OT or DFT+U for k-points

[Sheyoung Wang (W0728)]

Dear developers,

I know that k-points are not supported when using OT or DFT+U.

The problem is that for many systems that are not trivial, the OT scheme always achieves SCF convergence, whereas the standard diagonal algorithm (no matter which parameter combination is used for mixing) fails even if the calculation is started from electron densities that achieve full convergence using the OT scheme. DFT+U is also necessary to consider for some systems, in some cases far more cost-effective than hybridized generalized functions such as PBE0 and HSE06.

If this problem could be solved, it would greatly improve the applicability of the CP2K software and make more efficient use of computational resources.

Therefore, I would like to know if there is a specific reason why OT and DFT+U are unable to support k-points or, more simply, if it has not yet been realized.

Thanks in advance for your help and clarification.

[Jürg Hutter]

Hi

OT and DFT+U (at least in the main version of CP2K) have not been implemented for k-points.

Such implementations would be possible but haven't been done yet, or have been done but have
not been submitted to the main branch of CP2K.

regards
JH

________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of Sheyoung Wang (W0728) <grow...@outlook.com>
Sent: Monday, April 28, 2025 8:04 AM
To: cp2k
Subject: [CP2K:21419] OT or DFT+U for k-points

--
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...@googlegroups.com<mailto:cp2k+uns...@googlegroups.com>.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/2a01ba6a-7300-43d2-8d22-a40159dcb915n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/2a01ba6a-7300-43d2-8d22-a40159dcb915n%40googlegroups.com?utm_medium=email&utm_source=footer>.