Machine learning-based approach to discover nanocomposite films as biodegradable plastic replacements

The accumulation of plastic waste in the natural environment is of greatest concern as it contributes to the destruction of ecosystems and harms aquatic life. So in recent years, materials scientists have been trying to identify natural alternatives to plastic that can be used to package and manufacture products.

Researchers at the University of Maryland, College Park recently devised a new approach to finding promising biodegradable plastic alternatives. The method they proposed was natural nanotechnologycombines cutting-edge machine learning techniques and molecular science.

“My inspiration for this study came from a visit to Palau in the Western Pacific in 2019,” study co-author Professor Po Yen Chen told Tech Xplore. “The effects of plastic pollution on marine life there (such as floating plastic films fooling fish and sea turtles mistaking plastic waste for food) were very alarming. This led me to my own expertise. It motivated me to apply this to this environmental problem and focused my efforts on finding solutions.'' Founding a lab at UMD. ”

Traditional methods employed to date to find sustainable plastic alternatives are time-consuming and inefficient. And in many cases, negative results are obtained, such as identifying materials that are biodegradable but do not have the same desirable properties as plastics.

The innovative approach to identifying plastic substitutes presented in this recent paper relies on a machine learning model developed by Chen.

In addition to being faster than traditional materials search methods, this approach may be more effective in discovering materials that can actually be used in manufacturing and industrial settings. Chen worked closely with his colleagues Teng Li and Liangbing Hu to apply machine learning techniques to discover all-plastic alternatives.

“We combined automated robotics, machine learning, and molecular dynamics simulations to accelerate the development of environmentally friendly all-natural plastic alternatives that meet important performance criteria,” Chen explained. “Our integrated approach combines automated robotics, machine learning, and active learning loops to accelerate the development of biodegradable plastic alternatives.”

First, Chen and his colleagues compiled a comprehensive library of nanocomposite films derived from various natural resources. This was done using an autonomous pipetting robot that can independently prepare laboratory samples.

The researchers then used this sample library to train Chen's machine learning-based model. During training, the model gradually became adept at predicting material properties based on its composition, through a process known as iterative active learning.

“The synergy of robotics and machine learning not only facilitates the discovery of natural plastic substitutes, but also enables the targeted design of plastic substitutes with specific properties,” Chen said. Stated. “Our approach significantly reduces the time and resources required compared to traditional trial-and-error research methods.”

This recent study and the approach introduced therein may facilitate future exploration of environmentally friendly plastic alternatives. The team's model will soon be used by teams around the world to produce natural nanocomposites with tunable and advantageous properties.

“By combining robotics, machine learning, and simulation tools, we have established a workflow that accelerates the discovery of new functional materials and enables customization for specific applications,” said Chen.

“Our integrated approach lowers the design barrier for green alternatives to petrochemical plastics while remaining environmentally safe. We also focus on functional materials that are green, environmentally friendly, and biodegradable.” It also provides an open and extensible database with a focus on

In the future, the innovative approach developed by Chen could help reduce plastic pollution around the world by facilitating the transition of multiple sectors to more sustainable materials. In their next study, the researchers plan to continue their efforts to address the environmental problems caused by petrochemical plastics.

For example, we want to expand the range of natural materials that manufacturers can choose from. Additionally, we plan to expand the possible applications of the materials identified by the model and enable these materials to be produced on a large scale.

“We are currently working on finding biodegradable and sustainable materials suitable for post-harvest fresh food packaging, replacing single-use plastic food packaging, and improving the shelf life of these post-harvest products. “There are,” Chen added.

“We are also researching ways to manage the disposal of these biodegradable plastics, such as recycling them or converting them into other useful chemicals. The initiative is an important step in making our solutions not only environmentally friendly, but also economically viable alternatives to traditional plastics, making a significant contribution to global efforts to reduce plastic pollution. .”

© 2024 Science X Network