Getting back to science, at least for the moment, I was puzzled by a press release from RPI, with the eye-catching headline Solar power game-changer: 'Near perfect' absorption of sunlight, from all angles. The article describes work published in Optics Letters (that I haven't been able to put my hands on yet), developing new anti-reflection coatings to enhance the absorption of light by silicon solar panels:
An untreated silicon solar cell only absorbs 67.4 percent of sunlight shone upon it -- meaning that nearly one-third of that sunlight is reflected away and thus unharvestable. From an economic and efficiency perspective, this unharvested light is wasted potential and a major barrier hampering the proliferation and widespread adoption of solar power.
After a silicon surface was treated with Lin's new nanoengineered reflective coating, however, the material absorbed 96.21 percent of sunlight shone upon it -- meaning that only 3.79 percent of the sunlight was reflected and unharvested. This huge gain in absorption was consistent across the entire spectrum of sunlight, from UV to visible light and infrared, and moves solar power a significant step forward toward economic viability.
That all makes sense. What puzzles me is the second claim made in the release, which has to do with the problem of angle dependence. Most materials are reflect more light at glancing angles than normal incidence. If you look at a pane of glass straight on, for example, it reflects about 4% of the light striking it, which is why you only see a faint image of your reflection. If you look at a window from the side (put your head next to the frame), on the other hand, the reflection will be much stronger.
The release notes this, and then goes on to make a very strange claim:
This same is true of conventional solar panels, which is why some industrial solar arrays are mechanized to slowly move throughout the day so their panels are perfectly aligned with the sun's position in the sky. Without this automated movement, the panels would not be optimally positioned and would therefore absorb less sunlight. The tradeoff for this increased efficiency, however, is the energy needed to power the automation system, the cost of upkeeping this system, and the possibility of errors or misalignment.
Lin's discovery could antiquate these automated solar arrays, as his antireflective coating absorbs sunlight evenly and equally from all angles. This means that a stationary solar panel treated with the coating would absorb 96.21 percent of sunlight no matter the position of the sun in the sky. So along with significantly better absorption of sunlight, Lin's discovery could also enable a new generation of stationary, more cost-efficient solar arrays.
There's nothing particular wrong with the angle-dependent reflectivity idea, but the claim that this eliminates the need for sun tracking strikes me as really odd. I can believe that increased reflectance at large angles plays a role in this process, but that's not the only thing going on, and, indeed, it's not something I've heard used as an explanation for the need to have solar panels track the sun.
The other effect is purely a matter of geometry, and it's easy to illustrate with a flashlight or laser pointer. If you take a light source, and shine it straight down onto the floor, you'll see a nice, round spot of light. If you tip the source at an angle, that spot stretches out into an oval shape, getting longer along the direction of the tilt.
That oval shape covers more area than the original circular spot. Which means that the intensity of the light (the amount of light hitting a given small area in one second) is lower for the light coming in at an angle than for the light coming straight on.
For solar panels, this means that the intensity of light hitting a horizontal panel changes as the day goes along. It's highest at noon, and drops off dramatically in the early morning and late afternoons. The amount of electricity that can be produced by a solar cell is directly proportional to the intensity (the electricity produced is some fraction of the energy hitting the panel), so horizontal solar cells generate more electricity in the middle of the day than in the morning or evening.
This geometric factor is the usual reason given for why you need solar panels to track the sun, and it will still hurt the efficiency of solar power generation even if the reflectance problem is solved by this magic new coating. The reflectance thing can't be helping the overall efficiency, but it's not the only reason for tracking systems, and I'm not sure it's even the main reason.
It's conceivable that the increased efficiency gained by lowering the reflectivity of the panels (and eliminating the tracking motors) is enough to offset the losses due to geometric factors. But it's not immediately obvious to me that that would be the case.
Which is why, as I say, I'm puzzled by this claim.
- Log in to post comments
I'm also curious: does the underlying PV sunlight->electricity process work regardless of the wavelength of the sunlight?
Adding a coating to a green plant that makes it absorb more wavelengths won't result in more energy, it will just result in a cooked plant. And I understand that heat dissipation is a big problem for PV.
I believe the material is the one discovered earlier this year of the blackest material on record.
It an array of carbon nanotubes, growing like a grass stalks from the surface. The surface roughness and thickness of tube "stalks" means that it can absorb near optimally from larger angles.
Strange press release. I'm not sure it's such a good idea for 100% of the photons hitting your PV device to be absorbed. . . .heat dissipation is a HUGE issue, especially with high alpha/low eff devices. Those phonons have to go somewhere! (for an even bigger problem think of GaAs ~25%+ efficient cells in concentrator setups. . .tandem thermoelectric/hv can't come fast enough!)
If your device is hovering around 2-4% efficiency (total) but absorbs around 10^4 or 10^5 then your problem is electron/hole recombination, mobility, or interface problems. Fixing the interface recombination in CIGS is precisely what took them from 3-5% efficient to ~12% now. I don't know how they solved the long-term electromigration problems as I'm not THAT familiar with the nitty-gritty.
However, if this coating fixes the geometrical problem Chad laid out then I'm all for it. . .have to read the paper (and probably brush up on my optics to understand it!). It sounds like they generate/utilize negative refractive index-type effects...
p.s. I just got a copy of the article if anybody can't access it.
However, if this coating fixes the geometrical problem Chad laid out then I'm all for it. . .have to read the paper (and probably brush up on my optics to understand it!). It sounds like they generate/utilize negative refractive index-type effects...
No coating can fix the geometrical effect, as it affects the intensity of the light hitting the panel. Even if 100% of the photons hitting the surface are absorbed, there are fewer photons hitting the surface every second for light at large incident angles than for small incident angles.
...I think your explanation could be shortened dramatically with an appeal to the reduction to absurdity: If we use this material, and we don't have to sun-track, does that mean we can orient the panels at 90 degrees to the incoming light? No? That's absurd? 89 degrees? No? That's still absurd? When does that stop being absurd?
Oh, I see-- 0 degrees.
(Also, I wouldn't characterize it in terms of intensity, I'd characterize it in terms of radiative flux, which is directly dependent on geometry, but that's really just pedantry.)
I think this is another case of the badly written press release obscuring what is really happening. The press release includes the following sentence which ought to include the geometry argument:
But it's not clear from the press release whether this is what they mean, or whether they are referring to coatings with angle-dependent coefficients of absorption/transmission/reflection.
Where I think the ability to absorb photons from any positive incident angle will come in handy is for applications like roof-mounted panels where changing the orientation of the panels to track the sun is not an option. IOW, this will substantially improve the efficiency, and more importantly the duty cycle, of many home photovoltaic systems. But again, I may be reading something into the press release that isn't there.
I can't believe 3 comments passed without the observation that David St. Hubbins and Nigel Tufnel (with Derek Smalls, Viv Savage, and whatever drummer was around) already discovered the blackest material, which they used for the cover of "Smell the Glove"
Nigel-How much more black could it be? and the answer is none, none more black!
Proof! I tell you - expect the patent infringement claim any day now :)
It strikes me as typical PR department clueless writing. Shouldn't they give the scientists a chance to correct the copy before release? Nothing like your fellow scientists concluding you are an idiot, over something you had no control over.
Now presumably what they should have said, is that without treatment the output of the panels falls off faster than the cosine of the incident angle. That strengthens the benefit from mounting with a tracker, but as far as I am aware, few panels use trackers anyway. Most concentrating PV solutions need trackers, and reasonably accurate ones at that, as they are trying to concentrate the light, typically from 500 to a couple of thousand times depending upon the design, and I would think they are the only type of panel for which the case for a tracker is strong.
In any case, the point about only wanting to absorb wavelengths the active material can actually convert into electricity is a good one, the rest simply adds to the heat load. Additionally, nothing is said about the toughness of the coating. Can it withstand decades of exposure to the weather? Can it be cleaned? Is it cheap and easy to add to the manufacturing process? The maximum implied gain by this approach would seem to be on the order of 50% (100%/67%), so it is not exactly revolutionary, although any efficiency improvement of even a few percent is a welcome incremental gain for the cost effectiveness of solar, and is hence welcome new. But wildly exaggerated press claims, which are all too common do the industry a disservice.