Benchmark Dataset for Catalysis on 2D MXenes

2026-05-30

AI Summary

This work addresses the high computational cost of conventional density functional theory (DFT) calculations, which hinders efficient investigation of catalytic behavior and structural evolution of two-dimensional MXene materials under realistic conditions. By integrating first-principles calculations with machine learning, the study constructs a large-scale benchmark dataset comprising tens of thousands of DFT data points, providing the first systematic data foundation for catalytic studies of Ti₂CTᵧ MXenes. Advanced interatomic potential models—including EquiformerV2, MACE, MatRIS, and UPET—are trained on this dataset, achieving acceleration factors of 10³–4×10³ on CPU while maintaining high accuracy (force errors ≈ ±10 meV/Å, energy errors ≈ ±1 meV/atom). The models’ generalization capabilities are further validated on larger-system test sets, enabling efficient and high-throughput exploration of MXene catalytic properties.

Abstract

Merging first-principles calculations with machine learning (ML), we aim to accelerate the exploration of catalytic behaviour in novel materials. We focus on two-dimensional (2D) Ti$_2$CT$_y$ MXenes, whose versatile surface chemistry makes them particularly compelling candidates for catalysis. Resolving their composition and structure under realistic conditions exceeds the reach of standard density functional theory (DFT) due to computational cost. To address this challenge, we generate a comprehensive dataset of 50,000 DFT calculations for training and 10,000 for testing, encompassing both Ti$_2$CT$_y$ MXene configurations and molecular systems, along with an additional test dataset with 1000 genuinely new, larger systems to investigate how well models generalise. We train and validate widely used and competitive machine learning interatomic potential (MLIP) models, including EquiformerV2, MACE, MatRIS, and UPET, that accurately predict atomic forces and formation energies -- quantities that DFT must repeatedly compute for structural and catalytic investigations -- for these 2D materials. This combined DFT-ML framework achieves computational acceleration on the order of approximately $1-4 \cdot 10^3$ (on a CPU) while maintaining desired-level accuracy (approximately +/- $10$ meV/A for forces and approximately +/- $1$ meV for per-atom energies), paving the way for more efficient investigations of MXene catalytic behaviour. Moreover, we perform an extensive qualitative evaluation of the trained models, showcasing the importance of comprehensive simulation-based comparison beyond benchmark metrics. The dataset and the trained models with the code are available at https://huggingface.co/datasets/CatalystAnonymous/catalyst_mxenes.

Problem

• MXenes
• catalysis
• density functional theory
• computational cost
• 2D materials

Innovation

• machine learning interatomic potentials
• MXenes
• catalysis
• DFT acceleration
• benchmark dataset