Abstract

A new design paradigm for nuclear power plants is needed to complement the increasing adoption of low marginal cost variable renewable energy resources. The situation is reflected in the wholesale electricity price–duration curve with four distinct economic opportunities: (a) a hundred or so hours per year of high-value peaking power; (b) about 4000–5000 h of moderate electric prices; (c) about 2000 h per year when renewables set the marginal price at or near zero; and (d) about 1000 h of flexible ramping between the b and c regions. The current approach to the low-carbon energy transition reduces the need for baseload power and requires curtailment of conventional, nuclear, and even renewable generation, decreasing their capacity factors and increasing their fixed charges for electricity generation. Flexible low-carbon dispatchable power plants capable of daily cycling along with storage and time shifting of low-cost nondispatchable renewable power will be needed. Although nuclear power plants have demonstrated load-following capability, cycling can be limited by reactor kinetics (xenon poisoning) as well as by thermal stresses and fatigue considerations in the steam cycle. Storage of nuclear heat is hampered by the relatively low operating temperatures of existing nuclear reactors (but not advanced reactors) that lowers thermal to electric conversion efficiency, which in turn increases the required quantity of storage medium and the cost of storage. The quantity of storage medium can be reduced by integration of thermal energy storage with high-grade heat as in the liquid salt combined cycle (LSCC). The LSCC uses high-temperature gas turbine exhaust heat to increase the electricity output per unit of storage medium, uses the stored energy to add operating flexibility to a bottoming steam cycle, and substantially reduces the fuel heat rate. The low fuel heat rate improves economic competitiveness compared to alternative gas turbine-based power plants, especially when burning expensive fuels such as hydrogen. LSCC could be coupled to a nuclear power plant for time shifting both nuclear and renewable electricity and could support high utilization of a co-located hydrogen electrolysis plant. Further cost reduction could be achieved by using solid media for thermal energy storage, with the liquid salt used as a heat transfer medium.

1 Introduction

Because of their high-capital costs and low fuel costs, the traditional role of nuclear power plants has been the production of baseload electricity, while fossil plants with lower capital cost but higher fuel costs provided variable electricity to match demand. In a few countries, such as France, nuclear power plants also provided variable electricity to the grid. With the increasing role of wind and solar generation, there is a need for flexible, dispatchable generation to provide (1) a replacement for the variable electricity provided by fossil-fuel plants and (2) assured generation independent of wind or solar conditions.

In the summer of 2020, the Massachusetts Institute of Technology, Idaho National Laboratory, and the Electric Power Research Institute held a workshop on Separating Nuclear Reactors from the Power Block with Heat Storage: A New Power Plant Design Paradigm [1]. The authors presented papers on alternative energy storage options, motivated by the increasing abundance of zero-marginal cost variable renewable energy (VRE) as the preferred resource for decarbonizing the electric grid. A major interest at the workshop was the potential to couple energy storage with nuclear power to overcome the challenge posed by decarbonization:

  • lack of coincidence of VRE with peak demand and VRE’s limited dispatchability,

  • displacement of higher marginal cost generation, including baseload nuclear power, resulting in lower capacity factors and potentially stranded assets,

  • the need for flexible capacity to reliably back up VRE and serve peak load, and

  • the cost of alternatives including energy storage, renewable fuels, and carbon capture.

Two general approaches to integration of energy storage with nuclear power plants were discussed.

  • Thermal energy storage of either nuclear heat or nuclear electricity converted to heat, with stored energy subsequently used in a thermal power cycle. Several advanced nuclear power systems, such as the Terrapower/GE Hitachi Nuclear Energy Natrium reactor, are planning to install nitrate salt heat-storage tanks between the reactor and the steam cycle power block to enable baseload reactor operations with variable electricity to the grid. These thermal energy storage systems would be derived from concentrated solar power experience and use off-the-shelf steam cycles.

  • Chemical energy storage, for example, thermal or electrical electrolysis of water to produce hydrogen, ammonia, or synthetic fuels that could be converted back to electricity.

However, with expected future electricity markets, there are large economic incentives for lower cost heat-storage systems coupled to advanced power cycles designed to better match electricity systems with high levels of installed solar and wind capacity. This article proposes a novel nuclear hybrid system that includes both thermal and chemical energy storage, using the liquid salt combined cycle (LSCC) technology to discharge stored energy as electricity and economically provide flexible, zero-carbon, dispatchable power.

2 Challenges of Energy Transition

The transition to decarbonized energy resources involves a range of alternatives and trade-offs related to the variability of renewable energy, energy markets, flexibility, and reliability. Integration of nuclear power, energy storage, and a combustion-based discharge cycle introduces additional technical and integration challenges along with the need to balance capital and operating costs in developing a new power plant design paradigm.

2.1 The Challenge of Renewable Energy Variability.

Each day, as solar production begins, the CAISO (California Independent System Operator) grid experiences a large and rapid drop in net load that forces efficient natural gas combined cycle plants offline. As solar production wanes in the late afternoon, dispatchable plants must rapidly come back online. The two related grid operational issues—overgeneration with curtailment and steep ramps in the load served by CAISO—are phenomena that create the now famous “Duck Curve” [2] shown in Fig. 1. The Duck Curve highlights the challenges of overgeneration and ramping faced by the operator of the California Grid when 33% of the annual electricity was mandated from renewable resources, notably solar and wind.

Fig. 1
CAISO Duck Curve showing changes in net load with solar PV energy additions
Fig. 1
CAISO Duck Curve showing changes in net load with solar PV energy additions
Close modal

National, regional, and local governments throughout the world have established laws, mandates, and targets for still larger amounts of renewable electricity. For example, in 2018, California enacted SB-100 [3], which mandates that 60% of retail electricity (i.e., electricity delivered to customers) be from renewable resources, which was defined to exclude large hydro and nuclear power. As VRE increases, the “neck” of the Duck gets steeper, and the “belly” of the Duck gets deeper so that the need for baseload generation is effectively eliminated, as illustrated in Fig. 2 for a 50% renewable portfolio. Under such a scenario, power plants would not operate to serve baseload, but would be dispatched as a complement to renewable generation.

Fig. 2
Projected net load curve in California under a 50% renewable portfolio standard [4].
Fig. 2
Projected net load curve in California under a 50% renewable portfolio standard [4].
Close modal

Variable renewable resources are so abundant and inexpensive at certain times in many places that after displacing fossil generation, the renewables themselves must also be curtailed. For example, on April 21, 2019, before COVID-related load reductions, the CAISO curtailed almost 32 GWh of solar energy because generation exceeded demand [5]. As California further increases its renewable portfolio to 60% in 2030, additional curtailment will be needed. During 2021, CAISO curtailed 1504 GWh of utility-scale solar and wind generation [6], and in the first half of 2022, CAISO curtailed 2147 GWh.1 For both economic and environmental reasons, it is essential to store this otherwise curtailed energy for use when the natural variability of renewable resources demands backup from dispatchable generation.

2.2 The Challenge of Competitive Energy Markets.

As load increases, generation is generally dispatched in order of their marginal cost of energy (MCOE), from low to high. In the general form, covering thermal generation and energy storage, MCOE is expressed as follows:
(1)
where fuel heat rate is the ratio of fuel input to electricity output; fuel cost is the cost of fuel; stored energy rate2 is the ratio of electricity charged into storage to electricity output, the inverse of round-trip efficiency; electricity cost is the weighted average price of electricity during charging; and variable O&M reflects the marginal operations and maintenance cost proportional to electricity output.

For traditional thermal generation, the stored energy rate is zero, whereas for traditional energy storage, the fuel heat rate is zero. Hybrid energy storage described in this article uses nonzero rates of fuel and stored energy.

Because renewable resources have no fuel cost component, their MCOE is near zero, so renewables are dispatched before conventional power resources that incur fuel cost and sometimes significant operations and maintenance (O&M) expenses. High O&M expenses would cause, a zero-carbon nuclear resource to dispatch after the VRE, which has implications for capacity factor, plant design, and economics. Since nuclear resources have cycling constraints, they would need to be 'must take' resources, be idle during peak renewable seasons, or be retired.

The price of electricity in wholesale markets is set by the highest cost of generation needed to serve load. The price–duration curve in Fig. 3 shows electricity prices in a hypothetical power market with moderate quantities of wind and solar VRE.3 The curve illustrates four distinct economic opportunities for generating units:

  • High-value peaking power for several hundred hours per year.

  • Moderate value plateau for about 5000 h per year.

  • Ramping need for about 1000 h per year.

  • Zero-marginal cost for about 2000 h per year.

Fig. 3
Predicted price duration curve for an energy market with significant solar and wind resources, based on marginal cost of energy [7].
Fig. 3
Predicted price duration curve for an energy market with significant solar and wind resources, based on marginal cost of energy [7].
Close modal

This market environment illustrates the need for a low or zero-carbon power plant design paradigm capable of these missions:

  • Daily curtailment due to renewable over generation or uneconomic operation, with the potential to store excess renewable energy (region d).

  • Fast startup and flexible ramping to complement waning VRE (region c).

  • Competitive economic dispatch when marginal price is set by efficient dispatchable resources (region b).

  • Peaking capability (region a) to capture high value hours.

Consistent with Fig. 2, there is no need for traditional baseload generation, which with fuel and variable O&M costs, would operate at a loss in region d.

2.3 The Need for Flexibility and Reliability.

In competitive markets with high VRE penetration, nuclear, coal, and combined cycle power plants are uneconomic for many hours per year, during which time they may be run back to minimum load or shut down. Some combined cycle units now operate primarily in regions a and b of Fig. 3, with reduced capacity factors and frequent cycling. Restart of steam-based units can be lengthy and costly, impairs emissions compliance and heat rate, and may be impractical for large coal and nuclear units. Energy storage could increase the flexibility of thermal units by maintaining steam systems in a hot or warm standby condition, without burning fuel, to facilitate rapid return to power generation. Nevertheless, reduced capacity factor makes it challenging to justify investment in upgrades or new plants.

Regions a and c are characterized by fast startup, flexible ramping, and load-following for which simple cycle gas turbines, reciprocating internal combustion engines, and battery energy storage are suitable. Marginal cost and merit order depend on the cost of fuel or electricity and the efficiency of power production, but whether a plant gets built depends on other factors, including capital cost and greenhouse gas emissions.

2.4 The Capital Cost of New Generation and Storage.

The capital component of the cost of energy is expressed as follows:
where

AF=i(1+i)n(1+i)n1 is the capital recovery amortization factor over n years at the annual interest rate i. CF is the capacity factor, or annual delivered energy divided by the energy that could be produced at continuous full power operation.

For applications with low CF, the capital cost of energy (CCOE) tends to become more important than MCOE in determining an investment decision to build a new plant.

The amortization factor depends on the lifetime of the investment and the expected rate of return. The expected rate of return will be affected by

  • Perceived technology risk including risk of construction delays.

  • Market risk, including the potential for competitors to enter the market and reduce the market price paid for electricity and potentially reduce the anticipated capacity factor.

  • Resource risk, such as the cost and availability of fuel, which could be mitigated by a long-term supply contract. However low-capacity factor and uncertainty about when the plant would operate often results in short-term or spot-market fuel procurement. Recent events in Europe highlight the risk of relying on a single source of fuel supply, even with a long term contract. Energy storage could mitigate this risk by allowing renewable energy to displace and hedge fuel.

  • Investment term, which reflects both the technology’s lifetime and the investment horizon.

Cost and performance estimates for several kinds of dispatchable power plants are listed in Table 1. An investment decision will depend on the overall cost, considering both MCOE and CCOE, and the estimated capacity factor based on the expected market conditions. Capacity factor in turn depends on the MCOE and the competitive market environment, such as the price–duration curve of Fig. 3.

Table 1

Power plant technology comparison prepared by Sargent & Lundy and issued by the U.S. Energy Information Administration [8]

CaseTechnologyPower (MWe)Higher heating value (HHV) Heat rate (MMBtu MWh)Overnight capital cost ($Million/MW)O&M fixed ($/MW-year) and variable ($/MWh)CO2 emissions (m.t./MWh)
5Simple cycle-Aero1059.1241.17516,300 and 4.70.485
6Simple cycle F-class2379.9050.7137000 and 4.50.527
8Combined cycle H-class 1 × 1 × 14186.4311.08414,100 and 2.550.342
9Combined cycle + 90% carbon capture3777.1242.48127,600 and 5.840.038
12SMR60010.6086.19195,000 and 3.000.000
CaseTechnologyPower (MWe)Higher heating value (HHV) Heat rate (MMBtu MWh)Overnight capital cost ($Million/MW)O&M fixed ($/MW-year) and variable ($/MWh)CO2 emissions (m.t./MWh)
5Simple cycle-Aero1059.1241.17516,300 and 4.70.485
6Simple cycle F-class2379.9050.7137000 and 4.50.527
8Combined cycle H-class 1 × 1 × 14186.4311.08414,100 and 2.550.342
9Combined cycle + 90% carbon capture3777.1242.48127,600 and 5.840.038
12SMR60010.6086.19195,000 and 3.000.000

Figure 4 compares the total cost of energy (capital plus marginal) as a function of capacity factor, with natural gas as the fuel source for the simple and combined cycle power plants. This figure illustrates that less-efficient simple cycle gas turbines are more cost effective for low-capacity factor applications, such as peaking power, while more efficient combined cycles have lower cost of energy at high capacity factors. Despite low MCOE of small modular reactors, high-capital cost makes them uncompetitive for both peaking and baseload generation.

Fig. 4
Cost of energy as a function of capacity factor using cost and performance from Table 1. Assumes capital is financed at 8% for 20 years (40 years for case 12—small modular reactor (SMR)). Includes fixed and variable, O&M and assumes fuel at $2.50/MMBtu.
Fig. 4
Cost of energy as a function of capacity factor using cost and performance from Table 1. Assumes capital is financed at 8% for 20 years (40 years for case 12—small modular reactor (SMR)). Includes fixed and variable, O&M and assumes fuel at $2.50/MMBtu.
Close modal

2.5 The Need to Decarbonize.

The need to decarbonize generation creates an opportunity for nuclear power to compete with alternative dispatchable power plants using renewable fuels that are more expensive than natural gas, such as hydrogen, or plants integrated with carbon capture. As shown in Fig. 5, at high capacity factors, nuclear power begins to become competitive with combined cycles burning higher cost hydrogen fuel, but so does a natural gas combined cycle with carbon capture. The price–duration curve shown in Fig. 3 suggests that capacity factors would be limited to 60–70% and would trend lower as the amount of VRE increases. Operating regimes shown in Fig. 2 would challenge the plant design paradigm of both combined cycles with carbon capture and nuclear power plants.

Fig. 5
Overall cost of decarbonized dispatchable energy using costs from Table 1. Assumes natural gas at $3/MMBtu and hydrogen at $1.50/kg.
Fig. 5
Overall cost of decarbonized dispatchable energy using costs from Table 1. Assumes natural gas at $3/MMBtu and hydrogen at $1.50/kg.
Close modal

2.6 The Cost of Hydrogen.

Hydrogen can be produced at an estimated cost of $1.50/kg ($11.15/MMBtu) by steam methane reforming of natural gas, referred to as “Blue” hydrogen when coupled with carbon capture and storage [9]. “Green” hydrogen, produced by water electrolysis using renewable energy, has a projected levelized cost between $1.50 and $5.00/kg in 2050 [10], depending on the scenario, production conditions, and electricity price. Like the cost of generation, the cost of electrolytic hydrogen depends on

  • capital cost of electrolyzers, which is expected to follow experience–curve cost reductions, and the utilization factor (the percent of time the electrolyzer is used).

  • cost of electricity used to drive electrolysis, which depends on the price duration curve as shown in Fig. 3.

  • cost of compression for transport and/or storage, which can be reduced if hydrogen is consumed at the production location.

High-temperature electrolysis (HTE) is in the early stages of commercial deployment [11] and is a near-term option to reduce hydrogen costs. HTE uses steam as a source of high enthalpy water to reduce the electrical energy input and approach 90% hydrogen production efficiency. Steam for electrolysis could be continuously available from thermal energy storage or a nuclear steam supply system and has the potential to halve the cost of hydrogen [12], potentially creating an additional revenue stream for a nuclear power plant.

Reliance on low-cost excess VRE from region d of Fig. 3 as the primary electricity source would necessitate over-building electrolysis capacity. A more cost-effective approach could be the use of baseload nuclear power to produce hydrogen, rationalizing the investment in electrolyzer capital. However, even at high capacity factors, nuclear electricity would have difficulty compete against steam methane reforming with CCS in regions with good carbon dioxide sequestration sites, unless the price of natural gas increases.

2.7 Thermal Energy Storage Challenges.

Thermal energy storage using bulk storage media such as molten salt and well-understood thermal/mechanical systems is subject to Carnot efficiency limits. For example, a steam cycle with a conversion efficiency of about 40% (thermodynamic heat rate of 9000 kJ/kWh) would require a steam temperature greater than 500 °C.

Two-tank molten salt energy storage has been successfully employed in parabolic trough concentrating solar power (CSP) applications [13], operating between about 390 °C and 260 °C, with lower cycle efficiency. At these conditions, carbon steel tanks have been reliably used to store nitrate salt. Recent CSP projects employing central receivers heat molten salt to 565 °C, but hot tank failures have occurred in two projects. Higher temperature systems require stainless-steel tanks and piping, which are more expensive, experience higher thermal stresses, require higher quality and more costly salt to avoid corrosion from impurities, and may also result in thermal decomposition of the salt with nitrogen oxide air emissions.

The cost of storage medium is about $25/kWh-th for molten salt [14], resulting in the marginal capital cost of storage at around $62.50/kWh-e at SER = 2.5. The molten salt thermal energy storage system cost for an 8-h duration 100 MW-e (net) CSP plant is estimated to cost $65,276,000, or $77.845/kWh-e, including salt, tanks, pumps, piping, heat tracing, and a nitrogen ullage system to minimize oxidation.

2.8 Storage Cost Reduction.

It is evident that the storage medium dominates the capital cost of storage, so lower cost storage media would be attractive. The quantity of storage medium is proportional to SER, so increasing the electricity output per unit of stored energy is also critical. Cost reduction strategies for thermal energy storage include:

  • Increasing the hot tank temperature and/or the temperature range across which energy is extracted from storage to allow more energy to be stored per unit of storage medium. As noted earlier, 565 °C salt has already proved challenging; further increases may require unproven materials and fabrication methods.

  • Substituting a lower cost storage medium for the salt, such as sand, gravel, aggregate, or slag. In the two-tank system, molten salt functions as both a storage medium and the heat transfer fluid. Cost reduction strategies that separate the storage and heat transfer functions require a heat transfer fluid and must contend with heat transfer at three interfaces instead of two (heat transfer to/from storage medium in addition to heat transfer from charging source and to discharging sink).

  • Reducing the quantity of storage medium by integration with a thermal power plant as discussed in Sec. 3. Such hybrid storage approaches can improve the coupling of thermal energy storage with thermal generating resources to reduce the quantity of storage medium and the storage temperature.

Partially filling a molten salt storage tank with a lower cost solid fill material to create a thermocline in the tank has been demonstrated to reduce the capital cost about 40% [15]. However, this approach requires tanks with large ratios of height to diameter, which increases hydrostatic pressure and wall thickness. At a large scale, it appears more economical to use conventional large diameter storage tanks without solid fill.

The Crushed Rock Ultra-large Stored Heat (CRUSH) concept (Fig. 6) presented by the second author [16,17] is inspired by leach heaps used for extracting valuable minerals from crushed ore. In the charging mode, hot nitrate salt is sprayed on top of the crushed rock, transfers heat while draining through the crushed rock, and is collected below and returned to the heating source. During discharge, cool salt is sprayed atop the heap to be heated by stored energy while draining through the heap and is returned to the heat working fluid.

Fig. 6
Crushed rock ultra-large stored heat system
Fig. 6
Crushed rock ultra-large stored heat system
Close modal

CRUSH provides a two-dimensional thermocline. Like a storage tank with inert fill, the high-temperature front moves down during charging. If the nitrate salt is not fully cooled when it reaches the drain pan, it is pumped to the next section to heat more rock. The second dimension is along the heap axis, with the inlet spray and outlet collection of heat transfer fluid moving from one end to the other, thereby overcoming the length to diameter challenge. As envisioned, the overall system would be enclosed within an insulated structure to reduce heat losses and replace expensive tanks with lower cost structures. The capital cost of the heat storage could potentially be reduced to $2–4/kWh-th [1,16,17] while retaining the superior heat transfer characteristics of molten salt.

2.9 Storage Integration Issues.

Integration of thermal energy storage with thermal power plants could time-shift uneconomic production to more profitable peaking power applications and mitigate cycling challenges. In principle, excess power in the form of electricity or nuclear heat could be stored thermally and then converted back to electricity via a Rankine cycle using steam or an organic working fluid. Coupling thermal energy storage to thermal power cycles must consider both the charging and discharging conditions. Water reactors produce saturated steam at about 286 °C, close to the freezing point of solar salt (240 °C), so heat transfer would not be practical for charging molten salt. Instead, excess electricity could be stored as high-temperature heat [18] to enable baseload reactors to supply variable electricity to the grid. However, returning the stored energy to a low efficiency nuclear steam cycle would result in a high stored energy rate and consequently require a large quantity of storage medium.

Various reactor concepts using molten salt, liquid sodium, or gaseous coolants are now being advanced to provide the high-grade heat needed to operate high efficiency power cycles. High-temperature reactor coolant could also be directly coupled to thermal energy storage to supplement or supplant electric heaters. Direct heating of storage would also avoid the inefficiencies of heat-to-electricity-to-heat conversions.

3 Hybrid Energy Storage

Hybrid energy storage systems combine thermal energy storage with thermal generation to develop technical, economic, and operating synergies, such as:

  • Reduction of fuel heat rate (kJ/kWh of electric output) by using stored energy to displace fuel.

  • Reduction of stored energy rate (kJ/kWh of electric output) by using gas turbine power and exhaust heat to produce more electricity per unit of stored energy.

  • Reduction of the quantity of storage medium used (kg-storage/kWh of electric output) by coupling thermal storage with waste heat.

The first two synergies can reduce MCOE by displacing fuel that would be required in a conventional thermal plant with less-expensive VRE. SER is commonly less than unity in hybrid energy storage systems that reduce the MCOE.

The third synergy reduces the quantity of storage media and accordingly the size and the cost of the storage reservoir. For thermal energy storage, where the cost of the storage medium can be substantial, reducing the quantity of storage medium per kWh of discharge energy reduces the size and capital cost of storage media and storage tanks.

3.1 Liquid Salt Combined Cycle Hybrid.

The Liquid Salt Combined Cycle shown in Fig. 7 is a patented [1923] hybrid power plant combining conventional generation with thermal energy storage. LSCC integrates electrically heated two-tank molten salt thermal energy storage with combustion and steam turbines in a novel system that enables faster startup, while reducing stored energy rate, fuel heat rate, and the quantity of storage medium required.

Fig. 7
LSCC flow diagram comprises electrically heated liquid salt thermal energy storage system, a steam generator with heating provided by the hot salt. Charging and discharging occur at difference times, with gas turbine exhaust used to superheat steam and economize feedwater. Dashed lines indicate flow paths when the gas turbine is not operating [25].
Fig. 7
LSCC flow diagram comprises electrically heated liquid salt thermal energy storage system, a steam generator with heating provided by the hot salt. Charging and discharging occur at difference times, with gas turbine exhaust used to superheat steam and economize feedwater. Dashed lines indicate flow paths when the gas turbine is not operating [25].
Close modal

In the LSCC approach, evaporative heating duty is removed from the exhaust heat recovery system, which instead provides only sensible heating of water and steam, while the steam is produced from stored heat. This arrangement removes exhaust gas heat transfer constraints that limit steam flow in a conventional combined cycle and increases steam flowrate and power output. Typical power output of an LSCC system is about twice the power output of the combustion turbine, compared to about 1.5× for a conventional triple-pressure plus reheat combined cycle. LSCC also uses lower live steam pressure, which reduces the thermal stresses on exhaust heat recovery components during startup.

Importantly, the lower live steam pressure reduces the steam saturation temperature, which in turn allows more heat to be extracted from the molten salt. Although LSCC can use any thermal storage medium, a low-freezing point eutectic salt (53% potassium nitrate, 7% sodium nitrate, and 40% sodium nitrite), commonly known by the tradename Hitec® Heat Transfer Salt [24] is advantageous. The hot salt is heated to about 425 °C using efficient, flexible, and inexpensive, electric heaters, rather than 565 °C, which is the typical limit for solar salt. Under these conditions, the low-freezing point salt is stable, noncorrosive, nontoxic, nonflammable, and does not degrade with use, no matter how often the system is cycled, or how fast it is charged or discharged.

To appreciate the salt temperature selection of LSCC, it is important to understand that high steam temperature is needed primarily for managing moisture content at the steam turbine exhaust, rather than to improve efficiency. In high-pressure steam cycles, the exhaust moisture content is controlled by reheating the steam after partial expansion (with moisture separation for nuclear steam cycles). By reserving high-temperature gas turbine exhaust for superheating and using stored energy for evaporation at medium pressure, steam flow is maximized, and there is no need for high salt temperature. The lower salt operating temperature is compatible with carbon steel tanks and piping to reduce capital cost and makes it feasible to use low-freezing point salt, instead of solar salt, to reduce operating costs.

Figure 8 shows a conceptual general arrangement for a modular 92 MWe LSCC power plant with 19-h of storage (1746 MWh-e) in two tanks (hot and cold) of customary size. LSCC integration reduces the mass and volume of salt per MWh of AC energy by more than 80% compared to CSP4 plants to reduce the capital cost of storage medium to less than $25/kWh-e.

Fig. 8
General arrangement of LSCC plant: 92 MW × 18-h (1650 MWh) added to an existing LM6000 gas turbine at a two-unit peaking plant
Fig. 8
General arrangement of LSCC plant: 92 MW × 18-h (1650 MWh) added to an existing LM6000 gas turbine at a two-unit peaking plant
Close modal

LSCC can be applied to any gas turbine, including industrial, frame, and aeroderivative types, to boost output, reduce fuel heat rate, and make efficient use of stored thermal energy [26]. Table 2 compares the performance of simple cycle, combined cycle, and LSCC configurations for the General Electric 7FA.04 gas turbine. The LSCC approach displaces fuel to result in lower fuel heat rate and emissions than a triple-pressure reheat combined cycle, despite lower overall thermal efficiency.

Table 2

Performance comparison for F-class combustion turbinea

Net ratingsGE 7FA.04 gas turbine
Simple cycleTriple-pressure reheat combined cycleLiquid salt combined cycle
Power (MW)198.0305.0410.5
Fuel heat rate (MJ/MWh) (HHV)10.2576.6334.349
Stored energy rate (MJ/MWh)003.268
Thermal efficiency (HHV)b35.1%54.3%47.6%
CO2 emissions (m.t./MWh)0.5170.3340.219
Net ratingsGE 7FA.04 gas turbine
Simple cycleTriple-pressure reheat combined cycleLiquid salt combined cycle
Power (MW)198.0305.0410.5
Fuel heat rate (MJ/MWh) (HHV)10.2576.6334.349
Stored energy rate (MJ/MWh)003.268
Thermal efficiency (HHV)b35.1%54.3%47.6%
CO2 emissions (m.t./MWh)0.5170.3340.219
a

Simple and Combined Cycle Performance from Gas Turbine World. LSCC performance from Pintail Power LLC.

b

Thermal efficiency is 3.6 MJ/MWh divided by the sum of the fuel heat rate and stored energy rate.

The estimated installed cost for a 410 MW × 16-h (6560 MWh) LSCC system is $656 million ($100/kWh). When charged with electricity at $10/MWh (region d shown in Fig. 3 plus a transmission access charge) and fueled with hydrogen at $1.50/kg, the LSCC system would be competitive with large combined cycle power plants, while offering flexibility like the simple cycle power plants, as shown in Fig. 9.

Fig. 9
Hydrogen-fueled LSCC technology would be competitive with simple cycle gas turbines and combined cycles using costs from Table 1 and LSCC performance from Table 2. Assumes natural gas at $3/MMBtu and hydrogen at $1.50/kg.
Fig. 9
Hydrogen-fueled LSCC technology would be competitive with simple cycle gas turbines and combined cycles using costs from Table 1 and LSCC performance from Table 2. Assumes natural gas at $3/MMBtu and hydrogen at $1.50/kg.
Close modal

Flexibility is also provided by the charging system, which can rapidly add or drop load to compensate for variable wind and solar resources, and even provide single-cycle fast frequency response using solid-state power controls. Likewise, by inserting a SSS clutch (automatic overrunning clutch) on the steam turbine shaft, its generator could operate as a synchronous condenser to provide VOLT/VAR ancillary services during charging.

3.2 LSCC Nuclear Hybrid.

LSCC could be coupled to a nuclear power plant, as shown in the block diagram of Fig. 10, storing some electricity as thermal energy and some as hydrogen. Instead of using hydrogen produced by steam methane reforming with CCS, a fraction of the nuclear electricity would continuously drive a hydrogen electrolyzer to produce fuel for the LSCC system. When market prices are low, the LSCC would thermally store otherwise uneconomic nuclear power and low-cost renewable power; when prices are high, the LSCC would discharge zero-carbon power. When power needs are especially high, the electrolyzer could be curtailed to deliver additional power.

Fig. 10
Block diagram of annual energy flows for a hybrid power plant incorporating nuclear power, hydrogen electrolyzer, and the LSCC of Fig. 7
Fig. 10
Block diagram of annual energy flows for a hybrid power plant incorporating nuclear power, hydrogen electrolyzer, and the LSCC of Fig. 7
Close modal

Electrolyzers have high-capital costs [28], so this hybrid combination maximizes utilization of high-capital cost equipment (nuclear reactor and electrolysis plant), provides high-value peaking capacity, adds flexibility to support high VRE grids, and complements government and industry initiatives to couple nuclear power with Green Hydrogen.5 Higher electrolysis efficiency would increase the grid electricity delivered from the hybrid system, with a corresponding reduction in cost of energy. Low-cost renewable energy can also be used opportunistically to charge thermal storage and to produce hydrogen using the electrolyzer.

Figure 11 shows a hypothetical power delivery profile for the system shown in Fig. 10. In the example, a baseload nuclear power plant and an electrolyzer operate at 90% capacity factor (7884 h per year), to maximize their economic value and eliminate the need for cycling or load following. Load following is instead provided by the LSCC, which charges from both renewable and nuclear electricity and discharges using electrolytic hydrogen produced by nuclear power. The generation profile is superimposed on the price–duration curve shown in Fig. 3 to illustrate economic dispatch of power to the grid.

  • During the highest value peaking periods, the electrolyzer is turned off to deliver 1029 MW, the combined output of the nuclear and LSCC equipment.

  • In the higher value part of region b, the LSCC delivers 410.5 MW in addition to the 400 MW of nuclear electricity, while 200 MW drives electrolyzer. In the lower value part of region b, the LSCC is idle, while the nuclear plant delivers 400 MW of electricity to the grid and 200 MW to the electrolyzer.

  • Ramping in region c is handled by the LSCC charging system, to keep the nuclear plant and electrolyzer operating at full capacity.

  • In region d, the LSCC system can charge from both the nuclear power plant and the grid. Refueling outages would be scheduled during low-value periods, but the LSCC and electrolyzer could continue to operate, using grid power for charging of thermal and hydrogen storage.

Fig. 11
Hypothetical power output profile for hybrid NSSS/LSCC/electrolyzer power plant
Fig. 11
Hypothetical power output profile for hybrid NSSS/LSCC/electrolyzer power plant
Close modal

The overall cost of energy from the nuclear LSCC hybrid is estimated to be about $117/MWh under the following assumptions:

  • Capital charge factor of 0.068199, derived from a real after-tax weighted average cost of capital of 5.4% and a 30-year cost recovery period, consistent with EIA calculation of levelized cost [27].

  • Electrolyzer capital cost at $625/kW with 70% efficiency and fixed O&M of $25,000/MW.

  • Nuclear power plant cost obtained from Table 1.

  • The LSCC system at $100/kWh with fixed O&M of $30,000/MW.

  • Excess renewable energy sourced at $10/MWh.

Such a nuclear/electrolyzer/LSCC hybrid offers operational flexibility, security of energy supply, and peaking capacity and could be competitive with steam methane reforming at higher natural gas prices if carbon sequestration was not available.

Adding electrolyzers and LSCC technology to an existing nuclear power plant could time-shift low-value electricity and cost-effectively add zero-carbon dispatchable peaking capacity. The U.S. government is supporting demonstrations of several hydrogen electrolysis approaches at nuclear power plants operated by FirstEnergy Solutions, Xcel Energy, and Arizona Public Service. These demonstration projects address not only the technology but also operational and licensing aspects of coupling nuclear reactors and hydrogen production—necessary for hydrogen-fueled LSCC hybrids. Xcel Energy’s Minnesota demonstration site is a transmission hub already coupling a reactor to large wind resources, so it could also host an LSCC hybrid to integrate variable wind generation. If the capital cost of the nuclear unit has been fully amortized, the overall cost of energy from an LSCC-hydrogen-nuclear hybrid with the operating profile of Fig. 11 would be reduced to $42.50/MWh.

There are three other nuclear LSCC hybrid options that improve long-term economics.

  • High-temperature electrolysis. The electrolyzer can be replaced with a high-temperature electrolyzer, now being commercialized, that uses steam as a feedstock to reduce electricity requirements. Steam could be supplied from the nuclear plant or storage while the electricity could be sourced from the nuclear plant or renewables. Xcel Energy’s Minnesota demonstration is the first demonstration of high-temperature electrolysis coupled to a nuclear reactor.

  • High-temperature reactors. For advanced reactors with higher exit temperature working fluid, thermal storage can be charged directly with heat from the reactor to rather than from electricity. As discussed earlier, Terrapower and GE/Hitachi Nuclear plan to deploy their first Natrium reactor in Wyoming this decade with the traditional two-tank nitrate heat-storage system for variable electricity to the grid with baseload operations.

  • Seasonal swings or extended weather conditions. There are seasonal swings in renewables production and periods of extended low outputs. Nuclear power provides heat, electricity, and hydrogen through those events. There is the option to add low-cost salt heaters using hydrogen or biofuels to provide added heat to maintain the full capability of the LSCC system.

4 Conclusion

A new design paradigm integrates nuclear power with hydrogen electrolysis and thermal energy storage to improve integration with variable renewable energy. The hybrid liquid salt combined cycle uses liquid salt as a thermal energy storage medium in a novel arrangement that reduces fuel heat rate, reduces the cost of storage, and enhances dispatchability of thermal power plants. Reduction of the fuel heat rate by the LSCC hybrid makes the use of hydrogen fuel for decarbonization more economical. When coupled to a nuclear power plant, the LSCC can provide a thermal sink for storing electricity that is uneconomical compared to low marginal cost renewable energy. The nuclear power plant could also be coupled to a hydrogen electrolyzer to provide a secure fuel source for the LSCC system.

The LSCC can be built today with little technical risk because it uses off-the-shelf systems. At the same time, there is the potential for substantial future reductions in cost and improvements in performance. CRUSH has the lowest projected cost of any existing or proposed heat-storage system but is in the early stage of development and not yet ready for deployment with LSCC.

Footnotes

1

EnergyGPS, The State of CAISO Curtailments, July 1, 2022, Newsletter—The State of CAISO Curtailments (energygps.com)

2

For energy trading, it is convenient to express the stored energy rate in relation to the commodity cost. For engineering efficiency, this can be converted to GJ/MWh, kJ/kWh, MMBtu/MWh, or other equivalent units.

3

According to Ref. 7, this curve was developed for the PG&E balancing area of CAISO using a production cost model with 8.7% of load served by wind and 37% served by solar.

4

Twelve tanks (six hot and six cold) of comparable size are employed at the Solana Generating Station (https://www.powermag.com/solana-generating-station-maricopa-county-arizona/) to deliver 1500 MWh-e (net).

Acknowledgment

Dr. Conlon’s work was partly supported by NETL under Contract DE-FE0032016. The valuable review and comments by Dr. Mark A. Vince are gratefully acknowledged. Dr. Forsberg’s works was partly supported by the INL National Universities Consortium (NUC) Program under DOE Idaho Operations Office Contract DE-AC07-05ID14517.

Conflict of Interest

Conflicts of interest have been declared to the Editor and will be included in a Conflict of Interest Declaration section of the final paper. This article does not include research in which human participants were involved. Informed consent not applicable. This article does not include any research in which animal participants were involved.

Data Availability Statement

The datasets generated and supporting the findings of this article are obtainable from the corresponding author upon reasonable request.

References

1.
Forsberg
,
C.
,
Sabharwall
,
P.
, and
Sowder
,
A.
,
2020
, “
Separating Nuclear Reactors From the Power Block With Heat Storage: A New Power Plant Design Paradigm
,” MIT-ANP-TR-189, https://www.osti.gov/biblio/1768046-separating-nuclear-reactors- from-power-block-heat-storage-new-power-plant-design-paradigm10.2172/1768046, Accessed November 2020.
2.
California ISO
,
2016
, “
What the Duck Curve Tells Us About Managing a Green Grid
,” http://www.caiso.com/Documents/FlexibleResourcesHelpRenewables_FastFacts.pdf, Accessed 25 January 2014.
3.
California Renewables Portfolio
,
2018
, “
SB-100 California Renewables Portfolio Standard Program: Emissions of Greenhouse Gases
,” https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=201720180SB100, Accessed November 3, 2020.
4.
Lin
,
J.
,
2017
, Future of California's Energy Grid: Highly Renewable, Highly Flexible California Energy Storage Alliance, Yosemite Law Conference, October 22, 2017.
5.
California ISO
,
2019
, “
Wind and Solar Curtailment April 21, 2019
,” http://www.caiso.com/Documents/Wind_SolarReal-TimeDispatchCurtailmentReportApr21_2019.pdf, Accessed May 1, 2020.
6.
California Independent System Operator
,
2021
, “
ProductionAndCurtailmentsData_2021.xlsx
,” http://www.caiso.com/Documents/ProductionAndCurtailmentsData_2021.xlsx, Accessed February 24, 2022.
7.
Ruth
,
M.
,
Cutler
,
D.
,
Flores-Espino
,
F.
, and
Stark
,
G.
,
2017
, “
The Economic Potential of Nuclear-Renewable Hybrid Energy Systems Producing Hydrogen, NREL/TP-6A50-66764, April 2017
, https://www.nrel.gov/docs/fy17osti/66764.pdf
8.
EIA
,
2020
, “
Capital Cost and Performance Characteristic Estimates for Utility Scale Electric Power Generating Technologies. U.S. Energy Information Administration
,” https://www.eia.gov/analysis/studies/powerplants/capitalcost/pdf/capital_cost_AEO2020.pdf, Accessed February 2020.
9.
IEA
,
2019
, “
The Future of Hydrogen
,” https://webstore.iea.org/download/direct/2803, Accessed March 28, 2020.
10.
World Energy Council
,
2021
, “
Hydrogen Demand and Cost Dynamics
,” https://www.worldenergy.org/publications/entry/working-paper-hydrogen-demand-and-cost-dynamics, Accessed March 4, 2022.
11.
Hauch
,
A.
,
Küngas
,
R.
,
Blennow
,
P.
,
Hansen
,
A. B.
,
Hansen
,
J. B.
,
Mathiesen
,
B. V.
, and
Mogensen
,
M. B.
,
2020
, “
Recent Advances in Solid Oxide Cell Technology for Electrolysis
,”
Science
,
370
(
6513
), p.
eaba6118
.
12.
Pinsky
,
R.
,
Sabharwall
,
P.
,
Hartvigsen
,
J.
, and
O’Brien
,
J.
,
2020
, “
Comparative Review of Hydrogen Production Technologies for Nuclear Hybrid Energy Systems
,”
Prog. Nucl. Energy
,
123
, p.
103317
.
13.
Mehos
,
M.
,
Price
,
H.
,
Cable
,
R.
,
Kearney
,
D.
,
Kelly
,
B.
,
Kolb
,
G.
, and
Morse
,
F.
,
2020
,
Concentrating Solar Power Best Practices Study
,
National Renewable Energy Laboratory
,
Golden, CO
, NREL/TP-5500-75763, https://www.nrel.gov/docs/fy20osti/75763
14.
Kearney
,
D.
,
Kelly
,
B.
,
Herrmann
,
U.
,
Cable
,
R.
,
Pacheco
,
J.
,
Mahoney
,
R.
,
Price
,
H.
,
Blake
,
D.
,
Nava
,
P.
, and
Potrovitza
,
N.
,
2004
, “
Engineering Aspects of a Molten Salt Heat Transfer Fluid in a Trough Solar Field
,”
Energy
,
29
(
5–6
), pp.
861
870
.
15.
Thess
,
A.
,
Klasing
,
F.
,
Odenthal
,
C.
, and
Bauer
,
T.
,
2020
, “
Combining Hot Rock and Nitrate Salts to Lower Heat Storage Costs
,”
MIT/INL/EPRI Workshop on Separating Nuclear Reactors From the Power Block with Heat Storage: A New Power Plant Design Paradigm
,
Virtual, Aug. 12
.
16.
Forsberg
,
C.
,
2021
, “
100-Gigawatt-Hour Crushed Rock Heat Storage for Variable Electricity and Heat With Base-Load Reactor Operations
,”
ICONE28-64632, Proceedings of the ASME 2021 International Conference on Nuclear Engineering, ICONE 28
,
Aug. 4–6
.
17.
Forsberg
,
C. W.
, 2022, “
Separating Nuclear Reactors From the Power Block With Heat Storage to Improve Economics With Dispatchable Heat and Electricity
,”
Nucl. Technol.
208(4).
18.
Forsberg
,
C. W.
,
2019
, “
Variable and Assured Peak Electricity From Base-Load Light-Water Reactors With Heat Storage and Auxiliary Combustible Fuels
,”
Nucl. Technol.
,
205
(
3
), pp.
377
396
.
19.
Conlon
,
W.
,
2017
, “
Dispatchable Solar Hybrid Power Plant
,” U.S. Patent No. 9,816,490 B2, Nov 14, 2017.
20.
Conlon
,
W.
,
2018
, “
Dispatchable Combined Cycle Power Plant
,” U.S. Patent No. 10,113,535 B2, Oct 30, 2018.
21.
Conlon
,
W.
,
2020
, “
Dispatchable Combined Cycle Power Plant
,” U.S. Patent No. 10,808,685, Oct 20, 2020.
22.
Conlon
,
W.
,
2021
, “
Dispatchable Storage Combined Cycle Power Plants
,” U.S. Patent No. 10,982,570, Apr. 20, 2021.
23.
Conlon
,
W.
,
2022
, “
Dispatchable Storage Combined Cycle Power Plants
,” U.S. Patent No. 11,359,521, June 14, 2022.
24.
Coastal Chemical Co., LLC
,
2020
, “
HITEC® Heat Transfer Salt
,” Revised December 2020.
25.
Conlon
,
W. M.
,
Venetos
,
M. J.
, and
Hume
,
S.
,
2022
, “
Liquid Salt Combined Cycle
,”
ASME Turbo Expo 2022
,
Rotterdam, The Netherlands
,
June 13–17
.
26.
Conlon
,
W.
,
2019
, “
Decarbonizing With Energy Storage Combined Cycles
,”
POWER Mag.
,
163
(
12
), pp.
36
39
.
27.
LucidCatalyst
,
2020
, “
Missing Link to a Livable Climate: How Hydrogen-Enabled Synthetic Fuels Can Help Deliver the Paris Goals (2020)
,” https://www.lucidcatalyst.com/hydrogen-report, Accessed January 7, 2022.
28.
U.S. Energy Information Administration
,
2021
, “
Levelized Costs of New Generation Resources in the Annual Energy Outlook 2021
, https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf, Accessed February 10, 2022.