Charged particle deposition on Atomic style data file and what are best Ti-Si and Ti-O potentials?

LAMMPS LAMMPS General Discussion
, ,
1

Hi,
I am trying to deposit TiO2 on Si substrate. ( simulate a magnetron sputtering on Si substrate)
First, I have first formed Si substrate found the equib lattice via NPT runs, removed half plane in the top surface and added 10A buffer zones to the data file in both z directions, removed the periodicity in z in the input script and ran it for NVT. Now I have the Si surface, but I used atomic style for generation of the surface so my current data file is in atomic format. Now in magnetron sputtering particles are charged so I have to include that also.

Data File

LAMMPS data file via write_data, version 29 Sep 2021, timestep = 3100000

7303 atoms
3 atom types

0.01887559500250191 54.25337321499684 xlo xhi
0.01887559500250191 54.25337321499684 ylo yhi
-10.00000 65.00000 zlo zhi

Masses

1 28.0855
2 47.867
3 16.000

Atoms # atomic

1 1 1.3407728263786889 1.3583167141708739 1.3925263426717569 0 0 0
2 1 6.7274841734057009 1.3185113025147779 1.3931498266753020 0 0 0
3 1 8.1276281904510981 0.1533973093788366 2.7763599479603318 0 0 0
4 1 10.7926763723013419 2.7796469604562137 2.7725943886935775 0 0 0

----------

I have two questions:

1) My data file initially had 1 atom type so I added 2 and 3 for Ti and O later before I start the deposition. How can I start a charged particle deposition on this existing data file? what is the input stricpt modification to address this change? Otherwise deposition of Ti and O directly will be just neutral particles falling down right? I think I also have to define atomic charges for Ti and O explicitly.

2) I have been trying to find the correct potentials for these three atom types. Especially for Ti-Si and Si-O pairs. Need your help if it is true so far in my script below and what are the best potentials for these pairs?

Attahched is my input script:
with the data file above…

---------- INPUT SCRIPT ---------------------

clear
units metal
dimension 3
boundary p p s
atom_style atomic # Should I change it here?

---------- Create Atoms ---------------------

#lattice diamond 5.431
#region box block 0 5 0 5 0 5 units lattice
#create_box 1 box
#create_atoms 1 box

#mass 1 28.0855
#----------------------------------------------

read_data SiSurface.data

mass 2 47.867 #Ti
mass 3 16.000 #O

region surface block 0 55 0 55 33 65 units box
region mid block 0 55 0 55 17 33 units box
region bottom block 0 55 0 55 0 17 units box

group surface region surface
group mid region mid
group bottom region bottom

group Si type 1
group Ti type 2
group O type 3

----HELP!------ Define Interatomic Potential !!! ---------------------

I need Ti-Si and O-Si interaction potentials here, which one?

pair_style hybrid tersoff buck/coul/long 2.5

pair_coeff 1 1 tersoff 1 Si.tersoff Si

I am getting ERROR for this line!

ERROR: Incorrect args for pair coefficients (…/pair.cpp:805)
Last command: pair_coeff 1 1 tersoff 1 Si.tersoff Si

pair_coeff 2 2 buck/coul/long 2.5 31120 0.154 5.250 # Ti-Ti
pair_coeff 3 3 buck/coul/long 2.5 11783 0.234 30.220 # O -O
pair_coeff 3 2 buck/coul/long 2.5 16958 0.194 12.590 # O -Ti

kspace_style ewald 1.0e-4

neighbor 1.0 bin
neigh_modify every 1 delay 10 check yes

---------- Define Settings ---------------------

#compute eng all pe/atom
#compute eatoms all reduce sum c_eng

---------- Run Minimization ---------------------

reset_timestep 0

min_style cg
minimize 1e-12 1e-12 5000 10000

thermo 1000

thermo_style custom step pe lx press temp

#!!!PLEASE NOTE THAT DAMP PARAMETERS ARE IN TIMEUNITs here in PicoSec, it is a good practice

to have damping at every 100 TIMESTEPS so here 0.1Ps=100fs which is 100 timesteps in metal

#units!!!

velocity all create 300 3243434

#fix 1 all npt temp 300 300 0.1 iso 0.0 0.0 1.0

fix 1 surface nve
fix 2 mid nvt temp 300 300 0.1
fix freeze bottom setforce 0.0 0.0 0.0

fix dep Ti deposit 3000 2 1000 342535 region surface vz -1.0 -0.8
fix dep O deposit 3000 3 500 342535 region surface vz -1.0 -0.8

dump 1 all custom 1000 Diamond_dumps/Ti_depon_Si.dumps.* id type x y z

run 100000

Thank you very much in advance.

2

Quick note:

image
image1364×882 267 KB

I found this for TiO2, can I use it for Ti and O interations? or this is just for the molecule?

But here, the file that I pulled from Lammps potentials folder, showing the Ti mass wrong. ( it is Al mass listed there)

3

You have two options:

  1. modify the data file so that it conforms to the required format for atom style charge. this is more work at the beginning, but simpler for restarting and reusing the resulting data file after equilibration etc.
  2. use fix property/atom with with the options q ghost yes to add the ability to store charges after the fact. this is trivial to add but requires special care when restarting or rereading data files as the charge data will be stored in a separate section instead of in the Atoms section.
4

This information should be provided in the corresponding publication. If not, or for additional questions, you should contact the authors of the paper, which also happen to be the contributors of the potential file.

5

Thank you Alex. I will try the first one if I can. As I read, data file with charges contains the charge before the coordinates:

so I will replace this
Atoms # atomic

1 1 1.3407728263786889 1.3583167141708739 1.3925263426717569 0 0 0
2 1 6.7274841734057009 1.3185113025147779 1.3931498266753020 0 0 0

with this format

Atoms # charge

1 1 0 1.3407728263786889 1.3583167141708739 1.3925263426717569 0 0 0
2 1 0 6.7274841734057009 1.3185113025147779 1.3931498266753020 0 0 0

inserted the third element in the row after id-type-Charge(0 for the substrate)-x-y-x-nx-ny-nz

But what about the potentials to use for Ti-Si and O-Si interactions? any idea?

Thank you.