hi

HI users,

In my system, there are two systems.
I want to have special_bonds for one type of molecules and for another no special bonds…

can i do in LAMMPS???

thanks in advance

karteek kumar

HI users,

please consider using a more descriptive subject line.

In my system, there are two systems.
I want to have special_bonds for one type of molecules and for another no
special bonds..

can i do in LAMMPS???

special_bonds is a global setting.

can you provide some more details
*why* you would need this and for
what kind of potentials?

axel.

Yes. I've been wondering about this too. Unfortunately the
special_bonds command seems to apply to all of the molecules in your
system. (In most other ways LAMMPS is incredibly modular.)

Fortunately you can get around this limitation manually:
Suppose molecules of type A neglect 1-4 interactions ("special_bonds 0
0 0"), and molecules of type B include 1-4 interactions
("special_bonds 0 0 1"). To combine these two molecules in a
simulation, you could neglect 1-4 interactions ("special bonds 0 0
0"), and put them back in for molecule B by explicitly adding bonds
between every pair of 1-4 atoms in molecule B which were neglected.
(1-4 atoms are strictly defined as atoms which participate in a
dihedral interaction.) You can use or Lennard-Jones potentials
("bond_style fene"), or tabulated bond potentials ("bond_style
table"), to generate any pairwise force you want between these 1-4
atoms (to generate the force you neglected when you set "special_bonds
0 0 0"). If there's a more elegant way to do this, I'd love to know.

I hope this helps!

Andrew

Hi Axel sir,

Dreiding force field uses full-scale 1-4 interaction, when dreiding is combined with other force field… where 1-4 is scaled by different factor like for OPLS its 0.5…

how can I overcome this problem

thanks in advance

karteek kumar

Hi Axel sir,

Dreiding force field uses full-scale 1-4 interaction, when dreiding is
combined with other force field.. where 1-4 is scaled by different factor
like for OPLS its 0.5...

how can I overcome this problem

karteek kumar

If this is true, then you can use "special_bonds 0 0 0.5" and then add
bonds to the 1-4 atom pairs belonging to the molecule which uses the
Dreiding force field. Choose the bond_style and bond_coeffs carefully
so that the bonded force between these atoms half of what the pairwise
force would have been. (Compensate for the missing half.)

Or, if you prefer, you can use "special_bonds 0 0 1" and add bonds
between the 1-4 atom pairs in the molecule which uses the OPLS force
field. In that case use a "bond_style table" potential to subtract
half of what the pairwise interaction would have been between them.
Either should work. Be careful to only do this for pairs of atoms
which participate in dihedral interactions.

Good luck. (Please let us know if you find a better solution.)

Andrew

Hi Axel sir,

Dreiding force field uses full-scale 1-4 interaction, when dreiding is
combined with other force field.. where 1-4 is scaled by different factor
like for OPLS its 0.5...

how can I overcome this problem

you should not mix and match force fields like that.
this is almost always a "very bad idea(tm)".
this usually leads to inconsistent behavior
and you end up wasting your time and that
of the computer(s) you run on.

axel.