Silvera-Goldman Potential

Hello,

I am currently working on a project where the aim is to study the elastic properties of solid hydrogen during a cooling process. While searching through the LAMMPS folder, I found found text files discussing the Silvera-Goldman Potential in the tools/i-pi/drivers directory. I would like to use this potential in my simulation as it has been shown to be more accurate in tests with hydrogen than other potentials. Is it possible to implement this potential in LAMMPS?

Thank You,
Ben

Hello,

I am currently working on a project where the aim is to study the elastic
properties of solid hydrogen during a cooling process. While searching
through the LAMMPS folder, I found found text files discussing the
Silvera-Goldman Potential in the tools/i-pi/drivers directory. I would like
to use this potential in my simulation as it has been shown to be more
accurate in tests with hydrogen than other potentials. Is it possible to
implement this potential in LAMMPS?

certainly. LAMMPS is designed to be easily extended. in this case,
you'll even have a reference implementation (in fortran) available.
to implement a new potential in LAMMPS, please start to read here:

http://lammps.sandia.gov/doc/Section_modify.html

if you are successful, consider documenting your effort and
contributing the new potential style back to the LAMMPS community (a
large part of LAMMPS are contributions from other users).

axel.

You can also look at pair_mean.cpp, pair_reax.cpp, pair_kim.cpp

as examples of pair styles (written in C++) which wrap external

libs to do the heavy lifting.

Steve

You can also look at pair_mean.cpp, pair_reax.cpp, pair_kim.cpp
as examples of pair styles (written in C++) which wrap external
libs to do the heavy lifting.

steve,

if you look at the (cleanly written and well commented) code in
tools/i-pi/drivers/SG.f90 you'll see that interfacing to ~150 lines of
fortran code would be overkill in this case and writing a new pair
style based on , e.g. morse or buck likely to be less effort.

axel.

Hi,

I appreciate the responses and discussion. Unfortunately, I am somewhat inexperienced when it comes to programming. Do you have any recommendations as to where I should begin when trying to modify LAMMPS for a new potential.

Thank You,
Ben

Hi,

I appreciate the responses and discussion. Unfortunately, I am somewhat
inexperienced when it comes to programming. Do you have any recommendations
as to where I should begin when trying to modify LAMMPS for a new
potential.

i already pointed you to the part of the LAMMPS documentation that
covers this topic. there are some talk slider from previous LAMMPS
workshops and tutorials elsewhere on the LAMMPS that discuss how to
modify LAMMPS as well. short of actually doing the programming for
you, that is the best help you can get. taking a similar pair style
and adapting it for a different potential, is one of the more easy to
do modifications/extensions to LAMMPS.

axel.