[gmx-users] pressure of the liquid in NPT and NVT

[Mark Abraham]

----- Original Message -----
From: Vitaly Chaban <vvchaban at gmail.com>
Date: Wednesday, November 3, 2010 6:10
Subject: [gmx-users] pressure of the liquid in NPT and NVT
To: gmx-users at gromacs.org

> Hi all,
> 
> Please suggest me what is the reason for the below situation.
> 
> The 1728 TIP4P waters system was simulated in NPT under 1bar
> maintained with Parrinello-Rahman barostat at 300K. Some results are
> below:
> 
> Energy                      Average       RMSD     Fluct.      Drift  Tot-Drift
> -----------------------------------------------------------------
> --------------
> Total 
> Energy               -58121.6    344.637    344.476   0.182549    36.5101
> Temperature                 299.913    4.11684     4.1138 0.00274337   0.548679
> Pressure 
> (bar)              1.39811    326.506    317.633   -1.30948     -261.9
> Box-
> X                       3.74431 0.00844567 0.00843154 -8.45584e-06
> -0.00169118
> Density 
> (SI)                984.768    6.66431    6.65307  0.0067005    1.34011
> 
> ==========================================================
> 
> The density is stable and the system looks like equilibrated one.
> 
> The output GRO was then renamed to conf.gro and the average Box-
> X from
> the previous run was set as the cubic box side length. 

Unless the box was equilibrated at NPT at the start of the NPT run, or the run was very long, the average value over the whole run is not all that likely to be one from the equilibrium range. Take the last frame.

> The barostat
> was switched off with no other changes in MDP. The same 
> properties for
> NVT are below:
> 
> Energy                      Average       RMSD     Fluct.      Drift  Tot-Drift
> -----------------------------------------------------------------
> --------------
> Total 
> Energy               -58075.8    309.575    308.343  -0.477939   -95.5887
> Temperature                 299.916    4.08901    4.08678 -0.00233712  -0.467428
> Pressure 
> (bar)             -18.5469    286.376    285.574    0.37081    74.1628
> 
> 
> So the pressure in NPT is ~1.4 and in NVT is ~(-19). How to 
> explain so
> big difference in average values? Is it possible to get a value of
> around 1 in the NVT?

We had a thread about this last week, and numerous more in the past. These observations are strongly dependent on the length of time over which you observe the average pressure, and the size of the system. Note that your fluctuations dwarf your average... See http://www.gromacs.org/Documentation/Terminology/Pressure_Coupling

Mark

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101103/284363fc/attachment.html>