reg_unknown number of entries in pair/local

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,

I am trying to obtain force acting on atom i due to atom j in a gold system during the deformation process. I did used a unit cell (simple system) to check whether pair/local calculates the forces for all the pairs within the system. I am not sure why the number of entries are more than the combination expected from the system.

units metal
atom_style atomic
boundary p p p

lattice fcc 4.08 orient x 1 0 0 orient y 0 1 0 orient z 0 0 1 origin 0.05 0.05 0.05
region sim_region block 0 1 0 1 0 1 units lattice
create_box 1 sim_region
create_atoms 1 box

pair_style eam/alloy
pair_coeff * * Au_GLJ10_3.eam.alloy Au

dump 1 all atom 100 dump_min.atom
thermo 100
thermo_style custom step pe press vol etotal
min_style cg
minimize 1e-15 1e-7 70000 100000
print “MINIMIZATION DONE”
undump 1

write_data aftermin.txt

dump 2 all atom 1000 dump_for_deform.atom
variable tmp equal “lx”
variable L0 equal ${tmp}
variable strain equal “(lx - v_L0)/v_L0”
variable p1 equal “v_strain”
variable p2 equal “-pxx*0.0001”

variable srate equal 1.0e9
variable srate1 equal “v_srate/1.0e12”

fix 1 all nve
fix 2 all deform 1 x erate ${srate1} units box remap x
thermo_style custom step temp vol density pe ke etotal pxx pyy pzz lx ly lz
thermo 1000
timestep 0.001

compute 1 all property/local patom1 patom2
compute 2 all pair/local dist fx fy fz
dump 1 all local 1 tmp.dump index c_1[1] c_1[2] c_2[1] c_2[2] c_2[3] c_2[4]

run 1
undump 2
unfix 1
unfix 2

The output from dump file is

ITEM: TIMESTEP
1
ITEM: NUMBER OF ENTRIES
108
ITEM: BOX BOUNDS pp pp pp
0 4.08
0 4.08
0 4.08
ITEM: ENTRIES index c_1[1] c_1[2] c_2[1] c_2[2] c_2[3] c_2[4]
1 1 2 2.885 -0.0711631 -0.0711631 0
2 1 3 2.885 -0.0711631 0 -0.0711631
3 1 4 2.885 0 -0.0711631 -0.0711631
4 1 1 4.08 0.0463516 -0 -0
5 1 4 4.99696 0.0165693 0.00828464 0.00828464
6 1 1 5.76999 -0.000857057 0.000857057 -0
7 1 3 4.99696 -0.00828464 0.0165693 0.00828464
8 1 1 4.08 -0 0.0463516 -0
9 1 3 4.99696 0.00828464 0.0165693 0.00828464
10 1 1 5.76999 0.000857057 0.000857057 -0
11 1 2 4.99696 -0.00828464 -0.00828464 0.0165693
12 1 1 5.76999 -0 -0.000857057 0.000857057
13 1 2 4.99696 0.00828464 -0.00828464 0.0165693
14 1 1 5.76999 -0.000857057 -0 0.000857057
15 1 2 4.99696 -0.00828464 0.00828464 0.0165693
16 1 1 4.08 -0 -0 0.0463516
17 1 2 4.99696 0.00828464 0.00828464 0.0165693
18 1 1 5.76999 0.000857057 -0 0.000857057
19 1 1 5.76999 -0 0.000857057 0.000857057

I would like to know why there are more than " 1 2 " atom combination in the output.

Suggestions please.

Regards,
Santhosh

2

Dear all,

I am trying to obtain force acting on atom i due to atom j in a gold
system during the deformation process. I did used a unit cell (simple
system) to check whether pair/local calculates the forces for all the pairs
within the system. I am not sure why the number of entries are more than
the combination expected from the system.

​because minimum image conditions don't apply for LAMMPS. it creates real
copies of the periodic images of your atoms up to the communication cutoff
and all pairs (including interactions with periodic images) within the
cutoff are considered.

axel.​

3

Thank you Axel.

Is there any other way in LAMMPS to determine the force on atom i due to atom j in the system?

4

Thank you Axel.

Is there any other way in LAMMPS to determine the force on atom i due to
atom j in the system?

​these *are* the forces that are computed and added to atom i due to atom j
and all its replicas within the cutoff. that is exactly what it should be.
you *do* have periodic boundaries, so you *must* include the contributions
from periodic replica. if you don't want periodic replica, use a larger
system, but the number of contributions and the total contributed force
will be the same, as you will have the same number of neighbors within the
cutoff either way.

axel.

5

Thank you axel.