Monte Carlo barostat

[lar... at lbl.gov]

Dear all,

Given the terrific MC capabilities of cp2k I was wondering is there is any 
way of effectively implementing a MC barostat in an otherwise AIMD 
simulation. Another way could be to run a MC simulation with just volume 
changes every certain number of MD time steps. I often use meta-GGAs and 
the stress tensor analytical calculation is not implemented, therefore 
making impractical to use the NpT ensemble.

I have the feeling that hybrid MC may be the answer but I am not sure what 
parameters are controlling what. Although there is an example in 
tests/MC/regtest/ hmc.inp there is really no explanation of how it would 
run (how often MC kicks in, etc.). Any help pointing me out to the critical 
parameters (e.g NSTEP and NMOVES in the MC routine vs. STEPS in the MD 
routine, etc) that control this type of simulations would be highly 
appreciated. I'll also be happy to hear other possible ways of doing this. 

Thanks!
Luis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20151208/c2c18bef/attachment.htm>