[gmx-developers] Intermolecular parameters for graphene

Tue Nov 25 06:58:06 CET 2014

Hi Alex,

I'm one of the authors of the article mentioned by Gerrit. As you also
observe, there is no easy way of combining graphene and biological
compounds, each community has its own tools which only covers a part of the
possible applications. F.e. we were interested in mechanical properties, so
our FF tries to reproduce existing experimental data, but ignores
electrostatics... although electrostatics are of great importance in
biological molecular systems which often carry partial charges. To my
knowledge (which is likely already obsolete due to the high pace of
research related to graphene and CNTs), there isn't a FF which
comprehensively reproduces all kinds of physical and chemical properties of
these compounds. We mention specifically in the article that our FF makes
compromises in order to keep a simple functional form and be fast, which is
a must for any FF that hopes to be used for simulations on biological time
scales. In particular when used in GROMACS which is fast by default :-)

I beg to disagree with your statement on the Morse potential: it is present
in GROMACS for a long time, although no FF for biological compounds use it
at the moment. We also chose it because it is a pairwise potential, making
it simple to integrate in other (bio oriented) simulation packages like
CHARMM.

In my opinion, the integration of a graphene/CNT FF in GROMACS is a
multidimensional problem, having to consider the speed, the ease of
implementation, the compatibility to existing (bio) FFs, as well as the
ability of reproducing experimental data. Sadly the great majority of the
(many) graphene/CNT FFs ignores the first two or three aspects, making them
a poor choice for this task... How do the proposed FFs fare in these
directions?

Cheers,
Bogdan
On Nov 19, 2014 9:23 AM, "Alex" <nedomacho at gmail.com> wrote:

> Hi there,
>
> Sorry, I do not have institutional access to this journal.
> As a general comment, I am not at all doubting that there is work on the
> subject of developing Morse-like, or any other popular potential forms. For
> instance, the standard bond-order potential is also of the Morse type with
> bond coordination terms. Of course,the work you mentioned could be used as
> basis (additional to what I cited) for parameterizing the FFs already
> implemented within GROMACS).
>
> As far as I know, no bonding (intramolecular) interactions in GROMACS are
> based on Morse-like expressions. As of right now, people who want to do
> large-scale simulations of graphene or BN just go to LAMMPS, and those who
> want to do biomolecular simulations involving those membranes within
> GROMACS, are sort of out of luck. I am just generally hoping there's some
> interest on your end in actually having something built-in. I am not really
> here to promote my paper, I just like GROMACS. :)
>
> Also, could someone rename the thread subject? I really meant
> intramolecular (bonding) interactions.
>
> Thanks,
>
> Sasha
>
>
> On Wed, Nov 19, 2014 at 1:07 AM, Gerrit Groenhof <ggroenh at gwdg.de> wrote:
>
>> Hi,
>>
>> THere has been some work done on FF parameterization recently by the
>> Grater group, see:
>>
>>
>> http://pubs.rsc.org/en/content/articlelanding/2014/cp/c3cp55340j#!divAbstract
>>
>> Gerrit
>>
>>
>> On Nov 19, 2014, at 9:41 AM, Alex wrote:
>>
>> Dear developers,
>>
>> I really hope to get your attention.
>> I may be wrong, but GROMACS still doesn't have a reasonable built-in
>> parameterization for graphene as part of any FF. There are some tutorials
>> online, but they are all quite flaky, based on the parameterizations of
>> organic molecules (as expected). Given the few rounds of chatting with the
>> reviewers I've had for my last paper on the subject, the parameterization
>> is sorely desired. So, maybe it could be a good idea for you to look into
>> it, as people do want a proper model of graphene to simulate it in water in
>> the presence of all the other things that GROMACS simulates so well.
>>
>> For a DREIDING-like model, here are possibly useful papers containing
>> pretty much complete descriptions of the parameter sets:
>>
>> 1. J. H. Los et al* Scaling properties of flexible membranes from
>> atomistic simulations: Application to graphene*, Phys Rev B. 80, 121405
>> R 2009 .
>> (this model is parameterized by differentiating the LCBOPII bond-order
>> potential, which fails at describing the phonon spectrum of graphene. No
>> dihedral constraints, see below Fig. 1)
>>
>> 2. A Shakouri, T Y Ng and R M Lin *A new REBO potential based atomistic
>> structural model for graphene sheets *Nanotechnology 22(29), 295711,
>> 2011.
>> (no dihedrals here either, based on the second-generation bond-order
>> potential, which fails in the harmonic spectrum)
>>
>> 3. A. Smolyanitsky Molecular dynamics simulation of thermal ripples in
>> graphene with bond-order-informed harmonic constraints, Nanotechnology
>> 25(48), 485701, 2014.
>> (this is my work, the parameterization does have dihedrals, and it is
>> based on the phonon-optimized second-gen bond-order potential, which is
>> described in ref #6 of this work).
>>
>> For the most part, all of these parameterizations do not directly
>> correspond to the "parent" potentials (except around 0 K), as spring-like
>> FFs will generally present a relatively stiff model, but I think this info
>> could be a small step forward. I hope this is useful. If you have any
>> questions on the numbers, please do let me know.
>>
>> Thanks,
>>
>> Sasha
>>
>> --
>> Gromacs Developers mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers
>> or send a mail to gmx-developers-request at gromacs.org.
>>
>>
>>
>> --
>> Gromacs Developers mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers
>> or send a mail to gmx-developers-request at gromacs.org.
>>
>
>
> --
> Gromacs Developers mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_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-developers
> or send a mail to gmx-developers-request at gromacs.org.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20141125/cb129604/attachment.html>