[CP2K-user] Energy gap inconsistent with experiment

[Fabian Ducry]

Hi,

Have you checked the literature for band gap values obtained from DFT with 
the PBE functional for your compound? Often there is a mismatch between 
experiment and simulation.

Here are som pointers how you can improve your input

 - check and converge the cutoff: 
https://www.cp2k.org/howto:converging_cutoff
 - increase the accuracy: EPS_SCF 1.0e-6 or better 1.0e-7, EPS_DEFAULT 
1.0e-12
 - make sure your coordinates are ok. The forces on the atoms are large, 
this system is far from a minimum.

Best,
Fabian

On Monday, 25 May 2020 15:08:57 UTC+2, Ref wrote:
>
> Hello everyone,
>
> I tried to calculate the DOS of YBCO, but in the end, on the resulting 
> graph, the formed value of energy gap  is tens of times larger than the 
> expected result (obtained from experimental data - it should be about 20 
> meV). I used standard potential GTH, functional PBE and  normal parameters. 
> I performed geometry optimization for a primitive cell, and then multiplied 
> it to size 3 x 3 x 3 in a graphics editor with some doping. (about 190 
> atoms are in the final system.)
> What could be associated with such a strong discrepancy? What general 
> recommendations could you give to improve the result?
>
>
> I've attached the input file that I've used. 
>
> Thanks in advance for your assistance!
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200525/c84ff76f/attachment.htm>