a set of FePt Tersoff potential parameters

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,

I am trying to do some research on FePt based on an analytic bond-order potential(ABOP). ABOP is similar to Tersoff potential in the form. I want to change the ABOP to Tersoff potential. The corresponding parameters are easily calculated. There are three types(Fe-Fe, Fe-Pt, Pt-Pt) in ABOP, but eight types(Fe-Fe-Fe Fe-Fe-Pt Fe-Pt-Fe……) in Tersoff potential. How to correspond between three types and eight types?

Parameter Fe-Fe Fe-Pt Pt-Pt

D0 (eV) 1.50000000 2.64759104 3.20000000

r0 (Å) 2.29000000 2.36130052 2.42000000

β(Å−1) 1.40000000 1.45616698 1.61000000

S 2.06931090 2.26243642 2.29559060

γ 0.01157510 0.05633499 0.18546490

c 1.28987160 0.35073555 0.06090710

d 0.34132190 0.16902364 0.08000000

h -0.2600000 0.45035775 0.45500000

2μ(Å−1) 0.00000000 1.58564770 0.95780361

R (Å) 3.15000000 4.20000000 3.75000000

D (Å) 0.20000000 0.20000000 0.20000000

Thanks a lot,

Hao Dong

2

Aidan may have a suggestion. If you look
at the tersoff doc page in LAMMPS and some
of the example tersoff potential files in the
potentials dir, the format of the file is carefully
documented and correspond to the listed formula.

So you should be able to figure this out.

Steve

3

Hao Dong,

ABOP has different bond order term than Tersoff, particularly the three
body term and the angular term. I think you may have to modify Tersoff to
a new pair_style abop.

Best,
Ray

4

Hi Hao Dong,

Actually, after more careful examination, I think you can use the Tersoff
in LAMMPS for your ABOP potential. The differences in 3-body and angular
term can be manipulated by some conversion. For the 3-body, just set m to
1 and for the angular, set h_ters = - h_abop. Other conversions are more
straightforward.

Best,
Ray