[molpro-user] F12 methods

[Hans-Joachim Werner]

Dear Christian,
I am sorry for the trouble. The problem "Cannot find default basis VDZ- 
F12 for atom O"
is already fixed in our development version, and we will make a patch  
for 2008.1 as soon
as possible. For the time being please use for avdz, avtz (or vdz-f12,  
vtz-f12):

  mp2-f12,df_basis_exch=vtz
  ccsd(t)-f12,df_basis_exch=vtz

and for avqz:

  mp2-f12,df_basis_exch=vqz
  ccsd(t)-f12,df_basis_exch=vqz

The problem is that the mp2-f12 program does not know a default for  
the fitting basis for making the Fock and exchange matrices.

So far, I cannot reproduce the unreasonable energy you get for the  
avqz basis. Could you
please send your output? Which machine/compiler(version)/blas did you  
use? I get the following
results (using df_basis_exch as above):

                RHF               CABS Singles    Reference energy      
Ecorr(MP2)    Ecorr(MP2-F12) Ecorr(CCSD(T)-F12A) Ecorr(CCSD(T)-F12b)
avdz      -152.089672143873   -0.042471103247   -152.132143247120   
-0.438860307002 -0.586010337082   -0.584157512835    -0.569798542897
avtz      -152.127977677257   -0.009353087469   -152.137330764726   
-0.537069859004 -0.593727514886   -0.593285632319    -0.583798866845
avqz      -152.138817231767   -0.001917209877   -152.140734441645   
-0.572172044482 -0.598592928193   -0.597654392195    -0.592260886107

vdz-f12   -152.123795554166   -0.013293840277   -152.137089394443   
-0.482269701797 -0.593338690191   -0.589941424263    -0.577029504611

Best regards
Joachim Werner
-- 
Prof. Hans-Joachim Werner
Institute for Theoretical Chemistry
University of Stuttgart
Pfaffenwaldring 55
D-70569 Stuttgart, Germany
Tel.: (0049) 711 / 685 64400
Fax.: (0049) 711 / 685 64442
e-mail: werner at theochem.uni-stuttgart.de




Am 19.10.2008 um 06:22 schrieb Christian Evenhuis:

> I am trying out the new F12 methods but I am not having much luck.
>
> 1. if I use the VNZ-F12 basis sets I get the following error:
> 		1PROGRAM * CCSD (Closed-shell coupled cluster)     Authors: C.
> Hampel, H.-J. Werner, 1991, M. Deegan, P.J. Knowles, 1992
>                CCSD-F12 implementation by  H.-J. Werner, 2007
>                Density fitting integral evaluation by F. R. Manby,
> 2003,2007
>                Cannot find default basis VDZ-F12 for atom O
>
> 2. The MP2-F12 energies for AVNZ series is
> 	AVDZ	-152.71791751
> 	AVTZ	-152.73109952
> 	AVQZ 	-149.60581350
> The contribution to the energy are listed after the input deck.
>
> 3.  For CCSD(T)-F12 / AVNZ  I get the following error
>              ? Warning
>              ? Encountered below-occupied eigenvalues in virtual
> block. This may indicate serious deficiencies in the orbital or
>                 JKfit-basis sets (basis_exch). Offending values will
> be ignored, but PT2 is ill-conditioned!
>              ? The problem occurs in f12_common/Mp2Singles()
>              FockProj BlockDiagonalize() eigenvalues:
> (norb=                   314 )
>              E(Occ):   -20.5969651851479       -20.5474946885548
>               -1.38499649063043       -1.33164908493971
> -0.749957814812005
>              -0.701772636068004      -0.612407775875517
> -0.558616202902750
>              -0.531801726963611      -0.483450539393589
>
>     Changing to RI_BASIS=AVTZ/ORBITAL helped a little but did not
> resolve the issue
>
> Thanks,
>
> Chris Evenhuis
>
>
> INPUT DECK:
>
> *** water dimer
> geomtyp=xyz
>
> geometry={noorient;
> 6
> lp
> O   0.0640488861  -1.5542001572   0.0000000000
> H  -0.0275603110  -0.6109761929   0.0000000000
> H  -0.8291328873  -1.8994223998   0.0000000000
> O  -0.0539423286   1.4751039841   0.0000000000
> H   0.3402097464   1.8881478835   0.7547208224
> H   0.3402097464   1.8881478835  -0.7547208224
> }
>
> basis=BASIS
> {hf
>  accu,14
> }
>
> {METHOD-F12
> }
>
>
> CONTRIBUTION TO MP2-F12 AVDZ ENERGY
>
> MP2 singlet pair energy:              -0.369580050529
> MP2 triplet pair energy:              -0.202617031749
> MP2 correlation energy:               -0.572197082278
> MP2 total energy:                   -152.712931523922
>
> SCS-MP2 correlation energy:           -0.569568880390   (PS=
> 1.200000  PT=  0.333333)
> SCS-MP2 total energy:               -152.710303322034
>
> Norm of t2 vector:      0.35747963      P-energy:    -0.57219708
>
>  Reference energy                   -152.138817231768
>  F12 singles correction               -0.001917209877
>
>  F12 singles corrections added to reference energy
>
>  New reference energy               -152.140734441644
>
>  MP2-F12 singlet correction            1.955823920706
>  MP2-F12 triplet correction            1.151294106399
>  MP2-F12 total correction              3.107118027105
>
>  F12 corrections for ansatz F12/3C(FIX) added to MP2 energy
>
>  MP2-F12/3C(FIX) singlet pair ener     1.586243870177
>  MP2-F12/3C(FIX) triplet pair ener     0.948677074650
>  MP2-F12/3C(FIX) correlation energ     2.534920944827
>
> !MP2-F12/3C(FIX) Energy             -149.605813496817
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user