ZnO nanoindentation simulation


I have been trying to simulate nanoindentation experiment on a ZnO thin film (~120000 atoms). I created the lattice data file and replicated it and equilibrated the structure to create atoms in the simulation box. I am using buckingham/coulomb potential for the simulation. The simulation runs without errors for a certain number of runs but then lammps gives an error msg that it “failed to allocated ____ bytes.”

We have earlier tried simulations with a much higher number of atoms (on a different system) without having the memory problem, I am unable to figure out what might be the problem with the code. Does the calculations pertaining to buckingham potential require higher memory? Pls help.

3d indenter simulation

units metal
dimension 3
boundary s s s
atom_style charge
echo both

read_data znoequlbtd.data

pair_style buck/coul cut off 2.5
#kspace_style ewald/n 1.0e-4
pair_coeff 1 2 529.70 0.3581 0.0
pair_coeff 2 2 9547.96 0.219 32.0
pair_coeff 1 1 0.0 0.0 0.0

timestep 0.02

Energy Minimization

minimize 1.0e-4 1.0e-6 1000 10000

define groups

region 1 block -1.7771 83.6513 -5.3171 0.0 -0.1581 50.7263 units box
group lower region 1
group mobile subtract all lower

region indentvol block 10 70 50 85.0753 -5 55 units box
group local region indentvol

#dump 1 all atom 10 dump.indentag.lammpstrj

log indentzno.txt

initial velocities

compute new mobile temp
velocity mobile create 300 482748 temp new

fix 1 mobile nvt temp 300.0 300.0 0.05
fix 2 lower setforce 0.0 0.0 0.0

compute allrdf all rdf 100
fix out1 all ave/time 1 1 10 c_allrdf file tmp_load.rdf mode vector
compute localrdf local rdf 100
fix out2 local ave/time 1 1 10 c_localrdf file tmp_load_local.rdf mode vector
compute csp_all all centro/atom 12

assigining velocity to the indenter in y direction—loading

variable y equal “115.2-stepdt15” #115.2 = yhi+indenter radius+0.2,
print “y is $y”

indenter position and radius at onset of loading

dump 5 all custom 50 dump.indentcspall.lammpstrj id type x y z c_csp_all
fix 4 mobile indent 1000.0 sphere 40 v_y 25 30.0 units box

thermo 10
thermo_style custom step temp fnorm v_y f_4[1] f_4[2] f_4[3]

run 70
print “y is $y”

znoequlbtd.data (3.97 MB)

well you ran out of memory. This is usually
b/c something went badly wrong with your simulation.
Look at the log files, print out thermo into frequuenly, visualize it, etc.
You have to debug what is going wrong yourself.