Calculating Airebo-m Rebo, Morse, and Torsion energy terms for subset of atoms same as with all atoms

Hi all,

I would like to calculate the energy terms for the airebo-m potential for a subset of atoms. This is what I try in my script:

group mid2800 id 201:3000

compute en mid2800 pair airebo/morse
variable REBO equal c_en[1]
variable LJ equal c_en[2]
variable TORSION equal c_en[3]

compute energ all pair airebo/morse
variable REBO2 equal c_energ[1]
variable LJ2 equal c_energ[2]
variable TORSION2 equal c_energ[3]

There is a total of 3200 atoms in a straight carbon nanotube, and I made this group with the middle 2800. Then when I calculate the energy terms as described in the documentation for airebo, for the two different groups, the energies end up the same:

Step Time Temp c_px PotEng Fmax v_REBO v_LJ v_TORSION v_REBO2 v_LJ2 v_TORSION2 v_REBO2
        3 0 10 7.0574824e+08 -23384.286 1.1477917 -24439.676 -36.67048 1092.0596 -24439.676 -36.67048 1092.0596 -24439.676
      100 0.0485 10.566012 -6.697875e+08 -23385.109 1.2894551 -24439.851 -38.058452 1092.8004 -24439.851 -38.058452 1092.8004 -24439.851

I attached the files I used to test this. I'm using LAMMPS 16 March 2018. If anyone knots how to calculate the airebo energy terms for a subset of atoms, or what I am missing here, I'd really appreciate the help.

Thanks,
Kyle Rego

3200nanotube.data (201 KB)

knotscript.txt (2.01 KB)

Hi all,

I would like to calculate the energy terms for the airebo-m potential
for a subset of atoms. This is what I try in my script:

group mid2800 id 201:3000

compute en mid2800 pair airebo/morse
variable REBO equal c_en[1]
variable LJ equal c_en[2]
variable TORSION equal c_en[3]

compute energ all pair airebo/morse
variable REBO2 equal c_energ[1]
variable LJ2 equal c_energ[2]
variable TORSION2 equal c_energ[3]

There is a total of 3200 atoms in a straight carbon nanotube, and I made
this group with the middle 2800. Then when I calculate the energy terms
as described in the documentation for airebo, for the two different
groups, the energies end up the same:

​yes, please note what it says at the beginning of the documentation of compute pair:

​Define a computation that extracts additional values calculated by a pair style, and makes them accessible for output or further processing by other commands. The group specified for this command is ignored.

to explain, ​this compute actually does not perform a computation, but it gives access to data, that was tallied during force computation (over all atoms).

axel.​

​axel​