Chapter 2: Basic Principles of Solar Technology: 2.1 Photovoltaic Technologies
Chapter 2: Basic Principles of Solar Technology: 2.1 Photovoltaic Technologies
Chapter 2: Basic Principles of Solar Technology: 2.1 Photovoltaic Technologies
2.1 Photovoltaic Technologies The fundamental principle of converting the 5.2 kWh of solar energy that on average falls every day on every square meter in our geographical region1 into electrical energy is relatively simple. The basic idea is that the atoms in a semiconductor placed in the sunlight will absorb photons from the suns radiation. If these photons are of high enough energy, an electron in the valence band will use the absorbed energy to move up into the conduction band of the semiconductor, which allows it to move freely through the semiconductor to an electrical contact from where it is driven by a small voltage through wires as current. The energy the electron needs to move from the valence band to the conduction band is called the band gap of the semiconductor. For this to work in an efficient manner there must be two kinds of semiconductor in the solar cell. One, the n-type semiconductor, is doped with impurities such that it has an excess of electrons. The ptype semiconductor has a deficit of electrons, creating "holes" where it is missing electrons. When the two are placed in contact to create a p-n junction, electrons from the n-type semiconductor diffuse into the p-type semiconductor, leaving holes behind, until enough electrons accumulate on the p-type side of the boundary and enough holes accumulate on the n-type side of the boundary that an opposing electric field is created and equilibrium is reached. Photons are then absorbed within the p-type semiconductor, freeing electrons which will flow through the p-n junction to the electrical contact, while the hole goes the other way. (See Figure 2.1) The electric field at the junction is what drives the electrons around the circuit. The electron will travel into a circuit and do work before recombining with the hole. Since the first solar cell was built in 1883 out of selenium and gold (with an efficiency of 1%), the variety of technologies for converting solar radiation to electricity has greatly expanded and improved, to the point where efficiencies of 40% are now being achieved, with even higher efficiency possible in the near future. (See Figure 2.2) 2.1.1 Crystalline Silicon Cells These conventional cells are generally made from layers of silicon a few hundred micrometers in thickness. Silicon for bulk cells is refined and grown into lightly p-type doped crystalline ingots that are then sliced into extremely thin wafers. However, the size that the crystalline wafers can be cut to is still very thick when compared to thin film solar cells, and when considering the vast areas the wafers must cover it adds up to a highly intensive use of silicon. To make the wafers into solar cells, n-type dopants (often phosphorous) are diffused across the surface, creating the p-n junction.
1
National Renewable Energy Laboratory. National Solar Radiation Database: Typical Meteorological Year 3, Burbank Airport. 2005. http://rredc.nrel/solar/old_data/nsrdb/1991-2005/tmy3/.
2-1
Figure 2.1: Incident light is absorbed by the solar cell (some is reflected) and generates an electron and a hole, which migrate to the cell's electrical contacts. Source: Wikimedia user Cyferz. (Image licensed under GNU Free Documentation License)
Figure 2.2: Over the past thirty years the number of photovoltaic technologies and their effectiveness has greatly increased. Source: National Renewable Energy Laboratory. (Image in public domain)
2-2
The vast majority of solar cells sold commercially are crystalline silicon cells. As the second most abundant element on Earth, silicon is an almost inexhaustible resource. However, the supply of refined silicon for semiconductors is far from inexhaustible. For the past couple of years the industry has been facing a shortage of refined silicon due to growing demand that has kept the price of silicon solar cells from decreasing at their normal rate. New factories that are currently beginning production are expected to increase the supply of refined silicon. In the meantime, competing thin film technologies have recently been making a significant appearance, the shortage having made them economically competitive. 2.1.1.1 Mono-crystalline Silicon Cells Mono-crystalline cells are made from a single large crystal wafer of silicon. These cells have high efficiency, but are expensive due to the demanding production process. These crystals are grown using the Czochralski process in cylindrical ingots, and thus do not completely cover all the area of a square solar cell. 2.1.1.2 Poly-crystalline Silicon Cells Less expensive than mono-crystalline cells, wafer. Source: Armin Kbelbeck. (Image licensed under GNU Free poly-crystalline cells are also less efficient. Documentation License) They are made of silicon wafers cut from square cast ingots of silicon. The method of casting a poly-crystalline wafer of silicon as opposed to a mono-crystalline wafer requires much less precision and expense. 2.1.2 Thin Film Technologies In order to counter the processing, materials, and handling costs associated with crystalline silicon cells, much research has gone into perfecting methods of making solar cells with semiconductors only a few micrometers in thickness. The hope is that these cells will be able to achieve reasonable efficiency while using very little silicon and employing roll-to-roll processing. These cells have lower efficiency than crystalline silicon cells but frequently have costs that are low enough to make them competitive. 2.1.2.1 Amorphous Silicon Cells Amorphous silicon (a-Si) - the non-crystalline form of silicon - can be deposited onto a conductive substrate in a layer a few micrometers thick to create a thin film solar cell. The deposition process of applying a-Si allows it to be less than 1% of the thickness of a
Figure 2.3: A polycrystalline silicon
2-3
crystalline cell.2,3 Amorphous silicon cells are often built using two or three junctions to increase the amount of the solar spectrum they can utilize. Typically alloys of a-Si and germanium are used to create the additional junctions in the multi-junction cells. These cells are lighter, use much less material and are less energy intensive to produce than bulk silicon cells. However, the cell efficiency of amorphous silicon is much lower than crystalline silicon due largely to the increased recombination of the electron-hole pairs that results from the lower carrier mobility.4 Early versions of thin film amorphous silicon cells have been used for decades to power calculators. 2.1.2.2 Cadmium-Telluride Cells The crystalline compound cadmium-telluride (CdTe) is an effective solar cell material - it is a very strong absorber of light and has a band gap almost perfectly tuned to match the solar spectrum. To create a p-n junction for solar cells a layer of cadmium sulfide is added to the CdTe. Because of its effectiveness a CdTe solar cell uses only about 1% of the semiconductor material that bulk silicon cells use.5 CdTe solar cells are generally somewhat less efficient than bulk silicon cells, but have lower costs associated with them due to the smaller amount of material used and inexpensive production methods. While sales of low-priced CdTe cells have sharply increased, the soaring consumption of the very rare metal tellurium has pushed the price of that element up sharply.6 It remains to be seen how the production of CdTe solar cells will be affected in the future by supply constraints. CdTe is a toxic carcinogen and some concerns have been raised about the danger of solar cells made with CdTe. These concerns have been countered by noting that the Cadmium contained in one square meter of a CdTe cell is less than that within a size-C NiCd flashlight battery and that the CdTe is very well sequestered by the encapsulation of
2
T. Sderstrm, F.-J. Haug, V. Terrazzoni-Daudrix, and C. Ballif. J. Appl. Phys. 103, 114509. June 11, 2008. Optimization of Amorphous Silicon Thin Film Solar Cells for Flexible Photovoltaics. http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JAPIAU000103000011114509000 001&idtype=cvips&gifs=yes.
3
J. Poortmans, G. Beaucarne. IMEC. May, 2006. Advanced Industrial Multicrystalline Silicon Solar Cells and Epitaxial Solar Cells. http://www.imec.be/wwwinter/energy/advanced_multi_and_epi.shtml.
4
G. J. Lee, J. Park, E. Kim, Y. Lee, K. Kim, H. Cheong, C. Yoon, Y. Son, and J. Jang. Opt. Express 13, 6445-6453 . August 2005. Microstructure of Femtosecond Laser-Induced Grating in Amorphous Silicon, http://www.opticsexpress.org/abstract.cfm?URI=OPEX-13-17-6445
5
First Solar. Online promotional material. http://www.firstsolar.com/material_sourcing.php. Accessed July 23, 2008.
6
United States Geollogical Survey. Mineral Commodity Summary. January 2008. http://minerals.usgs.gov/minerals/pubs/commodity/selenium/mcs-2008-tellu.pdf.
2-4
the cell.7 2.1.2.3 Copper-Indium/Gallium-DiSelenide Cells In one of the most immediately promising technologies, cells are made from a combination of copper-indium-diselenide and copper-gallium-diselenide (CIGS cells) and have recently been shown to be up to 19.9% efficient,8 the highest efficiency of any
Figure 2.4: Different wavelengths of the solar spectrum are absorbed by stacked cells with different band gap energies. (Image in public domain)
thin-film cell.9 The mixture of the two materials creates a more complex and effective heterojunction, where the junction is formed by semiconductors of dissimilar band gaps, rather than the conventional cell junction of p-type silicon and n-type silicon, which have identical band gap energies. The band gap of the semiconductor in CIGS cells can be varied by altering the ratio of indium to gallium, allowing the band gap to range continuously from 1.0 eV to 1.7 eV in the cell,10 which matches very well with the solar spectrum. CIGS is also a relatively easy material to work with, not requiring the complex
7
National Renewable Energy Laboratory. Cadmium Use in Photovoltaics: The Perceived Risk and the Scientific Evidence. http://www.nrel.gov/pv/cdte/cadmium_facts.html.
8
SolarBuzz. NREL Sets New CIGS Thin Film Efficiency Record. March 30, 2008. http://www.solarbuzz.com/news/NewsNATE50.htm
9
A. Noufi and K. Zweibel. National Renewable Energy Laboratory. High Efficiency CdTe and CIGS Thin Film Solar Cells: Highlights and Challenges. May 2006. http://www.nrel.gov/pv/thin_film/docs/wc4papernoufi__.doc
10
Ibid.
2-5
vacuum deposition process that some other photovoltaic semiconductors require. Large scale production and distribution of CIGS cells have begun only in the past few months. 2.1.3 Multi-Junction Concentrators Multi-junction cells layer two or more p-n junctions which each have semiconductors with a band gap tuned to absorb a different portion of the solar spectrum. Photons will hit the top semiconductor - the one with the highest band gap energy - and if the photon energy is above that semiconductors band gap energy, it will be absorbed and produce electricity. If the photon energy is too low, it will pass through the semiconductor to be absorbed by one of the lower semiconductors in the cell. Because the advanced engineering often required to craft multi-junction cells can create high costs, concentrators are used to gather sunlight from a wide area and focus it on a small solar cell, reducing the amount of photovoltaics needed. The concentrators are usually mirrors and lenses that track the sun and enhance the intensity of the light incident upon the photovoltaic cells to an intensity dozens of times higher than unconcentrated sunlight. 2.1.3.1 Gallium-Arsenide Cells A triple junction gallium-arsenide (GaAs) cell may use gallium-indium-phosphide (GaInP), gallium-arsenide, and germanium (Ge) p-n junctions. Combined with concentrators, these cells are by far the most efficient yet developed, achieving 40.7% efficiency in the laboratory.11 Cells with efficiencies as high as 36% are currently commercially available.12 Although GaAs cells are highly efficient, their high cost currently discourages their more widespread usage. 2.1.4 Emerging Photovoltaic Technologies There exists a wide variety of photovoltaic technologies in the research stages that have not yet entered into common commercial production. These technologies are the subject of cutting-edge research at universities, government laboratories and start-up companies that are seeking photovoltaic cells that can be manufactured more efficiently, have greater quantum efficiencies than available technologies, and are composed of abundant non-toxic materials. 2.1.4.1 Organic Cells Still mostly in the university research stage of development, organic/polymer solar cells are relatively new and have a long way to go before they begin competing with the more established thin film technologies. Organic cells function in a slightly different way than most other cell technologies: instead of semiconductor p-n junctions, organic cells utilize
11
U.S. Department of Energy. New World Record Achieved in Solar Cell Technology. December 5, 2006. http://www.doe.gov/news/4503.htm.
12
Spectrolab. CDO-100-IC Concentrator Photovoltaic Cell product data sheet. March 24, 2007. http://www.spectrolab.com/DataSheets/TerCel/C1MJ_CDO-100-IC.pdf.
2-6
electron donor and acceptor materials. Typical choices are polymers for the electron donors and fullerenes for the electron acceptors. When an electron-hole pair is created by the absorption of a photon in the donor, rather than separating and migrating to opposite sides of the cell, the electron and hole stay together as an exciton. The exciton diffuses through the cell until it reaches the acceptor where the electron is transferred to the acceptor material, creating a current through the acceptor. Because the exciton will only travel for a few nanometers before the electron and hole recombine, it is highly beneficial to have donor and acceptor blended together, forming a bulk heterojunction. Research is ongoing into methods of effectively fashioning ordered heterojunctions that would be more efficiently structured. The benefit of using of organic materials is that it allows for the simple highvolume low-temperature fabrication of flexible solar cells on plastic substrates. If the efficiency of the cells can be improved, organic cell technology will realize extremely low cost production of very versatile cells. Efficiencies of organic solar cells are currently around 5-6%, although quickly rising. Organic solar cells must also overcome the issue of environmental degradation, a problem that most organic molecules face when exposed to oxygen, water and solar radiation. There remain many opportunities for breakthroughs in organic solar cell design, given the newness of the technology and great diversity of materials and techniques available. 2.1.4.2 Dye-Sensitized Cells Dye-sensitized solar cells (DSSCs) are quite different in function from the other photovoltaic technologies in existence, and in fact bear a somewhat loose resemblance to photosynthesis. The material components of a DSSC are a thin film of electrolyte sandwiched between two electrodes (the top electrode being transparent to allow light into the cell) with a lattice of dye-coated nano-scale titanium-dioxide (TiO2) particles coating one of the electrodes. The DSSC works on the principle of splitting the functions performed by the semiconductor in other photovoltaic technologies. In a silicon cell, the silicon both generates the charge, and conducts the electrons and holes to the electrodes. In a DSSC, the incident photons excite electrons in the dye molecules. If given sufficient energy, the excited electrons will escape from the dye to the conduction band of the TiO2 particles and will then diffuse to the electrode, generating a current. The electrons return to the dye through the electrolyte. DSSCs have the ability to provide power under conditions where other solar cells cannot. With most solar cells, if there are not enough electrons flowing through the semiconductor, the few that are will be lost as they recombine with the holes. DSSCs do not have this problem of recombination (often a serious drag on the efficiency of solar cells even during periods of high intensity light) due to the separation of the electron producer and the electron carrier, and can thus function robustly under limited light; under cloudy skies, and even indoors. This property was recently exploited during an Antarctic expedition where DSSCs were used to provide a significant portion of
2-7
Figure 2.5: The basic structure of a dye-sensitized solar cell. Source: Wikimedia user Oldboltonian. (Image in public domain)
the expedition's power under the extreme conditions.13 The widespread availability of the constituent materials and a simple manufacturing process similar to inkjet printing means that DSSCs could provide one of the lowest cost options for solar cell technology. With efficiencies above 10% already achieved with relatively little research into the technology, DSSCs could become more prominent in the near future. As of now there is one company that has just begun selling DSCCs in the form of mobile recharging units for electronics. Besides this, the technology has yet to be commercially exploited. 2.2 Cost Projections for Photovoltaic Technologies The price of photovoltaic systems has declined with improved engineering, new technologies and economies of scale. Solar panels are now a wise economic investment in many parts of the country where there is abundant insolation, favorable rebates and/or high electricity prices. Photovoltaic installations remain cost ineffective in the short term in locations that do not have these features, such as the Claremont Colleges, which have a great deal of insolation and extensive rebates through the California Solar Initiative, but pay extremely low electricity rates. While simple economics is often not the most important factor to consider when determining whether or not to install a photovoltaic system, for large scale adoption of solar electricity generation, it is essential that solar energy reaches price parity with grid-available electricity. Although dependent upon uncertain energy prices, with such advances as roll-to-roll processing of thin film panels, solar panels may soon become cost competitive with coal and natural gas as electricity
13
G24 Innovations. Revolutionary Solar Technology Used to Power Antarctic Mission. http://www.g24i.com/press,revolutionary-solar-technology-used-to-power-antarctic-mission,86.html
2-8
Figure 2.6: The price of photovoltaic modules has steadily declined with increased production, with the exception of the recent silicon shortage. Image authored by Solar Energy Technologies Program, US DOE. Image in public domain.
sources. The cost per installed watt for residential systems is projected to decline to two thirds of 2006 prices by 2010 and to one third of 2006 prices by 2015.14 Figures 2.7 through 2.9 show NREL maps that detail the projected market penetration of residential photovoltaics, showing the cost difference between conventionally generated electricity and electricity generated from solar sources. Figures 2.7 - 2.9 indicate that advances in solar technology along with projected increases in electricity prices will make solar electricity cheaper than conventional electricity in much of the country within seven years. The moderate case assumes a 1.5% annual increase in electricity rates, while the aggressive case assumes a 2.5% annual increase. The 2006 cost per installed peak watt of photovoltaics is $8.50/watt, while the 2015 cost is projected to be $3.30/watt. Recent technological innovations, along with an unexpected surge in energy costs, suggest that cost parity between solar electricity and fossil fuel electricity may arrive sooner than expected. While the NREL projections above assume that the cost of photovoltaic modules will be $3.30 per installed watt in 2015, Southern California Edison has announced that it expects to pay $3.50 per installed watt for the 250 megawatts of
14
United States Department of Energy. Solar Energy Technologies Program. Solar Energy Industry Forecast: Perspectives on U.S. Solar Market Trajectory. May 27, 2008. www.earthday.net/files/doe.ppt.
2-9
Figure 2.7: 2007 residential PV and grid electricity price difference with existing incentives. PV is only cost competitive in a few locations. Image authored by Solar Energy Technologies Program, US DOE. Image in public domain.
Figure 2.8: 2015 residential PV and grid electricity price difference with moderate price increase and no incentives. PV is cost competitive in a quarter of the largest utilities. Image authored by Solar Energy Technologies Program, US DOE. Image in public domain.
Figure 2.9: 2015 residential PV and grid electricity price difference with aggressive price increase and no incentives. PV is cost competitive in almost half of the largest utilities. Image authored by Solar Energy Technologies Program, US DOE. Image in public domain.
2-10
solar panels it plans to place on rooftops in Southern California.15 Nanosolar, Inc. has begun selling CIGS thin film cells that are produced with nanoparticle ink printed on rolls of aluminum foil, are 14.5% efficient16 and have been rumored to be selling for around $1.00/watt.17 Nanosolar is currently selling only to select utilities, and has yet to reveal much data on their panels in order to extend the period of protection on their products. While the price of solar electricity is decreasing faster than anticipated, it is clear that the price of electricity - closely tied to spiking energy costs - will be increasing more than expected. It may not be long at all before solar panels are able to provide electricity at the same cost as fossil fuel power plants. 2.3 Solar Thermal Technologies Solar thermal technology is the perfection of the idea of warming a cup of water by sitting it in the sun. While photovoltaic technology attempts to achieve efficient conversion of solar energy into electrical energy, solar thermal technology converts it into thermal energy, a much simpler and efficient task. A solar thermal cell utilizes all of the energy of incident photons that it absorbs to heat up a fluid that can then transfer that heat to places where it can be used, such as a swimming pool, a boiler, an absorption chiller or a power plant turbine. Solar thermal systems can have several advantages over photovoltaic systems. Being a simpler technology, they are generally cheaper than photovoltaics, while at the same time being more efficient at converting energy. An advanced solar thermal panel converts solar radiation to thermal energy with around 60% efficiency, although there is a wide variation among the different models available. Solar thermal systems have a wider variety of application than photovoltaic systems, being able to generate steam that can be used to turn a turbine to create electricity, or generate hot water to heat a building, provide domestic hot water, run an absorption chiller to cool a building, or any other application which requires hot water. Some systems can perform multiple functions with the same array of panels. With the use of a heat transfer fluid, the energy that a solar thermal panel generates can be stored in the fluid for extended periods of time, while a photovoltaic panel requires expensive batteries for electricity storage. Thus, solar thermal systems can continue to provide some energy during the night or during cloudy periods.
15
Southern California Edison. Southern California Edison Begins Construction of World's Largest Solar Panel Installation Project. July 16, 2008. http://www.edison.com/pressroom/pr.asp?bu=&year=0&id=7083
16
Nanosolar, Inc. High-Performance Thin-Film Photovoltaics Using Low-Cost Process Technology. Presented at 17th International Photovoltaic Science and Engineering Conference, Tokyo, Japan, December 2007. http://www.nanosolar.com/cache/PVSEC17_ns_dft.pdf.
17
2-11
2.3.1 Unglazed Collector Unglazed collectors have no insulation; they are usually just pipes placed in the sun with a black coating to enhance absorption. Without insulation they cannot hold onto heat at temperatures much above that of the surrounding air. However, when only ambient temperature (or slightly above) needs to be reached, they can be extremely efficient at heating the water within the pipes. The greatest use for unglazed collectors is for heating swimming pools. 2.3.2 Glazed Collector These systems function well for heating water to moderate temperatures of 100-200 F (35-100 C) and are suitable for domestic hot water heating. Their improvement over unglazed collectors comes from added insulation that allows the water to be heated to higher temperatures without losing much of its energy to the air. The absorber - usually a copper pipe - is coated with a selective surface that collects solar radiation very well, but has low emittance at the operating temperature. Black chrome is a commonly used coating. Glazed collectors are mounted in flat glass-enclosed frames. 2.3.3 Parabolic Trough A cylindrical row of mirrors collects the sun's rays and focuses them on the heat collection element that runs along the length of the trough at the focal point of the mirrors. The heat collection element consists of a stainless steel tube within an insulating evacuated glass tube. The steel tube is coated with a solar-selective absorber surface while the glass tube is coated with an anti-reflective surface. A heat transfer fluid, such as water or molten salts is pumped through the steel tube and is warmed as it passes through the focal point of the trough. Parabolic trough systems are normally designed for utility-scale electricity generation, with the transfer fluid being used to generate steam for a turbine, but there are some roof-top scale systems available that can run absorption chillers, generate electricity or produce steam for buildings. 2.3.4 Vacuum Tube Collectors If the mirror were taken out of the parabolic trough system, leaving only the heat collection element, the resulting panels would be a vacuum tube collector. These panels feature a row of metal tubes coated with an especially absorbent surface and insulated by being placed within a glass vacuum tube. The process of gathering heat is much the same as with parabolic troughs, only the light that is absorbed by the heat collection element is unconcentrated. Because the vacuum tube effectively prevents any heat losses from convection or conduction, the system's efficiency is largely unaffected by changes in the ambient temperature, wind speed, insolation levels or other environmental factors that would otherwise alter the rate of heat loss. Vacuum tube collectors generally heat fluids to moderately high temperatures, up to 400F (200C),18 to provide space heating and
18
Thermomax, Inc. Sol 25 Plus Flat Plate Solar Collector product data sheet. 2005.
2-12
cooling or domestic hot water, but they do not achieve high enough temperatures for efficient conversion to electricity. 2.3.5 Linear Fresnel Lens Reflector Instead of having one mirror gathering light and focusing it onto one heat collection element, a linear Fresnel lens reflector system takes an ordered array of flat or slightly curved mirrors and focuses the collected light from the array onto one heat collection element that is perched above the array of mirrors. See Fig 2.10. The heat collection element can be hundreds of meters long in a large system. By reducing the number of heat collection elements needed and by making the mirror array structurally simpler, the linear Fresnel lens reflector system can achieve lower costs than the parabolic trough system. Such a system can achieve a $5 per installed peak watt capital cost.19
Figure 2.10: A compact linear Fresnel lens reflector featuring staggered mirror orientation. Source: Applied and Plasma Physics, School of Physics, University of Sydney.
The linear Fresnel lens reflector design can be improved by alternating the orientation of the mirrors such that one array of mirrors can focus light onto two or more heat collection elements. This method reduces the shading effect that the mirrors have on their neighbors, thus allowing the mirrors to be placed closer together. A 177 MW power plant in San Luis Obispo County utilizing this compact linear Fresnel lens reflector technology is currently in the permitting process and is expected to begin producing electricity in 2010.
http://www.thermomax.com/Flat_Plate_Solar_Collector.php.
19
Applied and Plasma Physics, School of Physics, University of Sydney. Compact Linear Fresnel Reflector Power Plant Technology. http://www.physics.usyd.edu.au/app/research/solar/clfr.html. Accessed August 6, 2008
2-13
Figure 2.11: Efficiency curves of flat panel solar thermal technologies. Source: SunEarth Inc.
2.3.6 Solar Stirling Engine Invented in 1816, the Stirling engine features a contained gas which expands and contracts when heat is applied from an external source, pumping a piston which can run an electric generator. The engine is similar to the familiar internal combustion engine, although with an externally applied energy source, rather than igniting an injected fuel. The engine can theoretically be very efficient, operates with almost no maintenance and can be powered by any source of heat. A small number of companies have recently started building Stirling engines that utilize concentrated solar power to heat the pressurized gas that runs the engine (commonly hydrogen gas). In most designs, a large dish mirror focuses sunlight onto the Stirling engine in a setup not unlike a satellite TV dish. The efficiency of solar Stirling engine systems at converting solar energy to grid electricity is the highest of any solar technology (those PV technologies with higher efficiencies are small-scale and not grid-connected), having recently achieved 31.25% efficiency.20 Southern California Edison and San Diego Gas & Electric have signed deals with a solar Stirling engine company to construct Stirling engine power plants with a combined capacity of up to 1.75 gigawatts.21 Construction on the SDG&E plant is
20
Sandia National Laboratories. Sandia, Stirling Energy Systems Set New World Record for Solar-toGrid Energy Conversion. February 12, 2008. http://www.sandia.gov/news/resources/releases/2008/solargrid.html
21
Stirling Energy Systems. Sterling Energy Systems Signs Second Large Solar Deal in California. Sept. 7, 2005. http://www.stirlingenergy.com/downloads/7-September-2005-SES-Press-Release-StirlingEnergy-Systems-Signs-Second-Large-Solar-Deal-in-California.pdf.
2-14
expected to begin in late 2009.22 Utility-scale systems such as these are currently the only applications for solar Stirling engines. There is one company that plans to produce smaller dishes that are 3 kW in capacity and 22 ft (6.7 m) high.23 This is a good deal smaller than those to be built for the power plants, but still quite large and heavy, making them unsuitable for roof-top systems.
22
Stirling Energy Systems. Application Filed for World's Largest Solar Energy Generating Plant. June 30,, 2008. http://www.stirlingenergy.com/downloads/30-June-2008-Application-Filed-for-WorldsLargest-Solar-Energy-Generating-Plant.pdf.
23
Infinia Corporation. Stirling Solar product specification sheet. http://www.infiniacorp.com/applications/501336rB_Prod_Spec_English.pdf. Accessed August 4, 2008.
2-15