A Free-Market Energy Blog

High Capital Costs Plague Solar (RPS mandates, cost dilution via energy mixing required)

By Robert Peltier -- October 7, 2009

Renewable energy generates a larger portion of the world’s electricity each year. But in relative terms, solar power generation is hardly a blip on the energy screen despite its long history of technological development.

In this Part I, we review the standard taxonomy of central solar power generating plants by focusing our attention on solar thermal technologies and demonstration projects. The technologies are reasonably well defined yet two formidable hurdles remain: large-scale energy storage technologies and first costs on the order of $5,000/kW, the same cost range as a Generation III+ nuclear plant.

 Future posts will explore a number of interesting commercial projects that have either recently or will soon break ground and the latest developments in hybrid projects that combine many of the available solar energy conversion technologies with conventional fossil-fueled technologies. Hybrid projects offer the opportunity for utilities to reduce fuel costs, while simultaneously helping utilities cope with onerous renewable portfolio mandates.

U.S. Solar Growth: Mostly for Swimming Pools

The U.S. solar industry saw a third straight year of record growth in 2008, but looks can be deceiving. The installation of 1,265 MW of all types of solar power last year brought total U.S. solar power capacity to 8,775 MW, according to the annual report from the Solar Energy Industries Association (SEIA), US Solar Industry 2008.

The report breaks down the new capacity as follows:

  1. 342 MW of solar photovoltaic (PV);
  2. 139 MWth (thermal equivalent) of solar water heating
  3. 762 MWth of pool heating;, and
  4. 21 MW of solar space heating and cooling.

Grid-tied PV grew at a rate of 81%, to 292 MW in 2008, compared to 161 MW in 2007. In essence, the amount of utility-scale solar electricity plants installed was miniscule but proponents remain hopeful that the next few years will see explosive growth in larger-scale projects.

So far, the utility-scale projects are relatively small (most such plants are located in the EU). No new concentrating solar power plants came online in the U.S. in 2008, but projects in the offing add up to more than 6 GW, the report said. Among these are projects planned for California’s Mojave Desert, Arizona, and Florida where, understandably, the sun shines year-round. States that led grid-tied PV installation were California (178.6 MW), New Jersey (22.5 MW), Colorado (21.6 MW), Nevada (13.9), and Hawaii (11.3 MW).

Government Quotas and Tax Favors Drive New Utility Projects

A more pragmatic view might find that the relatively high first costs can be a deal-breaker, as Lockheed Martin Corp did when it canceled its $1.5 billion dollar Starwood Solar I that was proposed for west of Phoenix. The 290-MW plant was canceled because Lockheed couldn’t secure financing of the project. The project was conceived as a means for Arizona Public Service to meet its renewable portfolio standard (RPS) of 15% by 2025.

Several other states added or expanded incentives or requirements for solar energy, including California, Hawaii, Maryland, Massachusetts, Missouri, and Ohio. To date, 28 states have renewable portfolio standards that require a certain amount of energy be generated from renewable sources, with 19 of these states mandating that a portion come from solar or distributed sources.

The Emergency Economic Stabilization Act of 2008 included an eight-year extension of the federal solar investment tax credit that has spurred U.S. market growth over the past three years, SEIA said. “This long-term extension will facilitate the long-term planning and investment necessary for the U.S. solar industry to reach its full potential.”

The organization said that the industry’s growth would be supported by provisions in the American Recovery and Reinvestment Act of 2009. These include a 30% grant program for commercial and utility-scale solar installations to be administered by the Department of Treasury, a Department of Energy loan guarantee program, and a 30% manufacturing investment tax credit to attract investors to the U.S. market. “Crafted wisely, other policies being debated at the national level—electric transmission infrastructure, national RPS, and global warming legislation—would also stimulate continued growth of the industry,” SEIA added.

Two General Solar Technologies Used in Utility-Scale Systems

In general, there are two alternative technologies used to harness the sun’s energy. The first, and most familiar, is to use photovoltaic (PV) panels. Hook up enough panels in series and parallel and just about any voltage and current can be produced (Figure 1).

In addition to the large space required for PV systems, the major drawback is that electricity doesn’t flow if the sun doesn’t shine. The price of a utility-scale PV electricity producer directly proportional to the price of the PV panels which are rapidly becoming a commodity. The price of a PV plant will also increase if motorized sun-tracking systems are included.


Figure 1. The 6.3 MW Mühlhausen Solarpark under construction in 2006. The cost of the photovoltaic plant was about $6,250/kW. The 57,600 PV panels are spread over 62 acres.

Our focus in this article is the second, and much more interesting, family of solar thermal or concentrating solar power (CSP) plants. Concentrated sunlight has performed useful work for humans for many years. There is one notable example of Auguste Mouchout, a French inventor, who in 1866 successfully powered a steam engine with sunlight, making him the first known person to construct a concentrating solar-powered mechanical device.

Concentrated Solar Projects Build on Familiar Technology

CSP plants have a marked resemblance to conventional steam plants. The obvious difference is the fuel source: A CSP system concentrates solar radiation to either heat an organic working fluid or to superheat steam, which then is expanded in a turbine-generator to produce electricity. In both cases, the working fluid is condensed after its expansion and returned to the collector to close the cycle.

Existing CSP plants use one of three alternative collector designs (Figure 2). In the central receiver approach, large mirrors track the sun and concentrate solar energy on a central tower to heat a working fluid. A working fluid also is heated in the third approach, but by parabolic dish reflectors that concentrate solar energy at the focal point of the individual dish.

Figure 2. Comparing solar thermal power technologies. Photos courtesy National Renewable Energy Laboratory


CSP: Steaming with Solar

Today, a typical CSP system requires several unique components to produce electricity: a concentrator, a receiver, a heat transfer system, and a power conversion device. There are several ways to combine these components to produce a useful solar energy – powered, electricity-producing plant, although the key enabling technology is the collector. Descriptions of the currently available options follow. All of the available CSP technologies require direct sunlight to function and are of limited use in locations with significant cloud cover.

Parabolic Trough Collector. The parabolic trough is considered the most proven technology of all the CSP options. More than 350 MW of electric power have been installed using this technology since the 1980s.

The parabolic trough design begins with a very large curved mirror. The parabolic shape is designed to concentrate solar energy and reflect it onto a single point. The mirror position follows the sun’s movement in the sky, using a motorized device. The cylindrical parabolic reflector is traditionally made of thick glass silver mirrors (4 to 5 mm, or 0.15 to 0.2 inches), but thin glass, plastic films, and other polished metals are also used.

A receiver tube located at the focal point of the parabolic mirror collects the concentrated solar heat energy. This metal tube uses special coatings to maximize energy absorption and minimize heat losses. Flowing inside the tube is a conventional heat transfer fluid (HTF), which absorbs the thermal energy from the concentrated sunlight. Another glass tube, kept under a vacuum to further reduce the heat losses, envelops the receiver tube.

Several receivers connect to a circulating HTF loop, which enables the enlargement of the system. Many loops combine to form a plant that can now leverage its scale to produce economical power. About 4 to 5 acres will generate 1 MW of energy.

The hot HTF produced by the combined loops, serving as the “fuel” source, next enters a steam generator to produce superheated steam. The cooled HTF returns to the CSP modules to be heated again in a closed loop. The generated steam is used to produce power in a conventional steam-bottoming plant. The maximum HTF temperature is ~750F, mainly due to the operational limitation of the synthetic heat transfer fluid (Figure 3).

Figure 3. Parabolic reflectors focus the sun’s energy on a glass tube filled with a heat transfer fluid in this demonstration plant at the Plataforma Solar de Almería in southern Spain. The heat transfer fluid is used as “fuel” to produce steam that can then produce electricity in a standard steam-bottoming cycle. Courtesy: DLR

The two main disadvantages of trough technology are the relatively low maximum HTF operating temperature, which limits the thermal efficiency of the steam turbine system, and the added complexity of the binary fluid steam generator. However, trough technology is well understood and has a good operations record on a relatively large scale. This experience base may give trough designs an advantage over other, more interesting CSP technologies that are still in their infancy.

Fresnel Collector. The Fresnel solar collector is a line-focus system similar to the parabolic trough, although it uses an array of Fresnel reflectors to concentrate the sun’s energy on a series of receivers. Normally, single-axis tracking flat mirrors fixed to a steel structure are used. Several frames are connected together to form a module, and the modules form a long row up to 450 meters (1,470 feet) long (Figure 4).

Figure 4. Fresnel reflectors concentrate the sun’s energy on a receiver at the Plataforma Solar de Almería, Spain. Courtesy: DLR

The receiver consists of one or more tubes located above the mirrors at a predetermined height. The metal tubes have an absorbent coating very similar to that used in trough technology to increase heat absorption. Inside the tubes flows water or a mixture of water and about 70% quality steam. At the exit of the tubes, water and steam are separated, and saturated steam is produced either for process use or to generate electricity using a conventional Rankine cycle power block.

The Fresnel collector approach does have several technical advantages over the more familiar parabolic lens CSP technology: It can generate steam directly, without the need for an intermediate HTF or binary steam cycle; the optical precision required of the Fresnel lens is less than for a parabolic lens; and the system more easily lends itself to factory production, which means less field construction is required.

However, Fresnel technology is much less mature, and the lower steam temperatures keep the steam turbine cycle efficiency low. In addition, the lower optical efficiency of the Fresnel receiver increases heat losses due to the absence of insulation around the receiver tubes.

Solar Power Tower. In this concept a boiler or gas turbine on top of a tall tower receives concentrated solar radiation from a field of heliostats, which are two-axis tracking mirrors (Figure 5). The heat transfer media could be water or steam, molten salts, or compressed air, although water is usually selected. The heated water temperature — close to 1,015F) — is higher than in any other line-focus system.
Figure 5. German Aerospace Center (DLR) scientists were ableto demonstrate a hybrid solar-powered gas turbine system (230 kW) at Plataforma Solar de Almería. The sun’s energy is focused on a point at the top of the tower to produce heat energy that may be used to generate steam or to heat air. This plant uses three receivers, connected in series, to gradually heat the compressor air of a 250-kW gas turbine to 1,475F. Courtesy: DLR/Steur

The power tower can be connected to a molten salt storage system, thus allowing the system to operate for periods of low or no incident solar energy. The main advantage of this technology is its ability to provide high-temperature superheated steam. The design does require very accurate alignment with solar rays, plus heliostat controls, to avoid potential damage to the receiver on top of the tower.

CSP Favored in the EU

Though solar thermal tower technology has been around since the 1970s, to date, only one plant in the world commercially generates electricity: Abengoa Solar’s 11-MW PS10 tower just outside Seville, in Spain’s Andalucía desert has been grid-connected since early 2007. Because the technology relies on heat from solar energy that is reflected by mirror arrays (heliostats) onto a tower-mounted receiver, installations tend to be site-specific, expensive, and high-maintenance.

For instance, as Abengoa says on its website, one of the foremost difficulties it has experienced with the 18.5-acre PS10 project is controlling its 624-heliostat field. The heliostats, vertical panels each measuring 1,300 ft2, must be protected from high winds, which could damage their structural integrity. If they are not focused properly, they could damage the receiver. And if they are dirty, they could reduce the already-low total efficiency (15.45%) of the system.

Yet, Abengoa, like several entities around the world, deems the technology “feasible and reliable.” Despite the economic crunch, and egged on by climate change concerns, it has developed and plans to inaugurate in the coming months a second-generation plant, the PS20, on land adjacent to the PS10. Like PS10, the PS20 will use water as the operating fluid, delivering 55 MWt of saturated steam at 500F to turn an electricity-generating turbine. But it will be much larger: Using 1,255 heliostats, PS20 is expected to produce up to 20 MW of power (Figure 6).
Figure 6Spain’s Abengoa Solar is wrapping up construction of a second-generation solar tower, the PS20, at its Solucar Solar Park at Sanlúcar la Mayor, just outside Seville. Thenew plant will consist of 1,255 heliostats and a 525-foot-tall tower. Courtesy: Abengoa Solar

Several announcements in the past year hint that even larger and more diverse solar tower projects are on the horizon. BrightSource Energy, for example, will this year begin construction on the first of three phases of the 400-MW Ivanpah Solar Power Complex in California’s Mojave Desert. The first two phases alone will require the installation of 100,000 heliostats; the entire facility will cover 3,900 acres.

Spanish company SENER, meanwhile, is building a 17-MW facility, with a mirror field measuring 75 acres, also in Andalusia (Figure 7). That plant, Gemasolar (previously known as “Solar Tres”), has been under development for seven years now at the hands of SENER and, more recently, Abu Dhabi firm Masdar. It should be operational by 2011. Unlike PS10, it will use molten salt heated to 1,050F to transfer heat.

Figure 7. SENER and Abu Dhabi firm Masdar are building the 17-MW Gemasolar (previously “Solar Tres”) facility in Fuentes de Andalucía, in the province of Seville. That plant, slated for completion by 2011, will use molten salt technology developed at the U.S. Department of Energy’s Mojave Desert Solar One and Solar Two projects. The companies say Gemasolar will be the “world’s first utility-grade solar power plant with central tower and salt receiver technology.” Courtesy: Torresol Energy

Molten salt technology was developed from U.S. Department of Energy pilot tests in the 1990s of the Solar One and Solar Two projects in the Mojave Desert. This method will give the Spanish plant a thermal storage advantage of 647 MWh (permitting 15 hours of turbine load) over PS10’s paltry 20 MWh with steam (or 30 minutes), according to the Fraunhofer Institute for Systems and Innovation Research. But that advantage will have a cost, the group said: Gemasolar’s total efficiency is 14% — worse than PS10’s 15.45%. Added to that, due in large part to the associated storage technology, Gemasolar’s generation costs will be €78.5 /MWh ($105.3/MWh) higher than that of PS10.

The upsurge in interest in solar tower technology is also evident in the more numerous reports of demonstration projects, real and planned. In June 2008, scientists at the German Aerospace Center put into commission a solar-hybrid gas turbine that uses concentrated solar radiation and biodiesel at a test plant on the CESA-1 solar tower of the Plataforma Solar de Almería in Spain.

More recently, German firm Kraftanlagen München has said it is completing construction of the federally sponsored Jülich power tower, a 1.5-MWe project based on the volumetric effect to increase efficiency. Solar radiation concentrated by a mirror field of 18,000 square meters is absorbed by the porous structure of a volumetric receiver and converted to heat: As ambient air cools the outer part of the receiver, it heats up to about 1,300F at the inner surface. That heat is then used in the conventional steam power process.

Meanwhile, nations like Namibia, Morocco, and India are awaiting an Australian test of a solar tower technology that uses a chimney-like updraft to drive turbines and has the potential capacity of about 400 MW. That technology is based on Spain’s 50-kW Manzaneres prototype, which operated for seven years in the 1980s before it was decommissioned.

Even China has gotten into the game. The China Academy of Sciences has designed and is looking to build and test its first solar tower by 2010 in suburban Beijing. The $14.6 million Dahan project, rated at 1 MW, would comprise 100 curved heliostats to concentrate radiation on a 100-meter-tall tower, using water as the heat transfer fluid (Figure 8).

According to project leader Dr. Zhifeng Wang of the academy’s Solar Energy Laboratory, the plant differs from existing tower plants in two ways. First, it has a two-stage thermal storage system: A high-temperature storage tank would use “conductive oil” to produce superheated steam while a second low-temperature tank downstream would use steam as the storage medium to produce saturated steam. Second, to make full use of the land on which the plant will be used, the design also calls for growing crops under the heliostats, according to Wang. If is the Dahan project is successful, it could be expanded to a 5- to 10-MW project by 2015.

Figure 8. This is an artist’s conception of the solar tower technology planned for use in suburban Beijing. Courtesy: China Daily


Portions of this article appeared in POWER magazine. Sonal Patel, Senior Writer, originally prepared several segments of this article.


  1. Brian Washom  

    What about PowerSmith at PowerSmithgroup.com? They appear to have a technology that eliminates HTF and makes feasible Direct Steam Generation when applied to hybrid and combine cycle power stations.


  2. Alp Onyuru  

    I am really surprised to see the author has left out the concentrating solar dish technology which is the technology to be used in Southern California to produce a total of 1,600 MW of grid connected power starting in 2011. This technology being developed by Stirling Energy Systems has achieved the highest efficiency of any solar thermal technology by surpassing the 30% mark last summer.
    I can only assume that the author did not do a thorough research on the available technologies.


  3. Robert Peltier  

    Patience, please. As mentioned at the beginning of the article, this post discusses CSP and related pilot projects. Part II covers actual projects (including Stirling Energy) and Part III focuses on hybrid projects with coal and combined cycle plants.


  4. Tomazo  

    Great work! Things don’t look much changed (other than some big projects were completed – economic or not) since I evaluated this about 30 years ago – so this is a good overview. Looking forward to parts II and III!


  5. High Capital Costs Plague Solar (RPS mandates, cost dilution via energy mixing required) Part III - Master Resource  

    […] Also, given the lower capacity factors, the amortized cost of transmission per unit of energy carried is almost four times as high given the wide difference in capacity factors. We explored this systematic problem earlier. […]


Leave a Reply