CatFlow: An Automated Workflow for Training Machine Learning Potentials to Compute Free Energies in Dynamic Catalysis DOI Creative Commons

Yun‐Pei Liu,

Qiyuan Fan, Fu‐Qiang Gong

et al.

The Journal of Physical Chemistry C, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 31, 2024

Dynamic effects of catalysts play a crucial role in catalytic reactions, necessitating the incorporation statistical sampling and understanding impact dynamic structures free energy calculations. However, complexity systems poses challenges effectively exploring vast configurational space effectively. In this work, we propose CatFlow, an automated workflow for training machine learning potentials (MLPs) to compute energies reactions. CatFlow combines constrained molecular dynamics (MD) simulation with concurrent MLPs sequential calculation well trained MLPs. By rapidly generating reliable MLPs, facilitates rigorous calculations, enabling determination reaction profiles end-to-end manner. We showcased capabilities by investigating activation O2 catalyzed Pt clusters demonstrated phase transition on activities reaction. offers efficient solution studying elementary processes. It reduces need human intervention provides researchers powerful tool investigate catalysis.

Language: Английский

Machine Learning Accelerated Interfacial Fluxionality in Ni-Supported Metal Nitride Ammonia Synthesis Catalysts DOI
Pranav Roy, Brandon C. Bukowski

Published: Jan. 1, 2025

Language: Английский

Citations

0
Dynamic evolution of metal structures on/in zeolites for catalysis DOI

Yuexin Wu,

Pengcheng Deng,

Lujie Liu

et al.

Chemical Society Reviews, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

This review explores metal migration dynamics on/in zeolite supports, analyzing mechanisms and driving factors under varied conditions. It highlights the crucial role of in stabilizing metals enhancing performances.

Language: Английский

Citations

0
Advancing electrocatalyst discovery through the lens of data science: State of the art and perspectives☆ DOI Creative Commons
Xue Jia, Tianyi Wang, Di Zhang

et al.

Journal of Catalysis, Journal Year: 2025, Volume and Issue: unknown, P. 116162 - 116162

Published: April 1, 2025

Language: Английский

Citations

0
CatFlow: An Automated Workflow for Training Machine Learning Potentials to Compute Free Energies in Dynamic Catalysis DOI Creative Commons

Yun‐Pei Liu,

Qiyuan Fan, Fu‐Qiang Gong

et al.

The Journal of Physical Chemistry C, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 31, 2024

Dynamic effects of catalysts play a crucial role in catalytic reactions, necessitating the incorporation statistical sampling and understanding impact dynamic structures free energy calculations. However, complexity systems poses challenges effectively exploring vast configurational space effectively. In this work, we propose CatFlow, an automated workflow for training machine learning potentials (MLPs) to compute energies reactions. CatFlow combines constrained molecular dynamics (MD) simulation with concurrent MLPs sequential calculation well trained MLPs. By rapidly generating reliable MLPs, facilitates rigorous calculations, enabling determination reaction profiles end-to-end manner. We showcased capabilities by investigating activation O2 catalyzed Pt clusters demonstrated phase transition on activities reaction. offers efficient solution studying elementary processes. It reduces need human intervention provides researchers powerful tool investigate catalysis.

Language: Английский

Citations

1