Synthesis, optical imaging, and absorption spectroscopy data for 179072 metal oxides

Area of Science:

• Materials Science
• Spectroscopy
• Data Science

Background:

• Optical absorption spectroscopy is crucial for materials characterization, particularly for solar energy applications.
• Existing materials science datasets lack comprehensive optical property data, hindering predictive modeling.
• Metal oxides are vital materials with tunable optical properties.

Purpose of the Study:

• To describe the largest publicly available curated dataset of metal oxide optical properties (near-infrared to UV-Vis absorbance).
• To provide complete synthesis and processing histories for 179,072 samples from 99,965 unique compositions.
• To facilitate the development of predictive materials models and serve as a benchmark for machine learning integration.

Main Methods:

• Data curation and compilation of optical absorbance, composition, and processing data for metal oxides.
• Dataset includes detailed synthesis and processing parameters for each sample.
• Data covers the spectral range from near-infrared to near-ultraviolet (UV-Vis).

Main Results:

• The dataset comprises 179,072 samples from 99,965 unique metal oxide compositions.
• It represents the most extensive collection to date (Dec 2018) of its kind.
• Provides complete materials science data, including synthesis and optical properties.

Conclusions:

• The dataset will enable the development of predictive models for material optical properties based on composition and processing.
• It serves as a benchmark for integrating machine learning in materials science.
• The resource aids in identifying optimal material compositions and synthesis routes for specific optical properties.