[CP2K-user] [CP2K:13176] bulk cell opt questions

[Jing Liu]

Dear Thomas

The test cases I have ran are just test cases, their purpose are to test 
the input for the real work. So ideal is to run them with smallest basis 
set, get the results as fast as possible, check if there is any abnormal 
happens (like the non-zero charge in this case). If everything worked then 
I can move to the real case with sufficient basis set and functional.

You did make a good point on energy comparison...I have to think about the 
valid reference system...

Thank you for the suggestions!

Best regards

Liu Jing

在 2020年4月27日星期一 UTC+2下午11:42:16，tkuehne写道：
>
> Dear Jing, 
>
> honestly I don’t fully understand what you are testing and what is the 
> rational behind it. 
> However, first of all you can’t compare the absolute energies (there are 
> not even 
> expected to be close), since you are comparing a pseudopotential 
> calculation with an 
> all electron reference. Also, I suspect that your reference is a gas phase 
> calculation, 
> which can neither compared to your periodic test no. 1 (because of pbc) 
> and also 
> not to test no. 2, which is not a proper isolated supercell due to too 
> small cell dimensions 
> along the x-direction.
>
> Best, 
> Thomas
>
>
> Am 27.04.2020 um 11:25 schrieb Jing Liu <liu... at gmail.com <javascript:>
> >:
>
> Dear CP2K users
>
> Recently I'm trying to run the cell optimization for a organic material (
> Tetraazapyrene). My goal is to optmize the cell parameters of this bulk 
> system. 
>
> However during the test run there are few things that I could not explain, 
> it's very likely that I missed something so if anyone could point them out 
> would be very appreciated.
>
> *The test cases are*:
>
> Test_1: using the cif file, set pbc x y z both in &CELL and &POISSON, 
> initial cell parameters are taken from cif file, no super cell
>
> Test_2: using the same cif file as Test_1, set pbc x y z both in &CELL and 
> &POISSON, initial cell parameters are taken from the same cif file as 
> Test_1, 2*2*2 super cell, set charge constraint to 0
>
> Reference: single TAPP molecule, geo_opt with KOALA at cc2 level
>
> *The problems are*:
>
> In Test_1: total energy of the converged run greatly dffered with respect 
> to the reference run. They of course should not be identical but also 
> should be close. Total energy of this run is: -340.2863 H. 
>
> In Test_2: Total energy difference increased, total charge is also wrong 
> despite the constraint was applied. Total energy of this is: -2722.1877 H, 
> total charge of this run is: 0 (for Mulliken) and 1.2 (for Hirshfeld)
>
> In Reference: Total energy is: -1061.3367 H.
>
> All the inp and out files are attached.
>
> -- 
> 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 c... at googlegroups.com <javascript:>.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/cp2k/73b9a873-5d94-4fff-a37d-61ba48721201%40googlegroups.com 
> <https://groups.google.com/d/msgid/cp2k/73b9a873-5d94-4fff-a37d-61ba48721201%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
> <hth.cif><hth_supercell.inp><hth.inp>
> <hth_bulk_cell_opt_supercell_reduced.out><hth_bulk_cell_opt_reduced.out>
> <koala_cc2_geo_opt_last_iter_reduced.out>
>
>
>
>
> ==============================
> Thomas D. Kühne
> Dynamics of Condensed Matter
> Chair of Theoretical Chemistry
> University of Paderborn
> Warburger Str. 100
> D-33098 Paderborn
> Germany
> td... at mail.upb.de <javascript:>
> +49/(0)5251/60-5726
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200428/603feff2/attachment.htm>