TECH

Thermal energy storage could let nuclear plants flex with the grid

The increasing penetration of intermittent renewable energy sources (RES) introduces significant volatility to electric grids worldwide. While RES are crucial for global decarbonization, their variable nature creates a mismatch between energy supply and demand, necessitating flexible grid balancing solutions. However, large-scale baseload units, particularly nuclear power plants (NPPs), are traditionally designed for steady-state operation. To illustrate this challenge, Poland serves as a highly relevant case study as it exemplifies the conflict between inflexible baseload generation and a highly dynamic energy market. RES have a fast-growing share in Poland’s energy mix, reaching 36.6 GW (∼51% of total installed capacity) by the end of October 2025, a 3.5 GW increase over the end of 2024. While this expansion aligns with broader climate goals under the “Fit for 55″ European Union package, which aims for a 55% reduction in greenhouse gas emissions by 2030 compared to 1990 levels, it has also led to electricity grid stability issues. Renewable energy curtailment periods occur more frequently during periods of low demand or constrained transmission capacity – in 2024, 731.4 GWh of renewable energy was curtailed, compared to 74.4 GWh in 2023 and 8.4 GWh in 2022. This trend has intensified further, as curtailed renewable energy totalled 1347 GWh by the end of November 2025.

This results in wasted potential of installed RES capacity, their reduced profitability and higher average carbon dioxide emissions of grid electricity, but also forces coal and lignite power plants, designed for baseload operation, to work as grid stabilisation units, further reducing their efficiency and increasing their carbon footprint. The problem is likely to grow, as the capacity of weather-dependent power sources is expected to increase by 58.8 GW (178%) by 2040.

In 2024, Poland used coal-fired and gas-fired units for grid stabilization. Coal and lignite plants generated 105 TWh, approximately 63% of national electricity demand. Coal-fired power units can adjust their output with ramp rates of 2–4% of installed capacity per minute. They also have a technical minimum of 40–50% of their installed capacity and start-up times of up to 5 h, limiting their ability to balance intermittent RES generation. Coal-fired units are expected to be decommissioned by 2049 (with approximately 30% phased out by 2030), so cleaner alternatives are sought to replace them, such as combined cycle gas turbines (CCGT) and NPPs. Simultaneously, various energy storing solutions are considered to allow the use of renewable energy at a time other than when it was generated. Pumped hydroelectric storage offers large capacities, but is strictly limited by geographical constraints. Hydrogen generation incurs severe round-trip efficiency penalties, while sensible or latent heat storage (e.g., using molten salts) requires costly intermediate heat exchangers and complex plant redesigns. Battery energy storage systems (BESS) provide rapid response, however their operational lifespan is limited due to calendar and cyclic aging. The distinct advantage of adopting a direct-steam accumulator approach is its low-intrusion nature and working fluid compatibility as it utilizes the exact same medium as already used in the secondary circuit of an NPP, avoiding the need for construction of intermediate heat transfer loops.

Modern nuclear power plants, similarly to coal units, are classified as thermal power plants. They use steam as a working medium in the Rankine cycle for power generation and offer comparable ramp rates usually between 2.5–5% of their power capacity per minute (depending on their power output at a given time and progress of the fuel cycle). Their technical minimum can be as low as 20% of their installed capacity (with load cycling operation between 50% and 100% of rated power), but the time to reach full power depends on how they were shut down and their operation before the shutdown (related to reactor poisoning). Nuclear power plants are considered a low-emission energy source with a median lifecycle emission factor of 12 g CO2eq/kWh, comparable to values for onshore wind turbines (11 g CO2eq/kWh) and lower than for utility-scale photovoltaic (PV) installations (48 g CO2eq/kWh).

According to Poland’s energy system projections, generation III + nuclear power plants with a combined installed capacity of 6.2 GW will be introduced by 2040. NPPs are anticipated to operate predominantly at baseload, with minimal grid stabilization roles, due to high capital costs and concerns about possible accelerated component degradation from frequent load adjustments. Periodic shortfalls in renewable energy source (RES) generation will be addressed through three primary measures: 12.9 GW of storage systems, 0.8 GW of gas-fired peaking plants, and a significant overcapacity of RES installations (90 GW) relative to projected average demand (35 GW).

The feasibility of using variations in the power output of the plant's nuclear reactor has been explored for the Slovenian power grid – the results demonstrate promising capability to compensate for daily solar power fluctuations, but also highlight control challenges such as managing boron concentration and control rod movements, potentially affecting reactor stability and waste generation. Another, revenue-oriented analysis for the U.S. Southwest power grid showed that flexible nuclear power plant operation allows for higher integration of wind and solar resources and provides frequency regulation, therefore improving the profitability of nuclear plants.

Integrating thermal energy storage (TES) with NPPs could improve operational flexibility to address grid demand fluctuations or RES oversupply, while maintaining stable reactor output and avoiding excessive load variations. This method leverages NPPs as dispatchable power sources by stabilizing thermal output through secondary steam circuit heat storage, decoupling electricity generation from direct reactor operation.

Nuclear power plants were engineered to do one thing exceptionally well: generate electricity at a steady rate, around the clock. That design philosophy made sense for decades, but it is increasingly at odds with a grid that now shares space with solar and wind resources whose output rises and falls with the weather and the time of day. Researchers at the University of Central Florida’s Center for Advanced Turbomachinery and Energy Research (CATER) argue that thermal energy storage, or TES, could be the key to reconciling the two.

In a study published in the Journal of Energy Resources Technology, Part A: Sustainable and Renewable Energy, the CATER team paired thermodynamic modeling with an economic analysis of how nuclear plants could operate in a shifting energy market. The paper, titled “Molten Salt Reactor Dispatch Optimization via Thermal Energy Storage: Coupled Thermodynamic Modeling and Rolling-Horizon Economic Analysis,” examines how a molten salt reactor equipped with TES could be dispatched more flexibly than conventional nuclear plants allow. The research team included doctoral research assistants Augusto Delavald Marques and Abhilash Prasad, postdoctoral scholar Shinjan Ghosh, associate professor Erik Fernandez, alumnus Marcel Otto, and CATER director Jayanta Kapat.

The core problem the researchers set out to address is a mismatch between reactor design and grid demand. “Nuclear power needs to be flexible while nuclear reactors operate at a steady pace because grid demand varies throughout the day,” Marques says. “This is a challenge for power plants because they must adjust quickly to meet grid demand.”

Traditional nuclear plants are not built to cycle rapidly on and off. When demand drops, operators face an unattractive choice: deliberately ramp down output, which can place mechanical stress on equipment, or keep producing electricity that the grid does not currently need. Both options carry costs, and both become more frequent as weather-dependent renewables claim a larger share of generation.

TES offers a third path. Instead of forcing the reactor itself to follow demand, a storage system can absorb excess heat produced by the reactor and regulate how much of that heat is delivered to the power conversion side of the plant. When electricity demand is low, heat accumulates in storage; when demand climbs, the stored thermal energy can be drawn down to run the plant’s generation equipment.

The reactor keeps operating at the steady pace it was designed for, while the plant’s electrical output flexes to match the grid. The flexibility has a financial dimension as well. Electricity prices fluctuate through the day as supply and demand shift, and the researchers’ rolling-horizon economic analysis was designed to capture how dispatch decisions play out under those changing conditions.

“By storing the energy when demand and prices are low and using it when demand and prices increase, it can increase the value of the electricity produced,” Ghosh says. In effect, TES lets a nuclear plant sell its output when it is worth the most rather than when the reactor happens to produce it. The study’s focus on molten salt reactors reflects where the researchers see the opportunity.

“As renewable power generation grows, the need for flexibility in nuclear plants has become more important,” Marques says. “The deployment of next-generation nuclear reactors in the US over the next few years, such as molten salt reactors, is an opportunity to incorporate TES and improve the integration of nuclear and renewable energy on the same grid.”

It is worth emphasizing what the study does and does not show. The findings come from coupled thermodynamic modeling and economic analysis, not from an operating plant, so the results represent model-based projections of how a TES-equipped molten salt reactor could be dispatched and how such dispatch might affect revenue. Real-world performance would depend on factors the models necessarily simplify.

For a country investing in both advanced reactors and large volumes of solar and wind, the CATER team’s analysis suggests that the heat leaving a reactor does not have to become electricity at the exact moment it is produced. Stored as thermal energy and released on schedule, it could give nuclear power the dispatch flexibility that modern grids increasingly demand, without asking reactors to do something they were never designed to do.

The integration of Thermal Energy Storage (TES) systems with nuclear power plants is gaining momentum within the scientific community. Researchers from institutions such as the Center for Advanced Turbomachinery and Energy Research (CATER) at the University of Central Florida (UCF), MIT, and various national laboratories have been exploring the use of these technologies to revolutionize the role of nuclear energy in modern power grids.

💡 The problem with traditional nuclear energy...Conventional nuclear power plants are designed to operate under a constant-load (baseload) regime. The reactor core functions most safely and economically when it generates a continuous, stable flow of heat.

However, as the share of intermittent renewable sources (such as solar and wind) in the energy mix grows, electricity demand and prices fluctuate significantly:

• Damaging cycles: Forcing a nuclear reactor to rapidly ramp its power output up or down—a process known as "load following"—subjects equipment to thermal stress and shortens its operational lifespan.

• Negative prices: During periods of peak renewable energy production, nuclear power plants are often forced to sell electricity at very low—or even negative—prices, as they cannot be shut down quickly.

Integrating Thermal Energy Storage (TES) with nuclear power plants allows steady-state nuclear reactors to adapt to fluctuating grid demands and high levels of renewable energy without constantly ramping their internal heat production.

How nuclear-TES integration works:

• Decoupling Generation: Traditional nuclear plants struggle to cycle rapidly on and off. TES acts as a thermal buffer in the secondary steam circuit or heat loop, separating continuous reactor operation from real-time electricity generation.

• Absorbing Excess Heat: During periods of low electricity demand or high solar and wind generation, surplus heat or power is diverted to charge the thermal storage medium (such as molten salt or steam accumulators).

• Releasing Stored Power: During peak demand hours or when renewable output drops, the stored thermal energy is sent to turbines to generate extra electricity alongside the reactor's baseline output.

Key benefits:

• Reactor Stability: Keeps the nuclear core running at a constant, optimal output, avoiding mechanical stress from frequent load-following.

• Economic Value: Allows plants to store energy when electricity prices are low and dispatch it when prices spike, improving financial returns.

• Renewable Integration: Helps grids absorb variable wind and solar energy, preventing renewable curtailment and acting as a dispatchable low-carbon resource.

mundophone