Grain Boundaries

[Eric Hoglund]

Hi all,
I am trying to model some tilt grain boundaries. Two questions.
First:
I have seen the transport tutorial for creating a GB and that I managed fine, since it is just slicing two different planes.  I am having issues with creating symmetric tilt boundaries. For instance the special case of a Σ3, where the grains are both in a (110) orientation with (-1 1 1)//(-1 1 -1). If I use the cleave tool for the two different {111} planes, then it returns two crystals of the same orientation. So, when I build the "interface" using the interface tool, their is no miss-orientation (ie I am joining two identical cells).  I have also though about taking the initial FCC crystal and rotating, but it only has one atom in the basis, so the function changes nothing (rotate operates on the basis vectors not lattice). Using repeat then rotating gives the correct orientation, but then wrapping or using the interface tool causes some odd overlap within the rotated cell (since the rotated basis is not contained within the un-rotated lattice).  Does anyone have any ideas, or has anyone here modeled grain boundaries before?
I saw this thread with a the paper by Feldman https://quantumwise.com/forum/index.php?topic=1541.msg7699#msg7699. The two twin grain boundaries he molded would be a great start for me.  I see that I can use the CSL model while constructing the interface. If I want say Σ3, should I cleave (-1 1 1)&(-1 1 -1) then scan a rotation of 69 degree to 72 degree?

Second question:
After creating and relaxing a GB I would like to define a portion of the GB as a cell, then run some phonon calculations.  How do I define the cell?

Thank you
Eric Hoglund
University of Virginia

[Daniele Stradi]

Dear Eric,

the interface builder does not take into account the atomistic details of the interface, but work by matching the two surface cells by rotating and straining them. In this sense, the interface cell for a Sigma3 boundary is the same as that for the pristine materials, since the cells of both surfaces are not strained and are rotated by a multiple of 60 degrees.
The only way to create a Sigma3 boundary with the interface builder would be to create two surfaces which are one the mirror image of the other, and then match them together. You can use the 'Mirror' plugin for that: Coordinate Tools -> Mirror.
To remove overlapping atom, you can use the 'Close Neighbors' plugin: Selectron Tools -> Close Neighbors

Regarding the second question, what do you mean by "a portion of the GB"? If you want to move only selected atoms, this can be done during the setup of the phonon calculation.

Regards,
Daniele.