Trouble achieving GS convergence with tolvrs

Running Abinit Ground state
1

Hello everyone,

I am doing a ground state calculation for a relaxed functionalized bilayer graphene system and the potential residual seems to have reached a plateau.

tolvrs_conv
tolvrs_conv1280×960 48 KB

This is odd to me since I have done the same calculation on a similar system and the convergence was reached for that case, albeit very slowly. Any ideas on how to get out of this plateau?

Files:
bi_graphene-C6H4COOH.abi (11.2 KB)
bi_graphene-C6H4COOH.abo (329.4 KB)
log.txt (1.1 MB)

2

Hi Tony,

As a general advice, if one of your tolxxx criterion gets stuck in the SCF, you might try out one of the other ones. I personally recommend in more complex systems like yours to convergence against the forces instead of the potential and you might avoid getting trapped in a local minimum for thousands of steps.
Cheers,
Bogdan

1 Like