[lammps-users] acks2/reaxff vs. qeq/reaxff

LAMMPS LAMMPS Mailing List Mirror
1

Hi LAMMPS developers and users,

I have installed the latest patch release dated 27 October 2021, mainly to explore the electric field effects on charge equilibration, with qeq/reaxff and with the new fix acks2/reaff. I have started with the examples available in the package, specifically in lammps/examples/reaxff/water. By comparing reaxff files (acks2_ff.water and qeq_ff.water), I see that those files only differ in two parameter values, specifically on line 37 (softness in acks2_ff.water) and a value on the H block.

My questions: if I want to use an arbitrary reaxff file (like those available in https://www.scm.com/doc/ReaxFF/Included_Forcefields.html) with the new fix acks2/reaxff, should I modify it in order to include a softness value? If so, how do I determine that value? and, should I also modify that value on the H block?

With kind regards,

Luis

2

This is a Q for Stan (CCd)

Steve

3

Thanks Steve.

With kind regards,

Luis

4

Hi All,

I got access to the papers referenced on https://docs.lammps.org/fix_acks2_reaxff.html, which has allowed me to answer my questions. Sharing here in case someone else is expecting closure on this:

An arbitrary ReaxFF force field file cannot be used with the fix acks2/reaxff command. A bond softness value and a type-dependent bond cutoff should be determined and written on those files, or a parameter filename should be created containing this information, as described in the documentation. Getting those numbers is not trivial, as an empirical fitting should be performed (details on section 2.3 of the O’Hearn et al. paper).

With kind regards,

Luis

1 Like