Hi all,
I was wondering what exactly is the difference between using the fix pour and fix deposit commands?
It says in the documentation that fix deposit is only a part of the MISC package and fix pour,GRANULAR package. Is there any way to replicate these commands when using reax potential?
Thanks
Hi all,
I was wondering what exactly is the difference between using the fix pour
and fix deposit commands?
both are indeed similar, but have differences. fix pour is specifically
needed because it can deposit extended particles that are either mono- or
poly-disperse. fix pour requires an atoms style supporting such particles.
fix deposit does not have that feature and requirement, but instead has
features fix pour doesn't have.
It says in the documentation that fix deposit is only a part of the MISC
package and fix pour,GRANULAR package. Is there any way to replicate these
commands when using reax potential?
the fact that the fixes are part of a specific package, does not mean,
that they cannot be used with any other package. fix pour is required for
pair styles in GRANULAR, since you cannot generate extended particles (as
granular particles are) with fix deposit, and fix pour can only generate
such particles.
fix deposit should work with reax/c, but some care may be needed, since
reax/c is structured internally a bit different than other pair styles in
LAMMPS. its pair style is more like an interface to a library of a
formerly standalone simulation code that could do reax/c only. over the
years, any limitations of the reax/c code within LAMMPS have been lifted,
though, and parts not used when run inside LAMMPS removed. thus before
using fix deposit on production calculations, i would make some tests,
where you compare results from fix deposit with results where you do small
chunks of a simulation and then deposit particles via create_atoms instead.
these two can be made to act similar, but the second variant *has* to work,
as it enforces the init() and setup() phases of a simulation for every
chunk.
HTH,
axel.
Also, fix pour relies on gravity, to pull the inserted particles
out of the poured region. It would make little sense
to use fix gravity in an all-atom MD model for ReaxFF.
Steve