We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Bioplastic design using multitask deep neural networks
AI summary Read the abstract
Researchers used deep neural networks trained on nearly 23,000 polymer chemistries to identify 14 biodegradable bioplastics — made from biological sources rather than petroleum — that could replace the seven most commonly used synthetic plastics, which account for 75% of global plastic production. This AI-driven approach accelerates the search for eco-friendly plastic alternatives that naturally break down rather than fragmenting into persistent microplastics.
Abstract Non-degradable plastic waste jeopardizes our environment, yet our modern lifestyle and current technologies are impossible to sustain without plastics. Bio-synthesized and biodegradable alternatives such as polyhydroxyalkanoates (PHAs) have the potential to replace large portions of the world’s plastic supply with cradle-to-cradle materials, but their chemical complexity and diversity limit traditional resource-intensive experimentation. Here, we develop multitask deep neural network property predictors using available experimental data for a diverse set of nearly 23,000 homo- and copolymer chemistries. Using the predictors, we identify 14 PHA-based bioplastics from a search space of almost 1.4 million candidates which could serve as potential replacements for seven petroleum-based commodity plastics that account for 75% of the world’s yearly plastic production. We also discuss possible synthesis routes for the identified promising materials.
More Papers Like This
Feasibility of Utilizing Machine Learning to Identify a More Sustainable Alternative to Polyester in Textiles
AI summary Read the abstract
Researchers used machine learning trained on the PI1M polymer dataset of over one million compounds to identify sustainable alternatives to polyester in textiles, predicting key material properties including glass transition temperature, density, melting temperature, oxygen permeability, and bulk modulus. The study found that artificial neural networks could screen polymer candidates for environmental compatibility while maintaining performance characteristics comparable to polyester.
Synthetic bacterial consortium for degradation of plastic pyrolysis oil waste: experimental optimization and neural network modeling
AI summary Read the abstract
Scientists engineered a team of bacteria that can break down the toxic, hard-to-recycle oily leftovers from melting down mixed plastics, eliminating nearly all the harmful chemicals within just a week. This matters because plastic waste and its byproducts are piling up faster than we can safely process them, and microplastics are already showing up throughout the human body—so finding ways to actually break down plastic waste (rather than just burning or burying it) could reduce our overall chemical exposure over time. The researchers also used AI to make this cleanup process faster and cheaper to fine-tune, which could help this approach scale up for real-
Multi-Objectives AI-based Optimization Model for Microplastics- free Water Bottles
AI summary Read the abstract
Researchers used artificial intelligence to optimize the composition of biodegradable plastic materials mixed with natural additives like cotton and cellulose, aiming to produce water bottle materials that break down in the environment rather than fragmenting into microplastics. The AI model balanced mechanical performance, cost, and environmental impact to identify optimal formulations.
The use of artificial neural networks in modelling migration pollutants from the degradation of microplastics
AI summary Read the abstract
Researchers used artificial neural networks to model the emission of additives from degrading microplastics, finding that machine learning could predict migration patterns from the vast range of polymer types, chemical structures, and environmental conditions involved. This approach could reduce the need for extensive laboratory testing by identifying high-risk scenarios for further investigation.
Deep learning-guided design of hydrolases for crystalline PET depolymerization
AI summary Read the abstract
Scientists used AI to design a super-enzyme that can break down PET plastic (the material in water bottles and clothing) into its original building blocks, even without needing to pre-treat or heat-soften the plastic first—something that's been a major roadblock for plastic recycling. This matters because PET plastic pollution takes centuries to break down naturally and sheds microplastics into our food, water, and bodies; this enzyme could enable truly circular recycling where old plastic becomes new, high-quality plastic again instead of ending up in landfills, oceans, or our bloodstreams.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.