[molpro-user] DF-DFT-SAPT fails with PBE0 hybrid functional

[Leonid Shirkov]

Dear Molpro Users and Developers!

I got the following error when using hybrid PBE0 with DF-DFT-SAPT.

 ? Error
 ? Fitting level of 3 not implemented for hybrid functionals, localise
the xc-potential with OEP or use non-hybrid functionals  ? The problem
occurs in sapt_interface

The output with the error for argon dimer is attached. The only way to
avoid it is to set fitlevel=0, but in this case the time of
calculations will be almost the same as if DFT-SAPT (witn no DF) is
used,
so there is no gain in time. Does this option just switch off the
density fitting?

The reason is motivated by the fact, that for some complexes DFT-SAPT
(with no DF) gives better agreement with a benchmark for PBE0 than for
LPBE0, another hybrid  functional used as a sample input which works
fine.
That is why the idea was to check if it works if DF is employed.

However, in  Phys. Chem. Chem. Phys., 2007, 9, 1680–1687  the authors
clearly report that they used both  PBE0 and LPBE0 (asymptotically
corrected) functionals with DF in Molpro.

Could anyone please shed some light on that?

Best regards,
Leonid
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ar-ar_df-dft-f3_avtz_min-1.out
Type: application/octet-stream
Size: 61908 bytes
Desc: not available
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20140518/8d7f91a2/attachment-0001.obj>