Photovoltaic technology has been growing steadily for the last ten years, but a new breakthrough is preparing to unleash a whole range of economic, environmental and social benefits for the general market.
Andrew Blakers and Klaus Weber, Centre for Sustainable Energy Systems, Australian National University, Canberra, Australia
The photovoltaic or solar cell industry is 15 times larger than it was a decade ago and the industry is likely to have a turnover of $100 billion per year within a decade. The potential of the photovoltaic (PV) industry is sometimes compared with that of the mobile phone industry in the 1990s, although its long term potential suggests it should be far larger.
Figure 1. Blakers and Weber: Sliver technology offers opportunities for higher efficiency and greater design flexibility (picture courtesy of Canberra Times)
In addition to environmental and economic benefits, stand-alone photovoltaic systems will provide important social benefits to the 1.5 billion people without electricity grid connection by providing electricity for lighting, TV, telecommunications, refrigeration, water pumping, grain grinding and many other applications.
For 30 years the PV industry has been dominated by a single technology. This technology features screen printed metal contacts on single or multi crystalline silicon wafers. There have been many improvements to the technology and wafer diameter has grown to 200 mm. However, there have been no fundamental changes. With the invention of the Sliver solar cell there is an opportunity to introduce a fundamentally different and better photovoltaic technology that will have higher efficiency, lower cost, greater design flexibility and will use far less silicon.
The photovoltaic industry
Pure silicon is used in the growth of silicon ingots that are then sliced to form single or multi crystalline silicon wafers with diameters of 10-20 cm and thickness of about 0.25 mm. The wafers are subjected to various semiconductor process steps to form a solar cell. The solar cells are connected together electrically and packaged behind glass to form a solar power module. Many modules are then grouped together to form a solar power system.
While the photovoltaic and integrated circuit industries both require pure silicon, the integrated circuit industry creates a far higher value product (per cm2) than does the PV industry, and can therefore afford to pay a much higher price per silicon wafer. The PV industry makes use of fallout pure silicon and wafers from the integrated circuit industry, obtained at a substantially reduced price. When upswings in the business cycles of both the PV and integrated circuit industries coincide, as is happening now, the PV industry feels the pressure with both the supply and price of silicon.
Figure 2. The structural make-up of a Sliver cell means that voltage in a series connected string can be built quickly
About 95 per cent of the world market for photovoltaics is supplied using solar cells made on single or multi crystalline silicon wafers. This dominance is being challenged by alternatives based on materials other than crystal silicon, such as amorphous/micro crystalline silicon, copper indium diselenide, cadmium telluride, organic materials and dye sensitized titanium dioxide. These are strongly absorbing materials that only require a few microns to absorb most of the solar spectrum.
However, silicon has many highly desirable attributes, including non toxicity, abundance, high and stable cell efficiencies and the sharing of infrastructure and technology with the integrated circuit industry (which is also based on crystal silicon). Many people are working on methods of manufacturing thin crystalline silicon solar cells so that the attributes of silicon can be retained while reducing the amount of silicon required by a factor of five or more. Most of the techniques involve deposition of thin layers of silicon onto glass or some other substrate. Unfortunately, these techniques result in low quality silicon and relatively low cell efficiencies.
The maintenance of high and stable cell efficiency is important to achieve low PV system costs. The cost of a PV system comprises the cost of the modules plus the balance of systems cost (transport, mounting, interconnection, power conditioning). Most of the balance of system costs depend on the area of module deployed and therefore depend inversely upon the module efficiency. As the cost of PV modules declines, balance of system costs become relatively more important, which favours efficient modules.
Figure 3. Their thinness means that Sliver cells have high power-to-weight ratio and high radiation tolerance
The two most important constraints on the PV industry are the supply of pure silicon in sufficient quantities at an affordable cost and the attainment of high cell efficiencies at an affordable cost. Sliver cell technology removes these constraints.
Sliver solar cells
An alternative means of creating thin crystalline silicon solar cells is the Sliver solar cell process, which was invented by Dr. Klaus Weber and Professor Andrew Blakers at the Australian National University. The Sliver solar cell process uses standard materials and techniques in novel ways to create thin single crystalline solar cells with superior performance and reduced cost. Standard silicon wafers (1-2 mm thick) are used as the starting material. Low cost micro machining techniques such as etching in a potassium hydroxide solution are employed to create a series of very narrow grooves (with a spacing of about 0.1 mm) that extend through the wafer. The grooves lead to the creation of a series of thin silicon strips, ‘Slivers’. The grooves do not extend all the way to the wafer edges so that a frame of uncut silicon remains, which holds the Slivers in place. The wafer is then processed using standard techniques to turn each of the Slivers into a solar cell.
At the end of the process, the Slivers are cut out of the wafer frame and laid flat. By doing so, a large increase in surface area, compared to the surface area of the starting wafer, is obtained. Typical Sliver solar cells are 10 cm long, 1.5 mm wide, 0.05 mm thick and have an area of 1.5 cm2. Individual Sliver cells can be assembled on glass by a variety of means, such as a ‘pick and place’ machine, and are then electrically interconnected to form a solar power module.
A conventional solar cell made on a 15 cm diameter round wafer will have an area of about 140 cm2 after the wafer is ‘squared up’ to better fit in a module. In contrast, a 15 cm diameter wafer will yield about 1000 individual Sliver solar cells with a combined area of 1500 cm2.
Further large savings in silicon can be made by leaving gaps between each Sliver cell in the module. Light that passes through the gap is reflected by a scattering reflector at the rear of the module and has a good chance of intersecting a Sliver cell on the way out. For example, a module with 50 per cent Sliver cell coverage only loses 20 per cent of the light compared with a module with 100 per cent coverage. Saving half the Sliver solar cells easily pays for this loss of light.
The principal advantages of the Sliver process are as follows:
- Silicon consumption is 1-2 kg/kW-rated module, which is a factor of seven to ten less than for conventional silicon solar cells
- The number of 15 cm diameter wafers required to create a
- 1 m2 Sliver module is only two to four compared with 60 to 70 wafers in a conventional module of the same area
- Sliver cells have a high efficiency potential (>22 per cent) because high quality single crystalline silicon is used and sophisticated wafer processing can be afforded.
Nearly half the cost of a conventional PV module is the cost of the silicon wafer, with the balance being shared between cell fabrication and cell encapsulation. A mature Sliver cell technology will virtually eliminate both the wafer cost and the cell fabrication cost while retaining very high cell efficiency. Sliver solar cells have the potential for large reductions in the cost of both PV modules and PV systems.
Features of Sliver modules
Some of the attractive features of Sliver solar cells and modules are as follows:
- The best Sliver cell efficiency achieved is 19.6 per cent. While this is a very respectable figure, it is clear that major improvements will be possible. Modelling indicates an efficiency potential of 22 per cent. Since very few wafers need to be processed per kW, compared with conventional solar cells, sophisticated cell processing can be employed to extract the maximum cell efficiency. The cost efficiency compromise that limits the performance of cost-effective conventional modules is essentially removed.
- Sliver cells are perfectly bifacial, utilising light equally well regardless of which surface the light impinges upon. This allows light reflected from the bright surfaces behind a module to be collected. In addition, novel module mounting configurations can be employed. Normally, photovoltaic modules are mounted facing the equator and elevated at an angle equal to the latitude angle. However, at low-mid latitudes a vertical bifacial Sliver module facing east to west can have a greater annual energy production than a module mounted at the latitude angle. The deployment of Sliver modules on sound barriers on motorways is one possible application.
- Conventional solar cells are too thick to be bent without fracturing. Sliver cells are thin and flexible and can be mounted on curved glass surfaces or on flexible plastic. Curved glass modules have major applications in architecture. While plastic Sliver modules can be rolled up for portability.
- Transparent modules are typically used in architectural applications, where pleasing module design is crucial. A typical application is a curved glass ceiling. Sliver modules can have any desired degree of transparency by choosing the Sliver spacing and omitting the rear scattering reflector. The light transmitted through a Sliver module is natural light, unlike in the case of some other transparent photovoltaic technologies.
- The fact that Sliver solar cells are narrow means that voltage in a series connected string can be built very quickly. For example, 1.5 mm wide Sliver cells build voltage at the rate of about five volts per centimetre. Series connection of Sliver cells can therefore yield high module voltages (50 to 1000 volts). This might avoid the voltage up-conversion stage of inverters, although there are issues to be resolved with respect to electrical safety.
- Sliver cells may find a large market in powering mobile phones, toys, watches, calculators and other small electronic products. The important attributes of Sliver cells that will open this market include their physical flexibility, the ability to build the required battery voltage using only a small panel of cells and their high efficiency.
- Since Sliver cells are thin they can have high power-to-weight ratio and high radiation tolerance. This will allow Sliver cells to be used in satellites, balloons and solar powered aircraft.
- The temperature sensitivity of performance of a Sliver module is lower than for a conventional module. This means that the efficiency advantage of a Sliver module over a conventional module increases in hot climates. As the efficiency and voltage are improved this advantage is expected to increase.
- The energy (and associated greenhouse gas emissions) required to produce a silicon wafer dominates the energy content of a PV module. The energy pay back time is important when calculating the potential for mass deployment of PV to reduce greenhouse gas emissions. Since Sliver modules use only 10 to 20 per cent of the amount of silicon used in a conventional PV module, the energy pay back time is short. Estimates show a pay back time of 1.5 years compared with four years for a more conventional module.
The Australian National University (ANU) is developing reflecting trough concentrator solar systems. A Combined Heat and Power Solar (CHAPS) system has been developed that produces both solar hot water and solar electricity. The combined thermal and electrical efficiency of the system approaches 70 per cent. A 300 m2 demonstration system has been constructed at ANU, and commercialization of the system is expected in 2005.
Figure 4. Sliver cells utilise light equally well regardless of which surface the light impinges upon
Theoretical and experimental studies have shown that Sliver solar cells have excellent prospects in mid-range photovoltaic concentrator systems operating at 10-50 times normal solar intensity. ANU is developing Sliver solar cells and receivers that are suitable for use at the focus of its trough concentrators.
A first-generation of Sliver cell technology has been successfully developed at the ANU. Origin Energy, a large Australian energy company, is investing $40 million in the commercialization of Sliver cell technology. A pilot plant capable of producing 7 MW per year (expandable to 25 MW per year) is nearing completion in Adelaide and the first product is expected in 2005.
The ANU is engaged in the development of second-generation Sliver solar cell technology. Compared with first-generation, the number of solar cell process steps will be reduced by up to 50 per cent. The amount of silicon required per kW can be reduced by a further 40 per cent while Sliver cell efficiency can be increased from 19 per cent to above 22 per cent. Second-generation Sliver module construction technology will be a radical improvement over first-generation technology. Streamlining of the cell fabrication process will allow the cost of a Sliver solar cell factory, which is already only one third that of an equivalent conventional solar cell factory, to be further reduced.
Figure 5. The pilot Sliver cell plant will be capable of producing 7 MW
Sliver cell technology is nearly ideal as a candidate for the replacement of current photovoltaic technology. It has the potential to substantially reduce costs and increase efficiencies, and open new markets for photovoltaic products. And yet there is still plenty of room for further improvement of the technology.