# [molpro-user] Density Fitting SAPT

Glen Jenness grj3+ at pitt.edu
Thu Aug 31 23:05:24 BST 2006

```Dear MolPro users,
I'm currently trying to use DFT-SAPT to study large systems (using
MolPro2006v.1), and I would like to use density fitting to aide in the
computation, but I can't seem to get it to work in SAPT when I do it on a
sample calculation.
When I define jkfit and mp2fit (using vtz and vdz as the basis) in the
basis section as suggested by the manual, the program will tell me that
the variables vtz and vdz are undefined and sets them equal to zero.
So, I went ahead and explicitely set basis_coul=vtz and similar for the
other basis, it'll tell me that the basis is undefined.  I've been able to
use this method when I did a non-SAPT density fitting of water dimer with
success.
My question is am I missing something when I go to implement this with the
SAPT module?
I've included my input file if that helps.
Thanks,
Glen Jenness
Jordan Group
Department of Chemistry
University of Pittsburgh

***,water dimer polar geom r=2.20A DF-DFT-SAPT test
geomtyp=xyz
geometry={
angstrom;
nosym;
6
w2-2.20
O2 -.000419  -1.037571  .000000
O1 -.000419  1.162429  .000000
H1 .896450  1.496879  .000000
H2 .098487  .210352  .000000
H3 -.494114  -1.353046  .756950
H4 -.494114  -1.353046  -.756950
}
basis=avdz
jkfit=vtz
mp2fit=vdz

ca=2101.2
cb=2102.2

dummy,o2,h3,h4
{rks,pbe0; accu,20; save,\$ca}
sapt; monomera

dummy,o1,h1,h2
{rks,pbe0; accu,20; start,atdens; save,\$cb}
sapt; monomerb

{sapt;intermol,ca=\$ca,cb=\$cb,fitlevel=3
dfit,basis_coul=jkfit,basis_exch=jkfit,basis_mp2=mp2fit,cfit_scf=3}

```