Dear Lammps users,
I want to minimize some tiny hydrocarbon structures,my in and data files are as follow:
variable sname index ch4
units electron
newton on
boundary f f f
atom_style electron
read_data data.${sname}
pair_style eff/cut 100.0
pair_coeff * *
comm_modify vel yes
compute effTemp all temp/eff
compute energies all pair eff/cut
variable eke equal c_energies[1]
variable epauli equal c_energies[2]
variable ecoul equal c_energies[3]
variable erres equal c_energies[4]
thermo 1
thermo_style custom step etotal pe ke v_eke v_epauli v_ecoul v_erres
group CH4 type 1 2
fix 1 CH4 setforce 0.0 0.0 0.0
min_style cg
min_modify line quadratic
compute 1 all property/atom spin eradius erforce
dump 2 all custom 1 ${sname}.min.lammpstrj id type q c_1[1] c_1[2] x y z fx fy fz c_1[3]
minimize 0 1e-6 100000 1000000
data:
Created by ZZH
15 atoms
3 atom types
-500.0 500.0 xlo xhi
-500.0 500.0 ylo yhi
-500.0 500.0 zlo zhi
Masses
1 12 # C
2 1.007940 # H
3 1 # e
Atoms # full
1 1 6.0 0 0.0 0 0 0
2 2 1.0 0 0.0 0.312 0.312 0.312
3 2 1.0 0 0.0 -0.6293 -0.6293 0.6293
4 2 1.0 0 0.0 -0.6293 0.6293 -0.6293
5 2 1.0 0 0.0 0.6293 -0.6293 -0.6293
6 3 0.0 -1 1.0 0 0 0
7 3 0.0 1 1.0 0 0 0
8 3 0.0 -1 1.0 0.214 0.22 0.22
9 3 0.0 1 1.0 0.2203 0.2203 0.2203
10 3 0.0 -1 1.0 -0.4279 -0.4279 0.4279
11 3 0.0 1 1.0 -0.4405 -0.4405 0.4405
12 3 0.0 -1 1.0 -0.4279 0.4279 -0.4279
13 3 0.0 1 1.0 -0.4405 0.4405 -0.4405
14 3 0.0 -1 1.0 0.4279 -0.4279 -0.4279
15 3 0.0 1 1.0 0.4405 -0.4405 -0.4405
log:
Minimization stats:
Stopping criterion = linesearch alpha is zero
Energy initial, next-to-last, final =
-9.01933835313498 -24.5130496217989 -24.5130496217989
Force two-norm initial, final = 5.5174006 2.5761346
Force max component initial, final = 1.6774393 1.676482
Final line search alpha, max atom move = 1.4392713e-09 2.4129125e-09
Iterations, force evaluations = 999 9668
The minimization fails with the stopping criterion “linesearch alpha
is zero”.This question always occurs in eff, which confuses me a lot, especially when I want to use my own potential form, it occurs with large probability. The potential function is continuous of the original eff, I don’t know why it happens.