Change Log


  • Update Defect, DefectEntry and DefectThermodynamics properties/methods to be even more efficient with calculations of formation energies and concentrations. Gives ~10x speedup in Fermi solving and concentration calculations (e.g. from 2 hours to 12 minutes for 2D chempot vs temp CdTe grid in thermodynamics tutorial).

  • Avoid unnecessary DeprecationWarning``s from latest ``spglib release.


  • Enforce shakenbreak>=2.3.4 requirement.


  • Make oxidation state guessing more efficient, semi-significant speed up in generation/parsing for tough cases.

  • Add bulk_site_concentration property to DefectEntry, giving the concentration of the corresponding lattice site of that defect in the pristine bulk.

  • Minor updates to ensure compatibility with recent pymatgen and ASE releases.


  • Remove spglib<=2.0.2 dependency (set to avoid unnecessary warnings), and update installation instructions accordingly.


  • Allow cases where the calculated host material is unstable wrt competing phases (above the hull), by downshifting to the hull and warning the user about this.

  • General updates to chemical potentials code; more robust (better error catches and messages, API key handling), more informative, cleaner outputs.

  • Updates to match recent changes in pymatgen object types (e.g.

  • Minor efficiency & robustness updates


  • Speed up eigenvalue parsing by using the faster doped site-matching functions rather than MakeDefectStructureInfo from pydefect

  • Minor efficiency & robustness updates.

  • Minor docs & tutorials updates

  • Minor tests updates


  • Electronic structure analysis by @adair-nicolson & @kavanase:
    • Adds DefectEntry.get_eigenvalue_analysis() method to plot single-particle eigenvalues and analyse orbital character and localisation; usage and examples shown on the docs Tips page and the advanced analysis tutorial.

    • Projected eigenvalues can be parsed from vasprun.xml(.gz) files (preferred, as more accurate with 4 decimal places; c.f. 3 in PROCAR(.gz); more convenient and only ~5% slower) or PROCAR(.gz) files in calculation directories (both with significantly expedited parsing compared to pymatgen methods). Compatible with spin-polarised, unpolarised and SOC calculations. Comes with update by @kavanase to easyunfold PROCAR.gz parsing.

  • More efficient defect calculation parsing

  • Add get_magnetization_from_vasprun and get_nelect_from_vasprun functions to doped.utils.parsing, as these attributes are not available from

  • Improve testing efficiency


  • General robustness updates:
    • Updated file parsing to avoid hidden files.

    • Sanity check in DefectsGenerator if input symmetry is P1.

    • Add NKRED to INCAR mismatch tests.

    • Re-parse config & spin degeneracies in concentration/symmetry functions if data not already present (if user is porting DefectEntry`s from older `doped versions or manually).

    • Avoid unnecessary `DeprecationWarning`s

  • Updated docs and linting


  • Update to match breaking change in pymatgen==2024.3.1 (released today), handling incar_params.


  • Refactor (phase diagram) facet to (chemical potential) limit in doped chemical potential functions, as this is more intuitive for most users.

  • Tests updates.

  • Minor efficiency/verbosity/robustness/docs improvements.

  • Update default KPOINTS for convergence/production runs in chemical_potentials based on testing.

  • Add optional projections of site displacements upon given vectors by @ireaml


  • DefectsThermodynamics class has been added to replace and greatly expand the functionality of the DefectPhaseDiagram object. See tutorials for functionality and usage (plotting, Fermi level / concentration analysis, dopability, transition levels (with/out metastable etc).

  • Overhaul supercell generation as discussed, now optimises directly off minimum periodic image distance (thanks to efficient optimisation algorithm) with some prudent constraints. Significantly reduces supercell sizes required in most cases.

  • Overhaul defect grouping as discussed, to use the distance between equivalent defect sites (with this controllable via the dist_tol parameter).

  • Add point symmetry and orientational/spin degeneracy parsing, automatically included in thermodynamics analysis (and customisable by user).

  • Many efficiency improvements (particularly in defect & input file generation, and symmetry functions).

  • Check and warning for large defect displacements far from defect site.

  • Site displacement (local strain) plotting by @ireaml 🙌

  • Auto determination of X-poor/rich facets.

  • More control over site selection for eFNV correction.

  • Clean, grouped parsing warnings for DefectsParser (in case many warnings…)

  • __repr__ methods for all doped classes for informative outputs.

  • Tests and tutorials updates.


  • Added DefectsParser class for parsing defect calculations:
    • Uses multiprocessing and shared bulk data to massively speed up parsing of many defect supercell calcs at once (e.g. from 17 min to < 3 mins for 54 defects in CdTe).

    • Automatically checks INCAR, KPOINTS, POTCAR and charge correction compatibility between all calculations, and warns the user if any are likely to affect formation energies.

  • Make csv input to CompetingPhasesAnalyzer more flexible, along with other code and docstrings updates.

  • Format point group symbol in formation energy plots.

  • Refactor elt/elt_refs to el/el_refs by @adair-nicolson

  • Charge states can now be automatically determined even when POTCARs are not setup by the user.

Updates reflected in the doped parsing tutorial.


  • Update finite-size defect corrections implementations:
    • pydefect used directly for eFNV correction (with optimisation for efficiency). Moreover, the fully relaxed defect structure (with defect site determined by doped) is used.

    • FNV correction now uses optimised version of pymatgen-analysis-defects implementation.

    • Updated corrections plotting (much nicer formats, more informative etc)

    • The actual energy error in the correction is now estimated, and the user is warned if this exceeds error_tolerance (optional parameter, 0.05 eV by default)

    • Bandfilling corrections no longer automatically calculated as (1) almost always not recommended and (2) will show an example of calculating these if needed using our code in pymatgen on the docs

  • Efficiency improvements in obtaining defect site info (Wyckoff positions)

  • Additional utils and functions for defect generation and manipulation.

  • (Many) updated tests.

  • Added functionality for robustly determining the point group symmetry of _relaxed_ defects 🔥


  • Update oxi-state handling to:
    • Use pre-assigned oxi states if present

    • Handle pymatgen oxi-state guessing failures (non-integer oxi states, inaccurate oxi states with max_sites, failures for extremely large systems etc)

  • Update default probability_threshold from 0.01 to 0.0075.

  • Account for rare possibility of user being on a non UTF-8 system.

  • Italicise “V” for vacancy in plotting.

  • SMTG-UCL -> SMTG-Bham

  • Tests and formatting updates.


  • Add supercell re-ordering tests for parsing

  • Ensure final _relaxed_ defect site (for interstitials and substitutions) is used for finite-size charge corrections

  • Consolidate functions and input sets with ShakeNBreak

  • Update defect generation tests

  • Use more efficient Wyckoff determination code


  • Sort defect entries in DefectPhaseDiagram for deterministic behaviour (particularly for plotting).

  • Tests updates (archive test plots, update extrinsic generation tests etc).

  • Avoid long stacklevel issue which cropped up in python3.8 tests for SnB

  • Update PDF figure savefig settings, and add _get_backend function.


  • Refactor _check_user_potcars() to DefectDictSet rather than DefectRelaxSet, and add write_input () method (which runs _check_user_potcars() first).

  • Update defect generation tests

  • Add troubleshooting docs page and update tips docs page


  • Update naming handling in DefectPhaseDiagram to be more robust/flexible, following failure case noted by @utf 🙌

  • Ensure package data files are correctly included in the package distribution, again noted by @utf 🙌

  • Updates to chemical potentials code.

  • Refactoring of site-matching code.

  • Tests updates and code cleanup.


  • Major overhaul to rebase onto the new pymatgen defects code (>v2022.7.25).

  • Add documentation (

  • Add DefectsGenerator class with major upgrade in functionality.

  • Add DefectsSet classes in


  • Cap numpy to 1.23 to avoid pymatgen dependency issues.

  • Update example workbook to use recommended CubicSupercellTransformation

  • Add/remove some TODOs


  • doped now installable from conda-forge! 🎉

  • Major overhaul of primary parsing workflow (in defect_entry_from_paths()):
    • Automatic charge-state determination (throwing warning when user specification doesn’t match auto-determined)

    • Automatic charge correction determination and application

    • Improved error handling and more informative warning messages

  • Add, and -> significantly improve test coverage

  • Add _convert_dielectric_to_tensor() function to be more flexible to user input

  • Remove old unsupported/deprecated code.

  • Add check and warning if multiple output files (vasprun.xml/OUTCAR/LOCPOT) present in bulk/defect directory.

  • Minor bug fixes, formatting, docstrings improvement, the usual

  • Add and remove TODOs


  • Start keeping a CHANGELOG

  • README updates to give step-by-step instructions on setting up MP API key, POTCARs for pymatgen and virtual conda environments for doped and ShakeNBreak

  • Major overhaul of vasp_input functions setup to be far more streamlined and customisable.

  • Major overhaul of chemical_potentials code; now with improved algorithm for selecting potential competing phases

  • Update of example notebooks

  • Add tests for parsing calculations, chemical_potentials and vasp_input

  • Add GH Actions workflows (for tests, GH releases and pypi packaging)

  • Adopt recommended versioning convention based on dates

  • General tidy up, docstring padding, formatting and TODO addition/removal

  • Ensure all inputs/outputs are JSONable, now recommending this for better forward/backward compatibility

  • Refactor dope_stuff to plotting and analysis to be more clear and PROfessional, yo

  • Refactor from hard-coded defaults / slightly-less-human-readable json files to yaml files with default settings.

  • Refactor defectsmaker output, more efficient, cleaner and informative