Supporting Information: High-Throughput Discovery of Electrides

High-Throughput Discovery of Electrides

This repository hosts part of the results of chapter 5 of the PhD thesis ("Theory and Discovery of Electride Materials) of Chengcheng Xiao at Imperial College London. Specifically, it contains the high-throughput screening results of electride materials performed using computational approaches.

The results are stored in the form of a pickled pandas DataFrame (https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.to_pickle.html) with pickle protocol 3 (https://docs.
python.org/3/library/pickle.html). The results (Electride_Screening.pkl) can be read using the following Python code:

import pandas as pd

mpdf=pd.read_pickle('Electride_Screening.pkl')

The keys used and their meanings are listed here:

- material_id: Identification string from the Materials Project.
- incar: VASP-INCAR file.
- poscar: VASP-POSCAR file.
- potcar: VASP-POTCAR (header only) file.
- kpoints: VASP-KPOINTS file for HTP screening.
- WITH_ELF_MAX: Has local ELF maxima with > 3 neighbors?
- MAX_ELF: The value of the identified ELF maxima.
- ELF_LAPLACE: Laplacian of the ELF at the identified site.
- DIST2ATOM: Distance from identified site to surrounding atoms.
- MAX_OCC: The occupation associated with the identified site.
- ATOM_TYPE: The type of atoms surrounding the identified site.
- NUM_ATOM: The number of atoms surrounding the identified site.
- MAG_SYS: Is the system magnetic?
- INTERSTITIAL_STRUCT: Structure file with interstitial site labeled as Z.
- SPIN_CHANNEL: If is a magnetic system, which spin channel is used?
- WITH_CHG_MAX: Has local charge maxima?
- MAXOCC_NORMALIZED: Occupation normalized to be between 0 and 2.
- Ze: figure of merit [(MAXOCC_NORMALIZED/2+MAX_ELF)/2].