[CP2K-user] [CP2K:18862] OT for added MOS

[Lorenzo Agosta]

Hi Jurg,

Thank you a lot for the nice clarification.
This solved a lot of doubts for me.

Best,
Lorenzo

Il giorno venerdì 26 maggio 2023 alle 10:50:31 UTC+2 Jürg Hutter ha scritto:

> Hi
>
> OT is a direct energy optimization method. As the energy doesn't depend on 
> the virtual states, they cannot
> be calculated directly within the optimization.
> However, you can calculate additional states (LUMO +++) after an OT 
> optimization. Several property options
> allow for such calculations (e.g. PRINT/MO_CUBES). Another possibility is 
> to restart an optimized structure
> and requesting a single diagonalization step.
>
> regards
> JH
>
> ________________________________________
> From: cp... at googlegroups.com <cp... at googlegroups.com> on behalf of 
> Lorenzo Agosta <pha... at gmail.com>
> Sent: Thursday, May 25, 2023 1:40 PM
> To: cp2k
> Subject: [CP2K:18853] OT for added MOS
>
> Dear developers,
>
> I know that is not possible to perform SCF calculations using the OT 
> scheme when extra molecular orbitals are requested.
>
> The problem is that for many not trivial systems the OT scheme is always 
> able to reach a SCF convergence, while the standard DIAGONALIZATION 
> algorithms ( no matter which combinations of parameters are used for the 
> mixing ) fail, even starting from a fully converged electron density 
> reached with the OT scheme.
> Strict diagonalization algorithms are also required in various subroutines 
> like KPOINT, XAS and GW for example.
>
> Having the possibility to calculate added MOS using the OT scheme would be 
> enormously beneficial in terms of applicability of the CP2K software and a 
> considerable gain in computational resources.
>
> So, I was wondering if there is a specific reason why extra MO cannot be 
> computed with the OT scheme, or more simply it has not been implemented yet.
>
> Thank you in advance for the help and clarifications.
>
> Kind regards,
> Lorenzo Agosta
>
> --
> 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+uns... at googlegroups.com<mailto:cp2k+uns... at googlegroups.com
> >.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/cp2k/1d8bf378-a519-464e-893d-4fdeced35373n%40googlegroups.com
> <
> https://groups.google.com/d/msgid/cp2k/1d8bf378-a519-464e-893d-4fdeced35373n%40googlegroups.com?utm_medium=email&utm_source=footer
> >.
>

-- 
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/50392121-58b0-4d08-9026-487b826bb251n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20230529/f74bb0ef/attachment.htm>