PERIODIC XYZ

[ashish dabral]

Hi ,

I have a a monolayer which is periodic in xy direction. I can solve for 
cell relaxation(or energy/force) 
for my structure as a periodic structure with a high enough gap for vacuum 
in the z direction to prevent
interaction between layers and therby isolating them.

Can I use the PERIODIC XY option instead of XYZ to improve computation time 
and memory usage?
 
I would like to know the difference between the two setting XYZ and XY 
within
the current scenario. Also, the Cp2k manual specifies that the settings for 
Poisson solver should be the same.

I am doing cell optimisation and have not specified any poisson settings 
anywhere. 

Do I need to specify the corresponding settings for Poisson solver? How can 
I correctly setup for aperiodicity in
z direction for an xy periodic layer?

Update: I tried specifying the periodicity with PERIODIC XY keyword in my 
inp file but my out file gives me same setting 
of   Solver Periodic
      Periodicity xyz

as my original file without the periodic keyword.

I read about the solvers and I found that (MT poisson solver for example) 
the contribution of image cells is removed.
I don't want to remove the contribution of xy images but only the z 
direction image. 
Can someone explain how can it be done in a computationally efficient way 
and if PERIODIC keyword either in &CELL section
or the DFT section can be used?

Thanks for any pointers that might help!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20160311/18580f04/attachment.htm>