[lammps-users] A few doubts on MD: Calculation of pair potentials

Dear Dr.Plimpton, users of LAMMPS
I received a valuable advice from the LAMMPS community on the use of PLUMED for free energy calculations.
However, i seek the following clarification. The use of pair potentials in MD are well known. Now, let us consider two elements, say U or Pu. Hypothetically, let us assume that the pair potentials are not known.
How can then we proceed further in finding the effective pair potentials for U-U, U-Pu and Pu-Pu.
I guess that this may not be the question to be submitted to LAMMPS mailing list, still I request the users to throw some light
on this as I would be using LAMMPS to calculate some thermodynamic properties of alloys.
Are their any methods within LAMMPS itself to find out effective potentials for systems when we do not have any experiemental information as to how they interact at high temperature when they are liquids.

Thank you
Suddhasattwa

Are their any methods within LAMMPS itself to find out effective potentials
for systems when we do not have any experiemental information as to how they
interact at high temperature when they are liquids.

No.

Steve

Dear Dr.Plimpton, users of LAMMPS
I received a valuable advice from the LAMMPS community on the use of
PLUMED for free energy calculations.
However, i seek the following clarification. The use of pair
potentials in MD are well known. Now, let us consider two elements,
say U or Pu. Hypothetically, let us assume that the pair potentials
are not known.

then you have a big problem.

How can then we proceed further in finding the effective pair
potentials for U-U, U-Pu and Pu-Pu.
I guess that this may not be the question to be submitted to LAMMPS
mailing list, still I request the users to throw some light
on this as I would be using LAMMPS to calculate some thermodynamic
properties of alloys.
Are their any methods within LAMMPS itself to find out effective
potentials for systems when we do not have any experiemental
information as to how they interact at high temperature when they are
liquids.

what you want to do relates to the astounding stories of
baron munchhausen ( http://en.wikipedia.org/wiki/Baron_M�%
BCnchhausen )

in short, you cannot compute something from nothing.

you will have to research ways to generate some parameters
for the system that you are interested in. if you are
lacking experimental data, you may use (or generate) quantum
chemical data as an approximation. however, particularly for
elements as heavy as you indicate, those calculations are
far from trivial.

however, i suggest to have a closer look into the literature
to see, if there would be some potentials or at least some
studies of similar systems and suggestions for how to generate
suitable parameters.

cheers,
   axel.