Dear all users.
I’m going to simulate Argon and Platinum substances that they have
same lattice structure (fcc) but different lattice constant. So I
think I must write two codes with different lattice scale and then
make a file for read_data command. I’m using unit lj command, but I
have problem about unitless density (rho*) when I use lattice command.
I saw LAMMPS User’s Manual (units command section) and I was confused
for unitless density because I think it needs M (rho* = rho
m^-1*sigma^3). If it is true, what mass must I use for
non-dimensioning? I mean for example in fcc command for platinum,
must I use Argon atom’s mass or Platinum atom’s mass? (Because I use
atom mass, sigma and epsilon of Argon for non-dimensioning)
Thanks for helping.
Mohsn Ghorbani.
Dear all users.
I’m going to simulate Argon and Platinum substances that they have
same lattice structure (fcc) but different lattice constant. So I
think I must write two codes with different lattice scale and then
make a file for read_data command. I’m using unit lj command, but I
have problem about unitless density (rho*) when I use lattice command.
I saw LAMMPS User’s Manual (units command section) and I was confused
for unitless density because I think it needs M (rho* = rho
m^-1*sigma^3). If it is true, what mass must I use for
non-dimensioning? I mean for example in fcc command for platinum,
must I use Argon atom’s mass or Platinum atom’s mass? (Because I use
atom mass, sigma and epsilon of Argon for non-dimensioning)
it seems to me, you are missing the point of reduced units. if you
want to model specific materials, you may be better off using one of
the other unit sets like "real" or "metal".
units essentially always mean a scaling factor relative to a reference
and that is true for reduced units as well. reduced units have the
benefit that you can more conveniently generalize the results for
simple materials, as you can say you study materials with a specific
relation between their potential parameters and the results are not
necessary specific for a given pair of materials. if you do look at
two very specific materials, there is not much of a benefit from
reduced units.
the important point is that you can have only one reference in any
system, so to model a two component system, you have to pick which
component is providing the reference (energy, length, mass) and
express the other in units of the first. or you can pick an arbitrary
third component, e.g. something where you identify epsilon with
1kcal/mol and sigma with 1 angstrom and the mass with 1 amu and -
ta-DAA - suddenly you have to enter your parameters in real units.
axel.
Also, the reduced density in the lattice command is number density,
not mass density, as the doc page clearly states.