[lammps-users] Three Questions

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,

I have three questions. Any ideas would be appreciated.

1)
We can compute the temperature in a group or in a region. Why couldn't we
compute the pe/atom or ke/atom for region as same as computing the
temperature.

2)
I want to simulate a silicon cube with periodic boundary conditions. I
want to use
Tersoff potential function.
Here is the input script. Is that right?

..
..
boundary p p p
...
lattice diamond 1
...
region box block -1 1 -1 1 -1 1 units box
...
...
newton on
pair_style tersoff
pair_coeff * * ../../../potentials/Si.tersoff Si Si Si
...

3)
In many books and papers the coefficients for Tersoff potential are not as
same as that in the LAMMPS. For example the non-dimensional paramter "n"
is 0.7 in them but in the LAMMPS is 21.97.
I don't understand this difference. Would you please help me?

Ali Rajabpour

2

(1) you can compute the pe/atom of a group, but groups
are static. Regions are dynamic (in terms of what atoms
might be in them). So why not compute pe/atom
of everything and extract what you want?

(2) Tersoff example looks OK. Did it work?

(3) Mail Aidan Thompson with any Tersoff questions.
I am sure the LAMMPS formulation agrees with the
papers if you read the doc page carefully and understand
the formulas.

Steve