Hello lammps-users,

I am running NEB calculations with Reaxff on hydrocarbons cracking in zeolites. I obtained the Rexff parameters from dr. Adri van Duin. However, my Neb calculations never seem to converge even after 3000 steps. I am using the following script:

units metal
atom_style charge
atom_modify map array
boundary p p p
atom_modify sort 0 0.0

read_data /rdsgpfs/general/user/za1417/home/WORK/Fau_AlH/NEB/Fau_LAMMPSNEB_ASE/data_initial


pair_style reax/c NULL
pair_coeff * * ffield.reax Al Si O H C
mass 1 26.981538
mass 2 12.011000
mass 3 1.008000
mass 4 15.999000
mass 5 28.085000


variable i equal part

fix fix_nve all nve

fix 1 all qeq/reax 1 0.0 10.0 1e-6 reax/c

group nebatoms id 145:156

group nonneb subtract all nebatoms

fix 2 nebatoms neb 1.0 parallel ideal

fix 3 nonneb setforce 0 0 0

dump dump_all all xyz 100 neb$

thermo 10

timestep 0.001

min_style fire

neb 0.0 0.01 3000 1500 50 final final.coord

print “end_of_lammps

Is there anything that I could improve? or is it that NEB doesn’t work very well with Reaxff. Is there any other method to calculate reaction barriers?

Best Regards,

this is very difficult to debug from remote. have you checked that you can complete a run of a simple minimization for this system. what are the remaining forces?
please note, that by using fix setforce, your system’s forces are no longer consistent with the energies. so if the part of the system where the forces are set to zero are not properly relaxed, it might confuse the minimizer by resulting in energy changes that do not correspond with the forces based displacements.