Alternative energy has a bright future

A new type of solar cell should have high enough efficiency - and low enough cost - to be used in 'power generating' windows. Sharp Electronics says that its microcrystalline technology, which can be applied by vapour deposition, has efficiencies approaching that of expensive crystalline devices - but at a cost that should eventually be little more than cheaper amorphous versions. This would allow cells to be formed on large areas of glass window using only very small amounts of silicon - none of it in the form of delicate wafers of zone-refined crystal. The first applications will be in premium-priced products that produce light from built-in LEDs. These Lumiwall panels are powered from batteries that were charged by the cells during the day. In future, the cells could be incorporated into the window itself. Sharp says that its tandem structured thin film solar cell technology and 'LumiWall' panels have been available for some time. However, this is the first time it has explained how the technology works - and demonstrated working product samples. The company is the world's largest manufacturer of photovoltaics, producing just over a quarter of the world total last year. While demand is increasing, crystalline photovoltaics are expensive. Eight square metres of modules, generating 1kW, fully installed on a house roof in the UK costs around £8,000 including VAT. Such an array would generate about 850kWh per year. Lifetime of the cells is unknown, though Sharp cells installed in the early 1950s in lighthouses off the Japanese coast are still working well. The cost of installing solar cells on a roof is expensive - and so are the cells. There is currently a shortage of electronics grade silicon and producers can extract a premium from chip makers. Furthermore, the wafers are extremely thin - 180 microns thick, with plans to reduce them to 150 microns - and very delicate. Eureka has seen the process for incorporating wafers into solar cell modules at the company factory at Wrexham in North Wales. According to manager Gordon Butler, the first month of production in March 2004 resulted in only 200 working modules. Even now that the manufacturing processes have been optimised, and production is up to 110MW per month - 2,400 panels per 24 hours in February 2006 - as many as 5% still need re-work. Examination of the machines reveals collections of broken wafers rejected by the robotics. Monocrystalline cells have conversion efficiencies of around 20%, but are the most expensive. Polycrystalline cells, mostly derived from offcuts from single crystal production, have efficiencies of 8-12%. Amorphous cells, on the other hand, can be vapour deposited onto glass or plastic. They use only 1% of the silicon required for crystalline cells, but have efficiencies of 4-6%. The key to the new microcrystalline technology is that it is also vapour deposited, so can be laid down on glass. It also requires 1% of the silicon required for crystalline cells, but achieves a 50% greater conversion efficiency than amorphous. The company also offers concentrator cells, which use Fresnel lenses to concentrate light onto small area crystalline cells. These use the least amount of silicon of all (about 0.2% of crystalline wafers) and achieve efficiencies up to 25% - but the arrays must track the sun. Rather than going into the low end of the market, however, Sharp has taken the decision to make products that it can sell at a premium with maximum added value - hence LumiWall. LumiWalls are scheduled to go into volume production in 2007, though probably not in Wrexham, which is already at full capacity producing conventional photovoltaic modules. Initial applications for Lumiwalls are seen in conservatories, back-lit advertising signs and bus shelters. Presently, most photovoltaic cells are sold to local authorities and companies who want to be seen to be green - such as the CIS Tower in Manchester, which has walls covered in photovoltaic cells. The economics of using photovoltaics to power isolated units, however, are much better, because they do away with hard wiring to mains supplies. Once installed, they require minimal maintenance and can last indefinitely. Apart from Sharp's polycrystalline 'Tandem' technology, a number of other techniques could allow mass manufacture of solar photovoltaic cells with acceptable conversion efficiencies on large sheets of glass or plastic. One, presented at the recent Oxford Venturefest event, is based on artificial photosynthesis. However, all are still at the laboratory stage, and the technical challenges to overcome in order to get into volume production will be far from trivial. It is likely though that Sharp, with its well-established flat screen technology, will soon be offering flat television screens powered by photovoltaics on their reverse surfaces - be they on patio doors, windows or larger panels. Sharp Electronics (U.K.) Eureka says: Solar powered photovoltaics may have finally broken through into main stream, large scale power generation Growing efficiency Semiconducting photovoltaic solar cells with efficiencies approaching 40%, and cells converting infrared radiation directly into electricity with efficiencies of up to 80% are not impossible. All current solar photovoltaic cell technology comes out of the space race. The first practical photovoltaic cells were developed in AT&T's Bell Laboratories in 1954 and were first used on the US Vanguard 1 satellite in 1958. The 34 cells produced just 10mW. Since then efficiencies have been steadily improved. The Fraunhofer Institute says that it has developed a new range of devices for space use with efficiencies just over 30%. More than this will be difficult because maximum efficiency is only obtained by taking energy from visible light photons in several bites. An alternative to semiconductors is dye solar cells, such as those shown by the Fraunhofer Institute for Solar Energy Systems at the Hannover Fair earlier this year. Unlike conventional solar cells, dye solar cells use an organic dye to convert light into electricity. They can be made using simple screen printing, which offers a wide variety of design possibilities. Present efficiencies are only 2.5% but Andreas Hinsch, project leader at Fraunhofer ISE, expects improved printing technology to lift this to 5%. "Now is the time to investigate a market that is customised to the particular advantages of dye solar cells. We are looking for partners to invest in first testing systems and demonstration projects." In Japan, an efficiency of 10.4% has been reported for a dye cell with an area of 1 cm². Infrared photons are less energetic than visible light photons, so it is possible to take all their energy in one bite. This technology also comes out of the space race, but this time from military satellites with small nuclear reactors that produce heat that needs to be turned into electricity. Early devices were only about 2-4% efficient but a team of Russian researchers in St Petersburg are reported to have achieved 80%. If this turns out to be true, it would be possible to burn fuel, heat up a radiator and turn it into electricity at much higher efficiency than is possible using an internal combustion engine and a generator. Pointers * Microcrystalline photovoltaic cells are based on vapour deposition and require about 1% of the silicon required for crystalline cells * Conversion efficiencies of 6-9% are around 50% better than amorphous silicon, and approaching those of polycrystalline * They can be laid down on large sheets of glass which just appear to be heavily tinted * It is possible to attach LEDs to the conducting tapes used to gather power from the cells so the same panels can be made light emitting at night and light gathering during the day