 °C - 150 °C), and transformation of RA to cementite (250 °C). The formation of transition carbides during LTT is unlikely, as the driving force is insufficient due to previous auto-tempering. The mechanical properties of materials such as yield strength (YS), tensile strength (TS), and hardness are given in Table 1 along with RA concentrations measured with X-ray diffraction (XRD). Differential scanning calorimetry (DSC) measurements also confirmed the absence of RA in T250 when other materials had a clear transformation peak at 280 ° C (Figure 1).