Problem in modelling ZBL potential properly for sputtering process

LAMMPS LAMMPS Mailing List Mirror
1

Dear Lammps users,

I am trying to simulate the sputtering process. my surface atoms are
Nickel and my incoming particle is Argon. I should model the interactions
between Ni atoms by EAM and interactions between Ar and Ni should be ZBL.
though i am able to model EAM but i am unabe to model the ZBL potential as
it is combined with tersoff potential in Lammps. please suggest me how i
could model EAM and ZBL potentials perfectly.

my input script is as follows

#*****Initilization******#
dimension 3
units metal
atom_style atomic
boundary p p f
newton on
#******Atom Definition*****#

lattice fcc 3.52
region mybox block 0.0 30.0 0.0 30.0 0.0 30.0 units box
create_box 3 mybox

region 1 block 0.0 28.40 0.0 28.40 1.5 8.79 units box
region 2 block 0.0 28.40 0.0 28.40 0.0 1.465 units box
region 3 block 0.0 28.40 0.0 28.40 8.79 18.29 side in units box

mass 1 58.71 #mass of Ni
mass 2 58.71 #mass of Ni
mass 3 39.94 #mass of Ar

create_atoms 1 region 1 units box
create_atoms 1 region 2 units box

timestep 0.0001
pair_style eam
pair_coeff * * Ni_u3.eam

group Nislab region 1
group fixwall region 2
group sputter region 3

fix 1 all nve
fix 3 fixwall setforce 0.0 0.0 0.0
fix 5 fixwall wall/reflect zlo EDGE zhi EDGE units box

thermo 100
thermo_style custom step atoms temp ke pe etotal press vol

dump 1 all atom 100 dump_thermo.lammpstrj

run 10000

fix 4 sputter deposit 100 3 100 12235 region 3 global 18.29 18.29 vx 0.0
0.0 vy 0.0
0.0 vz -155.50 -155.50 units box

pair_style tersoff/zbl
pair_coeff * * NiAr.tersoff.zbl Ni Ni Ar

neigh_modify every 1 delay 1

compute 1 Nislab temp
compute 2 sputter temp
compute_modify 2 dynamic yes
compute_modify 1 dynamic yes

thermo 10
thermo_style custom step atoms temp c_1 c_2 ke pe etotal press vol
thermo_modify lost ignore flush yes

dump myDump all image 5 dump.*.ppm element type center d 0.5 0.5 0.5
dump_modify myDump element Ni Ar Ni backcolor white boxcolor black pad 5

run 100000

2

Dear Lammps users,

I am trying to simulate the sputtering process. my surface atoms are
Nickel and my incoming particle is Argon. I should model the interactions
between Ni atoms by EAM and interactions between Ar and Ni should be ZBL.
though i am able to model EAM but i am unabe to model the ZBL potential as
it is combined with tersoff potential in Lammps. please suggest me how i
could model EAM and ZBL potentials perfectly.

first off, i don't think that the actual potential does matter that
much for a process where atoms hit other atoms with high kinetic
energy. but then again, you can implement any non-existing functional
form in LAMMPS either by writing your own pair style (if you want the
"perfect" perfect solution) or using pair style table for any pairwise
potential (where you should pay particular attention to the fact that
you need a good representation of the repulsive part of your potential
since atoms will get close).

axel.

3

I think Aidan might have a stand-alone

version of ZBL he has experimented with.

Steve

4

Thanks for your suggestion.
in pair coeff command we dont the provision for specifing only particular
type of atoms we want to model. so how can i model my system such that i
have only EAM potential for Ni surface and ZBL potential for AR and Ni
interactions.

5

Here is the src and doc for pair_style zbl. I am going to check it in to
the repository now.

Aidan

pair_zbl.tgz (4.75 KB)