[gmx-users] Themostat problem

[Anton Feenstra]

qiao rui wrote:

> Dear All:
> 
> 	I am trying to use Nose-Hoover thermostat for my simulation
> system, where I have ~1500 water molecules, ~50 Na+ and ~50~ Cl- ions. I
> used the following paramters:
> 	tau_t = 0.05 ps; t_ref = 300K for each group.
> 
> 	Using g_energy, I got the following statistics:
> 
> Energy             Average       RMSD     Fluct.      Drift  Tot-Drift
> -------------------------------------------------------------------------------
> T-SOL              300       4.91058    4.90895 -0.000912265   -0.437889
> T-Na               300.003   279.733    243.602   0.992376    476.343
> T-Cl               299.997    385.632    373.082    0.70422    338.027
> 
> 	It seems that the fluctuation of the temperature for Na+ and Cl-
> is huge. Is there some way to solve the problem?

Are you sure it is a problem? I'd imagine that for a small number (~50)
of particles, temperature could fluctuate considerably, this is not
necessarily a problem as long as the average is right (as it appears to
be for your case). If these ions are free in the solvent, I would personally
couple them together with the solvent to one bath, as heat exchange between
water and solvated ions is relatively fast (i.e. you shouldn't see
temperature differences building up between solvent and ions even when
they're coupled as one).

> 	Another question I have is: is it possible to just couple the x-
> and y- degree of freedom with the themostat instead of coupling all the
> three directions to the thermostat?

In principle you could do that, but I don't see the point. THere is one
problem that I see, is that you'd need to define kinetic energy and
number of degrees of freedom in two dimensions in stead of three. It
might be trivial, but I'm not sure... Also, I'm not familiar with the
details of the nose-hoover thermostat, there may be caveats there as well...


-- 
Groetjes,

Anton
  _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610           |
|             | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/       |
|             | "If You See Me Getting High, Knock Me Down"           |
|             | (Red Hot Chili Peppers)                               |
|_____________|_______________________________________________________|