QMMM Periodic

[mark]

Dear CP2K developers,

I have some doubts about my QMMM calculations (I'm simulating one quantum 
glucose molecule 
in a classical Ionic liquid + water solvent).

I set &PERODIC = NONE for QM CELL, in &QMMM (and I set no PBC in DFT 
section for the POISSON solver), &PERIODIC = XYZ in &SUBSYS and I set
ECOUPL=GAUSS for the QM/MM interaction.

I have several question about the PBC:

1- with this setting I'm decoupling the QM CELL from all the other periodic 
QM images (because I set &PERODIC = NONE for QM CELL ), keeping only the 
zero-cell electrostatic between QM and MM + 
the interaction between QM and MM repeated images up to the cutoff (because 
I set &PERIODIC = XYZ in &SUBSYS )?

2- with this setting MM atoms interact only with the thier images or also 
with the QM images (because I set &PERIODIC = XYZ in &SUBSYS )?

3- if I set &PERODIC = NONE for QM CELL and PERIODIC = XYZ  in DFT section 
for the POISSON solver, would it has sense?

4- if QM CELL hasn't PBC (as in this case) there is the possibility that 
the QM atoms go away from the QM box?

Can someone clarify this to me?

Thank you very much 
Kind Regards

Marco
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20130526/b756c5f3/attachment.htm>