Fig. 1: Machine intelligence-accelerated discovery of all-natural plastic substitutes with programmable properties.
From: Machine intelligence-accelerated discovery of all-natural plastic substitutes

a, An integrated workflow using robotics and AI/ML predictions is demonstrated to construct a high-accuracy prediction model, enabling the accelerated discovery of all-natural plastic substitutes. Inset photo shows a model-suggested all-natural substitute that can be fabricated in a large area (with the dimensions of 53 cm × 38 cm). b, By inputting various property criteria for specific plastic products, the prediction model can automate the inverse design of all-natural substitutes to replace transparent badge holders, clear file folders, transparent shopping bags, translucent lamp shades, transparent air pillows, non-flammable battery packages and UV-blocking chemical packages. Top row photos show non-biodegradable plastic products and their property criteria. Bottom row photos show the biodegradable, all-natural substitutes suggested by the prediction model.