A robotics- and machine learning-integrated workflow for discovering all-natural plastic substitutes

The global accumulation of nondegradable petrochemical plastic waste is driving the search for biodegradable substitutes derived from natural resources. However, discovering all-natural composite materials that match the optical, thermal and mechanical performance of commercial plastics remains challenging because traditional iterative optimization is inefficient. Here the workflow described in this Protocol accelerates the screening process by using automated robotics for liquid handing and machine learning to predict material properties. The procedure begins by commanding a pipetting robot to prepare a library of 286 nanocomposites. Basic physical properties of these products are used to train a support vector machine classifier that locates the decision boundary separating feasible from infeasible film compositions in the four-component system, thereby defining the feasible design space. Next, through 14 active learning loops coupled with data augmentation (for example, adding small random changes to the characterization data within the observed experimental error range), the robot fabricates 135 all-natural nanocomposites in a stagewise manner to construct an artificial neural network prediction model. The model is used to predict (1) the physicochemical properties of a nanocomposite from its composition and (2) which compositions are likely to have the desired, user-specified properties (for example, transparency or fire resistance). This protocol further details methods for model interpretation using Shapley Additive Explanations, validation via molecular dynamics simulations and model expansion to incorporate new building blocks. This hybrid approach takes ~2 months and enables the accelerated discovery of eco-friendly materials with programmable functions using components drawn from the generally recognized as safe database.

Authors

Institutions

Publication Details

Journal
Nature Protocols
Published
2026-11-16
DOI
https://doi.org/10.1038/s41596-026-01449-4
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A robotics- and machine learning-integrated workflow for discovering all-natural plastic substitutes

Po‐Yen Chen, Yang Li, Abhishek Sose, Po-Yen Chen
Nature Protocols
Machine Learning in Materials Science
article

A robotics- and machine learning-integrated workflow for discovering all-natural plastic substitutes

Po‐Yen Chen, Yang Li, Abhishek Sose, Po-Yen Chen
article en

Abstract

The global accumulation of nondegradable petrochemical plastic waste is driving the search for biodegradable substitutes derived from natural resources. However, discovering all-natural composite materials that match the optical, thermal and mechanical performance of commercial plastics remains challenging because traditional iterative optimization is inefficient. Here the workflow described in this Protocol accelerates the screening process by using automated robotics for liquid handing and machine learning to predict material properties. The procedure begins by commanding a pipetting robot to prepare a library of 286 nanocomposites. Basic physical properties of these products are used to train a support vector machine classifier that locates the decision boundary separating feasible from infeasible film compositions in the four-component system, thereby defining the feasible design space. Next, through 14 active learning loops coupled with data augmentation (for example, adding small random changes to the characterization data within the observed experimental error range), the robot fabricates 135 all-natural nanocomposites in a stagewise manner to construct an artificial neural network prediction model. The model is used to predict (1) the physicochemical properties of a nanocomposite from its composition and (2) which compositions are likely to have the desired, user-specified properties (for example, transparency or fire resistance). This protocol further details methods for model interpretation using Shapley Additive Explanations, validation via molecular dynamics simulations and model expansion to incorporate new building blocks. This hybrid approach takes ~2 months and enables the accelerated discovery of eco-friendly materials with programmable functions using components drawn from the generally recognized as safe database.

Nature Protocols
University of Maryland, College Park (US)
National Science Foundation, Air Force Research Laboratory, Army Research Laboratory
Responsible consumption and production
Machine Learning in Materials Science
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon