|
Friday, May 16, 2008 A Cool Trick for Solar CellsA technology developed by IBM to cool computer chips could be a boon for solar energy. By Duncan Graham-Rowe
By employing a trick normally used to cool high-performance computer chips, IBM researchers have found a way to make concentrated photovoltaic cells that are more efficient in converting the sun's energy into electricity. The researchers have shown that it is possible to increase the concentration of light on photovoltaic cells by about ten times without causing them to melt. This, they say, makes it possible to boost the amount of usable electrical energy produced by up to five times. IBM is not known for its work in solar energy, but that has changed recently, with the rising cost of fuel and the growing interest in renewable alternative energies, says Supratik Guha, lead scientist of photovoltaic research at IBM's T.J. Watson Research Center, in Yorktown Heights, NJ. "About a year and a half ago, we decided to start looking at photovoltaics," he says. The principle behind concentrated photovoltaic cells is to use a large lens to focus light onto a relatively small piece of photovoltaic semiconductor material. The benefit is that only a fraction of the semiconductor material is used, thereby reducing costs. There are a number of companies marketing such technologies, but one of the main challenges is in coping with the vast amounts of heat produced by the focused sunlight, says Guha. "You're really heating the chip up. As you raise the temperature of the chip, its efficiency drops, so you've got to keep the temperature down." There are generally two ways to do this: either by using passive heat sinks--metal blocks that draw the heat away from the cell--or, for higher-temperature systems, by using water cooling, in which water is pumped through a metal heat sink to draw the thermal energy away more efficiently. In many respects, this is a problem very similar to cooling computer chips--something with which IBM has a long history, says Guha. State-of-the-art chips now kick out about 100 watts per square centimeter, which is similar to what concentrated photovoltaic cells have to endure, he says. So Guha adapted an IBM-developed material currently used for chips to improve the heat transfer between the photovoltaic cell and a water-cooled heat sink. "If you place the chip on a copper heat sink, the interfacial heat transfer isn't good enough to keep the temperature down," says Guha. This is because microscopic indentations in both surfaces means that there will be relatively little surface contact between the faces. So photovoltaic companies tend to use various organic pastes to act as thermal interfaces. The problem is that such materials aren't particularly efficient at transferring heat. |
Audio »
Share »
Digg this
Add to del.icio.us
Add to Reddit
Add to Facebook
Slashdot It!
Stumble It!
Add to Newsvine
Add to Connotea
Add to CiteUlike
Add to Furl
Googlize this
Add to Rojo
Add to MyWeb
Favorite
Print
E-mail
Focusing on Solar's Cost
05/07/2008
Comments
amgillard on 05/16/2008 at 4:21 AM
7
- both require thin-film substrates of silicon
(either amorphous, poly-crystalline or mono-crystalline)
- both utilise nano-scale features (see http://www.technologyreview.com/Energy/19696/,
http://www.technologyreview.com/Energy/18415/ and
http://www.technologyreview.com/read_article.aspx?id=20163)
- both employ multiple layers of various materials (semiconductors and metals) to act as electrical pathways and improve efficiency
I would have expected there to be enough complementary overlaps in manufacturing techniques that there are significant benefits to incorporating the two technologies within the same development environment.
MakeSense on 05/16/2008 at 9:17 AM
66
For this reason, I am most impressed by technologies that move Silicon PV cells forward. We have Silicon in great abundance. Efforts to bring down costs of Silicon production, reduce waste, improve its quality and to use far less of it to produce the same or more energy deserve the greatest attention in the long run. When combining the best of the latest innovations, it seems certain that Silicon-based cells could compete on cost.
After a little research, I found the following paper online. Here's one of its conclusions:
"Today it appears unrealistic to expect that a PV system based on either CdTe, CIGS or
ruthenium-dye-sensitised cells could produce more than 10 000 TWh/year
corresponding to some 6 TWp in the year 2100 and an average growth of 60
GWp/year sustained over 100 years. According to most energy scenarios this would
correspond to less than 5% of the total energy supply. A potential that is a factor of
ten lower is perhaps more realistic, that is significantly below 1% of energy supply.
Amorphous silicon, with germanium, is much less constrained, and without
germanium it is not material constrained, in our sense, at all."
http://frt.fy.chalmers.se/PDF-docs/BA_thesisWoP.pdf
quseio on 05/16/2008 at 7:08 PM
4
MakeSense on 05/17/2008 at 8:44 AM
66
ebonfyre on 05/18/2008 at 10:35 AM
5
LinuxUser_147560 on 05/17/2008 at 1:03 PM
1
There are ways to take the heat and increase the velocity of the fluid via orifice or venturi for extraction by a low pressure steam system to drive an auxiliary generator. It's a sound principle that is currently used in today's co-generation power plants. Use the exhaust heat of a gas or coal fired engine to drive a secondary low pressure steam system.
The remaining or residual heat can then be dumped to atmosphere using cooling vane technology. Sometimes the best solutions are old tech.
wrm_10 on 05/27/2008 at 4:01 PM
1
cripdyke on 07/28/2008 at 2:57 PM
14
One thing slowing home adoption is that the cheapest techs will be bought up by major projects to sell to the grid (with tax and other incentives, it competes in many places). The more expensive technologies that cannot compete even with incentives are left for homow's to buy ... because they don't have to compete with the cost of a coal plant, they have to compete with the cost of delivered kw/h's at their meter, a much lower standard.
Also, although the info above is correct as conventionally calculated, the truth is even the worst techs on the market can compete with bulk/grid electricity in an absolute sense. But investors don't ask, can we sell this power at coal rates and pay off the plant before it dies? They ask, can we sell this power at coal rates and pay off the plant in 5 to 7 years so we can have 20+ years pure profit?
Many of the PV techs are rated 25 years but in fact will be generating more than 80% of their nominal output 50 years from now. That can still give 20+ years of pure profit...but to the kids of the investors, not to the investors themselves.
The fact that PVs costs are all up front skews the profit equation to make them appear less attractive than they might be in absolute terms and/or to long-term thinkers.
rodbennet on 08/19/2008 at 5:11 AM
2
I was wondering what the capabilities are of polymer based photovoltaics. I found some thread on this site at: discussion on polymer properties, specifically: polymer based phtovoltaics. But this doesn't help me much.
Thanks
Rod