a question about lattice parameters calculated by thermo_style

Dear all,

There is the possibility of having xla, yla, zla ( lattice dimensions) in the thermo_style command. But the initial lattice parameters are shown throughout the simulation. Is it possible for LAMMPS to calculate the new lattice parameters if any phase transition happens? I mean, is it concluded that no phase transition happens due to the constant lattice parameters throughout the whole simulation?

regards,
Nazanin

Dear all,

There is the possibility of having xla, yla, zla ( lattice dimensions) in
the thermo_style command. But the initial lattice parameters are shown
throughout the simulation. Is it possible for LAMMPS to calculate the new
lattice parameters if any phase transition happens? I mean, is it concluded
that no phase transition happens due to the constant lattice parameters
throughout the whole simulation?

what "lattice" are you referring to?

the one build from the lattice command?
that one has no meaning for the simulation
itself; it is simply a template for placing
atoms in space and conveniently determining
commensurate geometries.

axel.

More specifically, LAMMPS does not calculate
"lattice parameter" for some arbitrary config of
atoms. Rather you specify the lattice and its
parameters via the lattice command. These
lattice spacings can be printed out with the
thermo output. But they are exactly the ones
you generated with your lattice comand. They
have nothing to do with where the atoms have
moved at a later time.

Steve

Dear axel,
I have built a BCC structure which has lattice parameter a=b=c=3.015 by the lattice command. a phase transition to another structure with these parameters is expected : a=3.00 b=4.03 c=4.45 .
By these conditions, does LAMMPS shows the change in the lattice parameters whether the phase transition occure?
thank you,
Nazanin

thank you Steve, I have taken my exact answer.

no - what command are you expecting would do that?

Steve

I have used thermo_style custom xlat ylat zlat command to print the lattice parameters out. I expect new parameters to be shown in the log file which prints the thermo_style command’s results, if any phase transition happens.
If the thermo_style command is not able to show the change in lattice parameters after structural change, is there any other command to do it?

thanks,

I expect new parameters to be shown in the log file

which prints the thermo_style command's results, if any phase transition
happens

Sorry, your expectation is wrong. If you read the
lattice doc page it says nothing about re-computing
lattice parameters based on current atom configs.

What you are asking for is non-trivial. Detecting
a phase transition is not easy. In a large
system, there could be many regions of different
phases, at different orientations. The "lattice param"
is not a global quantity, so how would LAMMPS
come up with one number for it?

There are several computes that look at an atom's
local environment to calculate a lattice symmetry.
They all produce a per-atom (not global) values that
can be dumped in a dump file. See compute centro/atom,
cna/atom, coord/atom, displace/atom, etc.

Steve