[lammps-users] fix indent and energy minimization

To all:
I am having trouble while performing an indentation simulation into Au with the included EAM potential. However, I am trying to move the indenter, and then minimize the energy, and repeat. I noticed that on 8/3/2006 another user wrote to the list with a similar question ([lammps-users] energy minimization question updated). However, I am working in 3D.

I changed the sample file from 2D to 3D using the l-j potential, and the simulation ran just fine. However, when I tried to use the eam potential (after changing all units to the 'metal' format) I found that the indenter did not appear to interact with the sample. Otherwise, I have implemented the suggestions in the 8/3/2006 email using the eam potential without success. As with the case reported in the email, a dynamic simulation using eam works as it should.

Any ideas would be appreciated. The input file is attached.

# 3d indenter simulation with minimization instead of dynamics

dimension 3
boundary p p s
atom_style atomic
units metal
neigh_modify delay 5

# create geometry
lattice fcc 4.08
#lattice fcc 4.08 orient z 1 1 1 orient x 1 1 -2 orient y -2 2 0 origin 0 0 0
region mainblock block 0 102.0 0 102.0 0 65.28 units box
create_box 2 mainblock
create_atoms 1 box

# EAM potentials

pair_style eam
pair_coeff * * Au_u3.eam

# define groups

region bottomface block INF INF INF INF INF 4.08 units box
group lower region bottomface
group mobile subtract all lower
set group lower type 2

# initial velocities

fix 2 lower setforce 0.0 0.0 0.0

# minimize with indenter

thermo 10
dump dumpxyz all custom 10 dump.xyz tag type x y z z

minimize 1.0e-6 1000 1000

#INDENT DISTANCE, 0.00
fix 4 all indent 10000.0 sphere 51.0 51.0 80.28 20.0
fix_modify 4 energy yes
minimize 1.0e-6 1000 1000

#INDENT DISTANCE, 0.10
fix 4 all indent 10000.0 sphere 51.0 51.0 80.18 20.0
fix_modify 4 energy yes
minimize 1.0e-6 1000 1000

#INDENT DISTANCE, 0.20
fix 4 all indent 10000.0 sphere 51.0 51.0 80.08 20.0
fix_modify 4 energy yes
minimize 1.0e-6 1000 1000

#more steps...

I found
that the indenter did not appear to interact with the sample.

A common reason for this is not realizing there are 2 kinds of
distance units in many LAMMPS commands: box and lattice.
Lattice is the default. So if you specify the indenter with radius
10, that can be 10 Angs or 10 lattice constants, depending
on the form of the command you use.

Steve