[gmx-users] Free Energy of Liquid Water

André Farias de Moura moura at ufscar.br
Tue Oct 6 20:47:41 CEST 2015


Nathan,

I'm not pretty sure that I understood what you were planning to do with the
temperature, but this is what I would try:

(1) start from an equilibrated box of water

(2) fix the volume and temperature

(3) slowly decouple the electrostatic part of the interactions to get the
electrostatic part of the free energy

(4) slowly decouple the van der Waals interactions to get the dispersion
part of the free energy

(5) these two components may be summed to yield the free energy difference
between the bulk liquid and the corresponding ideal gas, both at the same
(P,T) point.

(6) other (P,T) point may be obtained now by applying standard equations
describing the free energy change of an ideal gas

I cannot tell how hard it should be to do steps 3 and 4 with the proper
precision, but it is doable and these are the limiting steps of the
modeling.

I hope it helps.

André

On Tue, Oct 6, 2015 at 12:01 AM, Nathan K Houtz <nhoutz at purdue.edu> wrote:

> Thank you, Professor Farais de Moura. That method of integrating to an
> ideal gas makes sense. However, I'm having trouble figuring out how to deal
> with the temperature. I hope my question isn't too basic, but I can't find
> any examples online and I know I am misunderstanding how it works. I want
> to integrate from about 200K to 1000K or so (above the critical
> temperature, which I believe is somewhere in the 700's of Kelvin for TIP4P
> water). To do this, I thought I would set the reference temperature to
> 1000, and increment temperature-lambda from 0.2 to 1, thinking that
> temperature-lambda would scale the absolute temperature. But after running
> a the simulations, it's evident that temperature-lambda does not affect the
> thermostat. Should I set the reference temperature to different
> temperatures for each run? What does the temperature-lambda affect in that
> case?
>
> Thanks for your help! Regards,
> Nathan H.
>
> ----- Original Message -----
> From: "André Farias de Moura" <moura at ufscar.br>
> To: "Discussion list for GROMACS users" <gmx-users at gromacs.org>
> Sent: Friday, October 2, 2015 9:26:28 AM
> Subject: Re: [gmx-users] Free Energy of Liquid Water
>
> Apart from stability/convergence issues, I guess that turning off all
> intermolecular interactions should take you to the ideal gas
> straightforwardly, but in a different (P,T) point as compared to your
> target. But if you managed to alchemically turn water into an ideal gas,
> then you just need to apply standard free change for an ideal gas along a
> (P,T) process to achieve your target state.
>
> (I have not found the reference, but I read a paper doing just that with
> Monte Carlo simulations a few years ago)
>
> you should be able to track the conversion of water into an ideal gas by
> means of the g(r) profiles, which should change from the typical TIP4P
> profiles to g(r)=1 for all distances ranging from zero to half of the
> smallest box length.
>
> On Thu, Oct 1, 2015 at 11:44 PM, Nathan K Houtz <nhoutz at purdue.edu> wrote:
>
> > Hi everyone,
> >
> > I would like to use Gromacs to do Thermodynamic Integration (TI) from
> > liquid water (TIP4P model) to an ideal gas, to find the relative free
> > energy. To do this, I believe  one generally integrates above the
> critical
> > point by increasing the temperature above the critical temperature and
> then
> > relaxing the pressure until the system is a diffuse gas. The mdp options
> > documentation is helpful, and I went through an ethanol solvation
> tutorial,
> > but there doesn't appear to be a "pressure-lambda" or a "volume-lambda"
> > option that I could use to do the second part. How can I get Gromacs to
> > calculate the dh/dl derivative while relaxing the pressure?
> >
> > In addition, all of the tutorials I found for thermodynamic integration
> > were for finding solvation free energies. The coulomb and VDW forces are
> > essentially changed from "completely on" to "completely off". But in my
> > case, I'd like to change the temperature and pressure between two nonzero
> > values. I don't want to begin my simulation at 0K and 0atm, but lambda
> > *must* go from 0 to 1. How can I define both starting and ending points
> for
> > the temperature and pressure (or volume, or density)?
> >
> > Thanks for your help!
> > Nathan H.
> > --
> > 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.
> >
>
>
>
> --
> _____________
>
> Prof. Dr. André Farias de Moura
> Department of Chemistry
> Federal University of São Carlos
> São Carlos - Brazil
> phone: +55-16-3351-8090
> --
> 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.
>



-- 
_____________

Prof. Dr. André Farias de Moura
Department of Chemistry
Federal University of São Carlos
São Carlos - Brazil
phone: +55-16-3351-8090


More information about the gromacs.org_gmx-users mailing list