This draft contains a metainfo schema of the Perovskite Database with ELN annotations.
The schema contains several normalizers that populate quantities in
Results. It also contains extra tools like:
formula_converte.py which converts formulas with soma abbreviations like
MA to the chemical representation of the cations: CH3NH3.
eqe_importer.py which reads exemplary eqe files and does some parsing, and derivation of extra quantities.
jv_reader.py is a very specific reader of J-V data of a lab for demonstration purposes which is used to populate quantities in the schema.
entry_writer.py is a tool that was used to convert the entries in the .csv of the original database to archive.json type entries.
Additionally, an Optoelectronic subsection has been created in
Results that also contains a more specific SolarCell generalized subsection. The goal of this subsection is to accommodate any solar cell data in the future regardless of the technology.
At the moment, I have doubts about whether or not we should display the PerovskiteSolarCell schema in ELN mode or not. The reason for it is that as we replicated the original schema of the database, some quantities are not ideal, e.g. many numeric quantities being
str type because the database for this field contained numbers separated by
'|' and so on.
An alternative would be to disable the ELN annotations in this schema and create solar cell base classes with ELN annotations that could be re-used in future versatile solar cell-based schemas.