Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR data

Abstract Effective designs of biodegradable polymers are highly desirable for achieving a sustainable society by decreasing environmental burden and replacing petroleum-based resources with biomass. Low-field NMR is one of the candidate techniques because it provides information on the higher-order...

Full description

Saved in:
Bibliographic Details
Main Authors: Ryo Fujita, Yoshifumi Amamoto, Jun Kikuchi
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:npj Materials Degradation
Online Access:https://doi.org/10.1038/s41529-025-00613-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849329381649940480
author Ryo Fujita
Yoshifumi Amamoto
Jun Kikuchi
author_facet Ryo Fujita
Yoshifumi Amamoto
Jun Kikuchi
author_sort Ryo Fujita
collection DOAJ
description Abstract Effective designs of biodegradable polymers are highly desirable for achieving a sustainable society by decreasing environmental burden and replacing petroleum-based resources with biomass. Low-field NMR is one of the candidate techniques because it provides information on the higher-order structure and dynamics of polymers quickly and conveniently. Although machine learning approaches such as Bayesian optimization (BO) and convolutional neural networks (CNNs) are significant, there have been almost no reports on effective material design based on low-field nuclear magnetic resonance (NMR) data. This study proposes a method for optimizing polymer process conditions using CNN-based features extracted from relaxation curves. This approach identified important features related to material properties while reconstructing denoised relaxation curves of polylactic acid. BO of process conditions using these features achieved an optimization rate comparable to using material property values, suggesting that effective material design is possible without directly evaluating a large number of properties. This might be potentially insightful for the feasibility of a framework to accelerate polymer development through low-field NMR with minimal property data.
format Article
id doaj-art-133abe5832384df28fef6be09aa97a35
institution Kabale University
issn 2397-2106
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series npj Materials Degradation
spelling doaj-art-133abe5832384df28fef6be09aa97a352025-08-20T03:47:17ZengNature Portfolionpj Materials Degradation2397-21062025-06-01911910.1038/s41529-025-00613-7Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR dataRyo Fujita0Yoshifumi Amamoto1Jun Kikuchi2Graduate School of Medical Life Science, Yokohama City UniversityRIKEN Center for Sustainable Resource ScienceGraduate School of Medical Life Science, Yokohama City UniversityAbstract Effective designs of biodegradable polymers are highly desirable for achieving a sustainable society by decreasing environmental burden and replacing petroleum-based resources with biomass. Low-field NMR is one of the candidate techniques because it provides information on the higher-order structure and dynamics of polymers quickly and conveniently. Although machine learning approaches such as Bayesian optimization (BO) and convolutional neural networks (CNNs) are significant, there have been almost no reports on effective material design based on low-field nuclear magnetic resonance (NMR) data. This study proposes a method for optimizing polymer process conditions using CNN-based features extracted from relaxation curves. This approach identified important features related to material properties while reconstructing denoised relaxation curves of polylactic acid. BO of process conditions using these features achieved an optimization rate comparable to using material property values, suggesting that effective material design is possible without directly evaluating a large number of properties. This might be potentially insightful for the feasibility of a framework to accelerate polymer development through low-field NMR with minimal property data.https://doi.org/10.1038/s41529-025-00613-7
spellingShingle Ryo Fujita
Yoshifumi Amamoto
Jun Kikuchi
Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR data
npj Materials Degradation
title Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR data
title_full Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR data
title_fullStr Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR data
title_full_unstemmed Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR data
title_short Bayesian optimization of biodegradable polymers via machine learning driven features from low-field NMR data
title_sort bayesian optimization of biodegradable polymers via machine learning driven features from low field nmr data
url https://doi.org/10.1038/s41529-025-00613-7
work_keys_str_mv AT ryofujita bayesianoptimizationofbiodegradablepolymersviamachinelearningdrivenfeaturesfromlowfieldnmrdata
AT yoshifumiamamoto bayesianoptimizationofbiodegradablepolymersviamachinelearningdrivenfeaturesfromlowfieldnmrdata
AT junkikuchi bayesianoptimizationofbiodegradablepolymersviamachinelearningdrivenfeaturesfromlowfieldnmrdata