[CP2K-user] [CP2K:15385] Unreasonable CELL_OPT result obtained for TiO2

[Yike Huang]

Dear  Matthias,

I have tested these two methods. 
For the first &DIAGONALIZATION plus &KPOINTS method, I find it somewhat 
strange that even if I switch on &OUTER_SCF, SCF will still exist with 
non-converged warning, and by now I have no idea about it. So I turn to the 
second.
For the second that to set MULTIPLE_UNIT_CELL both in &CELL and &TOPOLOGY, 
it works quite well and I find for accuracy of structure prediction (bond 
length, cell parameters, etc.), it is even not necessary to use DFT+U to 
tune electron delocalization property of Ti (both anatase and rutile TiO2), 
although for band structure calculation, DFT+U is usually critical.
I share information above here and hope it will be useful for other users.

Thank you again.
Yike
在2021年5月16日星期日 UTC+2 下午6:46:31<Matthias Krack> 写道：

> Hi Yike
>
>  
>
> Your simulation cell is small and you use no k point sampling. Either you 
> use k point sampling or you enlarge your simulation cell to at least a cell 
> edge length of 12 A in each direction. I suggest the latter option for 
> CP2K. Furthermore, you could reduce EPS_DEFAULT to 1.0E-12 and increase 
> REL_CUTOFF to 60.
>
>  
>
> HTH
>
>  
>
> Matthias
>
>  
>
> *Von:* c... at googlegroups.com <c... at googlegroups.com> *Im Auftrag von *Yike 
> Huang
> *Gesendet:* Sonntag, 16. Mai 2021 18:23
> *An:* cp2k <c... at googlegroups.com>
> *Betreff:* [CP2K:15385] Unreasonable CELL_OPT result obtained for TiO2
>
>  
>
> Hello, cp2k developers and users,
>
>  
>
> I try to optimize primitive cell of rutile TiO2 by running CELL_OPT 
> calculation at PBE+U(presently set to 0 eV)/DZVP level, but I find cell 
> parameters deviate from experimental values quite quickly during 
> calculation.
>
> I have tried changing STRESS_TENSOR to both NUMERICAL and ANALYTICAL but 
> it does not help, I also think it may be not the reason why I get this kind 
> of strange result.
>
> I attach my input script and CIF file below, are there any parameters in 
> input script I forget?
>
> In addition, anatase TiO2 CELL_OPT task also has such problem, but I 
> remember that primitive cell of Co3O4 can be successfully optimized.
>
>  
>
> Thanks in advance.
>
>  
>
> Yike Huang
>
> -- 
> 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/11f85e5a-b3f6-45b2-b164-11f12644f358n%40googlegroups.com 
> <https://groups.google.com/d/msgid/cp2k/11f85e5a-b3f6-45b2-b164-11f12644f358n%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20210516/f0051e17/attachment.htm>