The Thermal Energy Storage Solution


As renewable energy generation continues its rapid ramp-up — increasing more than 14 percent last year over 2010 in the United States according to the U.S. Energy Information Administration — overcoming the variability of this energy source will be critical. After all, the wind does not always blow, and the sun is not always shining.

thermal solar storage
Based on efficiency improvements of the past century, prospects are good for achieving efficiency levels that help thermal energy storage realize its promise as a reliable, site-flexible, relatively low-cost option.

Several approaches may be applicable to overcome renewable power variability: balancing production and demand through forecasting [1]; aggregating generation from different locations and renewable sources, e.g., solar, wind, geothermal and hydropower [2]; and, ultimately, storing electric energy and using it on demand. This article examines electricity storage methods that can overcome resource variation and permit baseload power generation from renewable energy. Thermal energy storage (TES), coupled to a concentrating solar power (CSP) plant, is introduced as an option not only for energy storage in CSP, but also as a means for grid storage. In such a setting, off-peak electricity is stored as thermal energy and dispatched later to produce high-value, peak-demand electricity.

Weighing the Storage Options

Renewable Energy
Figure 1. Scale of renewable energy sources relative to global energy consumption. Solar and wind energy are the largest resources, but their variability has hindered large-scale grid penetration.

Among the deployed renewable power sources, hydropower is the most direct in supplying electricity and providing dispatchable power. However, as shown in figure 1,the hydro energy resource is an order of magnitude lower than that of wind and solar. Wind technology has developed rapidly over the past decade with its competitive electricity cost, but wind’s variability does not allow it to provide baseload power. Although solar heat is the most abundant renewable source, the cost of capturing it is still high, and, as with wind, generation variability has hindered large-scale penetration.

The challenge for energy developers is to reduce the capital cost of renewable technologies and improve the energy conversion processes. Renewable energy’s generation variability and low energy density make it challenging to collect and match to demand. Energy storage can bridge this gap by providing energy continuously and on demand.

Electricity from the grid is usually produced on demand or as needed, without storage ability. However, energy storage is already used in transportation, mobile power and distributed generation (DG). DG can be connected to a grid, or just provide power to a stand-alone load. This article presents energy storage technologies in utility-scale power backup and for renewable grid integration.

As renewable electricity generation technologies are deployed on an increasingly large scale, storage will be required to abate variability. Several electricity storage methods are available, and many others are under development [3]. Of these methods, thermal energy can drive mechanical, chemical and electric processes for power generation, and is also easy to store and retrieve.

For variable generation like wind and solar power, electric storage provides a reliable and uninterruptible power supply, reduces the need for spinning reserve (online reserve capacity synchronized with the grid and available within a few minutes) and helps meet peak power demands from the grid. In short-term electric storage, discharging electricity from storage can smooth grid fluctuations and improve power quality.

Electric Energy StorageTable 1 lists three types of energy storage applications — power quality, distributed generation and bulk energy storage — and specifies their power and storage capacity ranges. Utility-scale power is of the same scale as bulk energy storage [3], defined by a power rating of 10 to 1,000 megawatts (MW) and a storage capacity of 10 to 8,000 megawatt-hours (MWh). Bulk energy storage has several benefits:

  • Addresses variability and defers transmission congestion
  • Firms up and enhances the value of renewable energy generation
  • Reduces the amount of central power generation capacity needed for peak and baseload demand, and
  • Reduces time-of-use energy costs

Storage MechanismsTable 2 compares different storage methods with their applicable scales: batteries (lead-acid, lithium-ion and sodium sulfur), reversible fuel cells, superconducting magnetic energy storage (SMES), flywheels, compressed-air energy storage (CAES) and pumped hydro storage (PHS). Energy storage typically converts electricity to another form of energy that is easily stored for later generation. The most direct methods of electricity storage store chemical or electrical potential in a battery or capacitor, or, in the case of SMES, transform electricity into an electromagnetic field. Other energy storage methods convert electric energy into forms of mechanical energy, as in flywheels, PHS and CAES. Although the major electric storage technologies focus on the above methods, this article adds thermal energy storage as a means to provide electric energy storage [4].

The storage capacity for each method depends on the amount of the storage medium and its cost. Batteries, flywheels and SMES need chemicals and/or metals as storage carriers, which add expense, especially when exotic materials are used. Their value for small-capacity power quality and distributed generation is high, and is not measured by the storage cost, but rather by the ability to deliver electricity in seconds or minutes and to prevent power glitches. For this reason, the need for power quality may justify high-cost storage methods at lower power capacities. On the other hand, utility-scale storage may directly compete with the cost of conventional spinning-reserve power generation and is cost-sensitive. The storage media for CAES and PHS are air and water, respectively, which are essentially free. Therefore, CAES and PHS are suitable for storing large amounts of electricity, and their capacity may be limited only by the size of the storage space (cavern or reservoir) that is geographically available.

Thermal energy is stored as a form of internal energy that includes both sensible (associated with a change of temperature) and latent (associated with a change of phase) heat. It is generally easy and inexpensive to store, but it does not convert directly to electricity as with the electrochemical or electromagnetic methods. Instead, TES relies on thermal electric generation to convert the thermal energy back to electricity. Unlike other electric storage devices, the downstream conversion of electricity to thermal energy can be 100 percent by Joule heating (whereby passage of current through an electrical resistance releases heat). However, the conversion of thermal energy back to electricity is limited by the thermal energy quality (which depends on temperature) and the associated power conversion efficiency. Therefore, one key to enabling thermal energy for electricity generation is to improve the thermal cycle efficiency. Based on efficiency improvements over the past century, there are several prospects for achieving high-efficiency TES that promise reliable and relatively low-cost electricity.

Comparison of the Storage Technologies and TES as a Means for Electric Energy Storage

In selecting storage technologies one must consider each storage capability at its power rating, storage capacity, performance in terms of total roundtrip efficiency, lifetime, reliability, ease of installation and cost. Power rating and storage capacity determine the scale of the storage application. The other factors affect economic benefits.

Figure 2. Storage power rating and energy capacity comparison for various storage types.

Figure 2 illustrates storage technologies on power rating and capacity scales. The technologies in the lower-left quadrant — batteries, electrochemical capacitors, flywheels and SMES — serve short-term electricity storage needs for power quality and distributed generation.

For long-term bulk energy storage and stationary generation, CAES and PHS are broadly accepted. TES can achieve the same power rating and storage capacity as CAES and PHS by expanding the plant size as needed. The power rating of existing and planned CSP plants ranges from 50 to 300 MWe. Similar to CAES and PHS, CSP plants with TES have an expected life of 30 to 40 years, which is more than three times longer than the lifetime of electrochemical methods, resulting in long-term benefits and lower life-cycle cost.

Storage Method Efficiency
Figure 3. Comparison of roundtrip efficiency for energy storage methods.

Figure 3 shows storage performance in terms of roundtrip efficiency for typical storage technologies. Energy storage efficiency is a metric that accounts for the net electricity delivered from the storage system relative to the amount of electricity put into the storage system. Electric storage efficiency often excludes electricity generation and transmission efficiencies from the original power sources. From the metrics of power rating and storage capacity in figure 2, utility-scale energy storage favors PHS, CAES and TES. This article will focus on comparing the performance of those three storage methods in renewable applications. PHS is a well-known, highly efficient, long-term and large-scale energy storage method. PHS stores potential energy in the form of water, pumped from a lower-elevation reservoir to a higher-elevation reservoir, and released through power-generating turbines during high-demand periods. However, PHS development requires suitable geological conditions, large investment and extensive environmental assessment.

The efficiency of CAES and PHS storage methods can be affected by the geological configuration and method of operation. In a conventional CAES arrangement, the air in a cavern or container is heated during compression and cools in a storage container. The compressed air cools further as it expands to drive a turbine, and fuel (typically natural gas) is burned to heat the air back to the turbine inlet temperature. If there is no compression heat recovery or heat recuperation from the turbine exhaust, the net energy efficiency of CAES can be 40 to 50 percent [5]. To save fuel and improve efficiency, TES may be incorporated in an adiabatic CAES system. In an adiabatic CAES configuration (that is, one in which no heat is gained or lost by the system), heat from the compressed air is transferred to a thermal storage device and then reused to preheat the turbine inlet air, thereby minimizing or even eliminating the need for natural gas heating. In this configuration, adiabatic CAES may achieve an energy-usage efficiency of 50 to 70 percent.

As shown in figure 3, TES can have a thermal-to-thermal conversion efficiency of 95 to 99 percent in a well-insulated container. The roundtrip efficiency of TES for electricity storage is limited by the thermal-power conversion efficiency, which usually ranges from 30 to 60 percent, depending on the power cycle. From the perspective of energy usage, TES for electricity storage can have an efficiency comparable to that of conventional CAES [4]. A further step to boost the electricity generation efficiency is to apply “exergy uphold” — a unique method that may be possible for CSP plants with TES. This entails charging thermal storage at a high temperature level, for instance above 400°C (752°F), and keeping the low-grade heat of below 400°C in CSP generation to cushion the low-end thermal energy. In this operating mode, TES for CSP keeps discharging the temperature higher than the cold storage temperature. By superposing electric heat on top of the CSP TES level, the stored energy from electricity maintains its high exergy level (i.e., its capacity to produce electric power). High-grade energy has the ability to convert back into electricity more effectively, and achieves high electricity-to-electricity efficiency through augmenting the prior low-end thermal energy residing in the CSP TES system.

Selecting an appropriate energy storage technology is a multifaceted process, involving economics, power requirements, storage capacity, reliability, lifetime and efficiency. A proper choice will add value to the electric grid operation and enable renewable energy to provide baseload power. Low-cost, high-performance TES is currently being developed under the U.S. Department of Energy (DOE) CSP Program. Last year, DOE launched the SunShot Initiative, which aims at lowering solar power generation cost to be on par with conventional power plants (approximately 6 cents per kilowatt-hour by 2020). The realization of the DOE SunShot goal can enable broad deployment of CSP power generation via high electric conversion efficiency and high-performance TES that can provide additional use for bulk energy storage. Unlike the geological requirements of CAES and PHS, TES is not restricted by site, making it possible to co-locate with photovoltaic (PV) or wind farms to alleviate transmission load with firming renewable power. Under this scenario, TES as a part of a CSP plant can serve broader renewable energy storage at virtually no additional cost, as the TES cost is paid for upfront by the CSP plant. TES can also be integrated into the grid for electricity storage. Figure 4 shows a scenario for mixed renewable and conventional power production with TES electricity storage. Under this grid structure, off-peak wind, PV and conventional electric generation can be stored in the CSP TES facility by electric heating, and the stored thermal energy then converts back to electricity to serve the peak demand.

Renewable Energy Grid Scenario
Figure 4. A scenario of renewable energy integrated with storage as baseload power.

Advances in CSP technology under the SunShot Initiative may drive up thermal power conversion efficiency to possibly more than 50 percent, making the electric energy storage efficiency potentially competitive with conventional CAES. Meanwhile, the SunShot Initiative should bring down TES costs significantly. By incorporating essentially free electric storage inherent in a CSP plant’s TES into the renewable power generation system, TES will increase the wind and PV penetration into the grid, and make renewable baseload power possible. In conclusion, TES can be one of several electricity storage methods that can address the variability of renewable energy generation, thus allowing renewables to achieve large-scale grid penetration.

Zhiwen Ma, Ph.D. (, is a senior engineer in the Concentrating Solar Power Group at the National Renewable Energy Laboratory (NREL), and previously worked on aircraft engines for GE Aviation. He now works on CSP system analysis, advanced power cycles and thermal energy storage.

Greg Glatzmaier, Ph.D. (, joined NRE L’s Solar Thermal Program in 1987, where he demonstrated a new concentrating solar technology and, with Coors Ceramics, developed a high-temperature materials synthesis process. In 1997 he launched a private R&D company to develop spacecraft fluid compressor designs and feedback controls for NASA and the U.S. Air Force. Back at NREL since 2007, Glatzmaier
now manages advanced heat-transfer and thermal-storage work.

Chuck Kutscher, Ph.D. (, joined NREL in 1978 and is a principal engineer and group manager of the Thermal Systems Group, leading R&D for parabolic troughs. He is chair of the World Renewable Energy Forum 2012, served as chair of the American Solar Energy Society for 2000–2001 and was general chair of its SOLAR 2006 conference. Kutscher edited the ASES report, Tackling Climate Change in the U.S., and writes a column on climate change for SOLAR TODAY.

This work was supported by the U.S. Department of Energy under contract No. DE-AC36-08-GO28308 with the National Renewable Energy Laboratory. The authors thank Allison Gray for her input, Devonie McCamey and Raymond David at NREL for publication and graphic support.

[1] Melinda Marquis, Steve Albers and Betsy Weatherhead, “For Better Integration, Improve the Forecast,” SOLAR TODAY, September/October 2011.
[2] Chuck Kutscher, “Does Utility-Scale Renewable Energy Require Massive Electric Storage?” SOLAR TODAY, March 2010.
[3] U.S. Department of Energy’s Office of Electricity Delivery & Energy Reliability, “Energy Storage Program Planning Document,”, February 2011.
[4] Zhiwen Ma, Greg Glatzmaier and Chuck Kutscher, Thermal Energy Storage and its Potential Applications in Solar Thermal Power Plants and Electricity Storage,” ASME ES2011, Washington, D.C., August 2011.
[5] Bent Sorensen, “Renewable Energy Conversion, Transmission, and Storage,” Academic Press, November 2007.