Nylon, Teflon, and Kevlar are just a few of the more well-known polymers that have changed the world. From Teflon coating for frying pans to 3-D printing, polymers are essential to the creation of systems that help the world function better.
Now, the hunt for the next billion-dollar polymer is always a challenging task. Researchers at Georgia Tech are leveraging artificial intelligence to tailor and revolutionise the future of polymers. Rampi Ramprasad's group is in the development and adaptation of AI algorithms that accelerate materials discovery.
Ramprasad's team has developed innovative algorithms to predict polymer properties and formulations instantaneously before their actual fabrication. First, it requires the definition of application-specific target property or performance criteria. In view of this, the ML models should be suitably trained with existing material-property data to predict the outcomes.They also are in a position to synthesise new polymers whose properties are forecast with the ML models.The best candidates matching a target property are selected for real-world validation by laboratory synthesis and testing. Integrated into the original data, results from these new experiments further refine predictive models, on an ongoing iterative basis.
Source: Pixels
While AI can speed up the process of discovering new polymers, it is also introducing new problems. AI relies on richness, diversity, and size of the initial datasets, hence the good quality data. Moreover, it is challenging to develop algorithms that will produce both chemically plausible and synthesizable polymers.
The hard part, though, comes after the algorithms have made their predictions: proving that the designed materials can be made in the lab and function as expected, then demonstrating their scalability—producing them in large enough quantities—beyond the lab for real-world use.While Ramprasad's group designs the materials, their fabrication, processing, and testing are carried out by collaborators at various institutions, with Georgia Tech among them.Using AI, Ramprasad and his team, along with their collaborators, have pushed forward important fields such as energy storage, filtration technologies, additive manufacturing, and recyclable materials.
Using the AI tools, the researchers observed that high-energy density and high-thermal stability can be concomitantly achieved in insulating materials made of polynorbornene and polyamide polymers. The polymers would be further tailorable to operate in very challenging environments, such as aerospace applications, while remaining environmentally sustainable.
