Hi,
I plan to use the Vashishta potential for silicon nitride by manually inputting the relevant parameters to the already existing Vashishta potentials (for example SiC).
I noticed that there are zero entries whenever the neighbor atoms are different. Why is that?
In the LAMMPS manual, it says, “Two-body parameters appearing in entries where the 2nd and 3rd elements are different are stored but never used. It is good practice to enter zero for these values.” This explains why there are zero entries for the two-body potential terms, but what about three-body terms? why are they zero?
Thanks,
J
Hi,
I plan to use the Vashishta potential for silicon nitride by manually
inputting the relevant parameters to the already existing Vashishta
potentials (for example SiC).
I noticed that there are zero entries whenever the neighbor atoms are
different. Why is that?
In the LAMMPS manual, it says, "Two-body parameters appearing in entries
where the 2nd and 3rd elements are different are stored but never used. It
is good practice to enter zero for these values." This explains why there
are zero entries for the two-body potential terms, but what about
three-body terms? why are they zero?
have another, more careful look at the documentation. i am not at all an
expert on the vashishta potential, but what i read from the documentation
is the following:
there are two-body and three-body terms. the file format requires to
provide provide entries for all triples of atom types, but the vashishta
potential implementation only uses a subset of those entries, thus unused
entries have to be present, but may be set to zero. the potential file
parser looks for matching the two-body terms only at the first two atom
types. the three-body terms are only discriminated by the middle atom type.
thus for any binary system with atom types A and B, you need to provide
entries for A-A, A-B, B-B, B-A, X-A-X, X-B-X. the A-A term is in the A-A-A
parameter set; the B-B term in the B-B-B; A-B is taken from the first A-B-X
entry, and B-A from the first B-A-X entry, the two-body terms for those
have to be the same (technically, Zi and Zj are reversed, but due to the
symmetric nature of the two-body interactions, it should not make a
difference). the three-body terms are taken from A-B-X and B-A-X. the
remaining entries are ignored.
this seems to match with the provided files. it is easy to test, whether
this is correct, by setting up a test system and comparing total potential
energies different settings turned on or off in the potential file.
axel.