QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: sayantanu on June 22, 2016, 12:17

Title: Electron Count Discrepancy
Post by: sayantanu on June 22, 2016, 12:17

Dear ATK Users,

I'm trying to calculate the molecular energy spectrum of a periodic system with 16 atoms in the unit cell. I have taken different geometries having different cell parameters, all having the same number of atoms in the unit cell. But the output electron counts are different for different geometries. Why is this happening?

Title: Re: Electron Count Discrepancy
Post by: zh on June 22, 2016, 13:43

Please show the output and the corresponding input files.

Title: Re: Electron Count Discrepancy
Post by: sayantanu on June 24, 2016, 08:25

Here are the input and output files for two different optimized geometries. For cell1 the number of electrons is correct considering the valence electrons of 16 oxygen atoms. But in cell2 it is 94.

Title: Re: Electron Count Discrepancy
Post by: zh on June 24, 2016, 10:24

It is meaningless to calculate the molecular energy spectrum for a crystalline structure. The constructed configuration of Oxygen atoms is not a molecular system.   

If the property analysis of molecular energy spectrum is applied to the bulk configuration, only one k point (may be the gamma point) is used to sample the k-points in the Brillouin zone for counting the number of electrons. This will cause a big discrepancy just as what was seen in his results. So the problem you met is the normal case for the improper calculations.

Title: Re: Electron Count Discrepancy
Post by: sayantanu on June 24, 2016, 14:13

Thanks a lot for pointing out the mistake.