Concentration-dependent diffusion coefficient

User Support
1

Hello! It is known that with increasing concentration the diffusion coefficient decreases. So is it possible to set up such a dependency in FESTIM? As far as I understand, it is only the temperature dependence of the diffusion coefficient that can be easily taken into account.

2

Would you mind sharing a reference of this known fact? :smiley:

Out of the box this isn’t possible. However implementing this shouldn’t be very hard, we already do this for heat transfer (with temperature-dependent conductivities). Would you mind opening an issue on the GitHub repository?

3

Sure!

  1. Mütschele T., Kirchheim R. Segregation and diffusion of hydrogen in grain boundaries of palladium //Scripta metallurgica. – 1987. – Т. 21. – №. 2. – С. 135-140.

  2. Svoboda J., Fischer F. D. Modelling for hydrogen diffusion in metals with traps revisited //Acta materialia. – 2012. – Т. 60. – №. 3. – С. 1211-1220.

  3. Zaika Y. V., Rodchenkova N. I. Computational algorithm for solving the nonlinear boundary-value problem of hydrogen permeability with dynamic boundary conditions and concentration-dependent diffusion coefficient //Computer research and modeling. – 2024. – Т. 16. – №. 5. – С. 1179-1193.

I’m interested in this assumption because it could explain why the TDS spectra of austenitic steel samples exposed in deuterium gas have such a late start of active desorption and a sharper end. After exposure of the samples in gas for 100 hours at high temperature, most of the interstitial sites are probably filled with deuterium so this could hinder deuterium diffusion in the lattice. When I try to model TDS spectrum of such sample, I get broad and long tail on the right side of the model spectrum (deuterium continues to desorb even after 1200 K), and this discrepancy could be due to the fact that in the experiment after most of the deuterium has left the sample, the remaining deuterium becomes easier to desorb (because of the concentration dependence of the diffusion coefficient).

4

Thanks! Please open an issue on GitHub describing how you would like to use (from a user perspective) this feature with FESTIM