[gmx-users] Creating a temperature gradient in water

[Alex]

Physics-wise, this is actually a pretty crazy scenario regardless of
implementation. A similar approach with crystals is something I have done
routinely, and it is considered far more palatable. With solids, restraints
are no longer required and you can generate your heat fluxes all you want.
Even then, fluxes at a given temperature T are accomplished via setting
thermostatted groups at T-dT and T+dT, where dT<< T. To give a sense of the
numbers, it'd be something like 295K and 305K for T = 300K, and even then
such things come under a lot of [well-deserved] scrutiny during reviews,
because with distances of tens to hundreds of nm, we'ra talking about
10^6-10^7 K/m in gradient.

Ibrahim is asking about not just an enormous gradient, but also a case with
phase difference between he thermostatted groups. I'd like to see a
visualization posted on youtube! : )

Good luck anyhow.

Alex

On Tue, Jan 10, 2017 at 12:05 PM, Mark Abraham <mark.j.abraham at gmail.com>
wrote:

> Hi,
>
> Yes, as Alex says, you can get something done, but whether it is a
> reasonable model of reality is another matter. Having the "bath" groups at
> even wider temperature ranges, and analyzing only a subset of the data
> could help mitigate end effects from whatever regime generates the
> gradient. I would read carefully, and plan how I'm going to show my
> simulation is a model, and not just a computation :-)
>
> Mark
>
> On Tue, Jan 10, 2017 at 8:00 PM Alex <nedomacho at gmail.com> wrote:
>
> > I would like the asker to seriously read into what Mark has suggested to
> > help reduce the number of published papers with various degrees of setup
> > nonsense.
> > Mark's restraint strategy readily accomplishes the gradient, but creates
> > two partial "pistons" at the thermal boundaries due to restraints. If I
> was
> > to go for this, I'd use a really wide box. A really wide one.
> >
> > Alex
> >
> > On Tue, Jan 10, 2017 at 9:57 AM, Mark Abraham <mark.j.abraham at gmail.com>
> > wrote:
> >
> > > Hi,
> > >
> > > Naturally one can only have a monotonic gradient over half of a
> periodic
> > > direction of a simulation cell, ie between two themostatted groups.
> Also,
> > > if you rely on the temperature coupling groups to implement that, then
> > > these are fixed at grompp time, so diffusion will tend to wreck the
> > intent.
> > > One could either use (flat-bottomed?) position restraints to keep the
> two
> > > groups localized, or periodically stop the simulation and use gmx
> select
> > to
> > > re-create index groups with the desired spatial locality. Obviously,
> the
> > > remaining water should not be coupled to a thermostat.
> > >
> > > Mark
> > >
> > > On Tue, Jan 10, 2017 at 5:44 PM Alex <nedomacho at gmail.com> wrote:
> > >
> > > > Erik,
> > > >
> > > > Even with PBC, something like Ibrahim's scenario is possible in
> > principle
> > > > (T_1 in the region at the center of the box, T_2 at the edges).
> > However,
> > > > can this geometry-based thermostatting for fluids be accomplished in
> > GMX?
> > > >
> > > > Alex
> > > >
> > > > On Tue, Jan 10, 2017 at 8:15 AM, Erik Marklund <
> > erik.marklund at kemi.uu.se
> > > >
> > > > wrote:
> > > >
> > > > > Dear Ibrahim,
> > > > >
> > > > > Do you use pbc? If so, how does that work with your gradient?
> > > > >
> > > > > Kind regards,
> > > > > Erik
> > > > >
> > > > > > On 10 Jan 2017, at 15:00, ibrahim khalil <
> > > > ibrahim.khalil.chem at gmail.com>
> > > > > wrote:
> > > > > >
> > > > > > Dear gromacs users,
> > > > > >
> > > > > > I have a simulation box containing nothing but water (TIP3P). I
> > want
> > > to
> > > > > > create a temperature gradient within water (ie. 300K in the left
> > side
> > > > and
> > > > > > 500K in the right side of the simulation box).
> > > > > >
> > > > > > I have successfully applied different temperatures to different
> > > groups
> > > > in
> > > > > > my other simulations (for example, different temperatures for
> water
> > > and
> > > > > > proteins).
> > > > > >
> > > > > > How can apply different temperatures within the same group of
> > liquids
> > > > (in
> > > > > > my case, water)?
> > > > > >
> > > > > > Thanks in advance.
> > > > > > --
> > > > > > Gromacs Users mailing list
> > > > > >
> > > > > > * Please search the archive at http://www.gromacs.org/
> > > > > Support/Mailing_Lists/GMX-Users_List before posting!
> > > > > >
> > > > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > > > > >
> > > > > > * For (un)subscribe requests visit
> > > > > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_
> gmx-users
> > > or
> > > > > send a mail to gmx-users-request at gromacs.org.
> > > > >
> > > > > --
> > > > > Gromacs Users mailing list
> > > > >
> > > > > * Please search the archive at http://www.gromacs.org/
> > > > > Support/Mailing_Lists/GMX-Users_List before posting!
> > > > >
> > > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > > > >
> > > > > * For (un)subscribe requests visit
> > > > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
> > or
> > > > > send a mail to gmx-users-request at gromacs.org.
> > > > >
> > > > --
> > > > Gromacs Users mailing list
> > > >
> > > > * Please search the archive at
> > > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > > > posting!
> > > >
> > > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > > >
> > > > * For (un)subscribe requests visit
> > > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
> or
> > > > send a mail to gmx-users-request at gromacs.org.
> > > >
> > > --
> > > Gromacs Users mailing list
> > >
> > > * Please search the archive at http://www.gromacs.org/
> > > Support/Mailing_Lists/GMX-Users_List before posting!
> > >
> > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> > >
> > > * For (un)subscribe requests visit
> > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > > send a mail to gmx-users-request at gromacs.org.
> > >
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at http://www.gromacs.org/
> Support/Mailing_Lists/GMX-Users_List before posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>