barostat

[Nichols A. Romero]

Hi,

Sorry for all my questions.

I am interested in NPT dynamics simulation. I would like to understand how this is implemented.
MD&NOSE is the NoseHoover Thermostat chains (for use with NVT), while MD&BAROSTAT
seems to have its own thermostat. So the barostat has its on thermostat, and NoseHoover
Thermostat chains (MD&NOSE) should not be used in conjuction with the MD&BAROSTAT?

Thanks for all your help.

--
Nichols A. Romero, Ph.D.
1613 Denise Dr. Apt. D
Forest Hill, MD 21050
443-567-8328 (C)
410-306-0709 (O)

[akohlmey]

On May 2, 2:40 pm, "Nichols A. Romero" <narom...@gmail.com> wrote:
> Hi,
>
> Sorry for all my questions.
>
> I am interested in NPT dynamics simulation. I would like to understand how
> this is implemented.
> MD&NOSE is the NoseHoover Thermostat chains (for use with NVT), while
> MD&BAROSTAT
> seems to have its own thermostat. So the barostat has its on thermostat, and
> NoseHoover
> Thermostat chains (MD&NOSE) should not be used in conjuction with the
> MD&BAROSTAT?

hi nick,

have you considered the fact, that you can use nose-hoover
chains also to control the pressure?

cheers,
axel.

[Nichols A. Romero]

Axel,

I know very little about the Nose-Hoover chains. (My background is not in MD, but solid-state DFT).
I thought the only option for controlling the pressure was the barostat.

How do I control the pressure with Nose-Hoover chains? If you can provide me with the appropriate
background reference that would help. I also simple don't see the appropriate key word to set the
pressure with the chains.

[akohlmey]

On May 2, 4:36 pm, "Nichols A. Romero" <narom...@gmail.com> wrote:
> Axel,
>
> I know very little about the Nose-Hoover chains. (My background is not in
> MD, but solid-state DFT).
> I thought the only option for controlling the pressure was the barostat.

simply put, a barostat is a 'thermostat for pressure'. ;-)

> How do I control the pressure with Nose-Hoover chains? If you can provide me
> with the appropriate
> background reference that would help. I also simple don't see the
> appropriate key word to set the

you set the temperature with &MOTION&MD&TEMPERATURE, right?

now the pressure would be &MOTION&MD&BAROSTAT&PRESSURE.

both have &NOSE as a subsection with the options to the N-H chains.

cheers,
axel.

> pressure with the chains.

[Juerg Hutter]

Hi

the implementation in CP2K is a newer variation of the
algorithm outlined in this paper

@article{Martyna_1994_a,
author = {G. J. Martyna and D. J. Tobias and M. L. Klein},
title = {Constant-pressure molecular-dynamics algorithms},
journal = {Journal of Chemical Physics},
volume = {101},
pages = {4177--4189},
year = 1994
}

For the true technical details you have to ask Chris,
who did the implementation.

regards

Juerg Hutter

----------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut...@pci.unizh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
----------------------------------------------------------

[CJM]

Hi:

As Juerg said, NPT is fully operational. For use of the thermostating
issues, please look in /tests/FIST/regtest and
look for input files with NPT or grep for the section NPT. This
should help.
CJM