[CP2K-user] [CP2K:12591] Hybrid DFT for TiO2 surface

[Patrick Gono]

Dear Siva,

The computation of the exchange integrals is very costly, and, naively,
scales as O(N^4) with the basis set size N. Hence, the ADMM method
introduces an auxiliary, smaller basis to compute the exchange integrals.
What this means in practice is that you can use large, high quality basis
sets (like the MOLOPT basis sets mentioned in your question), and then
specify a smaller auxiliary basis for your elements to enable the ADMM
method, and speed up the HF calculations.

I recommend you read the tutorial on hybrid functionals and the ADMM method
here:
https://www.cp2k.org/exercises:2017_uzh_cp2k-tutorial:hybrid#auxiliary_density_matrix_methods_admm

Apart from including the &AUXILIARY_DENSITY_MATRIX_METHOD section in the
appropriate place, you need to choose which auxiliary basis sets you want
to use for each element.
In this presentation:
https://www.cp2k.org/_media/events:2015_user_meeting:cp2k-uk-2015-ling.pdf
the ADMM basis sets are explained, and the input file structure is given as
well.

For your convenience, I attach an input file for TiO2 using PBE0 and the
ADMM method, as well as the basis files BASIS_ADMM and BASIS_ADMM_MOLOPT,
which contain the auxiliary basis sets.

As a side note, if you are using an older version of CP2K, you may need to
specify the auxiliary basis sets via:
AUX_FIT_BASIS_SET ...
instead of:
BASIS_SET AUX_FIT ...
in the &KIND section of your input.

Yours sincerely,
Patrick Gono

On Tue, Dec 3, 2019 at 2:13 PM Siva Chiriki <chsiv... at gmail.com> wrote:

> Hi all,
>
> I was unable to use ADMM calculations for hybrid DFT (PBE0) calculation on
> surface TiO2 surface containing V and O and H atoms types along with Ti.
>
> Simple PBE0 (PBE+HF) is working for water system provided in the tutorial
> section based on the HF-basis set. But as soon as if I want to use my
> surface then PBE0 is not working with mol_opt basis_sr_gth sets and also
> for Ti and V there are no HF_basis sets.
>
> As suggested by the tutorials section of hybrid DFT, I tried to use ADMM
> but it is saying that AUX_BASIS_SET not activated.
>
> Please can anybody suggest me input basis sets and potential files for
> doing PBE0 calculations for surface containing element types (Ti, V, H,O)
> without ADMM or please tell me the why ADMM is not working with PBE0.
>
> --
> 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 cp... at googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/cp2k/c4056ba1-f8d1-402f-9c4b-5f9b7f49bbb9%40googlegroups.com
> <https://groups.google.com/d/msgid/cp2k/c4056ba1-f8d1-402f-9c4b-5f9b7f49bbb9%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191204/e4c4407e/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: input.inp
Type: application/octet-stream
Size: 2774 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191204/e4c4407e/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BASIS_ADMM_MOLOPT
Type: application/octet-stream
Size: 189911 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191204/e4c4407e/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: BASIS_ADMM
Type: application/octet-stream
Size: 57524 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191204/e4c4407e/attachment-0002.obj>