I used fitfc command to calculate phonon properties. However, I got "unstable mode found, aborting" error message sometimes. My question is that how can I know whether the error is caused by calculation convergence or dynamically unstable structure? Does fitfc command parameters, such as -er, -ns, -ms, -dr, -fr etc, affect the calculation results? How to set these parameters to avoid unstable mode?
Many Thanks!
The best way to settle this is to run fitfc with the -fu option (find unstable).
Then you look at the output:
grep u vol_*/unstable.out
pick an unstable mode and force the code to generate it with the -gu=[index] option
rerun you ab initio code for the generated structure.
rerun fitfc
If the mode is truly unstable it will tell you at this point.
This is explained in more detail in fitfc -h
Dear avdw,
Thank you for your reply.
After I run fitfc with -fu option, nothing was found in the vol_*/unstable.out. I don’t know why.
By the way, large supercells or small atomic displacement (-dr) is preferable to avoid artifact unstable modes. Is this right?
The supercells should be big enough to contain twice the length of the longest pair. Beyond that, there should not be any benefits.
I would say that using a large -dr is beneficial to avoid spurious instability.
As to why there is no unstable.out file, you need to tell me what the code did output.