[CP2K-user] [CP2K:14801] Warning in cryssym: Symmetry library SPGLIB not available

[Tiziano Müller]

Dear Pierre,

thanks for chiming it, but to clarify for the KP-related statements:

* spglib is not needed to build the k-point grid [1]
* spglib is used to reduce the number of k-points by considering the 
time-reversal symmetry (when FULL_GRID is .FALSE. and SYMMETRY is .TRUE.)
* it could be used, but this is not implemented yet, to reduce the 
number of k-points further by considering the space group
* correct, some features are not yet implemented for KPs [2]

Yet, we consider KP to be ready for general use (within the 
aforementioned limits). If you find inconsistent results, please get in 
contact with us, either by opening an issue on [3] or email.

Best regards,
Tiziano

[1] 
https://github.com/cp2k/cp2k/blob/9e72863af27f901e325a68031569daed26b32cda/src/cryssym.F#L255-L303
[2] https://github.com/cp2k/cp2k/issues/225
[3] https://github.com/cp2k/cp2k/issues/

On 2/20/21 10:51 AM, Pierre-André Cazade wrote:
> Dear Geng,
> 
> You compiled cp2k without spglib. This is the error message that you are 
> getting.
> 
> Initially, Spglib was used to build the k-point grid. It is used to 
> identify the space group of the system and build the smallest k-point 
> grid based on it. Spglib comes with such a feature. However, Spglib is 
> not used, in cp2k v8, to conserve the space group throughout the 
> geometry (GEO_OPT or CELL_OPT) optimization. This has changed in the 
> last version of the development version (v9) where the space group is 
> conserved during optimization. It is important to note that in neither 
> version, the space group symmetry is used to reduce the computational 
> cost of energy and forces evaluation.
> 
> Coming back to k-points, if you have issues with spglib, you can use the 
> keyword "SYMMETRY F". This will disable the reduction of the number of 
> k-points based on symmetry and the call to spglib: 
> https://manual.cp2k.org/cp2k-8_1-branch/CP2K_INPUT/FORCE_EVAL/DFT/KPOINTS.html 
> <https://manual.cp2k.org/cp2k-8_1-branch/CP2K_INPUT/FORCE_EVAL/DFT/KPOINTS.html>
> 
> Finally, as a user, I would recommend you not to use k-points. Many 
> features are not compatible with them in cp2k, and they provide 
> inconsistent results. Use a supercell instead. In both the cell  and 
> topology sections use the keyword MULTIPLE_UNIT_CELL to do so.
> 
> Regards,
> Pierre
> 
> 
> On Sat, Feb 20, 2021 at 12:12 AM GENG YUAN <yuange... at gmail.com 
> <mailto:yuange... at gmail.com>> wrote:
> 
>     Dear CP2K users,
> 
>     I am currently performing some geometry optimizations for my
>     metallic systems. There is one warning message on my output file
>     although my simulations successfully completed with the geometry
>     optimization completed. Below is the warning detail:
>     **** WARNING in cryssym. F : 163 : : Symmetry library SPGLIB not
>     available ****
>     I tried to search and what I know now is the Spglib is a library for
>     finding and handling crystal symmetries written in C. Another thing
>     I notice is that this warning is related to the KPOINT setting. All
>     my simulations with KPOINT setting have this warning, while this
>     warning is gone for my simulations without KPOINT setting.
>     So my questions are: Why is this warning shown in my output file, is
>     my KPOINT setting wrong? Will it affect the final energy results and
>     how could I get rid of this warning if my simulations still need
>     KPOINT settting. I have attached my input file here for your
>     reference, any suggestions are welcome!
> 
>     Many thanks in advance and have a nice weekend!
>     Sincerely,
>     Geng
> 
>     -- 
>     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
>     <mailto:cp... at googlegroups.com>.
>     To view this discussion on the web visit
>     https://groups.google.com/d/msgid/cp2k/b4437744-a691-4841-bbe2-34fc7fd64855n%40googlegroups.com
>     <https://groups.google.com/d/msgid/cp2k/b4437744-a691-4841-bbe2-34fc7fd64855n%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 cp... at googlegroups.com 
> <mailto:cp... at googlegroups.com>.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/cp2k/CAFnf9goHrDu9y4HFyFkuNQbukv43aqy3gwWAckgzO1pTaXcf8w%40mail.gmail.com 
> <https://groups.google.com/d/msgid/cp2k/CAFnf9goHrDu9y4HFyFkuNQbukv43aqy3gwWAckgzO1pTaXcf8w%40mail.gmail.com?utm_medium=email&utm_source=footer>.

-- 
Tiziano Müller
University of Zurich
Department of Chemistry
Winterthurerstrasse 190
CH-8057 Zürich

Tel: +41 44 63 54234
www.chem.uzh.ch
tiziano... at chem.uzh.ch