DOS quantities (from API data) transformation for graphs
This issue has been created to track the changes in the transformation of the DOS data (coming form the API) needed to be drawn on graphs properly. It's something (the changes) that's been happening all the time (different transformations for different types of calculation) and has become a mess. This post will try to prevent that since now.
General case:
-
Energy (Y axis) =
dos_energies
/1.602176565e-19 -
DOS (X axis) =
dos_values
* 1.602176565e-19
Particular cases
Staging based on production 1.11.6
The values for VASP, AIMS, and exciting are O(1e19), thus the general case should apply.
Staging based on prod-032
This is meant as "intermediate" update until the latest production becomes usable.
AIMS
- Values are O(1)
- So this seems to be from the state where we were not sure about the relation to the cell volume
- For now: let's keep them as they are.
VASP
- Values are O(1e-49)
- For investigations see wiki page, outcome: in prod-032 a transformation between A^3 and m^3 has been done, without considering the cell volume as it was at prod-028, thus
- Thus the transformation should be
dos_values
/ (1.602176565e-19 * 1e-30)
exciting
- Values are O(1e-16)
-
🆕 Transformationdos_values
/ (1.602176565e-19 * (27.211^2)) - There is data for only 4 materials, all 2D
Previous notes (most of 2018)
Code AIMS (NOT IMPLEMENTED)
Info from AIMS manual: "The unit of output for the density of states is (eV · Vunit cell)^(−1), i.e., number of states per energy unit (eV) and unit cell volume."
and it seems that the volume is given in A^3
Probably the Parser does not do the normalization to fulfill the metainfo definition
-
Energy (Y axis) =
dos_energies
/1.602176565e-19 -
DOS (X axis) =
dos_values
* cell_volume[A³] * 1e30❓
Code VASP (NOT IMPLEMENTED)
-
Energy (Y axis) =
dos_energies
/1.602176565e-19 -
DOS (X axis) =
dos_values
(1.602176565E-19)^2***