Basically, i am trying to relax the structure by lammps. My system is the oxygen atom adsorbed on the alloy created by the ase.I have generated many structure.However, some of them work well while some get stuck when perform the minimize command. I have tried several different force field published, but the problem exist.
Any help will be appreciated
Joey
here is my in.file
“”“”“”
units real
atom_style charge
boundary p p p
read_data new.dat
neighbor 0.3 bin
neigh_modify every 1 delay 0 check yes
pair_style reaxff NULL
pair_coeff * * CHONSSiPtNiCuCoZrYBa.ff Co Cu Ni Pt O
compute 1 all pe/atom
#compute 1 all property/atom fx fy fz
fix 1 all qeq/reaxff 1 0.0 10.0 1.0e-3 reaxff
minimize 1.0e-5 1.0e-5 1000 1000
dump melt all atom 20 melt.xyz
“”“”“”"
the output
“”"
Reading data file …
orthogonal box = (0.0000000 0.0000000 0.0000000) to (8.0000000 8.0000000 54.000000)
1 by 1 by 1 MPI processor grid
reading atoms …
33 atoms
read_data CPU = 0.029 seconds
WARNING: Changed valency_val to valency_boc for X (src/REAXFF/reaxff_ffield.cpp:296)
CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
Your simulation uses code contributions which should be cited:
- pair reaxff command:
- fix qeq/reaxff command:
The log file lists these citations in BibTeX format.
CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE-CITE
Neighbor list info …
update every 1 steps, delay 0 steps, check yes
max neighbors/atom: 2000, page size: 100000
master list distance cutoff = 10.3
ghost atom cutoff = 10.3
binsize = 5.15, bins = 2 2 11
2 neighbor lists, perpetual/occasional/extra = 2 0 0
(1) pair reaxff, perpetual
attributes: half, newton off, ghost
pair build: half/bin/newtoff/ghost
stencil: full/ghost/bin/3d
bin: standard
(2) fix qeq/reaxff, perpetual, copy from (1)
attributes: half, newton off, ghost
pair build: copy
stencil: none
bin: none
Setting up cg style minimization …
Unit style : real
Current step : 0
Per MPI rank memory allocation (min/avg/max) = 13.91 | 13.91 | 13.91 Mbytes
Step Temp E_pair E_mol TotEng Press
0 0 -525.34838 0 -525.34838 -nan
“”“”"
then it is stuck here for hours
Please always report which LAMMPS version you are using.
If you have a NaN pressure, that means some of your forces have overflowed or have caused a division by zero or some other invalid math. This is usually caused by having a “bad” geometry.
There is not much that you can do about it but improving your geometry to avoid getting into an unphysical state.
Much thanks for your help ,my LAMMPS version is lammps-29Sep2021.I have double-checked my geometry. I thought it is ok. Here is my POSCAR file of the corresponding datafile.
“”"
Ni Pt Ni Cu Ni Pt Ni Cu Ni O
1.0000000000000000
7.9999999999999982 0.0000000000000000 0.0000000000000000
0.0000000000000000 7.9999999999999991 0.0000000000000000
0.0000000000000000 0.0000000000000000 54.0000000000000000
Ni Pt Ni Cu Ni Pt Ni Cu Ni O
6 1 1 1 13 1 2 1 6 1
Cartesian
2.0000000000000040 1.9999999999999993 20.0000000000000000
6.0000000000000044 2.0000000000000027 20.0000000000000000
2.0000000000000062 6.0000000000000027 20.0000000000000000
6.0000000000000053 6.0000000000000027 20.0000000000000000
0.0000000000000027 4.0000000000000018 22.0000000000000000
0.0000000000000027 -0.0000000000000018 22.0000000000000036
4.0000000000000053 -0.0000000000000018 22.0000000000000036
4.0000000000000062 4.0000000000000018 22.0000000000000036
2.0000000000000049 2.0000000000000027 24.0000000000000000
6.0000000000000036 2.0000000000000027 24.0000000000000000
6.0000000000000036 6.0000000000000027 24.0000000000000000
2.0000000000000053 6.0000000000000036 24.0000000000000000
4.0000000000000018 0.0000000000000000 26.0000000000000000
4.0000000000000062 4.0000000000000018 26.0000000000000000
0.0000000000000036 4.0000000000000018 26.0000000000000036
0.0000000000000018 0.0000000000000000 26.0000000000000071
2.0000000000000022 1.9999999999999993 28.0000000000000000
6.0000000000000018 2.0000000000000031 28.0000000000000000
2.0000000000000044 6.0000000000000036 28.0000000000000000
6.0000000000000027 6.0000000000000036 28.0000000000000000
0.0000000000000000 4.0000000000000018 29.9999999999999964
-0.0000000000000000 -0.0000000000000009 30.0000000000000000
4.0000000000000027 -0.0000000000000009 30.0000000000000000
4.0000000000000044 4.0000000000000018 30.0000000000000000
6.0000000000000018 2.0000000000000036 31.9999999999999929
2.0000000000000031 2.0000000000000027 31.9999999999999964
2.0000000000000036 6.0000000000000036 32.0000000000000000
6.0000000000000018 6.0000000000000036 32.0000000000000000
0.0000000000000000 0.0000000000000000 34.0000000000000000
4.0000000000000000 0.0000000000000000 34.0000000000000000
0.0000000000000018 4.0000000000000018 34.0000000000000000
4.0000000000000044 4.0000000000000027 34.0000000000000000
7.4929289321881347 6.0000000000000000 33.5070710678118644
“”
Sorry, but this forum is not a data or input validation service.
Thanks for your help. I’m sorry for the trouble i have made, I just want to proof that my input geometry is ok. Will there be anything else that would result in such problem?
Any help will be appreciated
This doesn’t prove anything. What matters is what you input to LAMMPS, not what you get from VASP.
Only what I have already mentioned. There are plenty of ways to make mistakes. It is rather common that people with limited experience in running LAMMPS claim that they have checked “everything” multiple times and that “everything” is correct, but then there are issues nevertheless. So my recommendation is to be more humble and check for issues more carefully.
There are many, many such discussions in the archives, so I suggest you look for discussions with others that had similar issues.
Thanks for you suggestions, the poscar and the datafile are derived from the same ase atom object. Sorry for the trouble I made to you. I will follow your suggestions.
Much thanks for your help again.