Slashdot Mirror
Nanotechnology Boosts Solar Cell Performance
Roland Piquepaille writes "Physicists from the University of Illinois at Urbana-Champaign (UIUC) say they have improved the performance of solar cells by 60 percent. And they obtained this spectacular result by using a very simple trick. They've coated the solar cells with a film of 1-nanometer thick silicon fluorescing nanoparticles. The researchers also said that this process could be easily incorporated into the manufacturing process of solar cells with very little additional cost. Read more for additional references and a photo of a researcher holding a silicon solar cell coated with a film of silicon nanoparticles."
The nanoparticles improve efficiency by 60% in the ultraviolet spectrum. The visible light spectrum is only nominally affected.
It's still pretty cool, though.
Also, he doesn't post the whole story (60% improvement in the UV spectrum) but rather the more sensational version (60% improvement!). That's pretty dishonest.
It's still something, because to knock an electron out, the minimum frequency of the photon has to be at least the difference between the conduction band (where you want that electron) and the lower-energy valence band (where the electron originally is.) So you have a minimum energy cut off point. Exactly where that is, depends on the material, but generally you won't get any power out of the infrared falling on that cell.
However, the downside is that photons with higher energy than that bandgap, well, the extra energy is essentially wasted.
So basically, say, if you used Germanium at 0.67 EV bandgap, you'd catch more photons than with Silicium at 1.11 EV bandgap, but get less useful energy (i.e., electricity as opposed to heat) out of each photon.
And the higher frequency the photon, the more you waste as heat. So you won't waste more in the visible spectrum (well, unless the photon had less energy than the bandgap, in which case it's completely wasted), but in the UV spectrum you waste a lot.
So reducing the waste in the UV spectrum is really where it counts the most. Sure, it would be neat to gain everywhere, but the UV range is where we waste the most.
Their talk about fluorescent particles, makes me think they're essentially converting an UV photon into at least one lower frequency photon. The question is what they do with the extra energy. At the simplest imaginable way, you'd get at least two low energy photons from one UV photon.
On the other hand, it seems to be a bit more than that, from that short summary linked to. From their claim that they improve voltage, not just current, and that something happens at the interface between the particles and the substrate, it sounds like essentially they created a bunch of new junctions there. I.e., that it's a new way to make a multi-junction solar cell.
Multi-junction cells aren't exactly new, but traditionally they've been very expensive so far. If these guys invented a cheap way to make one, kudos to them.
On yet another hand, it will be interesting to see on exactly what existing cells can their film be applied. On silicon or other semiconductors, ok, I can see how it would form an extra junction. Would it also work on, say, Dye-sensitized Solar Cells? There essentially their particles would come on top of the dye, and I'm not sure how well that works. It'll be interesting to find out, eventually.
A polar bear is a cartesian bear after a coordinate transform.
That's a whole 12 characters shorter, and leaves out the important words 'in the ultraviolet spectrum', which changes the meaning completely. Also, those emitted words are 27 characters long, so if they were properly included, his summary is actually more wordy than the original source.
It's almost word for word. And it's wrong.
In a detector you are correct but in a power device you use the doping gradient because bias voltages leak, defeating the purpose.s -selling-solar.html
--
Get solar power: http://mdsolar.blogspot.com/2007/01/slashdot-user
At the end of the blog Roland asked why they didnt use multiple sizes of silicon nanoparticles, this was my long winded reply:
I am a graduate student working on the synthesis of silicon nanoparticles for solar cells and other applications. While silicon nanoparticles have been syntheszed for over 20 years, and their are many ways of synthesizing them, it is still very difficult to control the size of the particles. Unlike CdSe based quantum dots where the size of the particles is determined by how long your let the reaction run for, 1 min for blue 30 min for red, and various time lengths for other colors, silicon nanoparticles are more complex.
Silicon nanopariticles while they are still quantum dots, since the energy levels are somewhat quantized emit very differently as well. Silicon is an indirect band gap semiconductor, however the blue emitting silicon nanoparticles emit light with a direct band gap transisiton, where as they red emitting silicon nanoparticles are controlled by more surface effects and emit in a low energy indirect band gap transition which is slower and allows for more energy loss in other modes.
Anyways, what it comes down to, is it is difficult to make various sizes of silicon nanoparticles. I would also like to add that this technique is not very promissing for several reasons, they epense and other problems with traditional silicon etched solar cells still exist. Cost, lack of flexibility, low effeciency, heavy, glass... This method does not take full advantage of multiple exciton generation which was just proved for silicon nanoparticles in ACS journal of Nano Letters this week. PbSe quamtum dots have shown to generate 7 exciton for just one photon, which in theory could be converted to 7 electrons from 1 photon...someday. But 2.6 excitons from silicon nanoparticles is still pretty good. Especially when I have a way to get the excitons into free electrons And silicon is a non-toxic cheaper alternative to the PbSe quantum dots.