Dear all,

I hope this message finds you well.

I would like to share a recent issue regarding the (100) surface calculation of a vanadium nitride structure using the DMol3 module.

The default cutting plane for the surface had u and v vectors set as u (0, 0.5, -0.5) / v (0, 0.5, 0.5). However, in order to preserve the original shape modelled previously, I modified the vectors to u (0, 1, 0) / v (0, 0, 1).
I will attach images and the xsd file to illustrate the difference between the two surface models.

The problem is that, unlike the structure extended into a supercell using the default vector settings, the structure modified with the new u and v vectors did not undergo successful geometry optimization.

Upon inspection, it seems that the main difference between the two structures is whether atoms are positioned exactly on the lattice points. Could there be any other reason causing such differences?

Best regards,
Rakhoon Choi