lammps-users Digest, Vol 70, Issue 46

Message: 7
Date: Sun, 18 Mar 2012 17:11:36 +0200
From: “David Furman” <sirok4@…24…>
Subject: [lammps-users] skew box too large
To: “‘LAMMPS mailing list’” <[email protected]>
Message-ID: 000001cd0519$6ab48870$401d9950$@com
Content-Type: text/plain; charset=“us-ascii”

Hello all users and developers,

I try to simulate a triclinic system with LAMMPS. I have a crystal structure
of super-cell (4X4X4 unit cells) obtained from XRD (XYZ file).

When I prepare the data file for lammps run, I use the tilt factors and cell
dimensions from Materials Studio lattice parameters for this file.

But, then when I try to minimize I get the error of box skew is too large.
How should I modify one of the tilt factors accordingly so the geometry

of the lattice stays the same ?

Thanks !

Hi David,

The error appears because in LAMMPS: “A prism region that will be used with the create_box command to define a triclinic simulation box must have tilt factors (xy,xz,yz) that do not skew the box more than half the distance of corresponding the parallel box length. For example, if xlo = 2 and xhi = 12, then the x box length is 10 and the xy tilt factor must be between -5 and 5. Similarly, both xz and yz must be between -(xhi-xlo)/2 and +(yhi-ylo)/2.” This is from region command description in LAMMPS manual.

Based on this I suggest you copy your box via replicate command so that, say in case of xy tilt you replicate a few times in x direction only or more times than in y direction, i.e. say, replicate 2 1 1 or replicate 3 2 1. At least this is what I usually do in this case. Hope this helps.

Cheers,
Manana Koberidze