Nitrogen marker experiments in austenitic stainless steel for identification of trapping/detrapping processes
Expanded austenite has been established as a hard and wear resistant layer obtained by surface modification while generally maintaining the excellent corrosion behavior of the soft austenitic base material. Loss of corrosion resistance is supposedly correlated with the formation of Cr-N precipitates, which strongly reduces the mobility of chromium atoms and prevents the continuous formation of a passivating chromium oxide layer.
Previous work uses a trapping/detrapping mechanism during nitrogen insertion at elevated temperatures to explain the peculiar diffusion behaviour, however limited to a single temperature of 400 °C. This work is now extended to a more comprehensive temperature range of 370 to 550 °C with the higher end of the temperature range comprising the zone where the actual formation of CrN precipitates occurs during the nitriding of 1 – 2 hours. Sequential 14N/15N ion implantation in a system equipped with a low-energy broad-beam ion source and a fast in situ XRD system was performed. At lower temperatures, the initially implanted isotope is diffusing in front of the later implanted isotope with a finite amount of intermixing and a surface concentration of the initial isotope reaching zero at.%. With progressively increasing temperature, the intermixing of both isotopes becomes more pronounced and the surface concentration of the initial isotope is increasing. A strong correlation between this trapping of the initially implanted isotope and the onset of CrN formation is observed. For the extreme case of a closed CrN layer near the surface nitrided at 550 °C, the initially implanted isotope is distributed closer to the surface than the second implanted isotope.
The results demonstrate that highly efficient trapping only occurs during the formation of CrN precipitates with no other intermediate trapping/detrapping processes observed. Isolated pairs of Cr-N atoms, already leading to a delayed transport, as reported in the literature, are not found in this study. However, the main experimental difficulty is the correct identification of the onset of CrN precipitation without in situ XRD as the initial nucleation leads to grain sizes below 5 nm, invisible in XRD. Their existence can only be inferred from the continuous time evolution of the intensity of the substrate reflections and of the expanded phase reflections.