[gmx-users] Very large Max Force [Energy minimisation] in water with ions
Justin Lemkul
jalemkul at vt.edu
Tue Nov 11 13:36:48 CET 2014
On 11/11/14 5:12 AM, Kester Wong wrote:
> Dear Justin and Erik,
>
>
> > Thank you for the feedback. I have also tried the following:
> >
> > i) separating OH- and Na+ ions
> >
> > ii) placing the ions much closer to water droplet
> >
> > iii) increase the spacing between each ions, i.e. using a larger box
> >
> >
> > Perhaps I should try having the ions in a larger box, then fill the box with
> > water molecules and then redo the energy minimisation and NVT?
> >
>
> Well, like Erik said, there really isn't a smoking gun and what I'm observing
> might be either causative or symptomatic, there's just no way to know at this
> point. I don't know what good any of those three options will do, because I
> don't understand what you're trying to accomplish. What was the logic of
> putting a lattice of OH- and random Na+ in vacuo above a water droplet? What is
> the goal of your simulations?
>
>
> Thank you for the valuable input, the goal is to simulate the ions in water droplet.
>
> Instead of having a box of water with the charged ions, I used the
> water/graphene system that has been NVT equilibrated for 20ns, and placed the
> ions on top of the droplet.
>
Adding ions within the droplet should be simple in the case of Na+, just use
genion. As for OH-, that's more complicated.
>
> If adding 20 OH- and 20 Na+ proves to be problematic, is there a way to add the
> ions incrementally without attracting a large force?
>
>
> The OH- ions were arranged in such a way "genconf -nbox 4 5 1" as I have tried
> random insertion, using gmx insert-molecules, and have found that the ions were
> too close together.
>
Generating a grid with genconf is undoubtedly worse than random insertion; you
have the lone pairs of the OH- nearly in contact with one another, hence massive
repulsion with no charges close by to screen the interaction. This is probably
one of the central problems in your simulation. The behavior of gmx
insert-molecules can be changed by supplying different values to -seed or -radius.
-Justin
>
> From my other previous calculations, adding a cluster of H3O+ and Cl- ions
> (ranging from 40 to 200 ions) on top of the water droplet worked fine with NVE
> and NVT.
> However, I have not been able to do so with OH- and Na+.
>
>
>
> Regards,
> Kester
>
> -Justin
>
> >
> >
> > Regards,
> >
> > Kester
> >
> > On 11/10/14 5:04 AM, Erik Marklund wrote:
> > > Dear Kester,
> > >
> > > Hm. No obvious smoking gun as far as I can see. You could try a two-step minimisation where you first do one minimisation with soft-core potentials and then one without.
> > >
> >
> > What's strange to me is the lattice of OH- at the top of the box and the sodium
> > ions that are floating out in the middle of nowhere. I don't understand the
> > construction of this system, but having highly charged stuff in close contact
> > and floating around in vacuum certainly looks like it could be trouble.
> >
> > -Justin
> >
> > > Kind regards,
> > > Erik
> > >
> > > On 7 Nov 2014, at 05:36, Kester Wong> wrote:
> > >
> > >
> > > Dear Erik,
> > >
> > > The starting structure consists of water on graphene that has already been energy minimised, and subsequently equilibrated (NVT) for 20ns.
> > > Using that (water droplet on graphene), I added the ions for energy minimisation.
> > > I followed a two-step minimisation:
> > > i) Minimisation with no constraints (-DFLEXIBLE)
> > > ii) Minimisation with constraints (SETTLES for TIPS3P water, and (-DCONSTRAINTS) for hydroxide)
> > >
> > > With the ions:
> > > I have tried placing the ions box lower toward water droplet, but the Max Force remains large upon energy minimisation.
> > >
> > > The input files can be accessed here:
> > > https://drive.google.com/folderview?id=0B7ym8d6G9-e2dG5pOFQ4SWVyZG8&usp=sharing
> > >
> > > The hydroxide parameter is taken from Gerrit Groenhof, based on the supplementary material in this paper:
> > > http://dx.doi.org/10.1016/j.bpj.2014.04.062
> > >
> > > In short, this model was developed very similarly to the polarisable version of hydroxide in the SWM4-NDP force field.
> > > As for sodium and graphene, the topology parameters were taken from CHARMM27; TIPS3P was used for the water model.
> > >
> > > Visualisation after the energy minimisation showed a distorted virtual site of the hydroxide ions, where one of the four virtual sites no longer conform to the constraint. So, I am guessing whether the extremely large Max Force originates from the force constants in the hydroxide topology?
> > >
> > > Thank you for your time and assistance!
> > > -Kester
> > > --------- 원본 메일 ---------
> > > 보낸사람 : Erik Marklund>
> > > 받는사람 : ">">
> > > 받은날짜 : 2014년 11월 6일(목) 20:35:59
> > > 제목 : Re: [gmx-users] Very large Max Force [Energy minimisation] in water with ions
> > >
> > > Dear Kester,
> > >
> > > The potential energy is highly positive in the first case and the force is enormous in the second case, so no wonder that they blow up. How did you prepare these systems?
> > >
> > > Kind regards,
> > > Erik
> > >
> > > Erik Marklund, PhD
> > > Postdoctoral Research Fellow, Fulford JRF
> > >
> > > Department of Chemistry
> > > Physical & Theoretical Chemistry Laboratory
> > > University of Oxford
> > > South Parks Road
> > > Oxford
> > > OX1 3QZ
> > >
> > > On 6 Nov 2014, at 04:31, Kester Wong > wrote:
> > >
> > >
> > > Dear all,
> > >
> > > I have been trying to energy minimise ~1980 water (tips3p) molecules with 20 OH- anions (with virtual sites and dummy atoms) and 20 Na+ cations, however, I always ended up getting a very large Max Force.
> > >
> > > Using emtol = 1000 and emstep = 0.0001, I get:
> > >
> > > Potential Energy = 3.79607808453231e+09
> > > Maximum force = 9.86741557582003e+02 on atom 42394
> > > Norm of force = 2.38846729392462e+01
> > >
> > > Using emtol = 700 and emstep = 0.0001, I get:
> > >
> > > Potential Energy = -6.43566327273198e+13
> > > Maximum force = 9.24568347101016e+25 on atom 42262
> > > Norm of force = 5.74744822148669e+23
> > >
> > > As expected, both structures blew up during NVT equilibration.
> > >
> > > Could this be due to the arrangement of the ions? I tried separating the anions and cations in two boxes (placed on top of the water droplet), but that also yield a very large Max Force.
> > >
> > > FYI, the water droplet on graphene was equilibrated for 20ns.
> > > Any input is greatly appreciated. Thanks.
> > >
> > > - Kester
> > >
> > > --
> > > 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.
> > >
> >
> > --
> > ==================================================
> >
> > Justin A. Lemkul, Ph.D.
> > Ruth L. Kirschstein NRSA Postdoctoral Fellow
> >
> > Department of Pharmaceutical Sciences
> > School of Pharmacy
> > Health Sciences Facility II, Room 629
> > University of Maryland, Baltimore
> > 20 Penn St.
> > Baltimore, MD 21201
> >
> > jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> > http://mackerell.umaryland.edu/~jalemkul
> >
> > ==================================================
> > --
> > 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.
> >
> >
> >
> >
>
> --
> ==================================================
>
> Justin A. Lemkul, Ph.D.
> Ruth L. Kirschstein NRSA Postdoctoral Fellow
>
> Department of Pharmaceutical Sciences
> School of Pharmacy
> Health Sciences Facility II, Room 629
> University of Maryland, Baltimore
> 20 Penn St.
> Baltimore, MD 21201
>
> jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> http://mackerell.umaryland.edu/~jalemkul
>
> ==================================================
> --
> 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.
>
>
>
>
--
==================================================
Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow
Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201
jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul
==================================================
More information about the gromacs.org_gmx-users
mailing list