Usage notes

• At different points in the documentation, a “unit cell” is referred to, this is the cell for which the density of states data was calculated, and any volumes, degeneracies and numbers of electrons provided should be consitent with this structure for the reported defect concentrations to be accurate. In most cases, inconsistent specification of these values will lead to incorrect Fermi energies.

• The reported Fermi energy and transition levels reported are referenced to 0 eV; the code expects that the input density of states data is zeroed on the valence band maximum.

• The code operates using the following units, and all user-input must be consistent:

  • energy: electron volts.

  • temperature: Kelvin.

  • volume: Angstroms ³.

• Concentrations are a special case, internally, the py-sc-fermi operates in the concentration of sites in the unit cell which are defective, but will report in cm ^-3. The documentation should always specify what kind of concentration is expected by a particular function, if it does not, this is a bug! Please report it.

• Although in principle the code could be patched to account for temperature-dependent defect formation energy data (if this is something you would be interested in, please let us know!), as this is a rare practice in the field, we currently only calculate defect concentrations with respect to the “0K” formation energies, following the model of `Zhang and Northrup`_https://doi.org/10.1103/PhysRevLett.67.2339