Anharmonic phonons with EPW/ZG

[achintya]

Hello,

This is regarding excersie 3 from M. Zacharias Tutorial EPW/ZG, page 20 of 52.
When i was trying to run ZG.x < ZG_2.in > ZG_2.out, i got an error saying

"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine A-ZG (1):
Final coords in file are missing ...
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

stopping ...
"
I was able to run previous 2 excises and compiled ZG with the provided ZG.f90.
I am not sure how i should tackle the problem. I couldn't go further becasue of this.

Please, if you have solved this issue, let me know.
Many Thanks
Achintya

[achintya]

Good Morning.

I would like to have your help on EPW/ZG calculations.

I was doing the tutorial on EPW summer school 2023 for the anharmonic phonons with EPW/ZG. For The exercise 3, I encountered a problem which is on page 20 for the Zr sample.
whenever i try to run the code for ZG such as

ZG.x < ZG_2.in > ZG_2.out

i see an error whis is pasted below

"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine A-ZG (1):
Final coords in file are missing ...
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

stopping ..."


I compiled the EPW code using the provided ZG.f90 from the tutorial. ZG.relax.out was okay also. I am using Macbook, Ubuntu WSL; but both cases I see the same problem.
Earlier 2 exercises were okay, it ran smoothly without problem.

Even though I was using ZG.relax.out from the provided file from the output director, the problem remains the same.
Could you help me where it went wrong. Any help would be . I am new to this field.

I look forward to your response.

Many Thanks
Achintya
University of Manchester

[mzach]

Hello Achintya,

Thank you for your post.

The problem seems weird. The error " Final coords in file are missing ..." is printed by the code when the string "Begin" in ZG-relax.out is not found.

>> ZG.relax.out was okay also

Can you verify that in your output "Begin final coordinates" exists and also the final atomic coordinates are printed? If the relaxation didn't finish successfully you might need to rerun ZG-relax.in, but now starting from the last coordinates printed in ZG-relax.out.

>> Even though I was using ZG.relax.out from the provided file from the output director, the problem remains the same.

Since you are using Ubuntu WSL maybe there is a problem with the file permissions or character/file encoding or something else?

Please let me know if you need further help.

With best wishes,

Marios Zacharias

[mzach]

Hello Achintya,

I have compiled QE on my laptop (Ubuntu 20.04.6 LTS) using gfortran and managed to reproduce the error you mentioned. Although ZG-relax.out has finished successfully, the error persists:

"%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine A-ZG (1):
Final coords in file are missing ...
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

stopping ..."

For some reason, the error is related to the quotation marks " printed at the beginning of each QE run ("P. Giannozzi et al ...). The code compiled with gfortran encounters an error while attempting to read those lines containing ". A quick workaround is to remove the quotation marks from file ZG-relax.out using:

sed -i 's/"//' ZG-relax.out

and then:

ZG.x < ZG_2.in > ZG_2.out

You might need to apply the same strategy for reading the forces from the output files in the next steps.

Best,

Marios

[achintya]

Hello Marios,

Thank you very much. Now it worked for me too.
I wish i saw this post before, it would have saved my lot of effort time.
Thank you so much
Best Wishes,
Achintya