[QE-users] Relaxation in silicon with a divacancy

[David Fernandes Machado]

Hello everyone,
I am currently studying a supercell composed of 510 silicon atoms, including a divacancy. I am conducting a relaxation calculation using the PBE pseudopotential. Since silicon is tetravalent, each divacancy causes three neighboring silicon atoms to lose a bond. At a total charge of 0, I anticipate two distinct equilibrium structures. * The three silicon atoms form two bonds, resulting in two short distances (~3 Å) and one longer distance (~3.5 Å) between them. * The three silicon atoms form only one bond, leading to one short distance and two longer distances between them.
To achieve these two structures, I initialize the relaxation calculation by adjusting the distances between the three atoms to resemble the expected configurations. To expedite the calculation, I perform the relaxation in two stages: the first with force convergence (forc_conv) set to 10−310^{-3}10−3 and a wavefunction cutoff energy (ecutwfc) of 10 Ry, and the second with forc_conv set to 10−410^{-4}10−4 and ecutwfc of 25 Ry.
However, despite trying various approaches, I consistently observe three long distances of approximately 3.4 Å at the end of the second relaxation step, even though the structure appears correct after the first step.
Has anyone encountered a similar issue? Any suggestions on how to resolve this problem would be greatly appreciated.
Thank you.
Best regards,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240820/5f3a7480/attachment.html>