Research supported through the new award began in July 2026, with UT’s work centered on the discovery, production, and characterization of advanced scintillators. The university’s effort is led by Mariya Zhuravleva, a professor in the Department of Materials Science and Engineering, with joint Materials Science and Engineering–Nuclear Engineering Research Professor Chuck Melcher and Nuclear Engineering Assistant Professor Sandra Bogetic serving as co-principal investigators. Their research focuses on materials that emit light when they absorb ionizing radiation. That light can be measured and converted into information about the energy, intensity, and sometimes the origin of incoming radiation, making scintillators essential components in systems used for nuclear monitoring, medical imaging, high-energy physics, industrial inspection, and homeland security.

Scintillators are deceptively sophisticated materials. When gamma rays, X-rays, neutrons, or other energetic particles enter a scintillator, they transfer energy to the material’s electrons. The resulting excited states return toward equilibrium by releasing visible or ultraviolet photons. A detector then captures those photons and transforms them into an electrical signal. The quality of that signal depends on several properties, including light yield, energy resolution, decay time, radiation resistance, density, transparency, and the ability to distinguish one type of radiation from another. An ideal scintillator would be bright, fast, stable, affordable, easy to manufacture, and capable of delivering highly precise measurements under difficult conditions. Researchers at UT aim to identify materials that move closer to that combination and could eventually outperform today’s leading detector crystals.

Much of the work will rely on the University of Tennessee’s Scintillation Materials Research Center, directed by Melcher. The center has developed a global reputation for growing and studying crystals whose internal structure determines how efficiently they interact with radiation. Its facilities support methods including the Czochralski process, in which a crystal is pulled from a precisely controlled molten material; the Bridgman technique, which solidifies a melt through a carefully managed temperature gradient; and micro-pulling-down growth, a method capable of producing small-diameter crystals rapidly while using relatively little material. Each approach offers different advantages for controlling composition, defects, size, and scalability. Those factors are critical because a promising scintillator is not useful as a detector technology unless researchers can consistently produce crystals with uniform performance.

The UT team will combine crystal growth with advanced characterization to understand how microscopic defects and chemical substitutions affect detector behavior. Small changes in a crystal’s composition can alter the energy levels available to electrons, influencing how much light is produced and how quickly it is emitted. Trace impurities may act as activators, creating efficient pathways for light generation, while unwanted defects can trap charge carriers, reduce brightness, blur energy measurements, or cause signals to persist after the radiation event has ended. By correlating a material’s atomic structure with its optical and radiation-response properties, researchers can determine why one formulation succeeds while another fails. This approach transforms scintillator development from trial and error into a more systematic search for materials with precisely engineered performance.

The research will also address a central challenge in materials science: scaling laboratory discoveries into practical manufacturing processes. A crystal that performs exceptionally well in a small experiment may be difficult to grow in larger dimensions, vulnerable to cracking, chemically unstable, or too expensive to produce. The UT researchers will therefore investigate not only new compositions but also the processing conditions needed to manufacture high-performance scintillators reliably. Temperature control, melt chemistry, growth speed, atmosphere, post-growth treatment, cutting, polishing, and encapsulation can all affect the final detector. Establishing repeatable production methods will be essential if emerging materials are to replace current state-of-the-art scintillators in real-world radiation-detection systems.

Collaboration beyond Knoxville will broaden that effort. UT’s scintillation researchers will work with scientists at Los Alamos National Laboratory and Lawrence Berkeley National Laboratory to develop processing strategies that can optimize the performance and manufacturability of promising materials. National laboratories contribute specialized equipment, radiation sources, computational tools, and expertise in applications ranging from nuclear safeguards to defense technology. The partnership also allows researchers to test materials against demanding operational requirements rather than evaluating them only under controlled laboratory conditions. According to the investigators, the collaboration will support advances in detector materials for radiation monitoring and nuclear energy systems, where faster, more sensitive, and more selective detectors could improve the ability to identify radioactive materials and track complex nuclear processes.

The scientific program is paired with an ambitious educational mission designed to expand the national pipeline of nuclear materials specialists. UT, Oak Ridge National Laboratory, and the Air Force Institute of Technology jointly organized the first Scintillator Summer School in 2025, bringing graduate students from universities across the United States to Tennessee for intensive instruction. The program is scheduled to return in August with continued support from the Department of Energy. Participants receive hands-on experience in synthesis, crystal growth, scintillation measurements, and detector characterization, while also learning how materials research connects to nuclear security and energy applications. Training in these techniques is uncommon because it requires access to specialized furnaces, radiation sources, optical instruments, and experienced mentors. By exposing students to both fundamental science and practical measurement challenges, the program is intended to prepare researchers who can move between universities, national laboratories, industry, and government agencies.

The NSSC grant arrives as governments and research institutions confront increasingly complex demands for radiation detection. Nuclear security systems must identify illicit or accidental releases of radioactive materials while minimizing false alarms. Nuclear energy technologies require sensors that can operate in intense radiation fields, elevated temperatures, and chemically aggressive environments. Medical and scientific instruments depend on detectors that can measure faint signals with exceptional timing and energy precision. In each case, the detector’s performance is limited by the material at its core. UT’s researchers believe that progress in scintillator chemistry and crystal engineering can enable capabilities that are difficult or impossible with existing materials. By linking fundamental discoveries to scalable processing, national laboratory partnerships, and specialized workforce training, the new five-year effort positions Tennessee at the center of a broader campaign to modernize the science of radiation detection.

Subject of Research: Advanced scintillator materials, crystal growth, radiation detection, nuclear science, nuclear security, and materials engineering.

Article Title: $25 Million Nuclear Security Grant Drives Search for Next-Generation Radiation Detector Materials

Web References: https://www.energy.gov/nnsa/national-nuclear-security-administration ; https://tickle.utk.edu/mse/faculty/mariya-zhuravleva/ ; https://ne.utk.edu/people/charles-l-melcher/ ; https://ne.utk.edu/people/sandra-bogetic/

References: University of Tennessee, Knoxville; Nuclear Science and Security Consortium; U.S. Department of Energy’s National Nuclear Security Administration; Scintillation Materials Research Center; Los Alamos National Laboratory; Lawrence Berkeley National Laboratory.

Image Credits: University of Tennessee

Keywords: Nuclear engineering, radiation detection, scintillators, crystal growth, materials science, nuclear security, nuclear energy, nonproliferation, detector technology, University of Tennessee, National Nuclear Security Administration, Nuclear Science and Security Consortium

The Genesis Mission was created through Executive Order 14363 in November 2025 with the goal of doubling American scientific productivity within a decade. Its proposed infrastructure links the Department of Energy’s 17 national laboratories with some of the nation’s most powerful supercomputers, advanced AI models, quantum platforms and experimental facilities. The first funding round totals $293 million and represents the initial commitment toward an estimated $5 billion investment. Rather than treating artificial intelligence as a stand-alone computational tool, the initiative is designed to place AI inside scientific workflows, where algorithms can help formulate hypotheses, control experiments, identify physical patterns and optimize technologies that are too complex to manage through conventional methods.

One of the OU-led projects is focused on enhanced geothermal systems, a technology that could substantially expand access to reliable, low-carbon electricity. Ahmad Ghassemi, a McCasland Chair Professor of petroleum engineering in OU’s Mewbourne College of Earth and Energy, will lead the project “AI-Enabled Rapid Analysis and Control of EGS Stimulation Using Machine Learning & Physics-Based Hydraulic Fracture/Natural Fracture Interaction Modeling.” Jie Cao of OU’s School of Computer Science and Hao Hu of the School of Geosciences will support the effort. The project has received $734,129 in Phase I funding and is aimed at improving the way engineers create underground heat reservoirs by combining machine learning with detailed models of rock deformation and fracture behavior.

Enhanced geothermal systems are designed to extract heat from hot rock formations that do not naturally contain enough permeability for fluid circulation. Engineers inject fluid deep underground to open or reactivate fractures, creating pathways through which water can move, absorb heat and return to the surface. The process is difficult to control because the subsurface is heterogeneous: rock layers vary in strength, existing fractures can redirect fluid, and small changes in pressure may produce unexpected fracture networks. Ghassemi’s team will use AI to interpret data generated during stimulation while physics-based hydraulic-fracture and natural-fracture models constrain the algorithms. This hybrid approach could allow operators to distinguish useful permeability from unwanted fluid leakage and adjust injection strategies in near real time. The Department of Energy estimates that enhanced geothermal systems could eventually provide approximately 90 gigawatts of capacity nationwide.

“Plentiful energy can become available from enhanced geothermal systems,” Ghassemi said. “The key is fracturing rocks at great depths to create permeability, which experience has shown to be a challenging, complex process.” By integrating geomechanics, fracture mechanics and machine learning, the project seeks to turn underground reservoir creation from a largely empirical procedure into a more predictable and controllable engineering discipline. If successful, the system could help reduce drilling risk, improve the longevity of geothermal wells and make it possible to develop resources in regions previously considered unsuitable for conventional geothermal power. Because geothermal plants can operate continuously, unlike intermittent wind and solar facilities, improved subsurface control could also strengthen the reliability of future energy networks.

The second OU-led award is directed toward quantum computing. Grant Biedermann, Homer L. Dodge Endowed Chair and associate professor of physics in OU’s Dodge Family College of Arts and Sciences, will lead “AI-Driven Control and Optimization of QuDit Platforms in Rydberg Arrays.” The project has been awarded $651,891. It will investigate how artificial intelligence can control arrays of individually trapped atoms, each of which can serve as a quantum information element with more than two accessible states. These systems are known as qudits, in contrast to conventional qubits, which encode information using two states. A qudit can potentially store more information per physical element, although controlling its many energy levels introduces additional technical challenges.

The platform uses Rydberg atoms, whose outermost electrons are excited to very high-energy states. In this condition, atoms become extremely sensitive to one another over comparatively long distances, allowing researchers to engineer interactions that can implement quantum logic. Optical tweezers—tightly focused laser beams—hold individual atoms in precisely arranged arrays. Yet the same sensitivity that makes Rydberg systems powerful also makes them difficult to operate. Laser intensity fluctuations, atom loss, imperfect positioning, unwanted interactions and decoherence can all degrade performance. Biedermann’s team plans to use AI as a real-time control layer that can learn from measurements, predict system behavior and optimize experimental parameters faster than a human operator or a fixed control sequence.

The work connects to the Genesis Mission challenge “Discovering Quantum Algorithms with AI” and brings together researchers from OU, Oklahoma State University and Los Alamos National Laboratory. Biedermann said the collaboration includes Los Alamos scientists Martin, Zlotnik and Meier, as well as Bilitewski of Oklahoma State University, whose research focuses on atom-based qudit simulation. The broader objective is not simply to make quantum hardware run more efficiently, but to create a feedback loop between algorithms and physical devices. AI could search for quantum operations suited to the specific imperfections of an experimental platform, while the hardware could provide data that guides the design of new algorithms. Such adaptive control may become essential as quantum processors grow in size and complexity.

A third OU scientist, Xuguang Wang, Robert Lowery Chair Professor and Presidential Research Professor in the School of Meteorology, is participating in a project led by Lawrence Livermore National Laboratory. The project, “Scalable Twin for Intelligent Turbulence and Cloud Heuristics,” or STITCH, will use AI to model turbulence and clouds for Earth-system prediction. Turbulence occurs across a vast range of scales, from microscopic eddies to atmospheric systems spanning kilometers, while clouds form, evolve and dissipate through tightly coupled interactions involving moisture, radiation, temperature and airflow. Because numerical weather and climate models cannot resolve every relevant process directly, they rely on parameterizations—approximations that represent unresolved physics. STITCH aims to develop more intelligent and scalable methods for these difficult calculations.

The project addresses the Genesis Mission challenge “Predicting U.S. Water for Energy,” an issue with direct consequences for the national power system. Water availability influences hydropower production and affects the cooling systems used by thermal and nuclear power plants. More accurate predictions of clouds, precipitation and atmospheric turbulence could help energy operators plan for changing water supplies, extreme weather and regional demand. The combined OU projects illustrate how the Genesis Mission is connecting disciplines that are often treated separately: petroleum engineering with computer science, atomic physics with machine learning, and meteorology with energy planning. OU Vice President for Research and Partnerships Matt Hulver said the university’s researchers were selected because of the strength and relevance of their work, emphasizing that energy dominance, discovery science and national security are deeply interconnected.

Together, the projects represent a shift toward AI-augmented science in which algorithms do more than analyze completed experiments. In geothermal engineering, machine learning may help steer physical processes kilometers beneath the surface. In quantum computing, it may continuously tune lasers and atomic interactions while a processor operates. In atmospheric science, it may learn improved representations of physical processes that conventional models cannot fully resolve. The central challenge will be ensuring that AI systems remain scientifically reliable, interpretable and stable when confronted with conditions outside their training data. By combining data-driven methods with physical laws, laboratory measurements and high-performance computing, the OU teams hope to build systems that are not only faster, but also capable of producing discoveries and technologies that can be tested in the real world.

Subject of Research: Artificial intelligence applications in enhanced geothermal systems, quantum computing with Rydberg atom qudit arrays, and AI-based turbulence and cloud modeling for Earth-system prediction.

Article Title: University of Oklahoma Teams Bring AI to Geothermal Energy, Quantum Computing and Climate Prediction

Web References: https://www.ou.edu/mcee/mpge/people/faculty/ahmad-ghassemi ; https://www.ou.edu/cas/physics-astronomy/people/directory/faculty/grant-biedermann ; https://www.ou.edu/ags/meteorology/people/faculty/xuguang-wang ; https://www.energy.gov/undersecretaryforscience/genesis-mission/genesis-mission ; https://www.energy.gov/sites/default/files/2026-07/GM-RFA-Awards-List.pdf

References: University of Oklahoma; U.S. Department of Energy Genesis Mission; Lawrence Livermore National Laboratory; Los Alamos National Laboratory.

Image Credits: Travis Caperton/University of Oklahoma

Keywords

Artificial intelligence, Genesis Mission, University of Oklahoma, enhanced geothermal systems, geothermal energy, quantum computing, Rydberg atoms, qudits, machine learning, turbulence modeling, cloud prediction, Earth-system science, Department of Energy, discovery science, energy technology