EPW Forum

Dear all,
Is there a way to obtain just the POP matrix elements from the code?
In the system that I am studying, the polar optical phonon scattering is the major contributor that inhibits electron mobility. So I just want to analyse the POP modes.

Thank You,
Ankit Sharma

Dear Ankit,

If you set longrange=.true. you will only include in your calculation the generalized Frohlich matrix element from https://journals.aps.org/prl/abstract/1 ... 115.176401
(bear in mind that this will only calculate the long range matrix element using the analytical formula for gL instead of doing the full interpolation of the matrix elements)

Carla

Hi,
I am trying to implement the Frohlich Electron-Phonon Vertex in the paper https://arxiv.org/abs/1510.06373.
For eigendisplacements in the matrix element expression, I am using the values generated after running DFPT (Quantum Espresso). Also since my analysis involves only optical phonons, I consider q points very close to the gamma points with constant energies as that at gamma point.
1.So, the question is, can I directly use the eigendisplacement values generated by QE since these values are divided by sqrt(mass) and then normalized and the Frohlich interaction expression again has the term 1/sqrt(mass)???

2. Also, what does normalization within a unit cell mean as in the paper mentioned above???

Thank You,
Ankit Sharma

Dear Ankit,

Sorry for the late reply, I hope it can still be useful. If you need to implement the Frohlich coupling separately, yes you can use directly the eigenvectors generated by QE.
In practical implementations the 1/sqrt(N) factor is factored out of the matrix elements and introduced when doing the BZ summations.

Best
Carla

Thank You Carla!