special bonds

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

I have been reading the LAMMPS documentation for the special_bonds command, and am unclear about something. For argument’s sake, the command:

special_bonds lj/coul 0.0 0.0 0.7

nullifies lj and coulombic 1-2 interactions & 1-3 interactions, and scales by a factor 0.7 lj/coul 1-4 interactions.

Considering a linear molecule that is 7 atoms long, for atoms separated by 1 and 2 bonds, there will be no lj/coul interactions computed. For atoms separated by 3 bonds, there will be lj/coul interactions scaled by a factor of 0.7.

What happens with the atoms in the same molecule that are separated by more that 3 bonds? Are the corresponding lj/coul scaled at all? or by the factor 0.7?

Thanks for any clarification.

2

Hello,

I have been reading the LAMMPS documentation for the special_bonds command,
and am unclear about something. For argument's sake, the command:

special_bonds lj/coul 0.0 0.0 0.7

nullifies lj and coulombic 1-2 interactions & 1-3 interactions, and scales
by a factor 0.7 lj/coul 1-4 interactions.

Considering a linear molecule that is 7 atoms long, for atoms separated by 1
and 2 bonds, there will be no lj/coul interactions computed. For atoms
separated by 3 bonds, there will be lj/coul interactions scaled by a factor
of 0.7.

What happens with the atoms in the same molecule that are separated by more
that 3 bonds? Are the corresponding lj/coul scaled at all? or by the factor
0.7?

those interactions are computed in full.

axel.

3

Thanks Axel,

According to OPLS, atoms separated by >3 bonds in the same molecule have lj and coul interactions scaled by 0.5.

special_bonds lj/coul 0.0 0.0 0.5 does not do this if I’ve correctly understood.

I have looked, and not found this type of pairwise interaction defined in LAMMPS. Has anyone tackled this sort of interaction before ?

4

Thanks Axel,

According to OPLS, atoms separated by >3 bonds in the same molecule have lj
and coul interactions scaled by 0.5.

special_bonds lj/coul 0.0 0.0 0.5 does not do this if I've correctly
understood.

i think you have misread something. you have here the correct setting for OPLS.

axel.

5

According to OPLS, atoms separated by >3 bonds in the same molecule have lj and coul interactions scaled by 0.5.

I doubt that, as it makes little physical sense. If you have a 1000-mer
polymer chain, how would the 2 ends know anything about each other?
Or even know they were in the same molecule?

Steve

6

Agreed, this does make little sense, thanks for the clarification.