[molpro-user] noncanonical orbitals in CCSD-F12

[Mariusz Radon]

On 02/16/2013 10:59 PM, Kirk Peterson wrote:
> Dear Mariusz,
> 
> I'm interested in what the real experts have to say, but we have used pseudo-canonical orbitals (from CASSCF) in CCSD-F12 before and they (seem to) work fine. The "trick" is to pass the orbitals through a single iteration of the HF program in order to get some of the information that Molpro requires for F12, e.g.
> 
> {multi;canonical,3100.2}
> {hf;maxit,1}
> 
> I'm not sure if this same procedure is valid for KS orbitals though.
> 
> best regards,
> 
> -Kirk
> 

Dear Kirk,

thank you very much for your reply.

I have tried to perform calculation according to your suggestion (i.e.,
something like: {ks,b3lyp}, then {hf;maxit,1}, followed by
{ccsd(t)-f12;orbital,ignore_error}).

Unfortunately, it seems that Molpro always performs at least 3 HF
iterations (even if maxit is set to 1 or 0, or -1)....
Consequently, the wave function provided for subsequent CC calcs is
neither KS nor HF determinant, which seems to give very
unreasonable final energy. Relative energies obtained from such
calculations are also clearly wrong (for my system I have a number of
reliable, non-F12 CC calculations to compare with).

But, intriguingly, when I carry out CCSD(T)-F12 right after DFT
(w/o HF in between), the relative energies look very good, and rather
close to CBS limit from extrapolated conventional calculations.
I wonder whether F12 routine in Molpro can somehow complete the missing
information, even if HF program was not called before F12?
Or am I obtaining "good" results by pure accident?


Best regards,
Mariusz


> 
> On Feb 16, 2013, at 10:44 AM, Mariusz Radon <mariusz.radon at gmail.com> wrote:
> 
>> Dear Colleagues:
>>
>> Can anyone tell me whether the current implementation of CCSD-F12 in
>> Molpro is compatible with use of noncanonical orbitals, in particular
>> Kohn-Sham orbitals from preceding DFT calculations?
>>
>> Many authors have used KS orbitals in coupled cluster calculations; they
>> can often be regarded as a cheap replacement of Brueckner orbitals. In
>> Molpro this approach works nice for "ordinary" CCSD, but how about
>> explicitly correlated (F12) calculations? Thank you in advance for your
>> reply.
>>
>> Kind regards,
>> Mariusz Radon
>>
>> -- 
>> Dr Mariusz Radon, Ph.D.
>> Coordination Chemistry Group
>> Faculty of Chemistry
>> Jagiellonian University
>> ul. Ingardena 3, 30-060 Krakow, Poland
>> http://www.chemia.uj.edu.pl/~mradon
>>
>> _______________________________________________
>> Molpro-user mailing list
>> Molpro-user at molpro.net
>> http://www.molpro.net/mailman/listinfo/molpro-user
> 


-- 
Dr Mariusz Radon, Ph.D.
Coordination Chemistry Group
Faculty of Chemistry
Jagiellonian University
ul. Ingardena 3, 30-060 Krakow, Poland
http://www.chemia.uj.edu.pl/~mradon