The Open DAC 2025 Dataset for Sorbent Discovery in Direct Air Capture

2025-08-05

AI Summary

The scarcity of high-performance adsorbents for direct air capture (DAC) under humid conditions hinders scalable deployment. Method: This work introduces ODAC25—the first large-scale, high-fidelity open dataset tailored for DAC—by integrating functionalized metal–organic framework (MOF) design, high-energy grand canonical Monte Carlo (GCMC) conformational sampling, and a synthetic generation framework. It synergistically combines density functional theory (DFT) single-point calculations, GCMC simulations, and machine-learned interatomic potential modeling to enhance both chemical/conformational diversity and accuracy in adsorption thermodynamics prediction. Contribution/Results: ODAC25 comprises nearly 70 million high-quality CO₂ adsorption data points under realistic humid conditions. It also releases state-of-the-art ML interatomic potentials capable of accurately predicting adsorption energies and Henry coefficients. This resource enables high-throughput screening and rational design of next-generation DAC adsorbents, establishing a foundational data and computational infrastructure for accelerated materials discovery.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Data Mining & Knowledge Management: Knowledge Acquisition from the Web
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Identifying useful sorbent materials for direct air capture (DAC) from humid air remains a challenge. We present the Open DAC 2025 (ODAC25) dataset, a significant expansion and improvement upon ODAC23 (Sriram et al., ACS Central Science, 10 (2024) 923), comprising nearly 70 million DFT single-point calculations for CO$_2$, H$_2$O, N$_2$, and O$_2$ adsorption in 15,000 MOFs. ODAC25 introduces chemical and configurational diversity through functionalized MOFs, high-energy GCMC-derived placements, and synthetically generated frameworks. ODAC25 also significantly improves upon the accuracy of DFT calculations and the treatment of flexible MOFs in ODAC23. Along with the dataset, we release new state-of-the-art machine-learned interatomic potentials trained on ODAC25 and evaluate them on adsorption energy and Henry's law coefficient predictions.

Problem

• Identifying effective sorbents for DAC from humid air
• Expanding MOF dataset for CO2, H2O, N2, and O2 adsorption
• Improving DFT accuracy and flexible MOF treatment

Innovation

• Expanded dataset with 70M DFT calculations
• Functionalized MOFs enhance chemical diversity
• Improved accuracy in DFT and flexible MOFs