Linesearch alpha is zero - During energy minimization

Dear all
I am trying to replicate the results of the following paper (10.1103/PhysRevLett.93.016001) to understand the process of athermal quasi-static shear deformation. I am using LAMMPS version (lammps-patch-17feb2022) to perform my simulations.
I am using pair_style harmonic/cut with value of K equal to 1 to mimic a soft-potential as mentioned in the paper. The packing fraction is 1.0. I am using conjugate gradient method to minimize the forces and energy of the system. While doing so , instead of reaching the force tolerance criteria as mention in the code , the simulation is getting stopped with the criteria linesearch alpha is zero. Can anyone give me suggestion to understand the mistake I am doing. Any suggestion to solve this problem is appreciated. I have also attached the simulation code I am using for the same.

Minimization stats:
  Stopping criterion = linesearch alpha is zero
  Energy initial, next-to-last, final = 
       72153.432573231   39512.8547171024   39512.8547171024
  Force two-norm initial, final = 178.05505 40.280413
  Force max component initial, final = 3.3438459 1.012577
  Final line search alpha, max atom move = 5.7464705e-12 5.8187437e-12
  Iterations, force evaluations = 62 108

units		si  
atom_style	sphere
dimension	2

region		box prism 0.0 200.0 0 200.0 -0.5 0.5  0.0 0.0 0.0
region		particle prism 0.0 200.0 0.0 200.0 -0.5 0.5  0.0 0.0 0.0

create_box	2 box
create_atoms 1 random 38205 3000 particle units box
create_atoms 2 random 47225 4000 particle units box

set type 1 diameter 1.0
set type 2 diameter 0.52

pair_style harmonic/cut
pair_coeff * * 1.0 1.0

compute fcl all property/local patom1 patom2
compute fc all pair/local dist force
compute peratom all stress/atom NULL pair
compute p all reduce sum c_peratom[1] c_peratom[2] c_peratom[3] c_peratom[4] 
variable press equal -(c_p[1]+c_p[2])/(2*vol) 
variable sigmaxy equal c_p[4]/vol

fix 		2 all enforce2d
thermo 100
thermo_style custom step ke pe v_press v_sigmaxy
dump            mydump all atom 100 stiffness_1.dump
dump             1 all custom 100 *.data id type diameter mass x y z vx vy fx fy
dump_modify  1 sort id
min_style cg 
restart 100 restart.*.stiff_1 
minimize       0.0 1.0e-10 1000000 10000000

Maybe the force tolerance criterion is too tight, given that the simulation box is fixed. Based on your input, the structure does not change much after minimisation step 2500. I have changed the minimisation settings to:

minimize       0 1e-4 10000 100000

and run it with version 7Feb24. This is what I get:

Minimization stats:
  Stopping criterion = force tolerance
  Energy initial, next-to-last, final =
      47889.2891004889   17383.9669160363   17383.9669160348
  Force two-norm initial, final = 619.94254 9.9842569e-05
  Force max component initial, final = 7.8956045 3.5728918e-06
  Final line search alpha, max atom move = 1 3.5728918e-06
  Iterations, force evaluations = 5361 10685

Nice sample, thought!

1 Like

Thanks you very much sir for your help