Researchers from MIT and Duke University have teamed up to significantly strengthen the resilience of plastics using machine learning, potentially transforming everything from everyday products to reducing overall plastic waste. In a groundbreaking study reported by MIT News, the team has pinpointed a novel class of crosslinker molecules known as mechanophores, specifically zeroing in on ferrocene-based compounds that have shown the ability to withstand more force before tearing.
Machine learning isn't something you typically think of as walking hand-in-hand with chemical engineering, but the researchers from MIT and Duke are definitely trying to push boundaries. Making use of a neural network, they were able to quickly sift through thousands of compounds to pinpoint the most promising candidates — a task that traditionally could take weeks of painstaking laboratory work or days of intense computational simulations. As MIT News reported, Heather Kulik, the Lammot du Pont Professor of Chemical Engineering at MIT, underscored the significance of their discovery, sharing that these molecules "can be useful for making polymers that would be stronger in response to force."
What sets these ferrocenes apart is their organometallic structure, featuring an iron atom nestled cozily between two carbon rings. By dabbling with various chemical groups attached to these rings, the team has been able to alter the mechanical properties of these molecules. Some of the properties looked for include how these compounds could break apart quickly under force, ultimately leading to stronger polymer materials.
The research is poised to not just incrementally but possibly revolutionize the way we perceive and use plastics. Incorporating one of the macho-sounding crosslinkers, m-TMS-Fc, into a polymer led to a material that was about four times tougher than counterparts using standard ferrocenes. This is no small feat considering the current environmental burden posed by plastic waste. "That really has big implications because if we think of all the plastics that we use and all the plastic waste accumulation, if you make materials tougher, that means their lifetime will be longer," Ilia Kevlishvili, MIT postdoc and lead author of the study, told MIT News.
Looking forward, the team aims to go further to explore mechanophores with additional desirable properties, like changing color or becoming catalytically active in response to force. Their machine-learning approach could be a game-changer in identifying and utilizing such materials for a wide range of applications, from stress sensors to drug delivery systems. With funding from the National Science Foundation Center for the Chemistry of Molecularly Optimized Networks, the researchers could well be paving the way for a new era in material science where plastics are not only durable but also smarter.
