[Pw_forum] MD simulations of polar liquids

[Axel Kohlmeyer]

On Wed, 17 Jan 2007, Nichols A. Romero wrote:

NR> I am studying liquid nitromethane in a VERY small unit cell. There are only
NR> 8 nitromethane molecules.
NR> Already a problematic systems because the predominat interaction is vdW. It
NR> is also at a very low density.
NR> I run an NVT simulation for about 8 ps. I wrote a seperate program that
NR> takes each nitromethane molecular
NR> and calculation the translational kinetic energy, rotation kinetic energy of
NR> the molecular (I recalculate
NR> the moment of inertia at each step) and outputs a temperature based on those
NR> quantities, e.g. T_trans=KE_trans/(3*N_mol)
NR> 
NR> Anyhow, if the system is properly equilibrated, I would think that the
NR> average T_trans and T_rot would be equal
NR> to the imposed temperature by the thermostat. However, I am finding that it
NR> is lower by 20-40% lower. My only
NR> ideas are that:
NR> 1. That this system requires ALOT more time to equilibrate.
NR> 2. That initially all the energy must be getting locked into intramolecular
NR> vibrations.

to get equipartitioning of the kinetic energy between all degrees
of freedom, you will have to run individual thermostats on all of
them (so-called massive nose-hoover chains). the transfer between
rotational and translational degrees of freedom is generally slow
and in smaller systems even more so. if you start from a classical
pre-equilibration it is quite likely that you will provide a high
excitation of the vibrational modes, due to bond lengths being 
different. rotations (i.e. orientation) should be less critical,
particularly at lower density. please note, that using massive
thermostatting will improve ergodicity, but also modifies your 
dynamics significantly (= win some, lose some ;-).

also, and that may be even more important here, with only 8 molecules
each molecule will 'see' the _same_ molecule on the 'both sides'
in each direction, due to PBC. which should - to some degree - 
force all rotations to be somewhat synchronized across the whole
'crystal'. i have not looked at anything that small, but you can
should be able to verify the impact of that effect from classical 
MD, where equilibration is not a problem.

cheers,
   axel.

NR> 
NR> Ideas anyone?

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.