Dear LAMMPS users,
The aim of my project is to study the behavior of molten slags (ceramics)
of varying composition. To study slags one has to take a mixture of
ceramics and melt them at very high temperature. Miyake potential which
belongs to the family of Buckingham potentials is used for Al, O, Si, Ca,
K, Na systems. While in the past I have worked with Stillinger-Weber,
Tersoff, ReaxxFF potentials, Miyake potential is difficult for me. I have
tried a lot, but when I try using more than one type of atom the simulation
box just blows up or loose atoms or nan temperature.
I have tried energy minimization, nve/limit (because I generate atoms
randomly), npt, nve, nvt. Have played around with the
composition, temperature nothing seems to work. I think I am making mistake
with the potential parameters. I know it is an advanced problem, and no one
in my research group seems to know how to deal with it. It would be kind if
any one of you take a look and help me solve this problem. I am attaching
my input file with this email and the citation for Miyake potential.
A. Miyake, "Interatomic potential parameters for molecular dynamics
simulation of crystals in the system K2O-Na2O-CaO-MgO-Al2O3-SiO2" Miner. J.,
20 (1998), pp. 189-194
Thanks
Sankha Mukherjee
input.in (3.54 KB)
Dear LAMMPS users,
The aim of my project is to study the behavior of molten slags (ceramics)
of varying composition. To study slags one has to take a mixture of
ceramics and melt them at very high temperature. Miyake potential which
belongs to the family of Buckingham potentials is used for Al, O, Si, Ca,
K, Na systems. While in the past I have worked with Stillinger-Weber,
Tersoff, ReaxxFF potentials, Miyake potential is difficult for me. I have
tried a lot, but when I try using more than one type of atom the simulation
box just blows up or loose atoms or nan temperature.
have you tried to first unoverlap your system by starting with a (flat)
soft potential which you raise over time using fix adapt? you can check out
the micelle example for how to do this. just translate the parameters in
reduced units to your unit set.
also, i suggest you start with fewer atom types and then test groups of
parameter for individual compounds, before doing as complex a system as in
your input.
I have tried energy minimization, nve/limit (because I generate atoms
randomly), npt, nve, nvt. Have played around with the
composition, temperature nothing seems to work. I think I am making
mistake with the potential parameters. I know it is an advanced problem,
and no one in my research group seems to know how to deal with it. It would
be kind if any one of you take a look and help me solve this problem. I am
attaching my input file with this email and the citation for Miyake
potential.
actually, it doesn't look like an advanced problem, but just using a basic
step-by-step procedure to avoid having too many uncontrolled degrees of
freedom. i would call this common sense... 
axel.