Dear LAMMPS users,
Currently I’m in the process of trying to generate configuration files for symmetric tilt grain boundaries of SiC and ZnS. What is making this more difficult than usual is there are only two rotation axes for the Zinc-Blende structure which correspond to a symmetric tilt grain boundary: <110> with (1-10) median plane and <001> with (110) median plane. I’ve attempted to do this via LAMMPS input script but have failed miserable (maybe I’m just not thinking hard enough). Does anyone know of external tools which can generate bicrystals of binary compounds or a way to do it in LAMMPS?
Thanks in advance,
Stefan
You can take an idea from the Tutorials on LAMMPS home-page- http://lammps.sandia.gov/ and see if it can assist you.
Or if you understand the geometry properly, then all it will take is a few sheets of paper and some free time to generate your geometry and create your own data file to input in LAMMPS.
Sagar
Conecptually, the steps to make a bicrystal are as follows:
a) define 2 regions, e.g. upper and lower
b) define a lattice for the upper region, rotated appopriately
c) create atoms in the upper region
d) repeat b and c for the lower region, using a different lattice orientation
e) use delete_atoms overlap to remove atoms at the boundary
that are too close
f) minimize the energy
The devil is in the details of the 2 lattice commands and
insuring that you observe periodicity and get atoms right
where you want them. There is no simple solution for
that other than visualizing your system, dumping atom coords,
starting with a tiny system, etc.
Steve
Sagar and Steve,
Thank you for the advice. What I’m finding difficult is orienting the crystals twice. If I choose for example the
<001> axis as my rotation axis, then I have to choose the median plane as the (110). Following the steps
similar to those in the tutorial guides I will end up with a boundary plan of (010). I’ve tried circumventing this
by choosing a1 and a2 to point along the (110) and then orient each crystal with respect to those axes. However, it
seems I’m not implementing it correctly. I’ll approach this as Sagar suggested by working out the details and putting together a script to generate the bicrystals.
Thanks again,
Stefan