PSI - Issue 54

7

Liese Vandewalle et al. / Procedia Structural Integrity 54 (2024) 180–187 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

186

a) b) Figure 6: a) TDS spectra of the Fe-C-V alloy austenitized at 950°C and tempered in the hydrogen containing atmosphere. Three different heating rates were used (1200°C/h, 900°C/h and 600°C/h). The corresponding Choo-Lee plot is given in the inset. b) TDS spectra of Fe-C-Ti alloys (heating rate 900°C) containing undissolved carbides, tempered in the same hydrogen containing atmosphere (Vandewalle et al. (2023)). In a previous study, the use of local equilibrium partitioning of the H atoms between lattice and trapping sites (see equation 1) allowed to estimate the E B related to the TiC trapping sites. However, in this case the equation can give additional insight in the trapping sites by using the 22 kJ/mol as binding energies and calculate the corresponding number of trapping sites present. ∗( − ) ∗( − ) = ∗ exp (− ) (1) Whith c L and c T are the lattice and trapped H concentration, respectively, n L the concentration of interstitial lattice sites, and n T the trap density, in mol/m³, E B in J/mol, A the pre-exponential factor (related to the entropic contribution), R the ideal gas constant and T the temperature in K. The trapped H concentration c T was determined by the hot extraction measurements while c L was assumed to be equal to the equilibrium concentration induced in the steel at 600°C for a H partial pressure of 0.05 bar, based on Sievert’s law (Nagumo (2016). For A and n L the same values as in the previous study of Vandewalle et al. (2023) were used. This resulted in a trap occupation of only 6.3*10 -8 to 3.9*10 -7 , depending on n L value. Hence, an extremely high amount of trapping sites, i.e. interface C-vacancies should be present. Alternatively, assumption of local equilibrium might not be valid here or trapping mainly occurred for very short times at a lower temperature when the sample moved upwards through the furnace for extraction. One could also suggest that H might not be linked to the undissolved carbides but rather to the ferritic matrix and trapping sites therein, i.e. grain boundaries. However, no uptake was reported for the sample austenitized at 1250°C, which would implicate the martensitic matrix being unable to capture H at elevated temperatures. 4. Conclusions This study evaluated H uptake due to trapping at elevated temperatures in vanadium carbides by subjecting a Fe C-V alloy to a quench and temper treatment, where the tempering environment was either air or a 5% H 2 -95% N 2 mixture, and comparison was made to the titanium carbides. Strong differences could be observed for both types of carbides. The titanium carbides showed a strong uptake of H due strong trapping (E B =80-90kJ/mol) at the carbon vacancies in the bulk of the carbides. This resulted in a desorption peak at high temperatures (> 500°C). Moreover, the undissolved TiC were found to be major trapping site providers but limited contribution by precipitated could not be excluded. For the vanadium carbides, the precipitated carbides showed to be unable to trap H at elevated temperatures in dilute H 2 atmospheres. A small H uptake was observed in presence of undissolved carbides. Contrary to TiC, this uptake corresponded to a low-temperature peak (at 180°C) in the TDS spectrum for which an E a of 27 kJ/mol could be obtained. Differences between TiC and VC might be linked to differences in undissolved carbide size and amount, or binding energies. Based on literature, the trapping in the undissolved vanadium carbides