phonon spectral function

[Roc2019]

Dear Sir,
Is it possible to obtain the phonon spectral function (includinng the weight of each phonon mode) like the attachment?

Thank you so much.

[Petr Khomyakov]

Please clarify how you define the weight of phonon bands and what is your system(s) of interest.

[Roc2019]

Dear Dr.  Khomyakov,

I mean the spectral function of phonon like  the Fig. 6 in the following paper

PHYSICAL REVIEW MATERIALS 4, 045403 (2020)

So, how can one obtain that picture by using the newest version 2019.12.sp1. ?  (for example, http://ollehellman.github.io/page/workflows/minimal_example_1.html), is it beyond the newest version?

Another, how can one  obtain the electron and phonon linewidth? (for example, the Fig.3 to Fig. 6 obtained by EPW  http://epw.org.uk/Documentation/B-dopedDiamond)

Thank you so much.

[Troels-Markussen]

Hi, It is currently not possible in QuantumATK to calculate the phonon linewidth due to anharmonic phonon-phonon interactions as  done here http://ollehellman.github.io/page/workflows/minimal_example_1.html . In principle you can calculate the electron linewidth due to electron-phonon coupling.  It will require some scripting by yourself, but we might make an actual implementation next year. The way to do it would be something like:

Code

# Calculate electron-phonon coupling.
# This requires a HamiltonianDerivatives and a DynamicalMatrix object.
# Also, specify the k-points at which to calculate the broadening.
# Manually specify the bands to consider (speed up computation time).

electron_phonon_coupling = ElectronPhononCoupling(
    hamiltonian_derivatives,
    dynamical_matrix,
    kpoints_fractional=kpoint,
    qpoints_fractional=MonkhorstPackGrid(30,30,30),
    electron_bands=[3, 4, 5],
    energy_tolerance=1.0*eV,
    initial_state_energy_range=[-100, 100]*eV,
    )

# Calculate the inverse life-time (broadening)
inverse_life_time = electron_phonon_coupling.inverseLifeTime()

The calculated inverse life time corresponds to the linewidth calculated with the EPW code. Regards, Troels

[Roc2019]

Thank you Dr. Markussen .