Issue with symmetry and 1/3, 2/3

[paulsaxe]

• molsystem version: 2023.11.5
• Python version: 3.11
• Operating System: MacOS

Description

It appears that there is a problem with symmetry in CIF files with 3-fold symmetry and atoms on the sites with fixed coordinates of 1/3 or 2/3. The CIF files encode these with a truncated decimal value, like 0.3333 and 0.6667. The truncation seems to cause problems with the symmetry operators resulting in duplicate atoms close to each other. The problem seems to be that 2/3 - 1/3 = 0.6667 - 0.3333 = 0.3334 which is not 1/3.

What I Did

Try to use CSD ABANUN ... and got a structure with too many atoms