The Chinese Solar Machine Layer by Layer Fire in the Library The Mystery Behind Anesthesia
Microcell: The solar cells made by Semprius are 600 micrometers on each side and can be combined with high-power optics. The cell itself (the black square at center) is mounted atop a ceramic base with electrical contacts on each side.
Semprius
Cells absorb sunlight concentrated 1,000 times without cooling.
A startup company hopes to bring down the cost of generating power with concentrated sunlight by using microscale solar cells that can utilize twice as much light as other panels, without the need for expensive optics or cooling systems. Panels made from the tiny cells, which the Durham, NC-based company Semprius developed using a novel microprinting technology, also offer significant savings on materials costs. In late January, the company announced a joint agreement with Siemens to develop demonstration systems based on its technology. Semprius plans to begin volume production of the modules in 2013.
Adding concentrating lenses to solar panels increases the amount of electricity they can produce. But photovoltaic concentrators add a great deal of expense to a solar installation. The optical systems themselves are expensive and bulky--the larger a cell, the larger its paired lens must be. More intense light also means that more performance-degrading heat must be dissipated using heat sinks or fans. Although the cost is partly offset by the efficiency of high-concentration photovoltaics, it limits the potential power of such concentrator systems. The two major suppliers of concentrated solar modules, Amonix and Emcore, both sell systems based on conventional-size cells that operate under 500 times concentration sunlight with costly cooling systems.
Semprius's solar modules contain arrays of square cells that measure just 600 micrometers on each side. These cells have three semiconducting layers--each of which is based on gallium arsenide and absorbs a different band of sunlight--and they are made using a combination of chemical etching and printing, which means fewer raw materials are wasted. They can operate under sunlight concentrated 1,000 times using cheap optical systems. According to the National Renewable Energy Laboratories, the efficiency of the resulting modules ranges from 25 to 35 percent and they can provide electricity for about 10 cents a kilowatt hour. The company expects the final costs, including installation, to be $2 to $3 per watt.
Last year, a study by researchers at Sandia National Laboratories in Albuquerque, NM, suggested that microscale solar cells might offer various cost and design advantages. "You reduce the amount of semiconductor you need, so there can be a big cost savings," says Gregory Nielson, head scientist on the Sandia project. "And you can do things with the optics that you can't do with larger cells."
Smaller solar cells are more efficient at dissipating heat. "When the cells are below a millimeter, they reject the heat so efficiently they'll be just as cool as a one-sun panel," without the need for any cooling systems, says Nielson. This is because the tiny cells have a much greater percentage of total area given up to heat-diffusing edges.
Normal PV can operate in diffuse light, but any time you concentrate, you need direct sunlight. Even things like wind on the outer concentrator can cause the focus to change and for power to be lost. Still, this ultimately sounds promising.
Also, even though the solar cells reject the heat, the heat has to go somewhere, presumably into the air around the panel. It seems like solar air heating would be an appropriate way of handling waste heat.
I've often wondered, given that at least 70% and as much as 90% of the incoming sunlight does NOT turn into electricity, if there was a way to cheaply combine these systems with solar hot water systems. I was actually thinking about some of the supposedly semi-transparent thin film systems being developed.
The chief savings, as I saw it, would be in the physical rooftop installations. You would need more "hot water" modules to optimize recovery, but perhaps only 30% more or so. The solar panels would fit on top of the units, waste heat or attenuated sunlight passing through the electricity generating area would heat the water. My principal concerns would be overall effectiveness (obviously) and whether or not one can effectively keep the solar electric elements cool enough (I doubt that printing thin cell on the surface of a solar hot water collector would work, but maybe I'm wrong. This is off the top of my head, non-engineering person who has supervised a few large construction jobs for my New York City co-op. 40,000 square feet of rooftop space for a 54 unit building, no shadowing by other structures (everything is six stories around here, and really cheap ways (stacked janitor's closets, and abandoned incinerator chutes (fire brick) to send piping/wiring straight to the basement should be good for something.
Anyway, this new system sounds neat. Is gallium arsenide the only option?
Manufacturing in the United States is in trouble. That's bad news not just for the country's economy but for the future of innovation.
aunderdown
77 Comments
Grid Level Prices?!
In a recent Semprius Press Release, a Siemens executive is quoted as saying the technology: "has the potential to deliver electricity at grid level prices to both industrial and utility scale customers". - roughly consistent with the 10 cents/KWH (NREL) estimate quoted in the article. If the technology delivers as promised, the solar industry will expand faster than anyone (well, at least some of us) imagined, and requirements for operating subsidies will diminish just as rapidly.
Reply