I'm trying to figure out how to define an potential with functional form I do not see in the manual (looking at 4.08).
I am looking to set a Sanderson potential namely
-Aij*exp{ - [ (Rij-Lij)/d ]**2}
Since I have the functional form I do not need a tabulated potential, but that is still an option. Is a tabulated potential less efficient in terms of computational speed.
Please advise whether there is a fairly easy way to set up the Sanderson potential in DL_POLY, or whether tabulated is better.
I will be using the Sanderson along with Buckingham and Coulombic potentials.
implementing it shall be the better option in terms of speed.
I cannot know how much the speedup will be for that form but is faster than the tabulated on modern hardware.
implementation is relative straight fwd. have look at how few others are implemented.
I suggest you to use the latest version 5.0 and you can contribute it back