My Community

Recently, we calculated the electronic structure of Cu2CdSnS4 using VASP 5.2 with the hybrid functional. However, an irregular occupation of electron was found at the Gamma point (see attached OUTCAR file for detail):

58 1.9237 1.00000
59 1.9237 1.00000
60 3.5799 0.86695
61 2.5755 1.00000
62 2.5755 1.00000
63 2.6523 0.13311
64 5.0434 0.00000
65 5.6845 0.00000

1. Why the 60th level is not listed in order?

2. Why the occupation number of 63th level is lower than that of 60th level, though the 63th level is lower than 60th level in energy.

When the PBE functional (not hybrid) is employed, this irregular
occupation disappear.

Any hints on the issue?

A hybrid-functional calculation uses original indexing of energy levels. The reordering is achieved via diagonalization (LDIAG=.TRUE.).