Interpretable Machine Learning Models for Practical Antimonate Electrocatalyst Performance
Authors:
- Melissa E. KreiderMKMelissa E. Kreider
- Gaurav KamatGKGaurav Kamat
- McKenzie HubertMHMcKenzie Hubert
- José A. Zamora ZeledónJZJosé A. Zamora Zeledón
- Lingze WeiLWLingze Wei
- Jesse MatthewsJMJesse Matthews
- Thomas F. JaramilloTJThomas F. Jaramillo
- Frank Abild‐PedersenFAFrank Abild‐Pedersen
- Michaela Burke StevensMSMichaela Burke Stevens
- 25 April 2024
- journal article
- research article
- Published by Wiley in Chemphyschem
- Vol. 25 (13) , e202400010
Abstract
Computationally predicting the performance of catalysts under reaction conditions is a challenging task due to the complexity of catalytic surfaces and their evolution in situ, different reaction paths, and the presence of solid‐liquid interfaces in the case of electrochemistry. We demonstrate here how relatively simple machine learning models can be found that enable prediction of experimentally observed onset potentials. Inputs to our model are comprised of data from the oxygen reduction reaction on non‐precious transition‐metal antimony oxide nanoparticulate catalysts with a combination of experimental conditions and computationally affordable bulk atomic and electronic structural descriptors from density functional theory simulations. From human‐interpretable genetic programming models, we identify key experimental descriptors and key supplemental bulk electronic and atomic structural descriptors that govern trends in onset potentials for these oxides and deduce how these descriptors should be tuned to increase onset potentials. We finally validate these machine learning predictions by experimentally confirming that scandium as a dopant in nickel antimony oxide leads to a desired onset potential increase. Macroscopic experimental factors are found to be crucially important descriptors to be considered for models of catalytic performance, highlighting the important role machine learning can play here even in the presence of small datasets.Keywords
Funding Information
- Basic Energy Sciences
- National Energy Research Scientific Computing Center
- Office of Science (DE-AC02-05CH11231)
- National Science Foundation
This publication has 59 references indexed in Scilit:
- Artificial Intelligence in Physical Sciences: Symbolic Regression Trends and PerspectivesArchives of Computational Methods in Engineering, 2023
- Symbolic regression for the interpretation of quantitative structure-property relationshipsArtificial Intelligence in the Life Sciences, 2022
- Bayesian Optimization of High‐Entropy Alloy Compositions for Electrocatalytic Oxygen Reduction**Angewandte Chemie International Edition, 2021
- LOBSTER: A tool to extract chemical bonding from plane‐wave based DFTJournal of Computational Chemistry, 2016
- Crystal growth experiments in the systems Ni2MSbO6 (M = Sc, In) using chemical vapour transport reactions: Ni2InSbO6 and NiSb2O6 crystals in the millimetre rangeCrystal Research and Technology, 2013
- Information Theory and an Extension of the Maximum Likelihood Principle1998
- Electronic factors determining the reactivity of metal surfacesSurface Science, 1995
- Genetic programming as a means for programming computers by natural selectionStatistics and Computing, 1994
- A powder neutron diffraction study of cation ordering in 6H Ba3Sb2NiO9Journal of Inorganic and Nuclear Chemistry, 1978
- Die Kristallstrukturen der hexagonalen Elpasolithe Ba3NiSb2O9 und Ba3CuSb2O9 – röntgenographische und spektroskopische ErgebnisseZeitschrift für anorganische und allgemeine Chemie, 1977