[gmx-users] Surface tension readings in NVT vs NPT simulations

[David van der Spoel]

On 2011-03-12 16.45, Denny Frost wrote:
> I have run NPT simulations using isotropic and semiisotropic coupling
> with the same results.  I have never done coupling in just one direction
> though, how do you do this?.  I have never used Dispersion corrections.
> It seems to me that this would help, rather than hurt though since, as
> Aldi said, it will make the system closer to experimental values.  I
> will give this a try and see what happens.  My question still remains -
> why do NPT and NVT simulations give such different values for surface
> tension?
> Denny Frost

You don't give any values so it is hard to judge.
- NVT may have completely wrong pressure
- Dispersion correction assumes a homogeneous system as regards the 
average disperson constant per volume, which you probably do not have. 
E.g. in an ice/water surface dispersion correction may induce melting.

The dispersion correction is *not* to bring your system closer to 
experiment but rather to correct for the use of a cut-off.

- Coupling in one direction: specify e.g.
ref-p = 0 0 1
compressibility = 0 0 4e-5
tau_p = 0 0 5
>
> On Sat, Mar 12, 2011 at 6:40 AM, aldi asmadi <aldi.asmadi at gmail.com
> <mailto:aldi.asmadi at gmail.com>> wrote:
>
>     David,
>
>     I have a question that is still related to your reply.  If the bulk
>     liquid NPT and the interfacial liquid-vapor NVT simulations are
>     performed using dispersion corrections to the pressure and energy,
>     while the intefacial liquid-liquid NPAT simulation don't use any
>     correction, can we say that all results are valid since we don't give
>     the same treatment for all systems?
>
>     In the NPT and NVT calculations, we apply corrections in order to
>     reduce the discrepancy between the calculated and experimental
>     properties (say density and surface tension) as small as possible.
>     Here we have more confidence that our molecules in systems behave
>     accordingly judging from the macroscopic values we obtain.  Meanwhile,
>     in the NPAT calculation, we don't use such correction meaning that the
>     property (say interfacial tension) is expected to deviate more from
>     the experimental value? This indicates that the system behaves
>     differently in comparison to the same simulation conducted with
>     correction?
>
>     Many thanks,
>     Aldi
>
>
>     On Sat, Mar 12, 2011 at 11:44 AM, David van der Spoel
>     <spoel at xray.bmc.uu.se <mailto:spoel at xray.bmc.uu.se>> wrote:
>      > On 2011-03-12 06.09, Denny Frost wrote:
>      >>
>      >> I am running MD simulations on Liquid/Liquid interfaces and
>     measuring
>      >> the interfacial tension between them.  I have found that the
>     readings in
>      >> NVT simulations are close to experimental values, but have a lot of
>      >> variation.  I run NPT simulations on the exact same system and
>     find the
>      >> results show very little variation, but the values are far from
>      >> experimental results.  Does anyone know why this happens?
>      >>
>      > Please be more specific. How do you do NPT simulations? This may
>     influence
>      > the result. To get good result I would suggest to do pressure
>     coupling only
>      > in the normal direction and to turn off dispersion corrections to the
>      > pressure.
>      >
>      > --
>      > David van der Spoel, Ph.D., Professor of Biology
>      > Dept. of Cell & Molec. Biol., Uppsala University.
>      > Box 596, 75124 Uppsala, Sweden. Phone:  +46184714205.
>      > spoel at xray.bmc.uu.se <mailto:spoel at xray.bmc.uu.se>
>     http://folding.bmc.uu.se
>      > --
>      > gmx-users mailing list gmx-users at gromacs.org
>     <mailto:gmx-users at gromacs.org>
>      > http://lists.gromacs.org/mailman/listinfo/gmx-users
>      > Please search the archive at
>      > http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>      > Please don't post (un)subscribe requests to the list. Use the www
>     interface
>      > or send it to gmx-users-request at gromacs.org
>     <mailto:gmx-users-request at gromacs.org>.
>      > Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>      >
>     --
>     gmx-users mailing list gmx-users at gromacs.org
>     <mailto:gmx-users at gromacs.org>
>     http://lists.gromacs.org/mailman/listinfo/gmx-users
>     Please search the archive at
>     http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>     Please don't post (un)subscribe requests to the list. Use the
>     www interface or send it to gmx-users-request at gromacs.org
>     <mailto:gmx-users-request at gromacs.org>.
>     Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se