PC GAMESS/Firefly-related discussion club



Learn how to ask questions correctly

Calculation of vertical ionization potential by DFT-FZOA

Fumihito Mohri
qzg00677@nifty.com

Dear Friends
I am asked for calculation of vertical-ionization potential (Ip) of
compounds for organic solar cell materials. I have already completed Ip
calculation by Hartree-Fock-Koopmans method. Calculated Ip value of a compound
that is one of the candidate materials is different from 7 % (overestimated).
This is not so bad. But, in order to get more accurate value, I consider that
combination of DFT and Frozen Orbital Approximation (FZOA) is more appropriate.
Then, I have come up with the following procedure.

 Ip=E(cation)-E(neutral)
1st step : Usual geometry optimization for neutral molecule. E(neutral) is
         obtained.
       $CONTRL   SCFTYP=RHF  MULT=1  ICHARG=0  DFTTYP=B3LYP....$END
2nd step: Single point calculation for E(cation), using the same geometry
         as 1st step and
       molecular orbitals which are stored in PUNCH file as $VEC data.

       $CONTRL SCFTYP=ROHF MULT=2  ICHARG=1 DFTTYP=B3LYP  MAXIT=1
           RUNTYP=ENERGY.....$END
       $GUESS GUESS=MOREAD NORB=XX (number of all MOs)  $END
        $VEC

 Here,I think that option of MAXIT=1 is essential. I have already tested this technique
  for H2O.
         E(neutral)=-76.4196837 au, E(cation)=-75.8666303 au,   Ip=15.05 eV
  I have disappointed because this Ip value is larger than HF-Koopmans value
  (13.5 eV). The observed Ip is 12.60 eV.  Can I get the vertical-Ip by the above
   technique?  
     
Thank you in advance,
Fumihito Mohri

[ Previous ] [ Next ] [ Index ]           Sun Sep 20 '09 6:38am
[ Reply ] [ Edit ] [ Delete ]           This message read 1416 times