Atomic-scale study of stacking faults in Zr hydrides and implications on hydride formation

Abstract

We performed atomic-scale ab initio calculations to investigate the stacking fault (SF) properties of the metastable ζ-ZrH zirconium hydride. The effect of H near the SF was found to entail the existence of negative SF energies, showing that the ζ compound is probably unstable with respect to shearing in the basal plane. The effect of temperature on SFs was investigated by means of free energy calculations in the quasiharmonic approximation. This evidenced unexpectedly large temperature effects, confirming the main conclusions drawn at 0 K, in particular the ζ mechanical instability. The complex behaviour of H atoms during the shear process suggested ζ-hcp → ZrH[Formula: see text]-fcc as a plausible shear path leading to an fcc compound with same composition as ζ. Finally, as shown by an analysis based on microelasticity, this ZrH[Formula: see text]-fcc intermediate compound may be relevant for better interpreting the currently intricate issue of hydride habit planes in zirconium.

Ab initio thermodynamics of complex alloys: The case of Al- and Mn-doped ferritic steels

“…Since the lattice parameter that minimizes the free energy of the system depends on temperature (thermal expansion), it would be necessary, in order to refine the thermodynamic description of the system at zero pressure, to calculate phonon frequencies for different lattice parameters and take those values that minimize the vibration free energy, which constitutes the quasiharmonic approximation. However, this kind of study has shown [44] that even at relatively elevated temperature (800 K) the free energy difference between the harmonic (adopted in this work and that consists in calculating F vib at fixed volume) and quasiharmonic approximation is only a few meV/atom, a degree of refinement of second 7 order with respect to other uncertainties in the ab initio input data. We therefore opted to work at constant lattice parameter (the 0 K equilibrium value) to preserve computational resources.…”