Perovskite Properties · foundry-ml
perovskite_opbandcenter_v1-1
DFT-calculated oxygen p-band centre energies across 2,912 perovskite oxides for catalytic property modelling.
What perovskite_opbandcenter_v1-1 is
perovskite_opbandcenter_v1-1 is a tabular materials science dataset compiled by Tianyu Ma, Ryan Jacobs, John Booske, and Dane Morgan, published via Foundry-ML and the Materials Data Facility in 2022. It provides computed oxygen p-band centre values for 2,912 distinct perovskite oxide compositions.
The oxygen p-band centre serves as an electronic descriptor for evaluating catalytic activity, particularly in oxygen-evolution reactions, and aids the study of low work function perovskite materials. By coupling quantum mechanical calculations with chemical compositions, the dataset enables computational researchers to train and evaluate regression and classification models for functional energy materials.
How it was made
The dataset tabulates material compositions alongside density functional theory (DFT) calculations. Each entry records the basic chemical composition as well as a representation with brackets identifying distinct sublattices.
The primary target features are O p-band centre energies reported in electronvolts (eV). These are provided at two levels of theory: standard Generalized Gradient Approximation (DFT-GGA) and Hubbard U-corrected Generalized Gradient Approximation (DFT-GGA+U).
Key facts
- Publisher
- Foundry-ML
- Release year
- 2022
- Dataset size
- 2,912 perovskite oxides
- Data format
- Tabular (Parquet)
- Inputs
- Composition, Composition with sublattices
- Targets
- O pband (GGA)_[eV], O pband (GGA+U)_[eV]
- Supported tasks
- Tabular regression, tabular classification
- DOI
- 10.18126/f3p0-2c5f
What you can use it for
- Oxygen-evolution catalyst screening
- Assessing perovskite oxides to predict catalytic activity descriptors for oxygen-evolution reactions in energy technologies.
- Electronic property regression
- Training machine learning models to predict GGA and GGA+U oxygen p-band centres directly from chemical composition and sublattice arrangements.
- Low work function perovskite exploration
- Screening perovskite candidates to identify compositions suitable for low work function material applications.
Explore the data
The schema, real rows, full-text search and a SQL filter — read in your browser, nothing downloaded.
Explore
Read the real rows without downloading anything
Reading rows…
LiveRead from foundry-ml/perovskite_opbandcenter_v1-1 at the moment you asked. Nothing is cached or stored — every row above came from that request.
How the columns are distributed
Over 2,912 rows of default/train
Composition
string_text- min
- 7
- median
- 12
- mean
- 12.80
- max
- 33
Composition with sublattices
string_text- min
- 13
- median
- 18
- mean
- 18.80
- max
- 39
O pband (GGA)_[eV]
float- min
- -6.33
- median
- -2.98
- mean
- -3.00
- max
- -0.81
O pband (GGA+U)_[eV]
float- min
- -6.33
- median
- -2.22
- mean
- -2.58
- max
- -0.58
Strengths and limits
Good at
- Provides paired DFT calculations using both standard GGA and Hubbard U-corrected GGA+U methods in electronvolts.
- Includes both standard chemical formulas and sublattice-annotated compositions across 2,912 perovskite oxides.
- Supplies ready-to-use electronic descriptors relevant to oxygen-evolution reaction catalysis without requiring direct quantum chemical simulation.
Watch out for
- Contains only a single default training split with no predefined test or evaluation splits.
- Property targets are restricted to calculated O p-band centre energies rather than direct experimental catalytic performance.
Questions
Can I look inside perovskite_opbandcenter_v1-1 without downloading it?
Yes. This page reads its schema, real rows and per-column statistics in the browser, with full-text search and a SQL filter. Nothing is downloaded to your machine.
What licence is perovskite_opbandcenter_v1-1 under?
Its authors declare cc-by-4.0. Read the terms on the original page before reusing the data.
What target variables are available in the dataset?
The dataset provides DFT-calculated oxygen p-band centre values in electronvolts (eV) using both GGA and GGA+U functionals.
How many materials are represented?
The collection contains 2,912 perovskite oxide materials.
What features are available as inputs?
Inputs include the general material composition and a detailed composition string using brackets to indicate distinct sublattices.
What machine learning task formats does this dataset support?
It is formatted for tabular regression and tabular classification workflows.
How to cite it
Ma, Tianyu; Jacobs, Ryan; Booske, John; Morgan, Dane (2022). Discovery and engineering of low work function perovskite materials. Materials Data Facility. https://doi.org/10.18126/f3p0-2c5f
Written from perovskite_opbandcenter_v1-1's own dataset card. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.