Firefly and PC GAMESS-related discussion club

 
Learn how to ask questions correctly  
 
 

 

Alex Granovsky
gran@classic.chem.msu.su

This will not work. CI optimization will reach a point where state 2
and state 3 become quasi-degenerated and then start oscillating
between points where the third state corresponds either to the
diabatic state #2 or to the diabatic state #3

Kind regards,
Alex Granovsky


On Fri Feb 27 '15 7:47am, Abhineet Agarwal wrote
------------------------------------------------
>Hi,
>Using the current code, I had hoped to find it out using istate=1 and jstate=3 as we would be trying to optimize the energy difference between these two states, wouldn't that help me to solve the problem. And if it does, I haven't been successful to locate a stationary point in 2000 or 3000 stationary point locations.

>The input parameters I have used are
> $CONTRL SCFTYP=mcscf RUNTYP=optimize inttyp=hondo
>         NZVAR=1 MAXIT=200 fstint=.t. gencon=.t.
>         EXETYP=run d5=.t. icut=11
> $END
> $moorth nostf=1 nozero=1 tole=0 tolz=0 $end
> $SYSTEM TIMLIM=5000000 mwords=250
> nojac=1 kdiag=0 $END
> $STATPT NSTEP=2000 METHOD=conic
> hssend=.t. $END
> $SCF dirscf=.F. diis=.T. soscf=.F. shift=.F.
> $end
> $BASIS GBASIS=N31 NGAUSS=6 ndfunc=1 $END
> $GUESS GUESS=MOREAD NORB=21 $END
> $MCSCF NUMFO=20 MAXIT=200 CISTEP=GUGA
> soscf=.t. fullnr=.f. istate=1 fors=.t.
> acurcy=5d-8 engtol=5d-13 micit=20 $END
> $p2p p2p=.t. dlb=.t. $END
> $smp csmtx=.t. call64=1 $end
> $DRT NMCC=17 NDOC=2 NVAL=2 FORS=.t. $END
> $GUGDIA nstate=3 itermx=2200 $END
> $GUGDM2 wstate(1)=1,1,1 $END
> $gugem pack2=.t. $end
> $trans cuttrf=1d-12 mptran=2 dirtrf=.t. aoints=dist altpar=.t. mode=111 $END
> $mcaver jstate=3 conic=2 multiw=.t. $end
>

>On Fri Feb 27 '15 2:54am, Alex Granovsky wrote
>----------------------------------------------
>>On Wed Feb 25 '15 2:32pm, Abhineet Agarwal wrote
>>------------------------------------------------
>>>Hi,
>>>I have started using Firefly a few months back. I know how to calculate conical intersection between two PES. But, I wanted to know if we could calculate conical intersection between 3 PES using firefly.

>>>

>>Hi,

>>there is no specific code for this purpose in current Firefly.
>>Most likely this feature will be implemented in one of future
>>releases.

>>Kind regards,
>>Alex Granovsky
>>