Dear Lammps user,
I work with crystal. I made the crystal in Material studio and exported it as a pdb file and made the data file in VMD (topo tools). I suppose that the simulation box size should be the crystal lattice constants. But in my data file the differences between the coordination of atoms in z direction is greater than the lattice constant ©. I mean atoms zmax-zmin> lattice constant © Please guide me if you face with a similar problem. Also I want to know can we choose xlo, ylo and zlo a negative number.
.
Kind Regards,
Sara
You can choose all the xyz lo/hi values for the box to be anything you wish. The box
size does not have to be a multiple of some lattice constant unless you want it
to be periodic in a particular dimension See the boundary command.
Steve
Dear Lammps user,
I work with crystal. I made the crystal in Material studio and exported it
as a pdb file and made the data file in VMD (topo tools).
PDB files contain only very limited information. nothing about
bonding, force field assignment and so on. this cannot be recovered by
TopoTools and it doesn't try. if you are not careful you can
incorporate lots of bogus information by importing a PDB file into VMD
as VMD will assume: a) that you have a molecular system and will try
to guess bonds, b) assume that your atom naming follows CHARMM force
field conventions, c) VMD will assume, that there are no periodic
interactions or bonds.
I suppose that the simulation box size should be the crystal lattice constants.
no.
But in my data
file the differences between the coordination of atoms in z direction is
greater than the lattice constant (c). I mean atoms zmax-zmin> lattice
constant (c) Please guide me if you face with a similar problem. Also I want
to know can we choose xlo, ylo and zlo a negative number.
you don't seem to understand periodic boundary conditions. you *can*
have atom positions outside the simulation box. with PBC, they will be
wrapped back into the principal simulation cell for the force
computation.
please do yourself a big favor and find a local tutor, that can help
guide you in how to set up and execute MD simulations. there are far
too many details and practical issues depending on the specifics of
your system, that you can obtain the necessary guidance from a mailing
list. nobody here has the time to do this and it is far, far more time
consuming to explain things via e-mail and without being able to look
over your shoulder and see what you are doing.
axel.