Dear Lammps Experts,
I do have a reactive system (full of reactive radical) and I want to equilibrate it. The force field I am using is ReaxFF. I wonder if there is any way to prevent reactions from happening while equilibration is in process? In other words I want to prevent reactions from happening while I am running nvt trajectory.
Please refer me to the LAMMPS documentation to learn more about it. I will be more grateful if you can briefly explain about it too.
Thanks.
Dear Lammps Experts,
I do have a reactive system (full of reactive radical) and I want to
equilibrate it. The force field I am using is ReaxFF. I wonder if there is
any way to prevent reactions from happening while equilibration is in
process? In other words I want to prevent reactions from happening while I
am running nvt trajectory.
this is a non-trivial thing to do.
Please refer me to the LAMMPS documentation to learn more about it. I will
be more grateful if you can briefly explain about it too.
the LAMMPS documentation will explain how you can "operate" LAMMPS, it
will not teach you how to do your research. This is not very different
from, say, the owner's manual of a car will tell you how to operate
the various features of a car, but it won't teach you how to drive.
you've got to go to driving school for that or learn from somebody
experienced and also study the rules (i.e. suitable text books).
thus you may get a few pointers from this mailing list (or the docs),
but the burden of figuring out how to get it right is on you and the
burden of how to train and advise you to do your research is on your
adviser.
with all that in minds, here is my suggestion: you could try to add
restraints to your system, i.e. additional forces that try to enforce
bonds being broken and, if needed angles, dihedrals being pushed from
their desired initial position. how to do that depends on how many
molecules/bonds/angles are affected. LAMMPS has a fix restraint, that
can add such forces individually. if you need more, e.g. all bonds,
then you could add a bond topology and assign (harmonic) bond forces.
but watch out, explicit bonds will lead to pairwise exclusions by
default, which must not happen with reaxff. so you'll have to turn
exclusions off using the special_bonds command.
axel.
Since REAXFF is a multi-body potential the suggestion Axel mad is a possible
path. However, even if you do what he suggests, you will not have an
equilibrated system at the end of your equilibration run since the
constraints will, when relaxed for your reactive run, result in an abrupt
change in energy.
What you might consider is equilibrating your run with a fully polarizable
force field and then use those coordinates and velocities as input to your
reax/c run. You will still not have a "properly" equilibrated reax system,
but you may experience less of an abrupt energy dislocation than that which
you will get with sudden bond relaxations.
That's my two cents, anyway.
Jim
James Kress Ph.D., President
The KressWorksR Foundation
An IRS Approved 501 (c)(3) Charitable, Nonprofit Organization
"ENGINEERING THE CURE" C
(248) 573-5499
Learn More and Donate At:
Website: http://www.kressworks.org
Facebook: https://www.facebook.com/KressWorks.Foundation/
Twitter: @KressWorksFnd
Confidentiality Notice | This e-mail message, including any attachments, is
for the sole use of the intended recipient(s) and may contain confidential
or proprietary information. Any unauthorized review, use, disclosure or
distribution is prohibited. If you are not the intended recipient,
immediately contact the sender by reply e-mail and destroy all copies of the
original message.