[Pw_forum] Significant figures of band energies

[EG Kim]

Dear Colleagues,

Would there be an input parameter in version 3.2.3 that controls the number of
significant figures of printed band energies?

Although the default number of decimal points (3 in the band structure output
and 4 in the standard output) is sufficient for plotting and a qualitative study
of the band shape, I oftentimes find it not enough.  When one calculates, using
a finite difference method, the effective mass (or the inverse of the curvature)
at the band edges in systems like organic semiconductors, for which bands are
quite flat, it becomes crucial to have better resolved energies for smallest
possible dk.  Things are further complicated by the fact that the dimensions of
the Brillouin zone are quite small (or the crystal unit cell is quite large in
general), imposing an upper limit on how large one can go for dk.

Regards,

EG Kim

--
+ -
| E.-G. Kim, Ph.D.
| School of Chemistry and Biochemistry
| 901 Atlantic Drive NW
| Georgia Institute of Technology
| Atlanta, GA 30332-0001
| U.S.A.
+ -
| +1.215.327.7315
| egk at gatech.edu
+ -