Polymers touch almost every aspect of modern life. Chances are that if you are reading this post you are in contact with at least one polymer-containing product.
Polymers are materials made of long, repeating chains of molecules. They have unique physical-chemical properties that depend on a very high number of variables, such as the type of molecules that form the chain, how they are bonded, and how the polymer chains interact with each other. For example, some polymers bend and stretch like rubber and polyester; others are hard and tough like Kevlar and Bakelite. Consequently, the physical-chemical properties of polymers are extremely important for diverse applications in our everyday lives.
However, synthesising polymers with very precise physical-chemical properties is a complex challenge. Currently, researchers start by selecting monomers, target properties, a target molecular weight, and a manufacturing process according to their experience and a literature search. They then synthesise the polymer and perform a wide range of analyses to determine its properties and verify if they match the initial target. This three-step process is repeated over and over, until the properties of the synthesised polymer match the desired requirements. This process is extremely time- and resource-consuming, because it is based on a high number of experimental variables, on the researcher’s “chemical intuition”, and on an inefficient trial and error approach.
The advent of machine learning is opening incredible opportunities in many fields of research and technology. Our idea is to employ machine learning to address this complex challenge in polymer chemistry and develop the first algorithms that will assist researchers and industries in designing a priori polymers with precise physical-chemical properties. The achievement of this scientific goal will have an unprecedented impact on the polymer industry and research, reducing time and costs for the development of new materials.