[QE-users] Supercells and Hybrid exchange-correlation

[jmasnaghetti]

Dear QE users,

I'm currently working on Silicon-Germanium random alloys using hybrid 
exchange functional and supercells using QE 7.2 and 7.3.1. Running some 
test with the two pure bulk crystal, i have found that the energies per 
ion are not consistent between the classic 2 ions cells and bigger 8 or 
16 atoms supercells if i use ultra-soft pseudos. On the other hand, the 
problem does not seem to arise when i run similar tests with Norm 
Conserving pseudos, for comparison energy per ions between the 2 atom 
cell and the 8 atom cell differs of 0.1 or 0.2 meV for the NC pseudo, 
while the same difference is of some tens of meV for the US pseudo. I do 
not think that this discrepancy is generated by some kind of convergence 
problem, and i try to keep the products between the number of k points 
and atoms consistent across the different cells to prevent errors given 
by inconsistent Brillouin zone samplings. I'll include some of the input 
files with this mail as a reference.

Thank you in advance,
Jacopo Masnaghetti, Università degli studi di Milano
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: ge-new-sc8.exx.in
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20241113/b456725c/attachment.ksh>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: ge-new.exx.in
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20241113/b456725c/attachment-0001.ksh>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Ge4pbe0.pbe-rrkj3.UPF
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20241113/b456725c/attachment-0002.ksh>