I am trying to optimize Pt-doped SnO2 (110) at a particular dopant site. In the first attempt, my calculation crashed with an error saying, "Error in routine bfgs (1): dE0s is positive which should never happen". So I decreased the mixing beta value from 0.7 to 0.1 and re-ran the optimization. This time, the output file again ended abruptly with an error, saying, "Error in routine bfgs (1):
step_old is NOT normalized". I have attached my input and output files here for reference.
I admittedly do not see much of a problem here. The only point I would try to change is the startingpot and starting_wfc flags. Do you need to start with the atomic part? Does a random wavefunction give issues? I would also force a restart mode from scratch. It seems as your run are finding some ghost bfgs calculations and thus have a nonsensical results.
I hope this helps,
Roberto
Shaheen Gulshanah I tried to run your input file and the calculation runs with no problem (I didn't finish it, but it did complete the fist BFGS step and started the new scf calculation).
I did reduce a bit the density cut off (why so large?) and I am using Quantum Espresso 6.8. Might be this was a bug of 6.0?
Cheers,
Roberto
@Roberto D'Agosta sir, first of all, thank you so much for taking your valuable time and running my input file, and for your suggestions. When I tried running it for the first time, I got the error message after 2 scf cycles. I chose the kinetic energy cut off to be 500 eV, as it was suggested in most literatures related to my work. And regarding the charge density cutoff, since I am using an ultrasoft pseudopotential, I found that it's better to take 10-12 times the value of ecutwfc for ecutrho. I will look into the startingpot and startingwfc flags as you've mentioned. I will try to re-run the calculation keeping in mind the pointed you've stated. I hope it works this time. Thank you again, sir!
Good morning dear sir, were you able to solve the problem? I am getting the same message
- Badges
- Science topic
- Similar topics
- Filing