Electronic Structure · foundry-ml
double_perovskite_bandgap_v1-1
Computed band gaps of 1,306 double perovskite oxide compounds for machine learning screening in optoelectronics.
What double_perovskite_bandgap_v1-1 is
Double perovskite oxides are candidate materials for optoelectronic and photovoltaic technologies, where tuning the electronic band gap is critical for device performance. This dataset provides density functional theory (DFT) calculated band gaps across 1,306 double perovskite oxide materials, published by G. Pilania and colleagues via the Materials Data Facility in 2022.
The dataset enables computational researchers in materials science, physics, and chemistry to train predictive machine learning models. By linking chemical composition and sub-lattice elemental site allocations to calculated band gaps, it accelerates screening candidate double perovskites without requiring computationally demanding ab initio simulations for every new composition.
How it was made
The dataset is structured as tabular records describing 1,306 double perovskite oxide compounds. Input features specify the material composition formula along with the specific chemical elements assigned to the two A sub-lattice sites (a_1 and a_2) and the two B sub-lattice sites (b_1 and b_2).
The target variable is the electronic band gap (gap gllbsc) calculated using density functional theory with the GLLBSC exchange-correlation functional, measured in electronvolts (eV). The dataset is packaged in a single training split intended for tabular regression or classification tasks.
Key facts
- Size
- 1,306 materials
- Material class
- Double perovskite oxides
- Target property
- DFT band gap (GLLBSC) in eV
- Input features
- formula, a_1, a_2, b_1, b_2
- Task categories
- Tabular regression, tabular classification
- Year
- 2022
- DOI
- 10.18126/lss6-o5x4
What you can use it for
- Photovoltaic material screening
- Rapidly evaluate candidate double perovskite oxide compositions to identify formulations with optimal band gaps for solar cell absorbers.
- Band gap regression modelling
- Train supervised regression models to predict continuous band gap values directly from sub-lattice elemental site occupancy.
- Optoelectronic property classification
- Classify compounds into distinct electronic regimes based on whether their calculated band gap falls into targeted optoelectronic ranges.
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/double_perovskite_bandgap_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 1,306 rows of default/train
a_1
string_label- Ba11%
- Sr10%
- Ca9%
a_2
string_label- Ca14%
- Sr13%
- La12%
b_1
string_label- Ta25%
- Zr17%
- Hf16%
b_2
string_label- Nb18%
- Ti18%
- Sn9%
formula
string_text- min
- 8
- median
- 10
- mean
- 9.75
- max
- 10
gap gllbsc
float- min
- 0.11
- median
- 4.07
- mean
- 4.01
- max
- 8.34
Strengths and limits
Good at
- Provides consistently calculated GLLBSC DFT band gaps for 1,306 double perovskite oxide compositions.
- Explicitly resolves elemental occupancy across both A and B sub-lattice sites to facilitate site-specific feature engineering.
- Structured in standard tabular formats suitable for immediate use in machine learning regression and classification pipelines.
Watch out for
- Restricted exclusively to oxide-based double perovskites, omitting halide or other non-oxide perovskite chemistries.
- Includes only a single training split with no pre-defined validation or test sets.
- Provides composition and site assignments without crystallographic structural coordinates or experimental verification.
Questions
Can I look inside double_perovskite_bandgap_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 double_perovskite_bandgap_v1-1 under?
Its authors declare cc-by-4.0. Read the terms on the original page before reusing the data.
What electronic property is targeted in this dataset?
The target property is the DFT-calculated electronic band gap in electronvolts (eV), derived using the GLLBSC functional.
What information is provided for each material?
Each row contains the overall chemical formula, the two elements occupying the A sub-lattice (a_1, a_2), the two elements occupying the B sub-lattice (b_1, b_2), and the calculated band gap.
Does the dataset include separate test splits?
No. The dataset is provided only with a train split, requiring researchers to partition their own cross-validation or testing subsets.
How to cite it
Pilania, G., Mannodi-Kanakkithodi, A., Uberuaga, B. P., Ramprasad, R., Gubernatis, J. E., Lookman, T. (2022). Machine learning bandgaps of double perovskites. Materials Data Facility. https://doi.org/10.18126/lss6-o5x4
Written from double_perovskite_bandgap_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.