issues in system equilibration

Dear LAMMPS users

I am trying to use LAMMPS for indentation analysis of copper by diamond indenter. I am using hybrid pair style, and fix move command to move a spherical rigid region as indenter.

The issue is the copper structure is fluctuating inconsistently. The forces are also inconsistent; the unloading forces are negative from starting.

I would appreciate some expert advice on following considerations.

  1. Using four atoms types 1-3 copper (for visualizing boundary w_mobile, w_thermostat, w_boundary) and 4 diamond

pair_style hybrid tersoff eam morse 2.5658

pair_coeff * * tersoff SiC.tersoff NULL NULL NULL C

pair_coeff 1*3 4 morse 0.087 5.14 2.05 9.025

pair_coeff 13 13 eam Cu_u3.eam

my intention is to apply tersoff for all interactions, then override Cu-C as morse and Cu-Cu as eam. So the remaining C-C will be tersoff.

  1. Creating velocity at 300K and applying nve ensemble

velocity w_mobile create 300.0 12345 temp tmobile

run 0

velocity w_mobile scale 300.0

fix 1 w_mobile nve

fix 2 w_thermostat nve

fix 3 all temp/rescale 1 300.0 300.0 0.1 1.0

fix 4 w_boundary setforce 0.0 0.0 0.0

fix 5 i_diamond rigid single



unfix 3

fix 3 w_thermostat temp/rescale 1 300 300 0.1 1.0

  1. Can a region be considered rigid if applying no fix on the region? Or should apply fix rigid.

Using fix rigid command causing atom lost error.

  1. Is fix move command adequate for the indentation purpose?

  2. Using

a. variable fz equal fcm(i_diamond,z) for force on indenter

b. variable depthz equal xcm(i_diamond,z) for position of indenter

I am using periodic boundary in x, y and non-periodic fixed boundary for z

· Version: LAMMPS 64-bit 16Mar2018


Mohit Sharma

Precision Manufacturing Lab

NTUST, Taipei Taiwan

Can you run w/out moving the indenter? Is the system stable,

are the forces correct? Can you move and visualize the

indenter moving w/out the substrate? Does it move correctly?

In other words, you can debug a complex model by implementing

it one step at a time.