Help- Why are atoms clustering during 3D Indent?

LAMMPS LAMMPS Mailing List Mirror
1

Hi Lammps contributors,

I am fairly new to LAMMPs (been using for ~2 weeks), and am incredibly impressed with the excellent manual and the accessibility of LAMMPs. This is truly an excellent piece of code.

I have been working on developing simulations of 3D indentation, in a dynamic simulation of a box of Ta, with a spherical indenter. For some reason, as soon as the fix indent command is applied, the atoms within the simulation box all cluster together within the box. I have attached a movie of the indentation simulation.

I simply don’t understand what would cause this. If you can figure out what is going wrong (is the system not properly equilibrated? Is the fix indent command incorrect? Perhaps the position of the indenter is not in the correct location?)

The code I use is quite simple:

units metal
dimension 3
boundary p p p
atom_style atomic
neighbor 0.5 bin
neigh_modify delay 5 check yes

lattice bcc 3.3025234375000040998
region cell block 0 10 0 10 0 10
create_box 1 cell
create_atoms 1 box
region lower block INF INF INF 2 INF INF
group lower region lower
group Mobile_Up subtract all lower

pair_style eam/alloy
pair_coeff * * Ta.set Ta
compute csym all centro/atom bcc
compute pe all pe/atom

--------------------- Equilibrate Temperature and Pressure

velocity Mobile_Up create 100 12345 mom yes rot no
fix 1 all npt temp 100 100 1 iso 0.0 0.0 10 drag 1.0
fix 2 lower setforce 0.0 0.0 0.0
thermo 1000
thermo_style custom temp press lx ly lz step etotal
timestep 0.0005
run 40000
unfix 1

#--------------------- Want constant volume, so change to canonical ensemble
fix 1 all nvt temp 100 100 1 drag 1
run 10000

#---------------------- Dynamic indentation
variable k equal 100.0/xlat
variable y equal “23.0ylat-stepdt0.02ylat”
fix 3 all indent $k sphere 10 v_y 10 4.0 side out units lattice

#----------------------- Dump to a new directory (in case overload of dumpfiles- can delete directory)
shell mkdir dump2
shell cd dump2
dump 1 all cfg 500 dump.Indent_*.cfg id type xs ys zs c_csym c_pe
run 40000

#----------------- Remove indenter, to (hopefully) observe difference between reversible and irreversible deformation
unfix 3
run 10000

echo both

Cheers!
Nathaniel

Faulty_3D_Indent.gif

2

fix 3 all indent $k sphere 10 v_y 10 4.0 side out units lattice
I simply don’t understand what would cause this. If you can figure out what is going wrong (is the system not properly
equilibrated? Is the fix indent command incorrect? Perhaps the position of the indenter is not in the correct location?)

Using lattice units for the command means your distances are multiplied
by the lattice spacing of 3.3 Angs.

Your system is 3d periodic, so I don’t understand how you are moving
the indenter into the system. You can’t have it in place at time 0
and impose a huge displacement on atoms it overlaps. Maybe
you want to start if with radius 0 and grow it slowly.

One way to verify that the indenter is the size and in the position
you want, is to start with it far away from a free surface and move it
into position slowly. If you don’t see it affecting the surface when
you expect, then you haven’t specified the parameters correctly.

See examples/indent/in.indent for an example.

Steve