Oak Ridge National Lab, Cleveland Clinic, and IBM Achieve First-Known Computations of Fusion Materials on a Quantum Computer

Researchers from Oak Ridge National Laboratory, Cleveland Clinic, and IBM have successfully calculated nine molecular configurations of a material critical for fusion energy production using a quantum computer. This milestone marks the first time quantum-centric supercomputing has been applied to model the complex chemistry of FLiBe, a liquid salt used for extracting tritium fuel. By addressing a long-standing computational bottleneck in fusion research, this achievement supports the U.S. Department of Energy’s Genesis Mission to accelerate clean energy discovery through advanced computing paradigms.

The collaboration utilized IBM’s quantum-centric supercomputing framework to determine the electronic structure and atomic behavior of FLiBe (fluorine, lithium, and beryllium), a leading candidate for tritium breeding in fusion reactors. The team calculated nine distinct molecular configurations to understand how the material binds tritium at a fundamental level, a task that is notoriously difficult for classical computers to scale. These results, published on arXiv, demonstrate that quantum processors can complement classical supercomputers to solve specific chemical energetics and stability problems that were previously only accessible through expensive physical experimentation or less accurate classical approximations.

This project leverages the same quantum-centric techniques recently used by Cleveland Clinic to simulate proteins containing 12,635 atoms, extending those methods into the realm of materials science. By integrating CPUs, GPUs, and QPUs, the researchers were able to break down complex problems into quantum circuits, allowing for a more precise identification of how atoms move and interact under the extreme conditions of a fusion reactor. Tom Beck of ORNL highlighted that this multi-pronged discovery cycle, involving seven DOE national labs and various industry partners, is essential for optimizing tritium production—a material so rare in nature that its efficient extraction is a primary barrier to viable fusion power.

Jerry Chow, CTO of Quantum-Centric Supercomputing at IBM, noted that these results provide mounting evidence of quantum computing's role as a practical scientific tool for chemistry and materials science. The ongoing collaboration aims to further refine the workflow by reducing data transfer times between quantum and classical resources and increasing the scale of molecular interactions. Ultimately, the team intends for this quantum-centric workflow to be used directly by the broader fusion energy ecosystem to design and verify new materials, potentially unlocking a sustainable fuel supply for future power plants.