In this interview, AZoM speaks with Dr. Steve Edkins, VP of Strategy and Operations at Citrine Informatics Inc., about the key role that artificial intelligence (AI) plays in advancing polymer sustainability.
Can you please introduce yourself and your role at Citrine Informatics Inc.?
Certainly, I'm Steve Edkins, VP of Strategy and Operations at Citrine Informatics where, over the last 6 years, I have used my background in materials and AI/ML to help chemical and materials companies operationalize AI within R&D. Before moving into my current role focussed on drivingCitrine'ss growth, I led our Expert Services team. In that role, I gained first-hand experience in using AI to increase the sustainability of polymeric materials, including improving recycling efficiency, using bio feedstocks, optimizing processing conditions to reduce greenhouse gas emissions, and implementing AI-driven predictive models to balance material performance and environmental compatibility.
Image Credit: Anggalih Prasetya/Shutterstock.com
How is AI currently being utilized to improve polymer sustainability?
AI is revolutionizing polymer sustainability in several ways. Our customers are using machine learning algorithms to predict the properties of new polymer formulations, which accelerates the development of high-performing and environmentally friendly materials by 2-3 x. Particular examples include: replacing 90% of fossil-derived inputs with bio-feedstocks in a thermoplastic in 50% of the expected development time; lightweighting automotive grades while maintaining mechanical properties and cost; redeveloping a medical device to reduce 44 tons of CO2 emissions annually.
Can you provide an example of AI aiding in the design of recyclable polymers?
Absolutely. It can be used differently depending on whether you recycle mechanically or chemically. In mechanical recycling, the tricky bit is that incoming batches of post-consumer recyclate vary significantly. AI can be used to rapidly figure out which additives to use to maintain a consistent final product, and optimize catalysts and solvent mixtures to improve yields in chemical recycling.
What challenges do you face when integrating AI into polymer research?
There is a general lack of understanding around which AI tools best fit a given task. Not all AIs are Large Language Models (LLMs) like ChatGPT, and neither are LLMs appropriate for many tasks in chemistry and materials. Another example would be the misconception that you need large amounts of data to start. It's important to know that AI tools, like the Citrine Platform, have been specifically designed for small datasets, 30, 50, and 100 datapoints. Companies sometimes go down a data rabbit hole and delay the benefit they would get from AI.
Are there any recent breakthroughs in AI-driven polymer sustainability you'd like to highlight?
We have recently been working on a project to design new PVC grades. We worked with academic collaborators to optimize performance, ecotoxicity, and global warming potential simultaneously. This work is about to be published, so I can't say much more. But it's exciting!
How do you see the role of AI evolving in the field of polymer sustainability?
I envision AI becoming an indispensable tool in polymer research. This will significantly shorten development cycles and lead to more sustainable materials entering the market faster.
What advice would you give to researchers looking to incorporate AI into their sustainability efforts?
The best time to get started is today. Start small by applying AI to a couple of projects with real business value. Build upon the success of those early projects to create a flywheel effect. Also, be prepared for an iterative process—AI integration is a learning journey that evolves with each project. Finding partners with deep expertise can go a long way to smoothing any bumps in the road on your AI journey.
About Dr. Steve Edkins
Dr. Steve Edkins is the VP of Strategy and Operations at Citrine Informatics Inc.. He earned a Natural Sciences (Physics, Chemistry, Materials Science) degree from the University of Cambridge and a Ph.D. in Condensed Matter Physics from the University of St. Andrews. Steve has also held postdoctoral research positions at Cornell and Stanford. He has spent the past years operationalizing AI to generate business value in the chemicals industry.
This information has been sourced, reviewed and adapted from materials provided by Citrine Informatics.
For more information on this source, please visit Citrine Informatics.
Disclaimer: The views expressed here are those of the interviewee and do not necessarily represent the views of AZoM.com Limited (T/A) AZoNetwork, the owner and operator of this website. This disclaimer forms part of the Terms and Conditions of use of this website.