geometry optimization for graphite

[Chenxing Yang]

Dear All,

I have run some simulations of geometry optimization for graphite with 
OPT-PBE using CP2K 4.1, and I have some problem on reproducing the correct 
crystalline structure of graphite.

For some reason, the bond length is slightly overestimated (1.448 Å), and 
inter-layer distance is underestimated (2.756 Å), where experiment results 
are 1.421 Å and 3.33 Å. It seems those 2 distances can be very well 
reproduced using OPT-PBE in the literature. Could you please help me to 
check if something is wrong in the input file?

Another problem I encountered is, if I simulate in a 3*3*3 supercell, the 
atoms is located same position of atoms in the adjacent layer, instead of 
centre of hexagon, but it is fine if the supercell is set as 2*2*2 and 
4*4*4. 

Here is my input file attached. Thank you very much!

Kind regards,

Chenxing Yang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170331/b86d7f5e/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: graphite.inp
Type: chemical/x-gamess-input
Size: 2309 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20170331/b86d7f5e/attachment.inp>