Band Gap Estimator

Predict electronic band gaps from crystal structures

Oxide materials - AnBnOn CIF Machine Learning

This tool provides approximate band gap estimates for uploaded crystal structures (CIF format). Key features:

  • Quick prediction of electronic band gaps
  • Developed for oxide materials
  • Requires complete CIF files with formula units
For research-grade accuracy, consider our premium reports.
Try it!
For More Accurate Results

Our premium reports offer detailed band gap analysis using advanced computational methods. Available in the Pricing section.

OxyHopper

Analyze oxygen migration pathways

Oxygen migration Diffusion analysis Void mapping

This tool visualizes and analyzes oxygen ion migration pathways in crystal structures (CIF format). Key features:

  • Determination of migration pathway dimensionality
  • Calculation of void and channel sizes
  • Identification of potential diffusion bottlenecks
For other ions, see BatteryMaterials.
Try it!
For More Accurate Results

Our premium reports offer detailed oxygen migration pathways analysis. Available in the Pricing section.

File Requirements

Band Gap Estimator
  • CIF file must contain complete crystal structure information
  • Must include number of formula units (_cell_formula_units_Z)
  • Best results for AnBnOn structures (training dataset focus)
OxyHopper
  • CIF file must contain complete crystal structure information
  • Must include number of formula units (_cell_formula_units_Z)
  • Supports only oxide materials

Theoretical Validation

Our methods show consistent agreement with experimental studies across various material classes:

Published Validations:
Learn More About Our Methods

For detailed information about our theoretical approaches and software packages, please read DOI: 10.1016/B978-0-12-823144-9.00062-5.

Academic Use

When using results from this service in publications or research, please include a citation reference to this page:

Band Gap Estimator
APA (7th Edition)
LLC "Materials Analyzer". (2025). Materials Analyzer - Band Gap Estimation. https://materials-analyzer.info
MLA (9th Edition)
Materials Analyzer - Band Gap Estimation. materials-analyzer.info. Accessed XX Month 2025.
Chicago Style (Notes)
Materials Analyzer - Band Gap Estimation, https://materials-analyzer.info (accessed Month XX, 2025)
Russian ГОСТ style
Анализ Материалов. Предсказание ширины запрещенной зоны [Электронный ресурс]. — Режим доступа: https://materials-analyzer.info (дата обращения: dd.mm.yyyy).
OxyHopper
APA (7th Edition)
LLC "Materials Analyzer". (2025). Materials Analyzer - OxyHopper. https://materials-analyzer.info
MLA (9th Edition)
Materials Analyzer - OxyHopper. materials-analyzer.info. Accessed XX Month 2025.
Chicago Style (Notes)
Materials Analyzer - OxyHopper, https://materials-analyzer.info (accessed Month XX, 2025)
Russian ГОСТ style
Анализ Материалов. Анализ миграций кислорода [Электронный ресурс]. — Режим доступа: https://materials-analyzer.info (дата обращения: dd.mm.yyyy).

We appreciate your support in acknowledging our service.

Dataset used in Band Gap prediction.

Number of materials of each major computational materials databases used to create this dataset.

About Our Machine Learning Model

The prediction model was specifically trained on AnBnOn structures. Results for other material classes may not be accurately predicted.

About the Databases
AFLOW

A high-throughput framework for computational materials science that automates calculations and provides a comprehensive database of material properties.

Richard H. Taylor, Frisco Rose, Cormac Toher, Ohad Levy, Kesong Yang, Marco Buongiorno Nardelli, Stefano Curtarolo. A RESTful API for exchanging materials data in the AFLOWLIB.org consortium, Computational Materials Science, Volume 93, 2014, Pages 178-192, ISSN 0927-0256
Materials Project

An open database of computed materials properties designed to accelerate materials innovation by providing researchers with powerful tools for materials design.

Horton, M.K., Huck, P., Yang, R.X. et al. Accelerated data-driven materials science with the Materials Project. Nat. Mater. (2025)
OQMD

The Open Quantum Materials Database contains DFT calculated thermodynamic and structural properties for hundreds of thousands of materials.

Saal, J. E., Kirklin, S., Aykol, M., Meredig, B., and Wolverton, C. "Materials Design and Discovery with High-Throughput Density Functional Theory: The Open Quantum Materials Database (OQMD)", JOM 65, 1501-1509 (2013)

ML models quality metrics on training data (90% of all structures) from the dataset.

Band Gap Estimator for Classification
Accuracy
98.219%

Proportion of correctly predicted instances

Precision
98.251%

Ratio of true positives to all predicted positives

Recall
97.709%

Ratio of true positives to all actual positives

F1-Score
97.979%

Harmonic mean of precision and recall

Band Gap Estimator for Regression
R² Score
0.9507

Proportion of variance explained by the model

RMSE
0.2695

Root Mean Square Error - standard deviation of residuals

MSE
0.0726

Mean Square Error - average squared difference

Model Performance Summary
Classification Model

The Band Gap Estimator for Classification demonstrates excellent performance across all metrics with accuracy at 98.219%. The high precision (98.251%) and recall (97.709%) values indicate a well-balanced model with strong predictive capabilities.

Regression Model

The Band Gap Estimator for Regression shows strong performance with an R² value of 0.9507, indicating that the model explains approximately 95% of the variance in the target variable. The low RMSE (0.2695) and MSE (0.0726) values confirm the model's accuracy in predictions.

Both models are performing exceptionally well.

ML models quality metrics on testing data (10% of all structures) from the dataset.

Band Gap Estimator for Classification
Accuracy
88.579%

Proportion of correctly predicted instances

Precision
86.569%

Ratio of true positives to all predicted positives

Recall
83.794%

Ratio of true positives to all actual positives

F1-Score
85.681%

Harmonic mean of precision and recall

Band Gap Estimator for Regression
R² Score
0.9026

Proportion of variance explained by the model

RMSE
0.37947

Root Mean Square Error - standard deviation of residuals

MSE
0.14724

Mean Square Error - average squared difference

Model Performance Summary
Classification Model

The Band Gap Estimator for Classification demonstrates great performance across all metrics with accuracy at 88.579%. The high precision (86.569%) and recall (83.794%) values indicate a well-balanced model with high predictive capabilities.

Regression Model

The Band Gap Estimator for Regression shows strong performance with an R² value of 0.9026, indicating that the model explains approximately 90% of the variance in the target variable. The low RMSE (0.37947) and MSE (0.014724) values confirm the model's accuracy in predictions.

Both models are performing well.