EPW Forum

Dear developers and users,

I'm attempting to use EPW to extract the mobility along armchair/zigzag directions of strained 2D TMD materials, and I'd like to achieve similar results to (Fei Guo et al 2019 J. Phys.: Condens. Matter 31 295502) and (Manouchehr Hosseini et al 2015 J. Phys. D: Appl. Phys. 48 375104).

When we use IBTE to calculate the matrix of electron mobility for a given temperature, do the diagonals correspond to the mobility along unit cell vectors defined in the scf calculation, or the general 'x' and 'y' directions? I'm using "ibrav=0" along with the "cell_parameters" card to mimic a hexagonal lattice. For example, one of my cell_parameters cards looks like this:

CELL_PARAMETERS {alat}
1.000000731 -0.000000000 0.000000000
-0.500000365 0.866026037 0.000000000
0.000000000 0.000000000 4.700875000

Is the (2,2) mobility value calculated by BTE then given along the (0,1,0) or 'y' direction, or the (-0.5, 0.86, 0) direction? I suspect it's the latter, but I'd like to clarify.

Thank you,
Manasa Kaniselvan

Dear Manasa,

It is slightly difficult to answer this as it depends on the version of EPW you are using and the option you chosen.

Easiest it to check for yourself. You can look in the EPW/src/printing.f90 subroutine. If the velocity vkk_all and occupation vector f_out are in crystal when printed then indeed they should be along your crystal vector directions. Look for the various print_mob subroutine and the compute_sigma one.

In addition, please bear in mind that 2D materials are not yet officially supported in EPW, see viewtopic.php?f=3&t=1311

Best wishes,
Samuel