[gmx-users] Energy minimisation problem

[Kukol, Andreas]

First try the EM/MD of 1 molecule in vacuum ('ethanol-gas') and check, if it looks reasonable.

You could also have a look at the definition of ETHH in the GROMOS96 forcefield, e.g. in ffG43a2.rtp. Although this is a united atom topology it might give you clues, if something is wrong.

Andreas

> -----Original Message-----
> From: gmx-users-bounces at gromacs.org [mailto:gmx-users-bounces at gromacs.org]
> On Behalf Of Jussi Lehtola
> Sent: 13 May 2008 11:36
> To: gmx-users at gromacs.org
> Subject: [gmx-users] Energy minimisation problem
>
> Hi,
>
>
>
> I have constructed topologies for methanol, ethanol, propanol and butanol
> using x2top with the atomic coordinates from NIST Chemistry WebBook. I have
> modified the topologies by putting the CH3s, CH2s and OHs in their separate
> charge groups.
>
> Now I have run into problems when trying to minimize the energy of the
> configuration. I have used a system with 5000 molecules, with the box size
> chosen to give an initial mass density of the order 500 kg/m^3 (which is
> less than the physical densities).
>
> The system behaves really badly: the potential energy sticks to high
> positive numbers (order of 10^6). I have tried with smaller densities with
> no success. Sometimes the minimization gets energies of -10^8, but the
> forces remain large. A few iterations after the potential energy attains
> again positive values, and then converges to machine precision.
>
> Increasing the step size doesn't help, neither does using another minimizer
> (steep or l-bfgs): the system stays non-bonded.
>
> Interestingly, if one generates the system using the -rot switch of genconf,
> the potential increases some 10 orders of magnitude compared to the default,
> -norot.
>
> What could be the problem?
>
> Attachments: ethanol coordinates & topology
> --
> ------------------------------------------------------
> Jussi Lehtola, FM, Tohtorikoulutettava
> Fysiikan laitos, Helsingin Yliopisto
> jussi.lehtola at helsinki.fi, p. 191 50673
> ------------------------------------------------------
> Mr. Jussi Lehtola, M. Sc., Doctoral Student Department of Physics,
> University of Helsinki, Finland jussi.lehtola at helsinki.fi
> ------------------------------------------------------