[CP2K-user] [CP2K:19421] Re: Questions on the use of the WFN_MIX function to obtain the open-shell singlet state

[Marcella Iannuzzi]

Dear John, 

The restart does not work if the previous calculation had ADDED_MOS, but 
the ATOMIC guess should work. 
You have to keep the MO_CUBES in the PRINT section of DFT
It works for me. 

Regards
Marcella



On Tuesday, October 24, 2023 at 4:51:09 PM UTC+2 John Zhang wrote:

> Dear Dr Lannuzzi
>
> Thank your for your response!
>
> Following you suggestions, I performed a few test calculations (without 
> the ADDED_MOS)
>
> 1. Starting from the ATOMIC guess with singlet spin multiplicity. The 
> calculation stopped at the beginning of "Performing wfn mixing" and the 
> error message looks like (I pasted only a few lines) 
> "PBLAS ERROR 'Array subscript out of bounds: JX = 7, DESCX[N_] = 4'
> from {0,7}, pnum=7, Contxt=0, in routine 'PDCOPY'.
> PBLAS ERROR 'Parameter number 4 had an illegal value'
> from {0,7}, pnum=7, Contxt=0, in routine 'PDCOPY'.
> {0,7}, pnum=7, Contxt=0, killed other procs, exiting with error #-4.
> "
>
> 2. if the calculation started using the quintet wave function, it 
> immediately stopped and the error message is "Number of occupied MOs on 
> restart unit larger than allocated MOs"
>
> I also checked the examples in the tests, but I cannot figure out what the 
> problem is. Any insight? Thank you again!
>
> Best wishes
>
> John
> On Tuesday, October 24, 2023 at 7:10:30 AM UTC-4 Marcella Iannuzzi wrote:
>
>> Dear John
>>
>> I think that your input might work if you remove the key ADDED_MOS from 
>> SCF.
>> Anyway, the input structure for wfn_mix has slightly changed in most 
>> recent version of the code. 
>>
>> Regards
>> Marcella
>>
>>
>> On Saturday, October 21, 2023 at 7:52:11 PM UTC+2 John Zhang wrote:
>>
>>> Hi
>>>
>>> I want to obtain an open-shell singlet state wave function for a C2 
>>> molecule using the WFN_MIX function. I first performed a quintet state 
>>> calculation (converged smoothly) and tried to use this quintet state wave 
>>> function as the initial guess in the following open-shell singlet state 
>>> calculation. In the WFN_MIX section, I tried to flip the spin of the 
>>> original HOMO and HOMO-1 orbitals (Also see the attached input)
>>>
>>>     &WFN_MIX
>>>
>>>       &UPDATE
>>>
>>>         RESULT_MO_INDEX    -1
>>>
>>>         RESULT_SPIN_INDEX  BETA
>>>
>>>         RESULT_SCALE       0.00000
>>>
>>>
>>>         ORIG_MO_INDEX      2
>>>
>>>         ORIG_SPIN_INDEX    ALPHA
>>>
>>>         ORIG_SCALE         1.0000
>>>
>>>
>>>         ORIG_IS_VIRTUAL    .FALSE.
>>>
>>>       &END UPDATE
>>>
>>>
>>>       &UPDATE
>>>
>>>         RESULT_MO_INDEX    -2
>>>
>>>         RESULT_SPIN_INDEX  BETA
>>>
>>>         RESULT_SCALE       0.00000
>>>
>>>
>>>         ORIG_MO_INDEX      1
>>>
>>>         ORIG_SPIN_INDEX    ALPHA
>>>
>>>         ORIG_SCALE         1.0000
>>>
>>>
>>>         ORIG_IS_VIRTUAL    .FALSE.
>>>
>>>       &END UPDATE
>>>
>>>     &END WFN_MIX
>>>
>>>
>>> but the calculation failed with an error message "If ORIG_IS_VIRTUAL is 
>>> set, the array unoccupied_orbs must be associated!"
>>>
>>> I think my usage of WFN_MIX is not correct. Could someone please help on 
>>> this issue? Thanks!
>>>
>>>
>>> Best 
>>>
>>>
>>> John
>>>
>>>
>>>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/2f786391-b95d-439d-ad73-0091758b564an%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20231024/31de5347/attachment.htm>