Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zinzani, Sofia, Baletto, Francesca, Rossi, Kevin
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2509.10985
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911153077092352
author Zinzani, Sofia
Baletto, Francesca
Rossi, Kevin
author_facet Zinzani, Sofia
Baletto, Francesca
Rossi, Kevin
contents Establishing a mapping between nanocatalysts structure and their catalytic properties is essential for efficient design. To this end, we demonstrate the accuracy of a general machine learning framework on a representative and challenging application: predicting the mass activity of Pt nanoparticles for the electrochemical oxygen reduction reaction, estimated via a microkinetic model. Accurate models are obtained when leveraging either a nanocatalyst's structure representation accessible at the computational level, namely the surface site generalized coordination number distributions, or one accessible experimentally, namely the nanoparticle's pair distance distribution function. Building on this result, we demonstrate that our machine learning model, in tandem with Bayesian optimization, efficiently identifies the Top-10 and Top-100 most active structures out of a large pool of candidates comprising more than 50000 different structures, after probing the activity only of a few thousand structures. These findings provide a robust blueprint for accelerated theoretical and experimental identification of active nanocatalysts.
format Preprint
id arxiv_https___arxiv_org_abs_2509_10985
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bridging Structure and Activity in Nanocatalysts via Machine Learning and Global Structure Representations
Zinzani, Sofia
Baletto, Francesca
Rossi, Kevin
Mesoscale and Nanoscale Physics
Materials Science
Chemical Physics
Establishing a mapping between nanocatalysts structure and their catalytic properties is essential for efficient design. To this end, we demonstrate the accuracy of a general machine learning framework on a representative and challenging application: predicting the mass activity of Pt nanoparticles for the electrochemical oxygen reduction reaction, estimated via a microkinetic model. Accurate models are obtained when leveraging either a nanocatalyst's structure representation accessible at the computational level, namely the surface site generalized coordination number distributions, or one accessible experimentally, namely the nanoparticle's pair distance distribution function. Building on this result, we demonstrate that our machine learning model, in tandem with Bayesian optimization, efficiently identifies the Top-10 and Top-100 most active structures out of a large pool of candidates comprising more than 50000 different structures, after probing the activity only of a few thousand structures. These findings provide a robust blueprint for accelerated theoretical and experimental identification of active nanocatalysts.
title Bridging Structure and Activity in Nanocatalysts via Machine Learning and Global Structure Representations
topic Mesoscale and Nanoscale Physics
Materials Science
Chemical Physics
url https://arxiv.org/abs/2509.10985