From chemRxiv

This work aims to address the challenge of developing interpretable ML-based models when access to 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 to predict high-quality adsorption energies for H, N and NHx (x = 1, 2 and 3) adsorbates. This is achieved using three specific modifications to typical DFT workflows including, (1) using a sequential optimization protocol, (2) developing a new-geometry based descriptor, and (3) re-purposing the already-available low-cost DFT optimization trajectories to develop a ML-FF. Taken together, this study illustrates how cheap DFT calculations and appropriately designed descriptors can be used to develop cheap but useful models for predicting high-quality adsorption energies at significantly lower computational costs. We anticipate that this resource-efficient philosophy may be broadly relevant to the larger surface catalysis community.

Simplify

Published on November 20, 2023

Copy BibTeX

Loading...

This is an AI-generated summary

Key points

Made machine learning models to study high-entropy alloy catalysts in a cheaper way.

Used low-cost DFT calculations to predict adsorption energies.

Developed GLaSS descriptor for improved model efficiency and accuracy.

Used XGBoost with Optuna for hyperparameter optimization.

Data from DFT calculations matched well with predictions from our models.

Models were cost-effective and accurate for catalyst investigations.

Abstract

Page 2

We've designed workflow to predict adsorption energies using descriptor based approaches, machine learning force fields and low-cost DFT calculations. We implemented modifications to typical DFT workflows to effectively develop machine learning models for low-cost, high-quality energy predictions.

Introduction

Page 2

We're utilizing machine-learning to cut down exhaustive DFT calculation computational costs in predicting adsorption energies for catalyst design, which revolutionizes computational catalysis. By starting with a smaller database, we're able to develop surrogate predictive models exploring broader catalyst design space.