Can we use defect finder to find the vacancy

[whyydsforever]

It is really a good job for this project.

I wonder if the defect finder can find the vacancy of the defects since the defect finder is based on the SOAP descriptor, while SOAP seems not support for the vacancy.

Thank you very much and I am looking forward to hearing from you soon

[jmmshn]

Hi @whyydsforever, it should work for vacancies.
The finder calculates a scalar value for each atom that indicates how much that particular atom differs from the bulk version.
The does some weighted averaging of those atomic positions. I've tested this on vacancies and it should work.

[whyydsforever]

What if the structure has many vacancies, does it still work? Or is there any methods to determine many vacancies in a structure.

[jmmshn]

Sorry this escaped my stack.
This is not really designed for multi-vacancies so it will only find a single vector for the "center"