[CP2K-user] [CP2K:20873] Re: Changing pseudopotential in each ionic step

[Frederick Stein]

Dear Sagnik,
My first guess is to change the parameters at the end of an MD step as core 
Hamiltonians etc. require a rebuilt after updating the coordinates (and 
cell vectors). In the update step, I would all relevant qs_kinds and 
successively update zeff_correction. Beware that you will loose the 
original values. If you need the original ones, you should keep a copy of 
the original values somewhere (potentially in the qs_kind objects as some 
kind of zeff_correction_original or similar.
Regards,
Frederick

Sagnik Mukhopadhyay schrieb am Donnerstag, 7. November 2024 um 11:40:41 
UTC+1:

> Dear Fredrick,
>
> By changing I meant during the simulation in each ionic step I want to 
> change it on the way. I am not intend to change the potential files, rather 
> changing the potential by changing the core correction in each step.
>
> Regards
> Sagnik
>
> On Thursday, November 7, 2024 at 2:33:37 PM UTC+5:30 Frederick Stein wrote:
>
>> Dear Sagnik,
>> I am not entirely sure what you mean by changing the pseudopotential. Do 
>> you really intend to change the pseudopotential files on the fly? What is 
>> your workflow to achieve that? Wouldn't it be possible to just modify the 
>> values locally?
>> Regards,
>> Frederick
>>
>> Sagnik Mukhopadhyay schrieb am Mittwoch, 6. November 2024 um 12:52:12 
>> UTC+1:
>>
>>> Dear CP2K community,
>>>
>>> CP2K reads the pseduopotential once to setup the QS. For subsequent 
>>> ionic steps it wont be read. In my code I am making a change in the 
>>> qs_mol_dyn_low(md_run.F) which alters the zeff_correction. However I am 
>>> struggling to implement the code so that cp2k reinitialize the QS for each 
>>> step, i.e reads the pseudopotenital or alters it for every ionic step.
>>> Below I have provided how the md_run.F might be connected to the 
>>> pseduopotential initialization.
>>>
>>> get_qs_kind_set(qs_environment.F) --> build_core_ppl(core_ppl.F) --> 
>>> core_matrices(qs_core_hamiltonian.F) --> 
>>> build_core_hamiltonian_matrix(qs_core_hamiltonian.F) --> 
>>> qs_forces(qs_force.F) --> qs_calc_energy_force(qs_force.F) --> 
>>> qmmm_calc_energy_force(qmmm_force.F) --> 
>>> force_env_calc_energy_force(force_env_methods.F) --> 
>>> qs_mol_dyn_low(md_run.F)
>>>
>>> Any suggestions how can this be done?
>>>
>>> Regards
>>> Sagnik
>>>
>>

-- 
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 visit https://groups.google.com/d/msgid/cp2k/10229fd5-a7b9-4de3-a122-652914719dc4n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20241107/b532c240/attachment.htm>