How to get k-resolved Eliashberg spectrum

General discussion around the EPW software

Moderator: stiwari

Post Reply
hellolori

How to get k-resolved Eliashberg spectrum

Post by hellolori »

Dear all,

As we know, Eliashberg spectrum reads,

Code: Select all

\alpha^{2}F(\omega) = 1/N_F/N_k/N_q \sum_{k,k',\nu} |g_{v}_{k,k'}|^2*\delta(\epsilon_k)*\delta(\epsilon_{k'})*\delta(\omega - \omega_{q,\nu})

I would like to know can I get k-resolved Eliashberg spectrum, \alpha^{2}F(k,k',\omega) from EPW directly by setting some tags? Where can I find the corresponding output?
Can any one help me?

Bests!
roxana
Posts: 172
Joined: Fri Jan 22, 2016 6:48 pm
Affiliation:

Re: How to get k-resolved Eliashberg spectrum

Post by roxana »

Hi,

The anisotropic Eliashberg spectral function is not printed on file. You can make some small changes to the code if you would like to print it on file in one of the following two ways:

1) In eliashberg.f90 "CALL evaluate_a2fij" after "CALL evaluate_a2f_lambda"
You will need to add the lines to print to file in the subroutine evaluate_a2fij

or

2) In the subroutine evaluate_a2f_lambda calculate the anistropic Eliashberg spectral function as it is done in evaluate_a2fij and print it to file from there.

Best,
Roxana
Roxana Margine
Associate Professor
Department of Physics, Applied Physics and Astronomy
Binghamton University, State University of New York
hellolori

Re: How to get k-resolved Eliashberg spectrum

Post by hellolori »

roxana wrote:Hi,

The anisotropic Eliashberg spectral function is not printed on file. You can make some small changes to the code if you would like to print it on file in one of the following two ways:

1) In eliashberg.f90 "CALL evaluate_a2fij" after "CALL evaluate_a2f_lambda"
You will need to add the lines to print to file in the subroutine evaluate_a2fij

or

2) In the subroutine evaluate_a2f_lambda calculate the anistropic Eliashberg spectral function as it is done in evaluate_a2fij and print it to file from there.

Best,
Roxana


Thank you very much. What I have done is direct modify the subroutine
"evaluate_a2f_lambda" to print the k-resolved Eliashberg spectrum. After that, I test the validity of my method by summing up the output and the resultant Eliashberg spectrum is in lines with the "prefix.a2f" generated by EPW. I did not modify the subroutine "evaluate_a2fij" as you mentioned above. I wonder whether the above two methods are equivalent?
roxana
Posts: 172
Joined: Fri Jan 22, 2016 6:48 pm
Affiliation:

Re: How to get k-resolved Eliashberg spectrum

Post by roxana »

Hi,

Sounds good! Yes, the two methods are equivalent.

Best,
Roxana
Roxana Margine
Associate Professor
Department of Physics, Applied Physics and Astronomy
Binghamton University, State University of New York
Post Reply