The ABINIT Users Mailing List ( CLOSED )

[Masayoshi Mikami <mmikami@yk.rim.or.jp>]

Dear Aldo,

Just one more thing: 
In my previous mail, I assumed that your xred in the input 
were already well optimized before trying "optcell 1". 
If not, please try ionmov 2(or 3) without unitcell optimization.

If "optcell 1" (or optcell 2) do not work to get converged 
results, your initial geometry might be far from the plausible
geometry. In such case, you might want to prepare dataset 
with different volumes, and for each volume you could try 
"optcell 3" with "ionmove 2 or 3". (Still, for each volume,
the initial xred should be optimized with ionmov 
before trying "optcell 3")
The equilibirium volume may then be obtained with fitting
such as Murnaghan's equation of state.

Best wishes,
Masayoshi