[gmx-developers] Defining an exponential functional form in tabulated potentials

[David van der Spoel]

On 2013-10-18 14:41, Reza Ghafarian Shirazi wrote:
> Thanks for your reply,
>
> It's hard to make any sense of the formatting of your equations
>
>     it's the standard morse equation used in many references
>     <http://lammps.sandia.gov/doc/pair_morse.html>. the buckingham and
>     Lennard jones have a very high degree of deviation and using them
>     makes no sense for my application.
>     I'm defining potentials using ab-initio calculations and at high
>     levels of theory and large basis sets and I need a good potential
>     function.
>
>> and you haven't said what over what quantity your A varies or what r
>> and r* are
>
>     r* is a parameter calculated using fitting, r is the radius, A and
>     should be defined as coefficients of g® and h®, and since I dunno
>     how to define and exponential term I can not define them. This is
>     the problem.
>
So this is basically a Morse potential for non-bonded interactions, right?

Since this basically has the same parameters as a Buckingham potential 
you could just hack it in gmxlib/nonbonded/nb_generic.c instead of the 
Buckingham and turn off the acceleration. Then, if you have it working 
and want to do faster simulations you can come back to the list to 
discuss this. (look for GMX_NBKERNEL_VDW_BUCKINGHAM in that source file).


>
>> Hi,
>>
>> It's hard to make any sense of the formatting of your equations, and
>> you haven't said what over what quantity your A varies or what r and
>> r* are... Do check out manual sections 4.2.13 and 6.9.
>>
>> Mark
>>
>>
>> On Tue, Oct 15, 2013 at 10:02 PM, Reza Ghafarian Shirazi
>> <rezagh at student.uef.fi <mailto:rezagh at student.uef.fi>> wrote:
>>
>>     Dear all,
>>
>>     I'm trying to define modified morse potential in Gromacs to be
>>     used as a forcefield using tabulated potentials.
>>
>>     My potential's functional form is as: ε Exp[-2A(r-r*)] -2 ε
>>     Exp[-A(r-r*)]
>>
>>     So I have to define g® = ε Exp[-2A(r-r*)]  and h®=-2 ε
>>     Exp[-A(r-r*)]. They have to be defined as values in a xvg table
>>     where the table is filled with values of r, f®, f'®, g®,….
>>     My question is how can I make vales of these functions, while
>>     there's a variable in the exponential term.
>>     I saw in the manual on the website
>>     <http://www.gromacs.org/Documentation/How-tos/Tabulated_Potentials#Constructing_the_Table(s)> noting
>>     that h® for buckingham is expressed as h®=AExp[-Br] so I was
>>     wondering how they make the table of h® values while B is a variable.
>>
>>     Solutions which cam ego my mind were:
>>     Should we use some mathematical approximations like Exp[-u]=1-u.
>>     or simply tabulated potentials does not support potentials more
>>     complex than lennard jones, and Buckingham is an integrated
>>     feature which doesn't need the xvv table.
>>     Any suggestions will be appreciated
>>
>>     regards
>>     Reza
>>
>>     --
>>     gmx-developers mailing list
>>     gmx-developers at gromacs.org <mailto:gmx-developers at gromacs.org>
>>     http://lists.gromacs.org/mailman/listinfo/gmx-developers
>>     Please don't post (un)subscribe requests to the list. Use the
>>     www interface or send it to gmx-developers-request at gromacs.org
>>     <mailto:gmx-developers-request at gromacs.org>.
>>
>>
>> --
>> gmx-developers mailing list
>> gmx-developers at gromacs.org <mailto:gmx-developers at gromacs.org>
>> http://lists.gromacs.org/mailman/listinfo/gmx-developers
>> Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-developers-request at gromacs.org.
>
>
>


-- 
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se