Scientists from the Georgia Institute of Technology are leveraging artificial intelligence (AI) to revolutionize the future of the polymer industry. Led by Rampi Ramprasad, the team developed and refined AI algorithms to accelerate the discovery of new materials. Their research was published in the journals Nature Reviews Materials and Nature Communications.
Image Credit: Ricardo Javier/Shutterstock.com
Two articles published this summer in the Nature family of journals highlight significant advancements and achievements in AI-driven polymer informatics research. The first article, published in Nature Reviews Materials, discusses recent developments in polymer design across three key application areas: recyclable plastics, energy storage, and filtration technologies.
The second article, featured in Nature Communications, details the use of AI algorithms to identify a specific subclass of polymers suitable for electrostatic energy storage. These materials have been successfully synthesized and tested in the lab, demonstrating the practical application of AI in material discovery.
In the early days of AI in materials science, propelled by the White House’s Materials Genome Initiative over a decade ago, research in this field was largely curiosity-driven. Only in recent years have we begun to see tangible, real-world success stories in AI-driven accelerated polymer discovery. These successes are now inspiring significant transformations in the industrial materials R&D landscape. That is what makes this review so significant and timely.
Rampi Ramprasad, Professor, School of Materials Science and Engineering, Georgia Institute of Technology
AI Opportunities
Ramprasad’s team has developed innovative algorithms that can quickly predict the properties and formulations of polymers before they are physically created. The process begins by defining application-specific target properties or performance requirements. Machine learning (ML) models are then trained using existing material-property data to predict these desired outcomes.
How does AI help in polymer discovery?
The team can also generate new polymers, with their properties predicted by the ML models. The best candidates that meet the specified property requirements are selected for laboratory testing and synthesis, followed by real-world validation. This process is iterative; the results from further testing are integrated with the initial data to refine the predictive models continuously.
While AI accelerates the search for novel polymers, it also presents distinct challenges. Accurate AI predictions rely on the availability of rich, varied, and large data sets, making high-quality data crucial. Additionally, developing algorithms that can produce polymers that are both synthesizable and chemically feasible is a complex task.
After the algorithms make their predictions, the real challenge begins: proving that the developed materials can be manufactured in the lab, perform as expected, and be scalable for practical use.
Ramprasad’s group focuses on designing these materials, while collaborators at other institutions, including Georgia Tech, handle the materials’ manufacturing, processing, and testing. Professor Ryan Lively of the School of Chemical and Biomolecular Engineering, a co-author of the research, regularly collaborates with Ramprasad’s team on these efforts.
In our day-to-day research, we extensively use the machine learning models Rampi’s team has developed. These tools accelerate our work and allow us to rapidly explore new ideas. This embodies the promise of ML and AI because we can make model-guided decisions before we commit time and resources to explore the concepts in the laboratory.
Ryan Lively, Professor and Study Co-Author, School of Chemical and Biomolecular Engineering, Georgia Institute of Technology
Ramprasad’s group and associates have made major strides with AI in various areas, such as energy storage, filtration technologies, additive manufacturing, and recyclable materials.
Polymer Progress
One significant achievement is the development of novel polymers for capacitors, which are critical components in devices such as hybrid and electric vehicles. In this effort, Ramprasad’s group collaborated with researchers from the University of Connecticut.
Currently, capacitor polymers typically offer either thermal stability or high energy density, but not both. By employing AI techniques, the researchers identified that insulating materials made from polynorbornene and polyimide polymers can achieve both high energy density and good thermal stability simultaneously.
These polymers can be further refined to maintain environmental sustainability and perform effectively in extreme conditions, such as those encountered in aerospace applications.
The new class of polymers with high energy density and high thermal stability is one of the most concrete examples of how AI can guide materials discovery. It is also the result of years of multidisciplinary collaborative work with Greg Sotzing and Yang Cao at the University of Connecticut and sustained sponsorship by the Office of Naval Research.
Rampi Ramprasad, Professor, School of Materials Science and Engineering, Georgia Institute of Technology
Industry Potential
Industry involvement in the Nature Reviews Materials publication underscores the practical potential of AI-assisted materials development. Scientists from General Electric and the Toyota Research Institute are co-authors of this research, further highlighting its industry relevance.
Matmerize Inc., a software startup recently spun out of Georgia Tech and co-founded by Ramprasad, aims to accelerate the adoption of AI-driven materials development in the industry. Their cloud-based polymer informatics software is already being utilized by companies across various sectors, including energy, electronics, consumer goods, chemical processing, and sustainable materials.
Ramprasad said, “Matmerize has transformed our research into a robust, versatile, and industry-ready solution, enabling users to design materials virtually with enhanced efficiency and reduced cost. What began as a curiosity has gained significant momentum, and we are entering an exciting new era of materials by design.”
Journal Reference:
Tran, H., et al. (2024) Design of functional and sustainable polymers assisted by artificial intelligence. Nature Reviews Materials. doi.org/10.1038/s41578-024-00708-8.
Gurnani, R., et al. (2024) AI-assisted discovery of high-temperature dielectrics for energy storage. Nature Communications. doi.org/10.1038/s41467-024-50413-x.
- 0 Comments
- Ai Process
- Artificial Intelligence