Theoretical Understanding of Dynamic Catalysis DOI

Pankaj Jangid,

Srabanti Chaudhury, Anatoly B. Kolomeisky

et al.

The Journal of Physical Chemistry C, Journal Year: 2024, Volume and Issue: 128(22), P. 9077 - 9089

Published: May 23, 2024

Dynamic catalysis is a phenomenon in which the catalytic properties of system change with time. It has been recently proposed as an alternative to current widely utilized static approach because potential significant improvements efficiency. restructuring active sites on surfaces also observed some nanocatalysts. However, microscopic mechanisms underlying processes remain not well understood. We developed new stochastic framework that allows us quantitatively describe dynamic and compare its approach. found fluctuations between different pathways might lead enhancements chemical reaction rates but only for specific ranges kinetic parameters. Our theoretical method can explain these observations from point view. show temporal efficiency depends reactions transitions while being independent number sites. argued effects are purely nonequilibrium, associated energy dissipation source In addition, stochasticity investigated. The clarifies important aspects processes.

Language: Английский

Square-pyramidal subsurface oxygen [Ag4OAg] drives selective ethene epoxidation on silver DOI
Dongxiao Chen, Lin Chen, Qiancheng Zhao

et al.

Nature Catalysis, Journal Year: 2024, Volume and Issue: 7(5), P. 536 - 545

Published: March 25, 2024

Language: Английский

Citations

25
From Characterization to Discovery: Artificial Intelligence, Machine Learning and High-Throughput Experiments for Heterogeneous Catalyst Design DOI
Jorge Benavides-Hernández, Franck Dumeignil

ACS Catalysis, Journal Year: 2024, Volume and Issue: 14(15), P. 11749 - 11779

Published: July 24, 2024

This review paper delves into synergistic integration of artificial intelligence (AI) and machine learning (ML) with high-throughput experimentation (HTE) in the field heterogeneous catalysis, presenting a broad spectrum contemporary methodologies innovations. We methodically segmented text three core areas: catalyst characterization, data-driven exploitation, discovery. In characterization part, we outline current prospective techniques used for HTE how AI-driven strategies can streamline or automate their analysis. The exploitation part is divided themes, strategies, that offer flexibility either modular application creation customized solutions. exploration present applications enable areas outside experimentally tested chemical space, incorporating section on computational methods identifying new prospects. concludes by addressing limitations within suggesting possible avenues future research.

Language: Английский

Citations

24
The future of computational catalysis DOI Creative Commons
Joachim Sauer

Journal of Catalysis, Journal Year: 2024, Volume and Issue: 433, P. 115482 - 115482

Published: April 8, 2024

The future of computational heterogeneous catalysis is shaped by machine learning in two different but equally important areas: (i) development atomistic potentials that closely approximate DFT and wavefunction based ab initio methods (MP2, CCSD(T)), are computationally more efficient, (ii) finding structure reactivity descriptors for predicting catalytic materials reactions. Machine Learning Potentials will enable improved sampling the potential energy surface (PES) reaction conditions, they not do better than data on which trained. Therefore, this perspective focusses ways improving quality PES beyond generalized gradient approximation (GGA) climbing Jacob's ladder functionals up to RPA using such as MP2 CCSD(T). It problem solving close collaboration with experiment has made methodology relevant remains an aspect science catalysis.

Language: Английский

Citations

14
Catalysis in the digital age: Unlocking the power of data with machine learning DOI Creative Commons
B. Moses Abraham, M. V. Jyothirmai, Priyanka Sinha

et al.

Wiley Interdisciplinary Reviews Computational Molecular Science, Journal Year: 2024, Volume and Issue: 14(5)

Published: Sept. 1, 2024

Abstract The design and discovery of new improved catalysts are driving forces for accelerating scientific technological innovations in the fields energy conversion, environmental remediation, chemical industry. Recently, use machine learning (ML) combination with experimental and/or theoretical data has emerged as a powerful tool identifying optimal various applications. This review focuses on how ML algorithms can be used computational catalysis materials science to gain deeper understanding relationships between properties their stability, activity, selectivity. development repositories, mining techniques, tools that navigate structural optimization problems highlighted, leading highly efficient sustainable future. Several data‐driven models commonly research diverse applications reaction prediction discussed. key challenges limitations using presented, which arise from catalyst's intrinsic complex nature. Finally, we conclude by summarizing potential future directions area ML‐guided catalyst development. article is categorized under: Structure Mechanism > Reaction Mechanisms Catalysis Data Science Artificial Intelligence/Machine Learning Electronic Theory Density Functional

Language: Английский

Citations

9
Recent developments in the use of machine learning in catalysis: A broad perspective with applications in kinetics DOI Creative Commons
Leandro Goulart de Araujo, Léa Vilcocq, Pascal Fongarland

et al.

Chemical Engineering Journal, Journal Year: 2025, Volume and Issue: unknown, P. 160872 - 160872

Published: Feb. 1, 2025

Language: Английский

Citations

1
Machine Learning Potential for Copper Hydride Clusters: A Neutron Diffraction-Independent Approach for Locating Hydrogen Positions DOI

Cong Fang,

Zhuang Wang, Ruili Guo

et al.

Journal of the American Chemical Society, Journal Year: 2025, Volume and Issue: unknown

Published: March 15, 2025

Determining hydrogen positions in metal hydride clusters remains a formidable challenge, which relies heavily on unaffordable neutron diffraction. While machine learning has shown promise, only one deep learning-based method been proposed so far, diffraction data for training, limiting its general applicability. In this work, we present an innovative strategy─SSW-NN (stochastic surface walking with neural network)─a robust, non-neutron diffraction-dependent technique that accurately predicts positions. Validated against copper clusters, SSW-NN proved effective where X-ray or DFT predictions are available. It offers superior accuracy, efficiency, and versatility across different hydrides, including silver alloy systems, currently without any references. This approach not establishes new research paradigm but also provides universal solution localization other fields constrained by sources.

Language: Английский

Citations

1
Gas-phase errors in computational electrocatalysis: a review DOI Creative Commons
Ricardo Urrego‐Ortiz, Santiago Builes, Francesc Illas

et al.

EES Catalysis, Journal Year: 2023, Volume and Issue: 2(1), P. 157 - 179

Published: Sept. 29, 2023

In this review we show how DFT gas-phase errors affect computational models of electrocatalytic reactions.

Language: Английский

Citations

22
Artificial Intelligence (AI) Workflow for Catalyst Design and Optimization DOI

Nung Siong Lai,

Yi Shen Tew,

Xialin Zhong

et al.

Industrial & Engineering Chemistry Research, Journal Year: 2023, Volume and Issue: 62(43), P. 17835 - 17848

Published: Oct. 12, 2023

In the pursuit of novel catalyst development to address pressing environmental concerns and energy demand, conventional design optimization methods often fall short due complexity vastness parameter space. The advent Machine Learning (ML) has ushered in a new era field optimization, offering potential solutions shortcomings traditional techniques. However, existing fail effectively harness vast information contained within expanding body scientific literature on synthesis. To this gap, study proposes an innovative Artificial Intelligence (AI) workflow that integrates large-language models (LLMs), Bayesian active learning loop expedite enhance optimization. Our methodology combines advanced language understanding with robust strategies, translating knowledge extracted from diverse into actionable parameters for practical experimentation article, we demonstrate application AI synthesis ammonia production. results underscore workflow's ability streamline process, swift, resource-efficient, high-precision alternative methods.

Language: Английский

Citations

17
Advances in fundamentals and application of plasmon-assisted CO2 photoreduction DOI Creative Commons
Zelio Fusco, Fiona J. Beck

Nanophotonics, Journal Year: 2024, Volume and Issue: 13(4), P. 387 - 417

Published: Feb. 1, 2024

Abstract Artificial photosynthesis of hydrocarbons from carbon dioxide (CO 2 ) has the potential to provide renewable fuels at scale needed meet global decarbonization targets. However, CO is a notoriously inert molecule and converting it energy dense complex, multistep process, which can proceed through several intermediates. Recently, ability plasmonic nanoparticles steer reaction down specific pathways enhance both rate selectivity garnered significant attention due its for sustainable production environmental mitigation. The excitation strong confined optical near-fields, energetic hot carriers localized heating be harnessed control or chemical pathways. despite many seminal contributions, anticipated transformative impact plasmonics in selective photocatalysis yet materialize practical applications. This lack complete theoretical framework on action mechanisms, as well challenge finding efficient materials with high scalability potential. In this review, we aim comprehensive critical discussion recent advancements plasmon-enhanced photoreduction, highlighting emerging trends challenges field. We delve into fundamental principles plasmonics, discussing works that led ongoing debates mechanism, introduce most ab initio advances, could help disentangle these effects. then synthesize experimental advances situ measurements plasmon photoreduction before concluding our perspective outlook field photocatalysis.

Language: Английский

Citations

8
Developing Cheap but Useful Machine Learning-Based Models for Investigating High-Entropy Alloy Catalysts DOI Creative Commons
Chenghan Sun, Rajat Goel, Ambarish Kulkarni

et al.

Langmuir, Journal Year: 2024, Volume and Issue: 40(7), P. 3691 - 3701

Published: Feb. 5, 2024

This work aims to address the challenge of developing interpretable ML-based models when access large-scale computational resources is limited. Using CoMoFeNiCu high-entropy alloy catalysts as an example, we present a cost-effective workflow that synergistically combines descriptor-based approaches, machine learning-based force fields, and low-cost density functional theory (DFT) calculations predict high-quality adsorption energies for H, N, NHx (x = 1, 2, 3) adsorbates. achieved using three specific modifications typical DFT workflows including: (1) sequential optimization protocol, (2) new geometry-based descriptor, (3) repurposing already-available trajectories develop ML-FF. Taken together, this study illustrates how appropriately designed descriptors can be used cheap but useful predicting at significantly lower costs. We anticipate resource-efficient philosophy may broadly relevant larger surface catalysis community.

Language: Английский

Citations

8