I need to build a fcc lattice with the 111-plane oriented along the x-axis. This is accomplished using

lattice fcc 4.00 orient x 1 1 1 orient y 1 -1 0 orient z 1 1 -2

The lattice needs to be fully periodic and the simulation box needs to be orthogonal

The command above is then followed by the usual

region box block 0 xdim 0 ydim 0 zdim units lattice

create_box 1 box

create_atoms 1 region box

Is there any ratio of the parameters xdim,ydim and zdim which makes the lattice periodic?

I’ve tried many different combinations and all of them gave me lattice defects at the faces of the simulation box

best regards,

frank.

i can answer my own question:

lammps doesn’t try to make your lattice periodic, the user has to provide the proper dimensions himself.

the following works in my case:

sqrt(3/2)*a along x*

sqrt(2)/2a along y

sqrt(3/2)*a along z

where a is the lattice constant

and of course integer multiples of these should be fine too

EDIT: multiple integers DON’T work. I have no clue why, but you can simply replicate the uni cell using e.g. ovito or a custom script

EDIT2:

forget it…

you can’t use Lammps to create your 111 oriented crystal…do it by hand better

Frank,

I don’t agree with your claim. It is fairly straightforward to construct the crystal you are looking for.

For your case, you just need to make sure that your box length in z direction is a multiple of 3 in units lattice, for example:

units metal

lattice fcc 4.00 orient x 1 1 1 orient y 1 -1 0 orient z 1 1 -2

region box block 0 2 0 2 0 3

create_box 1 box

create_atoms 1 box

mass * 1.0

write_data lattice.data nocoeff

Axel.