[QE-users] About superconducting Tc from the output of matdyn.x

[alpin novianus]

Dear all,
This is my first question in this mailing list. I am using QE 6.3 now, although I would like to upgrade to the latest versions soon.

I would like to clarify about matdyn.x and lambda.x results. For example, I am trying to compute H3S hydride superconductor (Tc ~200K at 200GPa).
(This is a preliminary attempt to get used to the calculations, convergence is not achieved yet in the below results but it is not the core of this question) 
My question concerns the output of the "lambda" file from matdyn.x calculation, as shown below.------------------------------------------------------------------------------------------------------------
  Electron-phonon coupling constant, lambda
 Broadening   0.0050 lambda       2.9575 dos(Ef)  3.7441 omega_ln [K]   1181.8314 Broadening   0.0100 lambda       2.0277 dos(Ef)  3.1220 omega_ln [K]   1211.1636 Broadening   0.0150 lambda       1.8563 dos(Ef)  3.1665 omega_ln [K]   1264.6372 Broadening   0.0200 lambda       1.7776 dos(Ef)  3.2436 omega_ln [K]   1299.6930 Broadening   0.0250 lambda       1.7561 dos(Ef)  3.2997 omega_ln [K]   1319.3964 Broadening   0.0300 lambda       1.7416 dos(Ef)  3.3125 omega_ln [K]   1329.9055 Broadening   0.0350 lambda       1.7180 dos(Ef)  3.2854 omega_ln [K]   1335.8834 Broadening   0.0400 lambda       1.6872 dos(Ef)  3.2343 omega_ln [K]   1339.8464 Broadening   0.0450 lambda       1.6540 dos(Ef)  3.1731 omega_ln [K]   1342.8883 Broadening   0.0500 lambda       1.6216 dos(Ef)  3.1109 omega_ln [K]   1345.4217
------------------------------------------------------------------------------------------------------------If my understanding is correct, the lambda and omega_ln values here are already the ingredients needed to be plugged into the McMillan formula. Once the convergence is satisfied, their values shouldn't change much even with different broadening values. 

>From the examples of Tc calculation with QE, there is another calculation step using lambda.x after this matdyn.x step that yields Tc value.
For example, exercise 1 page 6 of https://indico.ictp.it/event/8301/session/96/contribution/534/material/0/0.pdf in 2018, or example 3 step 7 for Aluminum in PHonon (https://gitlab.com/QEF/q-e/-/tree/develop/PHonon/examples/example03)
However, after reading the past messages in this mailing list, while there were multiple reports of errors encountered when using lambda.xI also read the message from Prof. Eyvaz Isaev that as of QE v4.2 (2010), the function of lambda.x was already incorporated in matdyn.xhttps://www.mail-archive.com/users@lists.quantum-espresso.org/msg06376.html

If that is still true, is it then no longer necessary to perform lambda.x run provided that convergence is achieved in the "lambda" file from matdyn.x output? 
Wouldn't it suffice to directly substitute the converged lambda and omega_ln values to the McMillan formula after matdyn.x calculation?

Thank you for your guidance,Best regards,Alpin N. TatanUniversity of Tokyo.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20220808/8ff1e784/attachment.html>