Hi
This question has been asked previously but I did not get an answer specific to my question.
I have a long linear polymer chain with N =100. On;y interactions are harmonic between bonded and shifted LJ between non bonded. As per LAMMPS website, I have to specify
special_bonds lj 0.0 1.0 1.0
If I use this I get ave neigh/atom to be 5.88. This could suggest that 2 1-2 neighbours, 2 1-3 neigh and 2 1-4 neigh and end effects make it 5.8
However, even though I do not have any coulumbic interactions if I change the command to
special_bonds lj/coul 0.0 1.0 1.0
The ave special neigh/atom is 1.98.
Question is how does the two key words differ when I do not have coulumbic interactions
Thanks
Vaidyanathan M S
Hi
This question has been asked previously but I did not get an answer specific
to my question.
I have a long linear polymer chain with N =100. On;y interactions are
harmonic between bonded and shifted LJ between non bonded. As per LAMMPS
website, I have to specify
special_bonds lj 0.0 1.0 1.0
If I use this I get ave neigh/atom to be 5.88. This could suggest that 2 1-2
neighbours, 2 1-3 neigh and 2 1-4 neigh and end effects make it 5.8
However, even though I do not have any coulumbic interactions if I change
the command to
special_bonds lj/coul 0.0 1.0 1.0
The ave special neigh/atom is 1.98.
Question is how does the two key words differ when I do not have coulumbic
interactions
you know this, but the neighbor list code doesn't know this and for
complex setups using hybrid pair styles it may be very difficult to
determine this reliably.
so all it knows is that in the first case you have the special bond
settings lj 0.0 1.0 1.0 coul 0.0 0.0 0.0 and in the second case lj 0.0
1.0 1.0 coul 0.0 1.0 1.0.
when the lj and coul settings are identical, no special neighbor
encoding is needed and for a factor of 0.0 the pair is completely
removed from the neighbor list and for a factor of 1.0 the pair is
kept as a regular non-bonded pair. if lj and coul disagree, those
neighbors have to be kept around and suitably encoded so that either
the factor 1.0 or the factor 0.0 can be applied.
does this make sense for you now?
axel.