deposition de ga and as using a matlab script and lammps.

LAMMPS LAMMPS Mailing List Mirror
1 
Dear all,

i created a matlab script for simulate a deposition of Ga over a GAAS sustrate.
after that i rebuild the simulation and deposit As (removing all Ga in the space).
i use the same lammps script, but i see As Atoms deposited in surface of simulation, move a lot. not in a good structure. 

i read the manual and some message in mailing list and see you speak about a command called: compute_modify Dinamic=Yes, reading about that, this option made atoms going out of one area are considered take the fix of new area, but have some limitacion i dont know if i can use in my simulation. i attach the script.

could i add this option (compute_modify new dynamic yes) to this simulation?

if i use the fix langevin in the base, could be used yet?

thank a lot
With best regards,
Bernardo

simulacion.in (1.3 KB)

2

Your questions and explanations are unclear. There could

be any number of reasons why you don’t get a good structure.

Steve

3

Dear all,

sorry i dont explain clearly.

i modify the code for include langevin, and compute_modify

i am begining to understand some of what do you tell me in previous mails.

due to i put velocity only in g_gas other atoms not update the temp and velocity.

now i incluye the compute temp to all atoms (is this right?)

and include the compute_modify dinamic yes (atoms moving out of one area take properties of new areas).

i make the simulation of deposition in steps, first step i simulate the deposition of Ga atoms in a GaAs sustrate.
after that i analize rebuild the structure and take only the atoms in the botton part of simulation.
include the new As atoms in the upper area and use the same lammps script for simulate the new atoms deposition (becouse i dont know how to use deposit yet).

i dont know if the script i add in the botton of this email is correct or i need to add some more.
Thank a lot.
With King Regards,

Bernardo

Actual code for simulation.

Initialize

clear
units metal
dimension 3
boundary p p f
atom_style atomic

Create atoms

read_data data.GAAS_progresivo

SW Potential

pair_style sw
pair_coeff * * parametros.sw Ga As

Define Settings

#base
region r_bottom block INF INF INF INF INF 3.0 units box
group g_bottom region r_bottom
#sustrato a 49 es el original
region r_middle block INF INF INF INF 3.1 49.9 units box
group g_middle region r_middle
#area intermedia de 50 a 64.9
region r_deposition block INF INF INF INF 50 64.9 units box
group g_deposition region r_deposition

area superior gas

region r_gas block INF INF INF INF 65.0 INF units box
group g_gas region r_gas

#parametros
velocity g_gas create 1000 492459 rot yes dist gaussian
compute 1 all temp
compute_modify 1 dynamic yes

fix nve_deposition g_deposition nve
fix fix_bottom g_bottom setforce 0.0 0.0 0.0
#fix nvt_deposition g_deposition nvt temp 600 800 0.1
#fix nvt_middle g_middle nve
#fix nvt_middle g_middle nvt temp 600 800 0.1
fix nvt_middle g_middle langevin 550.0 650.0 1000.0 639483
fix nvt_gas g_gas nvt temp 1000 1200 0.1
fix muro all wall/reflect zhi EDGE

thermo 5000
#thermo_style custom step temp pe etotal

thermo_modify lost ignore

Diffusion run

dump salidadump all atom 20000 dump.salidaprogresiva
#si el paso es mas grade da NaN (0.05 fallaba).
timestep 0.001

#restart 100000 deposicion.restart
run 1000000

undump salidadump
dump dumpfinal all atom 100000 dump.finalprogresiva

minimize 1.0e-4 1.0e-6 1000 10000

Your questions and explanations are unclear. There could

be any number of reasons why you don’t get a good structure.

Steve

4

Dear all,

sorry i dont explain clearly.

i modify the code for include langevin, and compute_modify

i am begining to understand some of what do you tell me in previous mails.

due to i put velocity only in g_gas other atoms not update the temp and
velocity.

now i incluye the compute temp to all atoms (is this right?)

and include the compute_modify dinamic yes (atoms moving out of one area
take properties of new areas).

i make the simulation of deposition in steps, first step i simulate the
deposition of Ga atoms in a GaAs sustrate.
after that i analize rebuild the structure and take only the atoms in the
botton part of simulation.
include the new As atoms in the upper area and use the same lammps script
for simulate the new atoms deposition (becouse i dont know how to use
deposit yet).

i dont know if the script i add in the botton of this email is correct or i
need to add some more.

as steve was alluding to (in his extremely polite way), it looks like
you are doing the second step before the first.

do you have any proof, that the potential/parameters you are using is
actually parameterized and suitable for the kind of deposition
simulation that you plan to do?
for the kind of model you have, the potential will have to handle
dangling bonds or open valences and handle chemical reactions. the
symptoms and problems you describe hint very strongly that the
potential is *not* suitable for it. in other words, it looks like you
are trying to fix the problem in the wrong place. the factors that you
discuss as being problematic should not be and they are not for
problems depositing metals, for example.

axel.