Machine learning accelerates the search for new sustainable materials

TSUKUBA, Japan, May 25, 2022 – (ACN Newswire) – – Researchers from Konica Minolta and the Nara Institute of Science and Technology in Japan have developed a machine learning method to identify sustainable alternatives to composite materials. Their results were published in the journal Science and Technology of Advanced Materials: Methods.

Composite materials are compounds made up of two or more constituent materials. Due to the complex nature of the interactions between the different components, their performance can greatly exceed that of simple materials. Composite materials, such as fiber-reinforced plastics, are very important for a wide range of industries and applications, including electrical and information technologies.

In recent years, there has been a growing demand for more environmentally friendly materials that help reduce industrial waste and the use of plastic. One way to achieve this is to substitute the constituent materials of the composites with recyclable materials or biomass. However, this can reduce the performance compared to the parent material, not only due to the characteristics of the individual constituent materials, such as their physicochemical properties, but also due to the interactions between the constituents.

“Finding a new composite material that achieves the same performance as the original using only human experience and intuition takes a very long time because you have to evaluate countless materials while considering the interactions between them,” explains michihiro okuyamadeputy director at Konica Minolta, Inc.

Machine learning offers a potential solution to this problem. Scientists have proposed several machine learning methods to perform rapid searches among a large number of materials, based on the relationship between material characteristics and performance. However, in many cases, the properties of the constituent materials are unknown, making these types of predictive research difficult.

To overcome this limitation, the researchers developed a new kind of machine learning method to find alternative materials. A key advantage of the new method is that it can quantitatively assess the interactions between component materials to reveal how much they contribute to the overall performance of the composite. The method then searches for replacement constituents with similar performance to the parent material.

The researchers tested their method by researching alternative constituent materials for a composite made up of three materials – a resin, a filler and an additive. They experimentally evaluated the performance of surrogate materials identified by machine learning and found that they were similar to the original material, proving that the model works.

“By developing alternatives that make up composite materials, our new machine learning method removes the need to test large numbers of candidates through trial and error, saving time and money.” Okuyama said.

The method could be used to quickly and efficiently identify sustainable substitutes for composite materials, reducing the use of plastic and encouraging the use of biomass or renewable materials.

More information

michihiro okuyama


Email: [email protected]

About Advanced Materials Science and Technology: Methods (STAM Methods)

STAM Methods is an open access sister journal of Advanced Materials Science and Technology (STAM), and focuses on emerging methods and tools to improve and/or accelerate materials development, such as methodology, devices , instrumentation, modeling, high throughput data. collection, materials/process informatics, databases and programming.

dr. Masanobu Naito

Publication of STAM methods Director

Email: [email protected]

Press release issued by Asia Research News for advanced materials science and technology.

Copyright 2022 ACN Newswire. All rights reserved.

© Japan Corporate News, source JCN Press Releases

Comments are closed.