[gmx-users] Nonsymmetric Potentials

[David van der Spoel]

On Tue, 26 Feb 2002, Mark T. Lusk wrote:

>
>I would like to use GROMACS for an assembly of elastic disks, and the
>associated potential is therefore not symmetric:
>
>V = 0   for   r > rcrit
>
>V = k/2(r-rcrit)^2   for    r < rcrit
>
>Can GROMACS be used for such potentials? I think that it may be possible to
>use table-based interactions for this potential, and I am hoping that
>someone can point me towards information on this topic.
That should be quite straightforward. Fill the tables according to the 
manual, i.e. you supply a table with 7 columns, x, f, f", g, g", h, h"
If we take dispersion (g) and coulomb (h) zero, you only have to compute
f and f", f you have given above, f" is just k. The only point is maybe 
the fact that the force has a discontinuous derivative, which makes the 
second derivative a step function. I don't know off hand how the cubic 
spline interpolation will handle this...

Take care that your charges are zero, and you LJ parameters 1.0, since the 
f function will be multiplied by the LJ parameters.

Groeten, David.
________________________________________________________________________
Dr. David van der Spoel, 	Biomedical center, Dept. of Biochemistry
Husargatan 3, Box 576,  	75123 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://zorn.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++