In this Q&A, the THL team shares the technical milestones, engineering challenges, and lessons learned from operating the molten salt pump Shaft Seal Test Facility for over 2,300 hours, placing the lab among a select group of molten salt research leaders.
In the race to realize next-generation nuclear reactors, molten salt reactors (or MSRs) are among the most promising contenders. But to prove their potential, like greater fuel efficiency, enhanced safety, and reduced waste, researchers must go beyond design simulations. They must build and operate real systems under extreme conditions.
That’s exactly what NERS graduate student Shuai Che did.
Che recently completed a groundbreaking experiment at the University of Michigan’s Thermal Hydraulics Laboratory (THL). Over the course of 2,300 hours, Che operated the Shaft SealTest Facility (STTF), a large-scale molten salt experiment designed to evaluate shaft seal performance in high-temperature pump systems. The work placed THL among an elite group of fewer than ten facilities worldwide that have successfully operated fluoride or chloride salts for more than 100 hours using over 10 kilograms of material.
The project was led at Michigan by NERS Assistant Research Scientist Dr. Adam Burak, with Che performing the system’s design, construction, and day-to-day operation. The experiment also benefited from the expertise of NERS Professor Xiaodong Sun, whose research focuses on large-scale thermal-hydraulic systems relevant to advanced reactors. The broader effort was directed by Professor Minghui Chen at the University of New Mexico, with contributions from his student Yuqi Liu, under the U.S. Department of Energy’s Nuclear Energy University Program (NEUP), award number 20-19752. The team also had frequent technical interactions with project partners from High-Temperature Systems Design, Flowserve, and Kairos Power.
In this Q&A with members of the THL team, Che and Burak discuss what it took to design and operate the facility, how the experiment advances molten salt research, and what comes next.
THL Team:
Molten Salt Reactors (MSRs) are among the most promising designs in the race for next-generation nuclear energy. These advanced reactors offer the potential for higher efficiency, better fuel utilization, and reduced nuclear waste—capabilities that could reshape how nuclear power supports a low-carbon energy future.
But realizing that potential takes more than design concepts and simulations. It takes experiments: real hardware, high-temperature salt, and thousands of hours of operation under demanding conditions. And that’s where things get difficult. While global interest in MSRs continues to grow, only a handful of facilities have ever taken the significant step up to large-scale molten salt experiments—especially those sustained over long durations. Most studies use just grams to a few kilograms of salt, often for short campaigns. Operating a system with tens or hundreds of kilograms of salt for thousands of hours is an entirely different challenge—one typically reserved for national laboratories or commercial developers.
The gold standard remains the Molten Salt Reactor Experiment (MSRE) conducted at Oak Ridge National Laboratory in the 1960s. MSRE demonstrated that molten salt reactors could work at scale, and it continues to serve as a benchmark for the field—even more than 50 years later.
Against that historical backdrop, the recent success at the THL marks a significant milestone. The SSTF, the latest system installed in the lab, was developed to study the performance of shaft seals in molten salt pump systems. Over the course of the experiment, SSTF ran continuously for more than 2,300 hours with 32 kilograms of FLiNaK, a common high-temperature salt mixture.
The achievement sends a clear message: even with relatively limited resources, integrated and focused academic teams, with deep technical expertise and strong partnerships, can take on complex engineering challenges that help move advanced nuclear technology forward.
THL Team:
The SSTF was built to investigate shaft seal performance for molten salt pumps under prototypic conditions. The system was centered around a vertical pump test stand featuring a bushing-type mechanical seal and a pump shaft directly exposed to molten salt.
The working fluid was 32 kilograms of FLiNaK, a eutectic mixture of lithium fluoride (LiF), sodium fluoride (NaF), and potassium fluoride (KF), which melts at around 454 °C. During the experiment, the salt was maintained at three operating temperatures—500, 525, and 550 °C—to evaluate seal behavior across a range of relevant thermal conditions.
The goal of this facility was to assess whether this type of mechanical seal could function reliably while exposed to a molten salt vapor. But more important was the facility’s ability to measure gas consumption—the amount of inert cover gas required to maintain a moisture- and oxygen-free environment around the seal.
This metric has major implications for commercial MSR design. In an operating reactor, the cover gas must exclude air and water vapor while also retaining any fission products that may migrate into the gas space, such as fission gases. A higher gas consumption rate means more exhaust to treat, and gas treatment—especially for radioactive species—can become a significant operational expense. The SSTF experiment helped quantify this key parameter under long-duration conditions, offering valuable data for future reactor designers.
THL Team:
Joining that group is a major achievement—not just for THL, but for university-led molten salt research as a whole.
Operating fluoride or chloride salts at scale and for long durations is fundamentally challenging. These salts can be highly corrosive if not handled correctly, demand precise inert atmosphere control, and operate at temperatures that severely limit materials choices. Very few systems worldwide have managed to combine both scale (more than 10 kilograms of salt) and duration (over 1,000 hours), and most of those have been built and run by national laboratories or commercial developers with substantial financial and technical resources.
For the THL to join that group is a strong validation of its technical capabilities. It demonstrates that with the right expertise, discipline, and team, a university lab can operate at the same level—both in complexity and scale—as some of the top molten salt facilities in the world.
It also reflects a broader shift: the molten salt research community has matured to the point that these types of ambitious, long-duration experiments are now within reach for universities. That’s encouraging—it shows we’re moving in the right direction. But it also highlights the growing need to go beyond benchtop tests and short-duration trials. To make real progress, we need larger experiments, running longer, with more components integrated under realistic conditions.
Facilities like SSTF help close that gap. THL’s inclusion in this small but critical group shows that it’s not just participating in this next phase of molten salt development—it’s helping to lead it.
THL Team:
Safety is number one. There are a lot of hazards associated with this type of work, and you need a thorough understanding of what they are and how to mitigate them. That means multiple layers of protection—passive and active systems to handle power loss, gas leaks, or thermal excursions—and the confidence that even if something goes wrong, it won’t lead to catastrophic consequences. Our experience with other salt test facilities DRACS and FLUSTFA was critical in shaping the safety systems and procedures that made this long-duration campaign possible.
The scale is particularly challenging. Operating a high-temperature molten salt system is always difficult, but once you go beyond about 10 kilograms of salt, you’re forced to move outside the glovebox—and that changes everything. On top of that, when you’re running a facility for 2,300 hours, the system needs to be truly stable. In a university setting, that means you have to be able to walk away from it. We simply don’t have the resources to staff it around the clock, so the safety mechanisms needed to accomplish that had to be designed in from the very beginning.
In addition to the safety and scale, stabilizing a 0.9-m-long shaft under high rotation speeds (up to 1,500 RPM) was a challenge during the experiments. The unstable rotating shaft could lead to high vibration and damage to the bushing-type mechanical seal. Two accelerometers were installed to monitor the operating conditions of the shaft. Once a high acceleration is measured, warning emails will be sent to the operator to shut down the experiments. In addition, a set of troubleshooting methods was developed to identify the sources that affected the shaft stability. The repairing process can be completed at high temperatures with the FLiNak salt remaining in the liquid state.
Burak:
Shuai ran the show. He designed the facility, built the facility, and operated it from start to finish. His fingerprints are on every aspect of the system—from the CAD models and heating configuration to the gas management setup and control logic. He handled day-to-day operations throughout the entire 2,300-hour campaign, including troubleshooting, data collection, and keeping the system stable and safe.
Without Shuai, this project simply wouldn’t have happened. Over the course of the project, he developed razor-sharp instincts for running and troubleshooting molten salt systems. His technical skill, persistence, and ownership were absolutely essential to the success of the experiment.
THL Team:
The key accomplishment of this work is the generation of publicly available data on the performance of a shaft seal for molten salt pumps for the first time in the US. Pumps are a critical challenge for MSRs, and these data help developers evaluate which sealing technologies are best suited to their specific designs. Just as importantly, it shows that world-class, high-impact experiments can be executed in a university setting.
THL Team:
The findings are still being worked out—after 2,300 hours of operation as there’s a mountain of data to go through. But one clear lesson was how much the small details matter over long durations. Minor issues that might be tolerable in a short-term test can compound quickly and become major problems over weeks or months. A short experiment might be able to limp to the finish line, but that wasn’t an option here—everything had to be dialed in from the start.
Another key lesson was how to safely perform maintenance and troubleshooting on a high-temperature system. Over a multi-month run, things break, things wear out, and some components need to be adjusted. But cooling down and reheating a molten salt system is extremely time-consuming and introduces its own risks. So we had to develop procedures to safely intervene while the system was still at temperature—a skillset that’s rarely practiced but absolutely critical for future reactor operations.
The design and operation of this test facility is a multi-disciplinary task. The knowledge related only to thermal hydraulics is not sufficient to design the molten salt test facility. The fluoride salt corrosion must be taken into account starting from the design phase of the molten salt test facility. The fluoride salt corrosion is driven by the impurities. Therefore, preparing the high-purity fluoride salts and maintaining the fluoride salt purity during the test are important.
THL Team:
For a system of this complexity, and the length of time we planned to operate it, we had to carefully think through every possible failure scenario—almost like planning for a full-scale reactor. We spent a lot of time identifying what could go wrong and how to mitigate the consequences, even for unlikely events. Key safety measures included wearing molten metal protection suits whenever the system was hot, implementing secondary containment in case of a leak, and having defined protocols in place if elevated levels of HF gas were detected. This is just a small sampling of the many scenarios we planned for to ensure both the system—and more importantly, the people around it—remained safe.
THL Team:
This facility is a platform for much more than a single experiment. It can be used for component testing or expanded into a full molten salt loop. Students working on it gain rare, hands-on experience with real molten salt systems—experience that directly prepares them to contribute at national labs or MSR companies. It also enables experimental validation of models and component designs under realistic conditions, which is essential for building confidence in MSR technologies. Because it operates at a scale relevant to national lab and industry efforts, it creates clear opportunities for collaboration, shared testing, and data benchmarking. It also serves as a valuable intermediate step between bench-scale experiments and full-scale demonstration facilities.
THL Team:
The THL is at the forefront of large-scale molten salt research, and several new projects are currently under consideration for funding. One area of focus is the development of advanced flow sensors specifically designed for molten salt systems—an important step toward closing the gap in flow measurement for MSRs.
THL’s expertise isn’t limited to molten salts. The lab also conducts significant research in LWR thermal hydraulics and operates world-class high-temperature helium facilities. Another exciting direction is in laser-induced breakdown spectroscopy (LIBS). THL has collaborated for years with experts in this area at NERS to measure impurities in helium and sodium systems. With its strong foundation in molten salt experimentation, applying LIBS to salt systems is a natural and promising next step.
Pictured in header: Shuai Che (left) and Muyue Li (right).