| 2:32:06 PM PST - Fri, Dec 9th 2016 |
|
Dear All,
I am doint dft calculation of a system with degenerate HOMO-LUMO, They are supposed to be degenerate, because of symmetry of ionic structure. I am aware that enforcing symmetry by "autosym" may cause some problems in evaluating matrix element. So, I tried two approaches to fix this issue, but they end up with different results. So, I would like to some help to have correct calculations.
The first approach is not to use symmetry by 'noautosym'. Calculations go without warings, but in the end result HOMO-LUMO are no longer degenerate, which is a concern for me.
The second approach is to use open shell wave function by 'odft' and symmetry by 'autosym'. Calculations come with waring "Warning - the HOMO and LUMO are degenerate and you are using symmetry..." and "Warning: spatial symmetry breaking in UKS" In the end results, HOMO LUMO are degenerate, which is good. But the total energy is different from non-symmetry calculations, as described in the first approach.
So, which one should I trust? I guess non-symmetric calculations are usually more reliable, but since HOMO-LUMO are non-degenerate in my case, I don't know what to believe any more... Thank you for your help.
Cheers
Jia
|