[lammps-users] Question about branched chain connectivity

LAMMPS LAMMPS Mailing List Mirror
1

Hi all: I am trying to specify a branched connectivity using FENE
bonds. But LAMMPS has difficulties in identifying 1-3 neighbors with
this connectivity and the simulation collapsed. So I guess LAMMPS only
supports linear connectivity, but not branched ones like two linear
chains joined at middle points. Does LAMMPS support branched
connectivity?

Thank you.

Best,
Ting

2

Hi all: I am trying to specify a branched connectivity using FENE
bonds. But LAMMPS has difficulties in identifying 1-3 neighbors with
this connectivity and the simulation collapsed. So I guess LAMMPS only
supports linear connectivity, but not branched ones like two linear
chains joined at middle points. Does LAMMPS support branched
connectivity?

yes. are you sure your special_bonds setting is correct?

axel.

3

LAMMPS can handle some extremely complex connectivities, including chains grafted onto amorphous silica.

Are you sure that you're specifying your model correctly? More likely than not there's an error in your input files—wrong unit set, or connectivity problems.

—AEI

4

special_bonds lj 0.0 1.0 1.0 works, but special_bonds fene doesn't. Thanks .

Best,
Ting

5

special_bonds lj 0.0 1.0 1.0 works, but special_bonds fene doesn't. Thanks .

what version of lammps are you using.

special_bonds fene

should be equivalent to

special_bonds lj/coul 0.0 1.0 1.0

which is not quite the same as

special_bonds lj 0.0 1.0 1.0

at least in the current version of LAMMPS.

hmmm... does your system have charges?

can you produce a simple test input that
reproduces this unexpected behavior?
if it still exists with the current version, that is.

thanks,
    axel.