[gmx-users] Re: Re: Re: Simulation of silicon oxide surfaces

[Michael Hirtz]

Hi Justin,

Thank you for answering my mail. I'm new to GROMACS, but of course I checked
the manual before posting here... :) I succeeded in generating some other
topologies of smaller molecules, but still my problems in generating a SiOx
substrate persist.

>> I had found the inorganic builder plugin before (you can already
>> download and install it to vmd), but still only end up with a .pdb
>> that of course cannot be translated by pdb2gmx because it doesn't
>> recognize the residues.

> Right, so search the manual for *topology* and you will find the
> information about the appropriate format.  If you want a GROMOS-
> compatible topology, >PRODRG2 might be of use, otherwise you'll have
> to make the topology by hand.

I used PRODRG before to generate topologies for some simple small molecules
but it doesn't help me now, because it doesn't support Si atoms and has a
limit of 300 atoms per molecule (also I could get around the last problem with
only processing a small unit cell and then multiply this cell by hand).

>> When I look at the .psf for the SiN membrane of the tutorial, it
>> seems that the whole substrate is introduced as one molecule. Of course
>> I could build something similar in a GROMACS topography and define all
>> necessary bonds, but what atomtypes should be used and what for the
>> other parameters ([pairs], [angles] and [dihedrals]).

> Again, search the manual for these relevant keywords, as they are
> explained in Chapter 5.

The problem is that even though I know how a topology file should look like in
general (in a way that I understand what the different sections deal with and
how the parameters are incorporated into the simulation), it doesn't help me
in finding the parameters. I can get the bond lengths and angles from the
structure of course, but what I lack is the force constants and atomtypes (of
course I have O and Si in the GROMACS96 ff but no atomtype for
siloxyl-oxygen/silison, although maybe the carboxyl-atomtypes could be
similar), and especially the vdW parameters for non-bonded interaction (of
course these are already there, if I use existing atom types instead of
introducing new types for silicon bound to oxygen and vice versa).

To sum it up: How could I obtain these force constants and vdW parameter for
non-bonded interactions? Or has anyone some reference at hand where these
parameters are published for SiOx-Substrates?

I guess that I don't mind at all if the intramolecular paramters are not
perfect because anyhow I'll fix the whole substrate by position restraints.
What will be important is the interaction nonbonded interaction between the
substrate and the other molecules in the simulation (hence the vdW parameters,
right?).

>> Or would it be feasible to use single position restraint Si and O
>> atoms (or maybe better SiO2 molecules) to build up the substrate?
>> Since I only need the surface and it doesn't matter whether it's
>> crystaline or amorphous I guess that would be also ok...

> You can add position restraints on whichever atom(s) you choose,
> provided that such restraints accurately suit your needs.  I don't
> know what you mean by "build up the substrate," however...

I ment to construct the substrate (i.e. silicon surface) by just positioning
single Si and O atoms (or SiO2 molecules) in an appropriate way instead of
modelling the substrate as one big molecule. Because I guess this structure
would not be stable in itself, I would suppose that one needs position
restraints to keep the substrate together in such a case.

Your thoughts on this would be much appreciated,

Thanks,
    Michael


>> Any comments/suggestions , anyone?

>> Thanks,
>>    Michael


>>> On Tue, Jul 15, 2008 at 9:37 AM, Michael Hirtz <hirtz at uni-muenster.de
>>>> wrote:
>>>> Hello,

>>>> I want to set up a silicon oxide surface with varying -OH surface group
>>>> for my simulation. Does anybody know a tool that could do that or maybe
>>>> even has a topology I could start on and alter manually?

>>>> Thanks for any sugesstions,
>>>>    Michael
>>>> --
>>>> http://www.defux.de
>>>> _______________________________________________
>>>> gmx-users mailing list    gmx-users at gromacs.org
>>>> http://www.gromacs.org/mailman/listinfo/gmx-users
>>>> Please search the archive at http://www.gromacs.org/search before
>>>> posting!
>>>> Please don't post (un)subscribe requests to the list. Use the
>>>> www interface or send it to gmx-users-request at gromacs.org.
>>>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>> --
>>> Diego Enry B. Gomes
>>> Laborat&#65533;rio de Modelagem e Dinamica Molecular
>>> Universidade Federal do Rio de Janeiro - Brasil.
-- 
http://www.defux.de