Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly  
We are NATO-free zone

Re: A correct way to avoid orbital reordering after SCF convergence


Remove $scf dem=.t. $end and set SCFTYP=none in $contrl. This will force CI calculation using MOREAD orbitals without SCF.

On Tue Dec 4 '12 12:23pm, Ivan Fedyanin wrote
>I'm trying to perform a simple CI calculation of benzene molecule, using three pi (# 17, 20, 21) and three pi* (22,23,24) active orbitals. This setup requires orbital reordering. My input looks like this:

> $contrl runtyp=energy coord=cart cityp=aldet $end
> $basis gbasis=N31 ngauss=6 ndfunc=1 $end
> $guess guess=moread norder=1 iorder(17)=18,19,17 norb=24 prtmo=.t. $end
> $cidet ncore=18 nels=6 nact=6 $end
> $scf dem=.t. $end
> $data ... $end
> $vec ... $end

>I've noticed that after guess reading orbitals are in correct order, but after SCF cycle they are sorted by energy again and thus the choice of the active orbitals in the following CI calculation is wrong. I've found that the use of $scf dem=.t. $end prevents this. The question is: is this the correct way, is there any other way to prevent the reordering or am I doing something wrong?

[ Previous ] [ Next ] [ Index ]           Wed Dec 5 '12 5:17pm
[ Reply ] [ Edit ] [ Delete ]           This message read 882 times