Solar, Storage

Energy storage: How solar can always meet demand

Issue 5 and Volume 11.

Thermal energy storage enables concentrating solar thermal power (CSP) facilities to operate in a more flexible manner that allows both the developer and the local utility to maximize the value of a plant’s output. Richard Baxter explains why, with storage, the sun always shines on CSP.

Thermal energy storage provides a number of significant commercial benefits to the operation of a concentrating solar power (CSP) facility. Storage allows the plant to be optimized to address the peak load profile of a utility; it provides the operational flexibility required for the CSP facility to be more competitive and create the most value as a project; and it is a zero emission solution to the eternal ‘variability’ question that dogs much renewable generation, particularly when it is grid-connected. In short, storage enables a CSP facility to operate as a reliable and flexible low-carbon resource replacement for fossil generation.

Thermal energy storage is therefore an important technology that will play a key role in moving solar power into the mainstream of the power generation mix. CSP technology is a real, near-term solution to achieving widespread solar power facilities in the tens and thousands of megawatt class, and energy storage is a key enabling technology to make these facilities operate successfully in today’s power market.

Thermal storage technologies are designed to improve the availability and dispatchability of a solar thermal power facility — thereby enhancing its overall value. In the long run, thermal storage will help integrate more solar power into the generation mix by enabling CSP facilities to shoulder a greater component of the daily power demand in many regions of the world.

Why add thermal storage?

Incorporating thermal storage into a CSP facility provides a number of benefits. One of the most important is that it allows the installation to be optimized to address the peak load profile of the local demand load. By decoupling the production and delivery of solar-derived electricity, this allows a time shift in the despatch of electrical power from the middle of the day, when peak production occurs, until the typical peak demand periods that occur after 4pm. In addition, the stored energy allows the facility to be treated as a dispatchable resource — improving the value of the electricity generated from the facility to the local utility.

Thermal storage can also increase the value of the power produced from the solar thermal facility. By adding storage (and optimizing the design of the CSP facility to incorporate the storage), a solar thermal plant can raise its average capacity factor from below 25% to 60% or above — depending on the size of the storage facility and additional solar collectors — and then sell that power at the most commercially advantageous time.

Another key benefit is that thermal storage is a zero emission solution to enhancing a CSP facility. Currently, some CSP installations such as the Kramer Junction facility in California incorporate a supplemental conventional gas-fired boiler to augment the steam generation capability of the CSP facility. This strategy can provide these ‘hybrid’ facilities with a capacity factor approaching 100%, bar scheduled maintenance, but at the cost of increasing carbon dioxide emissions. A thermal energy storage component offers the CSP facility improved operating characteristics, and with all of the facility’s output derived from the solar resource.

With these significant advantages, it is no surprise that overall interest in incorporating a thermal storage component into a proposed CSP facility is growing. ‘Adding storage to CSP operations creates direct value-added benefits to the utility’, says Mike Taylor, director of research at the Solar Electric Power Association. ‘It is no longer just a large solar plant designed to generate renewable megawatt hours, but a direct substitute for a natural gas plant that can operate 6-12 hours on peak. Not only can it provide firm capacity, but also long-term price stability — there aren’t any supply or market constraints on the sun. And costs are projected to decline over time. I don’t know a utility who wouldn’t be interested in a firm, renewable generating source with stable pricing, and a declining cost future’ Taylor enthused.

Integrating thermal storage

Adding thermal energy storage to a CSP facility is generally not an after-the-fact decision because of specific design requirements. These requirements include the storage holding tanks and additional collector capacity (more troughs or heliostats) necessary to match the storage capabilities and demand profiles.

The ratio of the thermal capacity of the collector field to the thermal requirements of the steam generator is called the solar multiple. Having a solar multiple of greater than one ensures that there will be additional solar energy available that can be directed to the storage tanks. In fact, over-sizing the collector array for the power generating component of the facility is required for most thermal storage operating strategies. Without any additional means to gather and store more solar energy, the facility would merely have the ability to time-shift the delivery of the solar power. Additional capacity, as well as enabling the time shift capability, also allows operators to increase the capacity factor of the installation.

In a typical solar thermal facility, sunlight is focused onto a tube filled with a heat-transfer fluid (HTF), typically water, oil, or molten salt — depending upon the needs and limitations of the CSP design. Unless the HTF is steam, in which case it can drive the turbine itself, the HTF fluid is then passed through a heat exchanger to create steam which drives a turbine to produce electricity. If thermal storage is incorporated, it is located so as to be charged from the most recently heated HTF, prior to the steam generator.

The storage medium in these facilities is generally a working fluid such as molten salt. Overall, the type and ease of incorporating thermal storage into the CSP facility depends upon a number of factors including design type, goals for operation, cost, and such like. For example, in a solar power tower design, the HTF and the storage medium are both generally a single circuit of molten salt. Solar power towers can raise the molten salt to very high temperatures, frequently to 565°C. In order to increase the efficiency of the facility, the working fluid is cycled in as short a circuit as possible.

However, if the CSP design is a trough facility, the HTF most frequently used is Therminol, a synthetic oil specially designed for solar thermal facilities, or its equivalent, and thus requires an additional heat exchanger to transfer the heat energy back and forth from the HTF to the molten salt storage medium. This additional heat exchanger and molten storage unit reduces the overall efficiency of the unit by an additional 7% overall. In addition, because of the lower operating temperature of Therminol, at 370°C, a molten salt storage unit when used with a trough facility requires three times the volume of a power tower’s molten salt storage capacity — which is operating at a much higher temperature — to equal the same electricity generating production capacity.

Molten salt

Molten salt is the storage medium most often used for thermal energy storage because of its well understood operating properties. It is low-cost, non-flammable and non-toxic, and there is already a significant body of experience utilizing it in the chemical industry as an efficient heat-transportation medium. Its operating temperature range is compatible with modern high-pressure and temperature steam turbines. The most basic and common molten salt is a binary mixture of 60% sodium nitrate (NaNO3) and 40% potassium nitrate (KNO3). During operation, it is maintained at 290°C in the cold storage tank, and can be heated to 565°C. The storage tanks are very efficient, with only a 1%-2% loss over a one day storage period. As a liquid, molten salt operates similarly to water (Figure 1).

A significant amount of effort is being made to develop additives or other mixtures of salts which would have a larger operating temperature range, especially towards the low end of the temperature scale. Because molten salts freeze at temperatures of 120°-230°C (depending upon the salt’s composition), care must be maintained to prevent the molten salt from freezing at night, especially in any of the piping system. This is the primary reason that molten salt is not typically used as the HTF for trough facilities. The size of the molten salt storage tank is dependent upon how long the plant designers want the facility to operate after the peak sunlight period, although six or more hours of additional power is common.

An example of a modern solar power tower project utilizing molten salt storage comes from the Solar Tres facility under development near Seville, Spain by Sener. Once operational, this 17 MW facility will incorporate 15 hours of thermal storage, giving the facility a 74% utilization factor, and the ability to produce 110 GWh annually.

Parabolic trough facilities are also now being developed that will utilize molten salt for thermal storage capacity. The 50 MW Andasol 1 facility, under development near Gaudix, Spain – also by Sener – will incorporate enough molten salt to provide over seven hours of storage capability to the plant. The molten nitrate salt storage system will cycle between 292°-386°C, and will be housed in two tanks, each 14 metres tall and 36 metres in diameter.

Steam circuit

A single pass through loop (from the collector field, to the storage unit, and through the turbine) utilizing steam as the HTF and storage medium was used in many earlier CSP designs, and remains popular today for its simplicity and low cost. There are, unfortunately, some challenges that must be overcome when using steam as the storage medium.

The economics of power generation using steam turbine thermal technology dictate that the system is most efficient when operated at as high a pressure and temperature as possible. However, the high-pressure vessels necessary to store the steam cost significantly more than their low-pressure equivalents. For that reason, thermal storage designs utilizing steam generally use multiple storage vessels at a lower pressure, rather than a single storage vessel at higher pressures.

For steam thermal storage systems, supplemental energy with which to generate electricity is generally limited to an hour or so, and is frequently used to provide a buffer from the vagaries of short-term cloud transients. Steam storage systems, therefore, have a limited ability to extend the operation timetable of the facility into the evening peak demand period.

Nonetheless, a good example of a thermal power facility with an integrated, steam-based thermal storage capability is the 11 MW PS10 Power Tower facility located near Seville, Spain. Construction, by Abengoa, was completed in 2006 and the facility generates steam at 250oC and 40 bar pressure as both the HTF and for storage purposes. During operation, some of the steam generated is directed to a four-tank thermal storage installation. The steam held in these tanks has a capacity of 20 MWh (thermal) for an effective operational capacity of 50 minutes at 50% of maximum turbine workload. These four tanks are operated sequentially in relation to their charge status. The PS10’s thermal storage system was designed to give the facility the ability to act as a buffer during short-term cloud transients, both to maintain operation and to protect the steam turbine from excessive thermal cycling. If thermal energy is needed during operation, steam is retrieved from the tanks at 20 bar to run the turbine.

Cost vs Value

Reducing the capital cost of thermal storage is key to the continued deployment of the technology. Colorado-based Xcel Energy estimated the cost premium of adding thermal storage to a trough CSP facility of some 25%-100%, depending upon a number of factors, the most important being how long the facility would operate. Xcel also believes that the mirror field must be doubled.

As part of the US Department of Energy’s increase in support of large CSP technology last year, thermal energy storage was identified as a key technology. As part of the overall cost reduction goals for CSP power, to 7-10 US cent/kWh by 2015 and 5-7 US cents/kWh by 2020, the DOE established goals of thermal storage costing less than $15/kWh thermal with round trip efficiencies at or greater than 93%. With reduced capital costs, the levelized cost of power from solar thermal facilities will inevitably be greatly reduced.

However, operating a power facility in today’s market requires a design that promotes economic value, not simply lower levelized costs — a lower production cost is only part of the answer. This is where thermal storage brings the most benefit to the project developers. Solar thermal facility designs are based on a multitude of metrics, including the hourly market rate they can sell their power for, the amount of solar insolation received at the site, seasonal variations, and such like. Solar thermal facilities need flexibility in order to operate to maximize the value of the electricity produced, not just produce the largest volume or lowest levelized cost of power.

Designing a facility to sell the largest amount of output does not necessarily make that design the one with the best return on capital. Depending upon the daily change in the wholesale cost of power for an area between off-peak and peak demand periods, a more effective strategy for a thermal facility may be for it to store all of the energy produced in the morning instead of directing some to the storage component and some to the steam turbine to produce electricity for immediate sale. According to Larry Stoddard, manager of Renewable Energy Consulting for Black & Veatch, ‘thermal storage allows project developers to maximize the value of the solar thermal facility’s output for time-of-day pricing verses the cost of producing that electricity.’

As with other energy storage technologies, thermal storage is not a stand-alone technology, but one designed to optimize the operating characteristics of another facility – in this case, a solar thermal power facility. Challenges obviously remain in the deployment of this technology, including improving designs, and gaining a wider body of experience to allay fears over its cost-effectiveness.

All of these points are currently being addressed, and the integration of thermal storage technology into solar thermal facilities has and will continue to benefit from molten salt technology’s widespread use in the chemical process industry. According to Dr Tom Mancini, programme manager, Concentrating Solar Power, Sandia National Laboratories, cost remains one of the key challenges for thermal energy storage technologies. However, he also believes that thermal storage will become more important as the utilities become familiar with CSP plants and start to value higher capacity factor developments.

As solar thermal power production continues to gain favour as a solution for large-scale power production, thermal energy storage will be seen as a key opportunity to extend the technology’s capability and the reach of these facilities to provide a larger component of our electrical power needs.

Richard Baxter is a senior vice president at Ardour Capital Investments, LLC. He is also the author of Energy Storage: A Nontechnical Guide from PennWell Publishing.
e-mail: [email protected]

The thermocline storage system

Reducing the capital costs of thermal storage systems is a major goal of project developers in order to advance the deployment of this technology. In a thermocline storage system, instead of maintaining a ‘hot’ and a ‘cold’ tank of a molten salt, a single tank is used. The working storage fluid maintains a temperature gradient from the top of the tank to the bottom.

A second avenue of capital cost reduction available with thermocline technology is that instead of tanks emptying as the storage medium is shifted from cold to hot and vice versa — as occurs in two-tank storage systems — the thermocline tank is filled with a substance that has a higher volumetric heat capacity than the fluid it is replacing. One of the issues this technology faces is finding a suitable filler material that will not degrade from the thermal cycling across the high operating ranges envisioned with future systems.

The US Department of Energy incorporated one of these systems in the Solar 1 solar power tower facility, with the tank filled with a mixture of silica sand and quartzite rock. Nonetheless, efforts have been focused on both molten salt storage and synthetic oils, especially with regard to the trough type systems that have become prevalent today. This concept has been extended by others where a supplemental mass is used to increase the unit’s energy storage capability. In particular, the German Aerospace Centre (DLR) has undertaken significant work investigating the cost and performance of utilizing concrete or ceramic materials.