[lammps-users] Atom position definition in a parallelepiped simulation box

Dear all,

Now I am dealing with a non-orthogonal simulation domain in my calculation. I wrote a data file for inputting my box size and atom position.

In the input file, I use xlo/xhi, ylo/yhi, zlo/zhi, xy, xz, yz to define the parallelepiped box which will have a1/a2/a3 three non-orthogonal basis vectors.

Then when I define the atom position in Atoms section of the input file, the ‘x,y,z’ parameter for each atom should be based on the a1/a2/a3 non-orthogonal coordinate or the x/y/z original global orthogonal coordinates?

Thanks in advance!


The x,y,z coords of atoms in a data file should
always be actual positions. They will be mapped
into a periodic box, orthogonal or not, as needed.