Workpiece deformation about system relaxation in NVT

Dear Lammps uses

After relaxation operation in NVT, the middle of the green workpiece arched upward, is this a normal phenomenon?And how to solve it?

The following commands were used in relaxation.

general settings

echo screen
units metal
dimension 3
boundary f f p
atom_style atomic

#reading model
read_data molten.data extra/atom/types 6

group nt type 1 2
group hw type 3 4
group bj type 5 6
group TOOL type 7 8

######################################################
timestep 0.0005 # 0.5 fs
pair_style tersoff
pair_coeff * * 2007_SiO.tersoff Si O Si O Si O Si O

delete_atoms overlap 0.1 all all compress yes
write_data comepare.data
neighbor 0.5 bin
neigh_modify delay 0 every 1 check yes exclude group TOOL TOOL

initial velocities

velocity hw create 300 23456 dist gaussian
velocity nt create 300 34567 dist gaussian
velocity bj set 0.0 0.0 0.0 units box
velocity TOOL set 0.0 0.0 0.0 units box
compute T_TL hw temp
compute T_NL nt temp

equilibrium

fix 1 bj setforce 0.0 0.0 0.0
fix 2 TOOL setforce 0.0 0.0 0.0
fix 12 bj nve
fix 13 TOOL nve
fix 3 hw nvt temp 300 300 $(100*dt)
fix_modify 3 temp T_TL

fix 4 nt nvt temp 300 300 $(100*dt)
fix_modify 4 temp T_NL

thermo 500
thermo_style custom step c_T_NL c_T_TL pe ke etotal press
thermo_modify lost ignore
shell mkdir ./relax
dump 1 all custom 2000 ./relax/stress.*.cfg id type x y z
run 100000 # equlibrium in 50 ps

Thank you very much indeed
Zhensong.


This will happen when the geometry is either inconsistent with force field parameters or the force field is not suitable for the kind of system you have set up.

Here is the problem. Have you read the publication describing this parameterization? Is it suitable for systems that are not bulk but have surfaces? The one that is shipped with LAMMPS is only suitable for bulk systems, for example. For this kind of system, you will usually need a potential with charge equilibration like COMB. Also, you have to pay great attention to how you cut the surface or else there can be significant reconstruction or instabilities.

Thanks a lot. I will think that over.