What methods and tools are used for building models of grain boundaries?
I mean initial models that are then optimized using molecular dynamics potentials or a different methods.

When I was involved in such things 10-15 years ago we were writing own codes to generate starting structures. Both big systems that emulated nano-sized policrystals


and small bicrystals to analyze particular boundary types


(I added pictures to clarify what type of things I'm writing about)

I wonder what tools (if any) are used nowadays?


3 Answers 3


If you are looking for generating synthesized microstructures for using it in Phase-Field simulation, the best choice is Dream 3D, where you can control the distribution of grains and their crystallographic orientations:

enter image description here

If you are looking for a framework to generate synthesized polycrystalline structures for molecular dynamics simulation, the best tool is Atomsk, that is compatible with LAMMPS, DL_POLY, VASP, QuantumEspresso, SIESTA, etc.:

enter image description here

  • $\begingroup$ Can Atomsk generate bicrystal models for analysing particular types of grain boundaries like CSL/special boundaries ($\Sigma 3$, $\Sigma 5$, etc.)? $\endgroup$
    – Mythreyi
    May 5, 2020 at 11:30
  • 1
    $\begingroup$ @Mythreyi Not directly as far as I know. $\endgroup$ May 5, 2020 at 19:48

Alone Programmer's answer already mentions ways to generate polycrystalline microstructure. To add to that, it is possible to generate models of bi-crystals to analyse particular grain boundaries directly using LAMMPS. Here is a detailed wiki on how to generate such models.

Apart from this, there was a Java based applet called GBStudio that made generating grain boundary models very simple. Unfortunately, this tool seems to be no longer available.


In 2008 I wrote a Python script to generate bicrystal models in a PBC box.
Since the boundary is periodic (due to PBC), it's always a coincidence site lattice (CSL) boundary.

enter image description here

The script is in repository named gosam and is documented here.

The math was based on H. Grimmer's paper "The generating function for coincidence site lattices in the cubic system", so it works only for cubic lattices.

From time to time (maybe once every two years) I get a question from someone trying to use this script. Unfortunately, I don't remember much and I can only answer simple questions.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .