Hydrogen (H2) is expected to play a critical role in modern industry, particularly in ammonia synthesis, petroleum refining, and low-carbon transportation. The safe storage of H2 remains a major challenge due to its low volumetric density under ambient conditions. Metal–Organic Frameworks (MOFs), highly porous crystalline materials, have emerged as promising H2 storage candidates owing to their high surface areas and tuneable pore structures. Molecular simulations such as grand canonical Monte Carlo or density functional theory are costly and limited in exploring large material spaces, motivating efficient predictive tools to accelerate discovery. Here, Machine Learning (ML) techniques are compared to an explainable artificial intelligence (XAI) approach using symbolic regression (SR), trained on 10,123 experimentally measured H2 adsorption datapoints from real-world MOFs. The best performing model achieved a goodness of fit of 0.9986 with lower computational demand, but reduced interpretability, addressed using XAI analysis and clustering. SR achieves a lower goodness of fit of 0.914 but produces a physically meaningful equation highlighting structural features driving high gravimetric efficiencies. These results demonstrate strong ML capability for predicting how MOF properties and environmental conditions affect H2 uptake. This offers engineers and researchers a practical means of screening potential MOFs for H2 storage applications, with the XAI analyses providing additional confidence in the predictions. They allow researchers to understand the physical reasoning behind each output, assess the reliability of individual predictions, and make fully informed decisions, enabling predictive models to be acted upon with confidence in real-world contexts.
Two‐Stage Material Screening (TSMS) is developed, an AI‐driven framework that integrates density functional theory (DFT) computations, an active‐learning‐guided experimental feedback loop, and mechanistic interpretation to enable rapid discovery and systematic evaluation of promising electrocatalysts.
Xueyu Hu, Yucun Zhou, Haoyu Li et al.· Advances in Materials· 0 citations
A comprehensive account of ML applications in NRR, covering curated experimental databases, feature engineering based on atomic, structural, and DFT‐derived descriptors, and ML‐guided insights into single‐atom, dual‐atom, alloy, oxide, nitride, and defect‐engineered catalysts are presented.
Increasing global CO2 emissions are driving efforts to develop advanced carbon capture materials. Metal-organic frameworks (MOFs) show great promise as CO2 adsorbents, yet maintaining performance under humid flue gas conditions remains a major challenge. Herein, we present an integrated computational high-throughput screening workflow that begins with a library of over 110,000 experimental MOFs or MOF-like structures, explicitly includes the effects of water in adsorption simulations, and incorporates machine learning-based stability analysis to identify promising candidates from among existing MOFs. Guided by this workflow, we synthesized a top-performing MOF that demonstrates exceptional tolerance to humid conditions, maintaining high CO2 uptake at elevated relative humidity. We further revealed key structure-property relationships and structural motifs that offer valuable design principles for next-generation MOFs for CO2 capture in humid conditions.
Jiayang Liu, Xiaoliang Wang, Xiyang Liu et al.· Journal of the American Chem...· 0 citations
Ionic liquids (ILs) represent one of the most promising and versatile solvent families for CO2 capture due to their structural tunability and low volatility. However, the sheer scale and complexity of the IL chemical design space make the discovery of high-performance candidates extraordinarily challenging. In this work, we introduce a SHAP-guided Domain Filter (SGDF) framework that leverages machine learning (ML) to address this challenge, enabling reliable large-scale screening across millions of potential IL structures. Starting from over 32 million cation–anion combinations, sequential SGDF gates based on melting point (T m), toxicity, viscosity, and CO2 solubility reduce the search space by more than 2 orders of magnitude, ultimately yielding 69,009 promising candidates. Experimental validation is performed on 12 ILs, including two model-selected candidates that exhibit the desired properties. Notably, the two model-selected ILs exhibit the highest CO2 absorption capacities, surpassing all other samples. Across the 12 ILs evaluated, the framework achieves high predictive accuracy (RMSE = 0.046, MAE = 0.030, R 2 = 0.781), showing excellent agreement between predicted and experimental results. These findings demonstrate that combining predictive modeling with experimental validation offers both mechanistic insight and a scalable, data-driven pathway for designing next-generation sustainable ILs for CO2 capture.
Yushan Chen, Yongsheng Chen· Journal of Physical Chemistr...· 0 citations
This work presents a machine learning-symbolic regression (ML-SR) strategy to develop a physically interpretable formula for predicting low pressure CO2 adsorption capacity in hypothetical metal-organic frameworks (hMOFs). Four ML models were trained on a small dataset of 1,000 samples, and five key descriptors-largest cavity diameter, pore limiting diameter, void fraction, gravimetric surface area, and number of hydrogen atoms-were identified through SHAP and feature importance analyses. Symbolic regression was then employed to derive a concise adsorption formula, Q=aA, where a represents an adsorption baseline (mmol/g) and A is a dimensionless adsorption number incorporating four structural descriptors. We interpret A as the ratio between an adsorption binding force and a diffusion driving force, revealing how pore topology and surface chemistry jointly influence adsorption. Validation against a comprehensive dataset of 137,652 hMOFs demonstrates that this formula achieves over 70% prediction accuracy for 62,448 structures, confirming strong applicability within defined structural and operational ranges. Unlike conventional black box ML models, the proposed physics-guided expression enables efficient prediction and provides clearer insight into adsorption mechanisms.
Yimin Shao, Sheng-Ling Ma, Shenghong Ju et al.· 0 citations
Metal organic frameworks (MOFs) have emerged as promising electrode materials for supercapacitor (SC) due to their high surface areas, tunable porosity, and redox active sites. However, the vast chemical space of MOFs leads to millions of possible structures, makes experimental trial and error discovery inefficient. This review provides a focused perspective on how density functional theory (DFT) and machine learning (ML) are enabling the accelerated discovery and rational design of MOF-based SC electrodes. Key insights from DFT are discussed in relation to three critical performance descriptors as electrical conductivity, electrochemical and structural stability, and redox activity. In parallel, recent advances in ML-driven screening are reviewed, covering the development and use of large-scale MOF databases, descriptor engineering strategies, and predictive model architectures. Case studies demonstrating the successful integration of DFT and ML for identifying high-performance MOFs are highlighted in the review. Finally, the current limitations are analysed, including the discrepancy between idealized computational models and real polycrystalline electrodes, intrinsic trade-offs between conductivity and stability, and the need for interpretable and physics-informed ML models. Overall, this review outlines a computational roadmap for the rapid discovery and optimization of next-generation MOF-based SC electrodes. This is the comprehensive review on ML-driven prediction of electrochemical performance in MOF based electrode for SC applications. It integrates DFT-calculated electronic descriptors with ML models to discover hidden structure-property relationships. It identifies critical data gaps, model transferability issues, lack of dynamic ion-transport modelling in current studies. It proposed a multi-fidelity active learning framework combining DFT, ML and experiments for accelerated MOF discovery. It outlines standardized database protocols and explainable AI strategies to guide future high-performance MOF design.
Achal Siddharth Fulmali, H. Panda· Journal of Materials Science...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
