Re: How to run NVT simulations without applying periodic boundary conditions?

From: Yinglong Miao (yimiao_at_indiana.edu)
Date: Thu Feb 21 2008 - 09:44:45 CST

Leandro-

It's right that molecules are interacting with the replicated images of
other molecules in the system by applying PBC no matter whether I wrap the
coordinates in the output or not. But there exists a system in a closed box
with constant volume in real experiments and it makes my calculations much
easier. That's why I think NAMD should provide the option to simulate this
case and it should be much easier to implement than NVT simulations with
PBC.

Thanks,
Long

On Thu, Feb 21, 2008 at 9:34 AM, Leandro Martínez <
leandromartinez98_at_gmail.com> wrote:

>
> In general if you are working with the atomic coordinates you should not
> do that. You should use
> periodic boundary conditions and DO NOT wrap
> coordinates. (The wrapping of the coordinates in
> the output is independent of the use of PBC).
> By doing that your water will look like dispersed, but actually all
> molecules are interacting
> with the images of other molecules
> and are moving as if the system
> was infinite. For example, if you want to compute
> diffusion coeficientes thats what you should do.
> If you restrain you molecules to be in some region,
> when they approach the boundaries their movement will
> not make physical sense (unless you want that for
> some particular reason).
> Leandro.
>
>
>
>
>
> On Thu, Feb 21, 2008 at 3:24 PM, Yinglong Miao <yimiao_at_indiana.edu> wrote:
>
> > For NVT simulations with PBC, molecules are wrapped from one side to the
> > other side to avoid evaporation. I am doing some calculations with the
> > atomic coordinates and the wrapping of molecules will mess up my
> > calculations. So I want to keep all molecules in the volume without
> > wrapping. Yeah, I will try applying boundary constraints and/or the constant
> > spherical volume as Jeff and you suggested.
> >
> > Thanks,
> > Long
> >
> > On Thu, Feb 21, 2008 at 3:10 AM, Leandro Martínez <
> > leandromartinez98_at_gmail.com> wrote:
> >
> > > Ah, ok, so you need to add some kind of restraint to avoid you
> > > molecules
> > > to leave the volume you want (why would you want that?). You can do
> > > in namd I think. Without adding restraints it is not possible to do
> > > what
> > > you want.
> > > Leandro.
> > >
> > >
> > >
> > >
> > >
> > > On Wed, Feb 20, 2008 at 10:55 PM, Yinglong Miao <yimiao_at_indiana.edu>
> > > wrote:
> > >
> > > > But you need to set the three cellbasisvectors for constant volume.
> > > > That indicates PBC. So I'm saying the NVT ensemble option for a single
> > > > closed box with constant volume may be missed in NAMD.
> > > >
> > > > Thanks,
> > > > Long
> > > >
> > > > On Wed, Feb 20, 2008 at 3:53 PM, Leandro Martínez <
> > > > leandromartinez98_at_gmail.com> wrote:
> > > >
> > > > >
> > > > > Just set the three axis and do NOT set constant pressure coupling------=_Part_3707_26606213.1203608685393--

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:47:34 CST