October 02, 2003
STMicroelectronics Projects 20 Fold Decline In Solar Photovoltaics Prices
A Reuters article on the CNN site has a story about an effort by semiconductor company STMicroelectronics to drive down photvoltaic solar cell cost per watt by a factor of 20.
Over a typical 20-year life span of a solar cell, a single produced watt should cost as little as $0.20, compared with the current $4.
The article does not provide a prediction of when this price reduction will take place. Also, the article quotes a cost per watt for conventional existing electric power of $0.40 per watt and so the projected future price of photovoltaics will be half the price of existing sources of electricity. However, it is not clear how those costs were calculated. Do they include the costs of, for instance, storing solar power in batteries to have electricity to use when the sun goes down? My guess is that those costs are not included.
Photovoltaics without sufficient energy storage capacity will not reduce the peak electric generation capacity needed by electric utilities but will reduce average demand. So the result will be that a portion of the expected savings will be offset by higher prices charged by electric utilities that can't amortize their physical plant over as many generated kilowatt hours of electricity. What is needed is not just cheaper photovoltaics but also cheaper batteries or other means of electric power storage. Toward that end, advances in nanotechnology may eventually yield materials that combine batteries and solar cells in a single integrated sheet.
A September 30, 2003 press release from STMicroelectronics that is about their photovoltaic solar cell research efforts does not provide cost projections and sounds more qualified in terms of what they expect to achieve. The press release outlines 2 approaches they are pursuing to try to drive down the cost of photovoltaics.
Semiconductor-based solar cells have the highest efficiency (defined as the electrical energy produced for a given input of solar energy) but there is little that can be done to either increase the efficiency or reduce the manufacturing cost. ST is therefore pursuing alternative approaches in which the aim is to produce solar cells that may have lower efficiencies (e.g. 10% instead of 15-20%) but are much cheaper to manufacture.
"Although there is much support around the world for the principle of generating electricity from solar power, existing solar cell technologies are too expensive to be used on an industrial scale. The ability to produce low cost, high efficiency solar cells would dramatically change the picture and revolutionize the field of solar energy generation, allowing it to compete more effectively with fossil fuel sources," says Dr. Salvo Coffa, who heads the ST research group that is developing the new solar cell technology.
The ST team is following two approaches. One of these, invented in 1990 by Professor Michael Graetzel of the Swiss Federal Institute of Technology, uses a similar principle to photosynthesis. In a conventional solar cell, a single material such as silicon performs all three of the essential functions, which are absorbing sunlight (converting photons into electrons and holes), withstanding the electric field needed to separate electrons and holes, and conducting the free carriers (electrons and holes) to the collecting contacts of the cell. To perform these three tasks simultaneously with high efficiency, the semiconductor material must be of very high purity, which is the main reason why silicon-based solar cells are too costly to compete with conventional means of producing electric power.
In contrast, the Graetzel cell, known as the Dye-Sensitized Solar Cell (DSSC), mimics the mechanism that plants use to convert sunlight into energy, where each function is performed by different substances. The DSSC cell uses an organic dye (photosensitizer) to absorb the light and create electron-hole pairs, a nanoporous (high surface area) metal oxide layer to transport the electrons, and a hole-transporting material, which is typically a liquid electrolyte.
"One of the most exciting avenues we are exploring is the replacement of the liquid electrolytes that are mostly used today for the hole-transport function by conductive polymers. This could lead to further reductions in cost per Watt, which is the key to making solar energy commercially viable," says Coffa.
The ST team is also developing low cost solar cells using a full organic approach, in which a mixture of electron-acceptor and electron-donor organic materials is sandwiched between two electrodes. The nanostructure of this blend is crucial for the cell performance because the electron-donor and electron-acceptor materials have to be in an intimate contact at distances below 10 nm. ST plans to use Fullerene (C60) as the electron-acceptor material and an organic copper compound as the electron-donor.
STMicroelectronics isn't the only company pursuing Graetzel cell development. Venture capital funded Konarka Technologies, based in Lowell, Massachusetts, is also pursuing development of the TiO2 Graetzel cell with non-liquid electrolyte.
According to Paul Wormser, president and COO, the cells will be manufactured at room atmospheric pressure and below 150°C on flexible plastic substrates, using a non-liquid electrolyte.
Another venture capital funded company, Nanosolar, is pursuing a nanotechnological approach to produce flexible plastic photovoltaics.
NanoSolar asserts that it has solved some of the thorniest problems inherent in working with organic materials. The company, applying technology licensed from Sandia National Labs, says it has brought an architectural approach to the process, using self-assembling nano-structures that should substantially improve the energy efficiency of its solar cells.
With venture capital funded start-ups and established firms pursuing the development of photovoltaics that have the potential to be much cheaper to produce than silicon semiconductors a large reduction in the cost of photovoltaics seems a likely outcome.
Batteries able to cheaply store mains level power could alternatively be used to distribute conventional power sources better.
For example, with sufficient storage capablility, you could run your base load power stations so they charge the storage at low requirement times and discharge at peak demand. This way a lot of expensive, inefficient oil and gas plants could be replaced by a few big efficient coal/nuclear plants.
One imagines that, in the same way you can have hot water systems that heat up off peak, with lower charges for the off peak power, you could install a power pack in the house that charges up using cheap off-peak power, and discharges it at peak demand. This will also function as a UPS.
Without batteries there is an alternative enabling technology for solar and wind viability. An improved power grid, possibly a global grid delivering along superconducting cables, could better spread demand. Over the whole world, wind, solar power, and demand are all reasonably stable.
Folks have been working hard for a very long time on making energy storage more cost effective, but the problem is very hard. If battery technology were easy to do well, we would have adequate and affordable electric cars instead of the expensive toys that people have today. And since people have been working on electric cars as long as there have been cars, the trend is not very promising.
On the solar cells, it's too bad they don't rate them in watt-hours produced per year, rather than "watts" - because "watts" is not a measure of energy (but rather power), and does not easily compare to other energy systems. And the storage problem is still real - a global grid is a very expensive way to try to solve the problem, and the terrorism and geopolitical issues would make the problems with middle eastern oil look trivial.
Also, one wonders if the organic cells will actually have a 20 year life. It's one thing to use organic dyes or other compounds... it's another to leave them in the sun for 20 years!
But still, any improvement in solar technology is useful - even if it has zero effect on the total energy production issue. Furthermore, if you ignore global warming (and the science on that is far from solid), the US has enough energy supplies just in coal to run for a very long time, and modern coal technology is very clean. Then there are the huge natural gas supplies around the world. Add to that Canadian oil sands and there is a heck of a lot of fossil energy around. And if you can figure out a way to get a the methane hyrdates in the ocean, the supply is virtually limitless. Of course, if one is worried about CO2, just build a bunch of nukes, using a streamlined process, identical design, and concentrated facilities, and you have pollution free energy.
John, What has been changing that is going to make better batteries and solar cells eventually achieveable is the general advance of materials science. Yes, people have been working on all sorts of problems for a long time. But scientists know a lot more about various materials and how to manipulate them. They have better sensors for watching them with experiments. They have faster computers for simulating them. The rate of advance of science and technology is accelerating. The foundation of knowledge that exists from which to do more experiments is continually increasing. So the odds of success keep rising.
As for coal: keep in mind as well the post I made about the researchers who were working on ways to inject carbon dioxide deep into the Earth and the other post I made about bacteria for getting methane from coal. If coal energy can be made almost entirely non-polluting then there is no environmental reason not to use it.
Organic dyes in the sun: yes, that one crossed my mind as well. If it can be made cheaply enough then it doesn't need to last 20 years to pay itself back. If it lasts for just 10 years then at the end of 10 years there will be much better materials to replace it with anyhow.
Randall, nice post by the way. Even though you're for this tech as I am you're still hard headed about it which is admirable. And Randall, one small point, stop referring to people by their first name! It comes off as condescending.
Gentlemen, I believe what a Minister of Oil of OPEC said: that solar power is not a viable alternative simply due to the price, and that his job is to adjust the price of the oil in such a way that other sources of energy are not cost-viable. The same man added that with proper funding the oil could be replaced in 7 to 10 years, but its business rationale that keeps this from happening.
Mr Moore seems very pessimistic about the future of solar power (I wonder if he is invested in oil and coal companies?). Current solar systems, though pretty expensive (about $50,000 with installation and equipment), can be made to be grid-tied where there is no storage, rather an exchange of electricity with your power company - the meter turns one way during peak sunlight hours and turns the other way during the night. The most expensive part of the system is the solar cells (well over half the price of the system). If you could slash the cost of the solar cells by up to the predicted 40 fold, then having a solar system would be much more attractive. So instead of $50,000 for the system, it might only cost $30,000, which would end up costing $125/mo for 20 years (which is about what my electric bill is every month). AND you are immune from the innevitable energy crisis price spikes (ala California). If you add a battery back up system for a few grand, you're immune from black-outs as well. Factor in state incentives (up to 50% rebated), and its a no-brainer.
As for the unit of power used, for some odd reason, the standard has become kilo-watt hours. So if you buy a 5KW solar array and the sun shines for 5 hours, then you make 25kW-hours worth of electricity (my house averages about 20kwh).
I would like to note that if you're considering solar, you might also consider a supplemental wind turbine or 2 that can generate electricity all day and night (assuming the wind blows). And since you already have all the grid-tie equipment in place, the added cost of a wind turbine is marginal and you can generate a few extra KW hours a day.
If we keep depending on coal and oil, we would need to build more and bigger power plants which, despite our best efforts, will still polute and will make the surrounding areas much less attractive to live in. Solar and wind are very quickly becoming feasible (fuel cells need some work, but I think will eventually depend on these technologies as a source of hydrogen from water).