Catalysis, Meet the Machine: From Models to Meaning
Leandro Goulart de Araujo *
-
IRCELYON, Institut de Recherches sur la Catalyse et l’Environnement de Lyon, UMR5256 CNRS-Université de Lyon, 69626 Villeurbanne, France
* Correspondence: Leandro Goulart de Araujo
Special Issue: Machine Learning in Catalysis: From Data to Discovery
Received: June 25, 2025 | Accepted: June 25, 2025 | Published: June 27, 2025
Catalysis Research 2025, Volume 5, Issue 2, doi:10.21926/cr.2502005
Recommended citation: de Araujo LG. Catalysis, Meet the Machine: From Models to Meaning. Catalysis Research 2025; 5(2): 005; doi:10.21926/cr.2502005.
© 2025 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.
Graphical abstract
Keywords
Artificial intelligence; green chemistry; catalyst synthesis; molecular simulations; reactor design
In recent years, machine learning (ML) has expanded beyond specialized fields in informatics and physical sciences to become a part of our everyday lives — from personalized recommendations on streaming platforms to image generation and video synthesis using simple text prompts on Generative Pre-trained Transformer (GPT) architecture, in platforms like ChatGPT and other artificial intelligence (AI) tools [1]. While the potential of ML in many fields is already well recognized, its application in the natural sciences, including chemistry and physics, is still in a relatively early stage. Encouragingly, the results so far are impressive, and the next decade promises even greater advances.
This is especially true in catalysis, a field marked by complexity at every level — from the atomic structure of active sites to the large-scale design of industrial reactors. Catalysis plays a pivotal role in enabling a sustainable chemical industry, being integral to the production of over 90% of industrial chemicals [2]. Thus, innovations in catalyst design, process optimization, and system modeling are critical to reshaping how we produce fuels, materials, and fine chemicals.
Traditionally, catalytic systems have been studied through mechanistic understanding, empirical exploration, and first-principles computational modeling [3,4,5]. While these approaches have generated profound insights, they often face limitations when dealing with the high dimensionality, nonlinear behavior, and multiscale interactions inherent in real-world systems [6,7]. ML offers a complementary pathway — one that does not require full mechanistic knowledge to make robust predictions. From estimating adsorption energies [8] and reaction barriers [9] to optimizing operating conditions and reactor configurations [10], ML is reshaping catalysis research across scales. The recent surge in publications on ML in catalysis reflects this growing interest (Figure 1).
Figure 1 Number of publications per year combining the keywords “machine learning” AND “catalysis” using data from Scopus. Language: English. Data retrieved on 24 June 2025.
This Focus Issue, Machine Learning in Catalysis: From Data to Discovery, brings together contributions that highlight how ML is accelerating the discovery of new materials and improving catalytic processes. However, the story goes beyond speed; it is also about interpretability, hybridization, and the generation of insights.
As shown in Figure 2, applications may include the use of surrogate ML models trained on DFT calculations for catalyst screening [11], graph-based learning for exploring reaction networks [12], reinforcement learning for process optimization [13], and the increasingly prominent use of physics-informed machine learning (PIML) and physics-informed neural networks (PINNs) [14,15]. These hybrid methods embed scientific laws — such as conservation principles, kinetic expressions, and transport equations — directly into ML architectures. The result is a class of models that can generate reliable predictions while remaining consistent with physical reality.
Figure 2 Schematic illustration of examples of machine learning applications in catalysis.
Throughout this issue, the readers may find diverse examples of how ML is being applied across the entire catalytic value chain — from molecular understanding to process-level integration. Still, several challenges persist: limited datasets, bias in reported results, the scarcity of negative data, and the need for standardized benchmarks. Nevertheless, as the catalysis community increasingly adopts open science practices and interdisciplinary approaches, ML is poised to transition from a promising technique to a core tool in scientific discovery.
If catalysis is the science of transformation, then ML is transforming how that science is practiced. We hope this special issue captures the breadth, depth, and dynamism of this evolving field — and encourages researchers to see not just patterns in data, but pathways to discovery.
Author Contributions
The author did all the research work for this study.
Competing Interests
The author declares no conflict of interest.
References
- Feng Y, Poralla P, Dash S, Li K, Desai V, Qiu M. The impact of ChatGPT on streaming media: A crowdsourced and data-driven analysis using twitter and reddit. Proceedings of the 2023 IEEE 9th Intl Conference on Big Data Security on Cloud (BigDataSecurity), IEEE Intl Conference on High Performance and Smart Computing, (HPSC) and IEEE Intl Conference on Intelligent Data and Security (IDS); 2023 May 6; New York, NY, USA. Piscataway, NJ: IEEE. [CrossRef] [Google scholar]
- Zybert M. Applied catalysis in chemical industry: Synthesis, catalyst design, and evaluation. Catalysts. 2023; 13: 607. [CrossRef] [Google scholar]
- Cheng GJ, Zhong XM, Wu YD, Zhang X. Mechanistic understanding of catalysis by combining mass spectrometry and computation. Chem Commun. 2019; 55: 12749-12764. [CrossRef] [Google scholar] [PubMed]
- Sabbe MK, Reyniers MF, Reuter K. First-principles kinetic modeling in heterogeneous catalysis: An industrial perspective on best-practice, gaps and needs. Catal Sci Technol. 2012; 2: 2010-2024. [CrossRef] [Google scholar]
- de Araujo LG, Oscar Conte L, Violeta Schenone A, Alfano OM, Teixeira AC. Degradation of bisphenol A by the UV/H2O2 process: A kinetic study. Environ Sci Pollut Res. 2020; 27: 7299-7308. [CrossRef] [Google scholar] [PubMed]
- de Araujo LG, Vilcocq L, Fongarland P, Schuurman Y. Recent developments in the use of machine learning in catalysis: A broad perspective with applications in kinetics. Chem Eng J. 2025; 508: 160872. [CrossRef] [Google scholar]
- de Araujo LG, Farrusseng D. Artificial intelligence-driven advances in photocatalytic hydrogen production. New J Chem. 2025; 49: 6888-6913. [CrossRef] [Google scholar]
- Wang Q, Yao Y. Harnessing machine learning for high-entropy alloy catalysis: A focus on adsorption energy prediction. NPJ Comput Mater. 2025; 11: 91. [CrossRef] [Google scholar]
- Singh AR, Rohr BA, Gauthier JA, Nørskov JK. Predicting chemical reaction barriers with a machine learning model. Catal Lett. 2019; 149: 2347-2354. [CrossRef] [Google scholar]
- Wu Z, Tran A, Ren YM, Barnes CS, Chen S, Christofides PD. Model predictive control of phthalic anhydride synthesis in a fixed-bed catalytic reactor via machine learning modeling. Chem Eng Res Des. 2019; 145: 173-183. [CrossRef] [Google scholar]
- Tamtaji M, Chen S, Hu Z, Goddard III WA, Chen G. A surrogate machine learning model for the design of single-atom catalyst on carbon and porphyrin supports towards electrochemistry. J Phys Chem C. 2023; 127: 9992-10000. [CrossRef] [Google scholar]
- Wang B, Gu T, Lu Y, Yang B. Prediction of energies for reaction intermediates and transition states on catalyst surfaces using graph-based machine learning models. Mol Catal. 2020; 498: 111266. [CrossRef] [Google scholar]
- Xu W, Wang Y, Zhang D, Yang Z, Yuan Z, Lin Y, et al. Transparent AI-assisted chemical engineering process: Machine learning modeling and multi-objective optimization for integrating process data and molecular-level reaction mechanisms. J Clean Prod. 2024; 448: 141412. [CrossRef] [Google scholar]
- Wu Z, Wang H, He C, Zhang B, Xu T, Chen Q. The application of physics-informed machine learning in multiphysics modeling in chemical engineering. Ind Eng Chem Res. 2023; 62: 18178-18204. [CrossRef] [Google scholar]
- Ngo SI, Lim YI. Solution and parameter identification of a fixed-bed reactor model for catalytic CO2 methanation using physics-informed neural networks. Catalysts. 2021; 11: 1304. [CrossRef] [Google scholar]
