Concentrating Solar Thermal Power

Concentrating solar thermal power is emerging behind wind as a significant potential source of renewable wholesale generation capacity. In mid-2008, the amount of concentrated solar thermal power technology installed is 431 MW, with several plants under construction – largely in Spain – and an estimated 7000 MW potentially coming on-line by 2012. The majority of current installations use solar trough technology, though one 11 MW tower technology plant went into operation in Spain in 2006.

Some 97% of installations are currently in the US (almost entirely the ‘old faithful’ solar energy generating systems (SEGS) plants operating in California’s Mojave Desert since the 1980s, though the 64 MW Nevada Solar One went into operation in 2007). The remaining 3% of plants are now in Europe, with Spain seen as the key location for the next round of developments.

In areas where direct insolation is high (cloud cover minimal), and there is a need for power – Southwest US, Mexico, Brazil, Chile, north and parts of southern Africa, the UAE, Israel, parts of China, Australia – this technology seems likely to have a future if the price is right, and the site is right. A flat site is needed (certainly for parabolic troughs), near power transmission (and ideally close to a load centre to avoid long-distance transmission), with water available for steam generation/cooling. The right market conditions are what transform technical potential into operating plant.

According to a recent report from Emerging Energy Research, 7010 MW of new projects have been announced to the end of 2012 – 84% of this project pipeline is in the two key markets, Spain (41%) and the US (44%). A further 10% is in the pipeline for the Middle East, 3% for Africa, and 2% for the Asia Pacific region.

The US market is being driven primarily by state Renewable Portfolio Standard (RPS) requirements, and the soon-to-expire Federal solar tax credit, which the industry is vigorously lobbying to have extended.

Following the election of Governor Schwarzenegger five years ago, California has increased its RPS to 20% in 2012 and 30% in 2020. Governor Schwarzenegger mandated a Solar Task Force to look at implementing 3000 MW of new solar power by 2015. In June 2007 New Mexico mandated a CSP-specific task force to outline the way towards the first commercial plants.

In 2004, Spain became the first country in the world to establish a dedicated long-term feed-in tariff (of 27 eurocents/kWh) for power from concentrating solar power plants of up to 50 MW, which puts the implementation of this technology on a very firm financial footing. This rate is payable for 25 years, increasing yearly with inflation minus 1%. The commercial boost provided by the feed-in tariff has been supported by further legislation allowing operators to use natural gas as back-up to keep CSP plants primed (in practice storage may be preferable – see below). Spain has a national target to install 500 MW of concentrating solar power by 2010. However, some 800 MW are already on-line, currently under construction or planned for development.

In Spain, the first tower plant is already working (by Abengoa 2007) and the first parabolic trough plant (Andasol 150 MW) is currently being built by ACS Cobra. According to José Nebrera of ESTELA and ACS Cobra, around 400–450 MW capacity is already under construction in Spain, while something in the region of 8000 MW is in the permitting process, currently (June 2008) waiting for the go-ahead from authorities.

A look at new facilities planned from 2008 until 2012 shows that parabolic troughs will continue to dominate, but other technologies are likely to come into play on a commercial scale later.

Utilities like it

Among the plants in the pipeline is Abengoa Solar’s 280 MW Solana project (Arizona, US). Fred Morse, US adviser to Abengoa Solar and chair of the SEIA (US Solar Energy Industries Association) concentrating solar power division, explains that utilities like concentrating solar thermal power because it’s based on steam and it’s large-scale, with commercial installations on a familiar scale of hundreds of megawatts (about half the size of a typical coal-fired plant). It offers stable, known and decreasing costs, and zero carbon – which also appeals to utilities. Furthermore, this technology provides a hedge against natural-gas price volatility and carbon caps (that may be introduced subsequently). Its ability to provide firm dispatchable power is of great value to utilities.

What utilities also find reassuring, says Morse, is the involvement of large multinational corporations, big construction and engineering companies – all with a lot of experience and big balance sheets. Large utilities view this as reducing the risk of project failure. Though the front end (solar) may be new territory for them, turbines are likely to be supplied by companies such as GE, Siemens or ABB – all very familiar names in the conventional power sector.

Current ownership of concentrating solar thermal power systems (and this largely means the established SEGS plants) is largely in the hands of the Carlyle Riverstone Group and FPL Energy in the United States. Acciona Solar is next, though will be overtaken by Abengoa Solar once its 2008 additions are in place. Iberdrola, ACS Cobra, Sunray Energy and Solar Millennium are all likely to be plant owners by the end of 2008 (Sunray already owns about 35 MW.) (Source: Emerging Energy Research)

In Nevada, the 64 MW Solar One parabolic trough system is the first multi-megawatt power station of its kind to be built in over 15 years (built/operated by Solargenix/Acciona Solar), while in Spain, the recently completed PS10 (Abengoa Solar) is the first commercial-scale solar tower in the world.

The trump card – storage

Generation from solar plant with storage can be shifted to match the utility system load profile. This has captured the imagination of the US utilities in particular, as it allows solar to provide power when it is most needed. Such ‘peaking’ power has a very high commercial value. In brief, heat collected by day can be fed into storage tanks – using a medium such as molten salt to hold the heat – and, when needed, that heat can be released to generate steam to run the turbines. Adding storage, and the extra collector field to serve it, pays off when a peaking power price can be assured or where there is a good feed-in rate available.

Potential and costs

Within Europe, it is Spain that has the largest potential for rolling out this technology, as the location that has the best solar resource.

Costs are tricky to compare. While construction costs appear to be broadly similar wherever a plant is located (with components/equipment likely to be coming from the same sources in this new field), the location can make a big difference to output. As with almost any wind or solar plant, the performance – the energy yield and therefore price per kilowatt hour – will depend on the resource at the selected site. Jose Alfonso Nebrera estimates the current cost per kilowatt hour in Spain (assuming insolation of 2100 kWh/m2) at 27 eurocents/kWh – identical to Spain’s current feed-in tariff. In North Africa (assuming 2600 kWh/m2) he says this would be nearer 17.5 eurocents/kWh. Rainer Aringhoff of Solar Millennium points out that the solar resource in the American South-west is higher even than in North Africa. A nominal 100 MW plant in Spain would be 200 MW in California or Arizona. The price per kilowatt hour for systems operating in the United States is estimated to be in the range of 14.5–17.5 US cents/kWh. Today, Southern California’s peaking power costs anywhere between 10–18 US cents /kWh. This figures suggest that CSP is already cost-effective in these markets.

In the case of both Spain and North Africa, Nebrera expects to see a cost reduction of 3% per year. And – at some stage in the next 10–15 years (depending on the price of gas and other factors) – Nebrera expects this decreasing price curve to intersect with an increasing price curve for conventional power as fossil-fuel generated electricity is impacted by scarce resources and an increasingly carbon constrained operating environment. Others are less optimistic, as this industry, like others, is reliant on world commodity prices – for instance for steel, or aluminium. Companies such as SkyFuel and Ausra, which are just reaching the commercial phase, promise advantages that include lower investment costs (such as by use of Fresnel lenses to concentrate the solar radiation instead of mirror troughs) and higher yield, and their first large installations will surely attract a good deal of interest. The same goes for the dish-Stirling technology from Stirling Energy Systems (SES) – the company has signed power purchase agreements with US utilities Southern California Edison and San Diego Gas and Electric for over 1800 MW from two fields, SES Solar 1 and 2.

Rainer Aringhoff is convinced that the ‘home’ of CSP – California – has the potential for a massive build-out of this technology – the potential to build 10 GW of CSP peaking power by 2020 at a rate of 800 MW/year from 2008 onwards. Two-thirds of this would be in the Mojave Desert, the other third in Imperial Valley. This assumes that California will not be importing coal and has a target of producing 33% of its electricity from renewable sources. The advantage with California is that the installations would be relatively close to the load centres of the large Californian cities.

Jackie Jones is Editorial Director of Renewable Energy World magazine e-mail: rew@pennwell.com

CSP defined

This technology is usually known as CSP – concentrating solar power – though CSP also includes concentrating solar PV. To try to avoid confusion we are using the term concentrating solar thermal power. Other names in frequent use are solar thermal electricity, STEG (solar thermal electric generation) and thermosolar power.

Unlike photovoltaics, which generates electricity directly from sunlight, concentrating solar thermal technologies use heat to generate electricity in much the same way as a conventional thermal power station. A series of mirrors or parabolic troughs focus the sun’s rays on a central receiver containing a mineral oil or other thermal carrier. As this liquid heats up (reaching temperatures as high as 400ºC–600ºC), it passes through a heat exchanger and generates steam, which is then used to drive a turbine.

Solar towers employ fields of mirrors to reflect light onto a central receiver atop a tower, while parabolic troughs, as the name implies, use long fields of mirrors curved to reflect light onto a central receiver which runs along between them. The famous 344 MW parabolic trough array in California, developed by LUZ Engineering between 1984 and 1992, was almost the first of its kind in the world and is still the largest ever built (a 55 kW trough power plant was built in Egypt in 1912).

Altogether, LUZ built nine plants at this site until the company went out of business in the early 1990s. The troughs continue to function well, however, and produce electricity reliably, with seven operated by FPL Energy and another two by Carlyle/Riverstone.

Similarly, the solar tower is also a mature technology, with the first prototypes developed over 20 years ago in California. Solar One and Solar Two, as they were called, ran until 1989 and generated 38 GWh of electricity.

Recent activity highlights

• In February 2008, Abengoa Solar signed a contract with Arizona Public Service Co. (APS), one of Arizona’s leading energy utilities, to build, own and operate the 280 MW Solana plant. The plant, scheduled to go into operation by 2011, will be located south west of Phoenix Arizona. It will sell the electricity produced to APS over the next 30 years for a total revenue of around $4 billion. The installation at the plant will include six hours of molten salt thermal storage to improve its load following characteristics.
• In April 2008, Northern Californian utility group Pacific Gas and Electric Company (PG&E) entered into a series of contracts with BrightSource Energy, Inc. for renewable solar power. The first three contracts are for a total of 500 MW of power to be supplied from three concentrating solar thermal power plants. PG&E also signed two contracts for options on an additional 400 MW of solar power, which would bring the total amount of power purchased under these five agreements to some 900 MW. Founder and chairman of BrightSource Energy, Arnold Goldman, was also the founder of Luz International (no longer in operation). Goldman and the Luz International team built the nine SEGS plants in California’s Mojave Desert between 1984 and 1990. BrightSource is currently developing a number of solar power plants in Southern California, with construction of the first plant planned to start in 2009. In June, BrightSource Energy and subsidiary Luz II dedicated their Solar Energy Development Center (SEDC) in the Negev’s Rotem Industrial Park, Israel.
• Ibereólica Solar is developing 1000 MW of thermosolar plants in Spain, in the autonomous regions of Extremadura, Andalucia, Castilla-La Mancha and Castilla y León. Ibereólica currently expects construction of its first eight thermosolar power plants (totalling 400 MW) by mid-2011.
• In May, Ibereólica Solar placed an order with Solel of Israel for more than 190,000 UVAC 2008 receiver systems to power eight 50 MW solar power plants Ibereólica is developing in southern Spain. Delivery of the receivers will commence in 2009.
• In March, Spanish engineering group Sener Grupo De Ingeniería S.A. and Masdar, Abu Dhabi’s alternative energy company, announced a 60:40 joint venture – Torresol Energy – to design, build and operate concentrating solar power (CSP) plants in the world’s sunbelt regions. This joint venture will commence work on three solar power plants in Spain with an approximate combined value of €800 million, one of which will be a CSP central tower receiver system. This technology will feature the first-ever commercial deployment of the industry-changing technology, and will set the standards for anticipated CSP projects across the sunbelt countries by 2012. Independently of Torresol Energy, Masdar is developing CSP plants in Abu Dhabi, and their Shams 1 plant is expected to be completed in the fourth quarter of 2010.
• Sener has been working for almost a decade in the development of solar thermal power technology. The company is presently designing and building, in a joint venture, three 50 MW parabolic trough plants with molten salt storage in Spain.
• Ausra has just signed a power purchase agreement with PG&E to build the world’s first compact linear Fresnel reflector (CLFR) plant at 177 MW in California’s Central Valley.
• Solel is to construct a 553 MW complex of parabolic trough power plants in the Mojave Desert to fulfill a 25-year power purchase agreement with PG&E.
• Solar Tres, a central receiver design based on the US demonstration plant of the late 1990s, is reported to be close to obtaining financing, making it the first baseload solar power plant with round-the-clock power generation during the summer.
• Iberdrola is building a 50 MW parabolic trough plant at Puertollano in southern Castile, with plans for others.
• Southern California Edison (SCE) has signed a power purchase agreement with eSolar for a total of 245 MW from a series of prefabricated power tower solar thermal plants in the Antelope Valley region of Southern California. This will be the first commercial effort using the technology. By 2012 a 105 MW of operating capacity is scheduled for completion, ramping up to 245 MW by 2013. The development follows a $130 million funding round for eSolar in April led by Idealab, Google.org and Oak Investment Partners.
• Meanwhile, Schott is building new manufacturing plant for solar receiver tubes, both in Albuquerque, New Mexico and in Spain and Australia’s Ausra has opened a manufacturing facility in Las Vegas, Nevada.

Finance highlights

• Stirling Energy systems received $100 million financing
• SkyFuel announced $17 million in finance
• BrightSource announced that it had secured $115 million in additional corporate funding from its Series C round of financing, bringing the total the company has raised to date to over $160 million.

Use of water

Usage is about the same as for any plant using steam, but as this technology is likely to be located in arid areas this is a potential concern. Water consumption can be reduced by 90% using dry cooling, but performance will suffer which is likely to increase the price by about 10%.

In dusty environments parabolic mirrors need to be washed to maximize performance, but the quantity of water required is less than that used for steam.