King’s Interdisciplinary Researchers Develop AI-Powered Program PySoftK!
A team of interdisciplinary researchers across King’s Faculty of Natural, Mathematical, and Engineering Sciences has developed a program called PySoftK, which uses AI to identify new polymer materials. Consequently, these new polymers could be used in various applications, including medical technology, pharmaceuticals, energy storage, and more. Therefore, this software will significantly advance the discovery of new polymers.
PySoftK will enable the acceleration of novel polymer development for various applications. This technology will pave the way for the creation of materials that will aid in addressing large-scale challenges in healthcare, biodegradable home and personal care products, and environmentally friendly energy storage systems. This was stated by Professor Chris Lorenz from the Department of Physics, who is the lead researcher on the technology.
Polymers, which are large molecules made up of smaller repeating molecules called monomers and can be naturally occurring or synthetic, are being studied in new software development that could change the way we investigate the relationship between their chemical structure and function. This development provides a robust dataset for researchers to train artificial intelligence (AI) to identify desirable polymer properties, potentially leading to the creation of designer polymers. Gore-Tex is an example of a designer polymer, which was developed as an improvement to traditional nylon and has a particular chemical property that allows it to be waterproof and breathable.
Furthermore, designer polymers are used in various fields such as medical ointments, paints, coatings, food packaging, biomedical imaging, and energy storage. These polymers possess diverse functions due to their physical and chemical properties that arise from monomer type and arrangement. To enhance the discovery of these materials, high-performance computers (HPCs) simulate and predict the behavior of polymers. This informs researchers on how to construct polymers with specific properties to perform certain tasks. Molecular-level simulations have contributed to our comprehension of the interplay between chemical composition and purpose within polymers of advancing complexity.
Recent progress in computing capabilities and algorithms has allowed researchers to study even more intricate systems and generate more precise projections through swift molecular-scale simulations. As a result, a quicker and more economical approach to material design can be attained, ultimately minimizing the need for prolonged experimentation.
