Neighbor command for kMC

LAMMPS LAMMPS General Discussion
,
1

I’m trying a kMC simulation to understand interdiffusion around Pd/V interface. I create an interface model whose size is about 60 x 60 x 120 A^3. I want to swap atoms neighbors, so I use “neighbor” command. However, atoms at the center of the models were swapped. It seems likely that “neighbor” command doesn’t work.
Could anyone tell me where the problem is?

---------------input file ---------------------

swap

units metal
dimension 3
boundary p p p
atom_modify map array sort 0 0.0
atom_style atomic
read_data data_000.lmp
variable e equal pe

shell "echo variable rand equal {RANDOM} > rand" include rand variable r equal round(random(18347,434572049,{rand}))

pair_style meam/c
pair_coeff * * …/…/…/library/library.meam Pd V …/…/…/library/PdV.meam Pd V

neighbor 0.0 bin
neigh_modify every 2 delay 2 check yes

variable t equal 873
fix m1 all atom/swap 1 1 $r $t types 1 2

thermo 1
thermo_style custom step temp press pxx pyy pzz lx ly lz etotal pe ke f_m1[1] f_m1[2]
dump 1 all cfg 1 snapshot_*.cfg mass type xs ys zs fx fy fz
dump_modify 1 sort id element Pd V pad 8
run 1000

2

You didn’t study the documentation carefully enough.
Fix atom/swap does what its documentation says it does and that is what you are observing.
Adding a neighbor command doesn’t change that. As its documentation says, using neighbor 0.0 bin will change the neighbor list skin distance from the default value to 0.0 and nothing else.

3

Thank you for replying.

Is there another available command for kMC simulations?

4

I don’t think so. LAMMPS is a molecular dynamics software and not a kMC code.

5

Thank you

6

Redirecting the command fix neighbor/swap used in this paper has not been released.

7

What is the purpose of making this statement here?

It does not contain any meaningful contribution to the discussion (which is two years old) and the question about making code from a paper available is a topic that should be discussed with the authors of the paper or the editors of the journal that published it.
As it happens, this code seems to be included into the following pull request that has been submitted for inclusion into the LAMMPS distribution Kinetic Monte Carlo fix by jtavenner-dev · Pull Request #4424 · lammps/lammps · GitHub, so you can access the command now, if you download a copy of that submitted branch from its LAMMPS fork.