9
$\begingroup$

How should I calculate lattice thermal conductivity in Quantum ESPRESSO? I have installed phonopy. My Linux distribution is Ubuntu.

Improve this question
$\endgroup$
4
  • 3
    $\begingroup$ I'm kind of new to this, but from what I've read you might need to use the ShengBTE code along with QE in order to find the lattice thermal conductivity. $\endgroup$
    – Anoop A Nair
    Jul 19, 2021 at 19:25
  • 1
    $\begingroup$ @AnoopANair yes. That’s what I will be trying to solve this. If I get it done successfully, I will post the steps here after as answer to this problem. $\endgroup$
    – epsilon02fft
    Jul 20, 2021 at 5:19
  • $\begingroup$ @epsilon02fft Did you have any luck? $\endgroup$
    – Nike Dattani
    Aug 24, 2021 at 0:47
  • $\begingroup$ @NikeDattani Yes I did. I was successful in performing phonopy and phono3py calculations for Si. The codes ran successfully, but I am facing further problems of mpirun and RAM limitations about which I will post soon. Apologies for not responding before, couldn’t login to my account and also got busy with uni official work. $\endgroup$
    – epsilon02fft
    Nov 21, 2021 at 10:23

2 Answers 2

Reset to default
3
$\begingroup$

Some years ago I found the following steps in ResearchGate: You need:

  • First, you need to create a QE input file (pwscf), using ibrav=0, lattice vectors in bohr units, atomic coordinates in crystal units.
  • Using the input file, run phono3py/third_order.py to create supercell within finite displacement method.
  • Then you need to write a script that will copy all header option, specially &control and &system to each supercell file and rename to scf.in and generate sequential folder namely, disp-00001..., ect.
  • Now run pwscf for each supercell and please do not forget to select force calculation in pwscf input file.
  • After this step you need to run phono3py to calculate second order (harmonic) and third order (anharmonic) force constants.
  • Finally, run phono3py to calculate lattice thermal conductivity using dense q-point grid and primitive cell.
Improve this answer
$\endgroup$
1
  • 1
    $\begingroup$ Thank you sir for your reply. I was successful to run these codes. Now I am working to solve ShengBTE code and the problem I am facing of mpirun which is occurring in 3rd step above (as your wrote) $\endgroup$
    – epsilon02fft
    Nov 21, 2021 at 10:25
1
$\begingroup$

there is this D3Q package that works seamlessly with Quantum Espresso.

I used it in several papers such as PRB 98, 220301(R) (2018). Easier than the QE-phonopy-shengbte chain. Quite demanding computationally, though.

https://anharmonic.github.io

G Fugallo et al., PRB 88, 045430 (2013)

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thank you for your reply. Yes, recently I got to know about D3Q package and I have started reading its manual as well as practice sample examples. $\endgroup$
    – epsilon02fft
    Feb 27 at 4:48

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .