Re: Pressure Problems, Excessively High Readings

From: Jeffrey Potoff (jpotoff_at_chem1.eng.wayne.edu)
Date: Wed Nov 09 2011 - 04:02:30 CST

On 11/8/2011 9:43 PM, Rajan Vatassery wrote:
> Dear NAMD-l,
> I am trying to get an equilibrated box of CH2Cl2. I have a cubic box of
> side length 95 angstroms, and I fill it with 4913 molecules. This works
> out to be a lower density by a factor of 2/3 than the 1.33 g/cm^3 that
> would be normal for CH2Cl2. I have minimized and equilibrated for more
> than a nanosecond after a few brief increments of heating up the system
> to 298 K. I have taken parameters from Chem. Eur. J. 2007, 13, 8139 –
> 8152, and also from Chemical Physics
> 72 (1982) 141-145 (separately). All relevant files and a few images are
> located at http://www.tc.umn.edu/~rajan/NAMD-l.html
> I get a very large pressure (350 bar) where the pressure really should
> not be this high (it should be 1.01 bar). This is similar to these
> peoples' problems:
> http://www.ks.uiuc.edu/Research/namd/mailing_list/namd-l/12038.html
> http://www.ks.uiuc.edu/Research/namd/mailing_list/namd-l/1596.html
> and a few others that I don't have at hand. In my case, the pressure
> increased from -130 bar to 350 bar upon heating the system (100 K to
> 298) and I don't think it's an error due to a pathological portion of my
> simulation because of the steady rise in this trend. I also tried
> removing the PME portion of the calculation ("PME no"), but this did not
> solve the problem. Could it be the case that my system is simply not
> equilibrated? I looked at the log file and the pressure is not really
> moving with each output step.
> This is a problem because I need to equilibrate this solvent box at 1
> atm and when I set the barostat to this setting, the system boils. It
> almost seems like a unit conversion problem or something simple like
> that. It looks like I'm not the only one experiencing it, although this
> doesn't necessarily mean I'm not making a simple mistake.
>

First, I would not trust either of the models you are using to be reliable.

Second, your parameters from the Chem Phys paper are clearly wrong. I
don't have that paper in from of me, but converting your rmin/2 values
to sigma gives a value about 6 A, which is completely unreasonable for a
Cl or a C atom. Similarly, and RDF for H-Cl interactions shows a peak
at 6 A, which is also unreasonable.

-- 
=================================================================
Jeffrey J. Potoff			  jpotoff_at_wayne.edu
Associate Professor			  Wayne State University		
Department of Chemical Engineering and Materials Science
5050 Anthony Wayne Dr			  Phone:(313)577-9357		
Detroit, MI 48202  	             Fax:  (313)578-5815
http://potoff1.eng.wayne.edu
=================================================================

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:24:31 CST