different regions with different temperature in MD

172 views
Skip to first unread message

Jun

unread,
Feb 2, 2009, 10:05:31 AM2/2/09
to cp2k
Dear all,

Is it possible to define different regions with different temperatures
in a MD simulation using cp2k? If yes, how to do that?
Many thanks in advance.

Jun

Teodoro Laino

unread,
Feb 2, 2009, 10:09:37 AM2/2/09
to cp...@googlegroups.com
Dear Jun,
it is possible to define different thermostatting regions but all of
them have the same temperature.
It would not work (most of the time) having different regions at
different temperatures unless the regions
are not fully adiabatic. In fact, different regions at different
temperatures ( in case of energy flow between
the different regions) would never thermalize to the temperature you
asked for..

Cheers
Teo

Jun Cheng

unread,
Feb 2, 2009, 10:28:23 AM2/2/09
to cp...@googlegroups.com
Thanks a lot for rapid reply, Teo.

What I am trying to simulate is that a few atoms or molecules in an equilibrated system are excited to higher vibrational levels (e.g. due to photo adsorption), and see how the system evolves afterwards. I am thinking of assigning the excited species an high temperature initially to mimic the situation of excitation by photo adsorption, and then seeing how the system reacts. So, it is not necessary to keep the excited species at the high temperature during MD simulations. Please let me know what do you think. Thanks.

Jun

Teodoro Laino

unread,
Feb 2, 2009, 10:37:42 AM2/2/09
to cp...@googlegroups.com
Hi Jun

then you don't need at all the different thermostatting region at
different temperatures.
You can specify velocities in the section &VELOCITIES in &SUBSYS.
These velocities act like restart quantities.

You can (with a small external program) determine the velocities of the
hotter atoms through a Boltzman distribution at temperature T1, and the
velocities of the other atoms at temperature T2 (T1>T2).
You can then input these quantities in the &VELOCITIES section and run
your MD.

Of course you may consider about running an NVE ensemble instead of the
NVT (which you were mentioning in your previous message). NVE makes
more sense to me in order to inspect heat transfer (but of course more
expert people can comment on this point).
Cheers
Teo

Jun Cheng

unread,
Feb 2, 2009, 10:43:15 AM2/2/09
to cp...@googlegroups.com
Many thanks for the helpful suggestions.

Cheers,
Jun

marci

unread,
Feb 2, 2009, 11:19:00 AM2/2/09
to cp2k
Dear Jun,

in addition to what already said, I would add that it
is also possible to define different regions and initialize a
different temperature in each of them by using the input section
CP2K_INPUT / MOTION / MD / THERMAL_REGION

The temperature of each region can be kept within a given range
using velocity rescaling (keyword TEMP_TOL), which is a brute force
technique to control
the temperature and does not correspond to any statistical ensemble.
This tool has been introduced to simulate temperature gradients, heat
flows
and thermal spikes. It it is not meant to work together with proper
thermostats.

Kind regards,
Marcella

Jun Cheng

unread,
Feb 2, 2009, 12:42:57 PM2/2/09
to cp...@googlegroups.com
Many thanks, Marcella.

Jun
Reply all
Reply to author
Forward
0 new messages