Special Bonds parameters for OPLS

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

Two Questions.

One.

I am desiring to use the special bonds command. I am only interested in interactions of atoms with 3 bonds and 2 atoms between them such as 1 and 4 in 1-2-3-4.
When I use this command, does it also take in to account the interactions down the line, say 2-5 with the three bonds and two atoms between? (Which I think it does and that’s why I want to use it, tell me if I’m wrong)

Two.

In terms of the actual parameters for OPLS, Does anyone have an idea of the parameters or can someone point me in the write direction to find them?

Thanks,

Michael Goytia
Graduate Student
University of Utah Chemistry Department

2

Hello,

Two Questions.

both of your questions are more about force field implementations in
general. i recommend you first read up on that in a suitable text book
and also find the matching original publications that are applicable
for the specific force field you want to use.

One.

I am desiring to use the special bonds command. I am only interested in
interactions of atoms with 3 bonds and 2 atoms between them such as 1 and 4
in 1-2-3-4.
When I use this command, does it also take in to account the interactions
down the line, say 2-5 with the three bonds and two atoms between? (Which I
think it does and that’s why I want to use it, tell me if I’m wrong)

please note that the special_bonds command does nothing else but
*change* the scaling factors for exclusions. In that respect LAMMPS
behaves like conventional classical force fields like CHARMM, AMBER,
OPLS and similar expect it and - if anything - it is more flexible.

Two.

In terms of the actual parameters for OPLS, Does anyone have an idea of the
parameters or can someone point me in the write direction to find them?

if you haven't done a survey of the relevant literature, this would be
the right time to do it. you'll get the list of parameters as a bonus.

axel.

3

Hello,

Two Questions.

One.

I am desiring to use the special bonds command. I am only interested in
interactions of atoms with 3 bonds and 2 atoms between them such as 1 and 4
in 1-2-3-4.
When I use this command, does it also take in to account the interactions
down the line, say 2-5 with the three bonds and two atoms between? (Which I
think it does and that’s why I want to use it, tell me if I’m wrong)

Yes. "special_bonds lj/coul 0.0 0.0 1" does that.

However, for OPLS force-fields, you want to use these settings:

special_bonds lj/coul 0.0 0.0 0.5

Two.

In terms of the actual parameters for OPLS, Does anyone have an idea of the
parameters or can someone point me in the write direction to find them?

The original parameters and force-field description are in this paper
J. Am. Chem. Soc. 1996, 118, 11225-11236
http://pubs.acs.org/doi/abs/10.1021/ja9621760

However I vaguely remember that the parameters may have been adjusted
over time, probably to fit a wider range of molecules. So, if you
want the most up-to-date force field parameters, try the BOSS manual
(near page 66)
http://zarbi.chem.yale.edu/software.html

This list has been converted to a machine-readable form, at the TINKER web-site:
http://dasher.wustl.edu/tinker/distribution/params/oplsaa.prm

Lastly, there is a molecule-builder program specifically for LAMMPS
which makes it easy to use the all-atom version of OPLS (OPLSAA).
There's an example showing how to build a long alkane chain (polymer)
using OPLSAA on this web page:
http://www.moltemplate.org/
http://www.moltemplate.org/images/alkane_chain_single/t=1ns.jpg
http://www.moltemplate.org/examples/alkane_chain_single/ch3group.lt
http://www.moltemplate.org/examples/alkane_chain_single/ch2group.lt
http://www.moltemplate.org/examples/alkane_chain_single/alkane50.lt
...and other examples of simple hydrocarbon molecules in the
"examples" which you can download online (click on the upper-left
corner).

Moltemplate allows you to build organic molecules relatively easily
without reading or understanding the force-field details (so please
use it with caution).

Cheers
Andrew

P.S. Probably the most difficult challenge of using
moltemplate.sh/oplsaa_moltemplate.py is to figure out the
atom-type-ID-number for each atom in your molecules. (For example
"@atom:80", "@atom:81" correspond to carbon bonded to different atoms,
and "@atom:85" corresponds to carbon-bound hydrogen). The sum of the
partial-charges should be zero (neutral).
If you want to use moltemplate with OPLSAA, be sure to check out the
README file:
http://www.moltemplate.org/examples/alkane_chain_single/README.TXT