[gmx-users] Preparing a silicon bulk

Justin Lemkul jalemkul at vt.edu
Thu Dec 18 18:45:33 CET 2014

On 12/18/14 12:38 PM, Kevin C Chan wrote:
> Dear Users,
> I wish to simulate a Silicon bulk using Gromacs. I have created the bulk
> using a plugin in VMD but I found difficulties when generating a top file
> from this pdb file. I have tried pdb2gmx and x2top but both does not work.
> Gromacs tells me that although we already have parameters for SI in maybe
> ffnonbonded.itp, we do not have SI in the residue topology database and also
> ffbonded.itp. How could it be and I do appreciate if anyone could share his
> experience of simulating a silicon bulk in Gromacs.

pdb2gmx is not the answer, since it was designed for linear polymers with 
limited branching.  The only option for materials like this is g_x2top.  That 
requires creation of a suitable .n2t file, the format for which is described in 
http://www.gromacs.org/Documentation/File_Formats/.n2t_File - the contents 
should be fairly simple (one or two lines?) for silica.


> Another thing, I would like to assign different charge to different SI atom
> in my bulk. How can I achieve this?

...if you want to do this, there is no automated method.  You could, in theory, 
assign different residue names to each Si unit and assign that charge, but then 
the parameters of O are likely to be contingent upon the type of Si to which it 
is attached.  This eliminates g_x2top, which requires uniformity.  And as said 
above, this is not a task suited for pdb2gmx.  So either you need to (1) use 
g_x2top and hack or (2) write your own sort of code/script to generate the topology.



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


More information about the gromacs.org_gmx-users mailing list