Radioactive waste management remains a major hurdle in the safe and sustainable use of nuclear energy. Among the various contaminants, radioactive iodine—especially isotope I-129—poses a significant threat due to its long half-life of 15.7 million years, high mobility, and severe toxicity to living organisms.
Innovative Use of AI in Material Discovery
A Korean research team has made a major breakthrough by harnessing artificial intelligence to identify a novel material capable of removing radioactive iodine from contaminated environments. Their discovery opens new avenues for nuclear environmental remediation. The team, led by Professor Ho Jin Ryu from the Department of Nuclear and Quantum Engineering, collaborated with Dr. Juhwan Noh of the Korea Research Institute of Chemical Technology’s Digital Chemistry Research Center.
From AI Models to Real-World Solutions
The research, published in the Journal of Hazardous Materials, reveals how the team used a machine learning-based approach to pinpoint effective iodate adsorbents within a class of compounds known as Layered Double Hydroxides (LDHs). These materials can incorporate a wide range of metal elements, making them excellent candidates for anion adsorption.
As per Science X Daily, traditional silver-based adsorbents have shown limited effectiveness against iodate (IO₃⁻). It is a form in which radioactive iodine primarily exists in aqueous environments. In contrast, the newly developed material—Cu₃(CrFeAl), composed of copper, chromium, iron, and aluminum—achieved over 90% iodate removal efficiency.
AI Accelerates Discovery by Reducing Trial-and-Error
The key to this success was the application of AI-driven active learning, which allowed the team to navigate a vast compositional space with remarkable efficiency. Instead of exhaustively testing every possible combination, they began with data from 24 binary and 96 ternary LDH compositions. Then, through predictive modeling, they focused their efforts on the most promising quaternary and quinary combinations. Remarkably, they tested only 16% of all possible materials to find the optimal one.
Towards Commercialization and Broader Impact
“This study shows the potential of using artificial intelligence to efficiently identify radioactive decontamination materials from a vast pool of new material candidates,” said Professor Ryu. “It is expected to accelerate research for developing new materials for nuclear environmental cleanup.”
The research team has already filed a domestic patent and is moving forward with an international application. They are also working to improve the material’s performance under diverse conditions and plan to commercialize the technology. Through collaborations between academia and industry, they aim to develop iodine-adsorbing powders and contaminated water treatment filters for real-world application.
