Problem with pressure

Dear All users

I am performing a npt ensemble on a confined system(xz=pbc),decane and gold(two wall of gold) at

300 K and 1 bar, I can not understand why my temperature was converged but my pressure was not converged(its average is -890 bar)?

There are 340 decane chains,

How can I solve my problem?

It can be related to my initial configuration?

Dear All users

I am performing a npt ensemble on a confined system(xz=pbc),decane and
gold(two wall of gold) at

300 K and 1 bar, I can not understand why my temperature was converged but
my pressure was not converged(its average is -890 bar)?

There are 340 decane chains,

How can I solve my problem?

It can be related to my initial configuration?

This is my input file, I forgot Adding this,

Hard to tell. Maybe you just didn't wait enough for full equilibration
to happen. You have a large # of molecules. Even if you were using a
united atom FF, 340 decane chains could take a bit to find their
"happy" state when confined between two walls. In molecular liquids
the molecules will experience conformational changes that make system
equilibration to take longer than in a solid system. The longer your
molecule the longer the equilibration time, especially if you are
getting there via MD.
Yet, could be something else.. If I was new to the field I would not
get right away involved with a system of this type. You could test
your decane parametrization/setup by trying to compute the liquid
density at the T and P of interest. If this test fails you have more
fundamental problems to worry about.
Always attack the problem from bottom to top. Make sure you know how
to handle the individual parts of your system, i.e., gold walls and
liquid decane independently before getting them together into a more
messy setup. If you have done this already disregard my comments.
Carlos

Dear Carlos

The how you get to equilibrium depends on your initial config and approach. The equilibrium state doesn’t. I can’t tell you what the exact problem is because you problem phase space is rather large.
Carlos

Hi Ali,

I suppose you try the following

compute 4 mobile temp
compute 5 all pressure 4

fix fxnpt mobile npt temp 300.0 300.0 1 x 1 1 1 z 1 1 1 couple xz

fix_modify fxnpt press 5

By doing this, you specify the barostat used by fxnpt (and the
barostat's temperature compute) to the mobile group other than to the
whole system (which includes the immobile walls). Let's see if this
works.

Cheers,
-Trung

I met with similar problem. I cannot solve it so far.

Dear Ali Alizadeh,

How to test the parametrization/setup by trying to compute the liquid density at the T and P of interest.

Thank you very much and best regards,

Liu Bo

Dear Carlos Campana,

How to test the parametrization/setup by trying to compute the liquid density at the T and P of interest.

Thank you very much and best regards,

Liu Bo