energetic Au deposition into polymer

Dear all,

My substrate are coarse-grained PMMA. The monomers from the same chain are are linked by FENE bonds and monomers from different chains interact via LJ. The system was equilibrated. The correspondent data file is attached (“pmma_final.dat”). I am depositing energetic atoms of Au. I divided the deposition regions into 16 different areas, so I have 1 particle deposited for a square nanometer. My script is bellow.
I am playing with langevin damping parameter and I am also using fix bond/break to simulate polymer damage. I am doing deposition in loops, so I can run nvt and also force bias monte carlo between deposition events.

When I use a damping parameters that does not break bonds (0.01), everything is ok. But when I use values higher than 0.02 and some bonds are broken (for my criteria, typically 2-3 broke bonds per deposition), I got this kind of message (after 2 or 3 loops):

ERROR on proc 0: Bond atoms 208 209 missing on proc 0 at step 511010 (…/neigh_bond.cpp:65)

Sometimes it happens just after deposition, sometimes during monte nvt or fb monte carlo.

This happens for 1 or more processors.
Maybe is bad dynamics, but I tracking thermo I cannot see any abrupt variation of temperature. And I tried also to increase number of steps for nvt and for fb-monte carlo.
I have also extended communication between atoms using a larger cutoff “comm_modify cutoff 120.0”

I would appreciated some tip.

Fernanda S Teixeira.

Gold deposition

clear
units metal
dimension 3
boundary p p m
neighbor 2.0 bin
atom_style bond
special_bonds fene
atom_modify map array
bond_style fene

pair_style lj/cut 12

timestep 0.001

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

read_data pmma_final.dat

pair_coeff 1 1 0.07 5.16 12
pair_coeff 2 2 0.2294 2.629 12
pair_modify shift yes mix geometric

---------- Define Interatomic Potential ---------------------

mass 2 197.0

bond_coeff 1 0.08 7.5 0.07 5.16

Potential Simples de Lennard-Jones

neigh_modify every 1 delay 1 check yes

thermo 500
thermo_style custom step atoms pe ke etotal lx ly lz vol temp press zlo zhi density
compute all all temp
thermo_modify temp all lost ignore flush yes

pmma_final.dat (203 KB)

Dear all,

I think I got what is happening. The bond is broken by fix bond/break and one of the atoms are sputtered from the surface and that is normal for deposition. So I have used

“thermo_modify temp all lost ignore lost/bond ignore flush yes” to ignore those atoms.

The atom is gone but I got now the error:

ERROR: Fix bond/break needs ghost atoms from further away (…/fix_bond_break.cpp:406)
*** Error in `lmp_g++’: double free or corruption (!prev): 0x0000000003569460 ***
======= Backtrace: =========
/lib64/libc.so.6(+0x7364f)[0x7f01fe64c64f]
/lib64/libc.so.6(+0x78eae)[0x7f01fe651eae]
/lib64/libc.so.6(+0x79b87)[0x7f01fe652b87]
lmp_g++(_ZN9LAMMPS_NS6ThermoD1Ev+0x34)[0xa66404]
lmp_g++(_ZN9LAMMPS_NS6ThermoD0Ev+0x9)[0xa66479]
lmp_g++(_ZN9LAMMPS_NS6OutputD2Ev+0x27)[0x1252e07]
etc…
.
.
Aborted

Fernanda S Teixeira