If I wanted to build one that absorbed every frequency of light including the non visible type, what materials would I use? With billions of stars in the night sky I don't see why it isn't possible to bolster the energy production from photovoltaics into "astronomical" levels.
IR and UV capable photovoltaics have been developed, but their focus is to broaden the wavelengths captured during the day, and also harness some of the residual IR heat over the night left from the sun's heating of the surroundings.
78% of Earths atmosphere. I'm not surprised it helped to add nitrogen to the process for those specified results...
The biggest advantage seems to be reducing the problem of cloud cover, as well as the other benefits. I'm not sure how to calculate Aqueous Id's question, but given that extended exposure is needed to collect images in the infrared, I'd say not a lot.
You think? The rate of cosmic rays reaching us falls off rapidly as the amount of cosmic ray energy increases but for 1 GeV (10^9 eV) particles, the rate is ~10,000 per m^2 per second and for 1,000 GeV particles (or 10^12 eV), the rate drops to only 1 particle per m^2 per second. Or roughly 20,000 particles equal to 1 GeV per m^2 per second. 1 GeV is equal to 1.6×10^−10 Joules So 20,000 GeV particles per second is equal to 0.0000032 Joules per m^2 Which is equal to 0.01152 Joules per hour per m^2 BUT 3.7×10^7 Joules is equal to ~$1 worth of electricity delivered to you in a typical US home, or about 10 kWh. So it would take ~3.2 BILLION m^2 (a collector roughly 35 sq miles at perfect capture) to generate $1 worth of electricity per hour from Cosmic Rays. YMMV, but I think sticking to our sun's direct energy is by far the more practical approach. http://www.auger.org/cosmic_rays/faq.html http://en.wikipedia.org/wiki/Orders_of_magnitude_(energy)
What adoucette just walked us through was the consequence of scale. Suppose the world was different and all of that cosmic power could be developed in a collector only 1 m[sup]3[/sup]. In that case we would be having problems with all RF detectors, which would be swamped by a billion times more galactic noise. I asked this specifically to point out that spectral power content is an elusive notion. If I could prove that a band in the spectrum is truly continuous, then this would imply that that there is an infinite amount of power available to an infinitely precisely-tunable collector. That is, I could keep tuning and I would keep finding more power, and this would go on without end. By the same token, if we step back to the idea of a quantized spectrum, in which radiation is present in some quantum (or quantifiable) frequency steps-- say I find N of these steps--then if at each frequency f[sub]i[/sub] there is an average power of P[sub]i[/sub], (for i = 1 to N), and if I find the average power over the band, P[sub]avg[/sub] over each spectral line P[sub]i[/sub], then total power P[sub]tot[/sub] = ∑ P[sub]i[/sub] = N ∙ P[sub]avg[/sub]. Suppose that we figure out that N is an incredibly large number. We have a HUGE amount of power, even enough to offset the fact that the CMB is so low in power at a given frequency. All that's left is: how do we collect it? And therein lies the problem. I'm asking you if anyone can build a device that collects, in parallel, all the individual frequencies of incident radiation and sums them together to produce this monster device that will cure all ills. It's tough nut to crack. Now imagine you want to solve the same problem, only applying this is the visible band. You will be working in a different technology (photovoltaics instead of antennas and receivers), and I suspect the solution will be even harder. So this was just my way of restating the nature of the question asked in the OP.
In the real world, I think these, at ~80% efficiency, are about the best PV cells that I've heard of: http://www.sciencedaily.com/releases/2011/04/110429121853.htm
Yes, this process seems to emulate the way surface area was increased in capacitors. It looks like great step forward. Now if technologies like this can find market demand, to push down production costs, consumers will probably come around to the viability of investing in them. It's kind of a circular problem, but it seems inevitable. I was off just now checking out Pierre Auger Observatory. I like their site design which shows an attention to quality. Their mission is outstanding, and the mere fact that they were sponsored speaks volumes about them. I was wondering about their detection capability, and I noticed they have a 3,000 km[sup]2[/sup] aperture, which fits nicely with your demonstration of the problem of scale.
That doesn't make a lot of sense. That's like claiming that you can cut a cookie up (effectively) an infinite number of times, and therefore you can end up with an infinite amount of cookie. All photovoltaics work basically the same way. Photons transfer their energy to electrons; the electrons move from valence to conduction bands, and in a well designed and doped semiconductor, this results in a net movement of charge (i.e. current.) One big problem with photovoltaics is that they have specific bandgap energies depending on substrate, doping etc. A photon can move an electron across the bandgap (in which case power is generated) or it can fail to do so. There's no middle ground; you can't move an electron halfway across the bandgap and get half the energy. It's all or nothing. Thus there is an energy of photon (given by its wavelength, shorter = bluer = more energy) that will result in current flow; photons below that energy will not. Photons with more energy will still cause that current flow but their additional energy is wasted because they won't push the electron any farther than the lower energy photon. Because of this, with a single junction cell you can get, at best, about 30% efficiency out of the cell in sunlight*. (Practical single junction cells top out at about 25%.) There are ways around this; you can layer several semiconductors so that lower energy photons are collected by a top layer, followed by higher energy photons on the next layer etc. Those multijunction cells do something like what you are talking about and can get efficiencies considerably higher than single junction cells; I think the record for these is currently around 42%. But in all those cases you are still dealing with photons. You can't "divide" a photon and get twice the energy out of it by tuning your bandgap energy just so. At best you can convert almost all the energy in the photon into another form of energy. (* - that's in sunlight. For monochromatic sources you can get 70-80% efficiencies by optimizing for that one energy.)
"Xu and colleagues were able to obtain a light-to-power conversion efficiency of 3.2 percent compared to 1.8 percent efficiency of conventional planar structure of the same materials." That's an 80% improvement over 1.8%, not 80% efficiency.
@billvon: Let me pose a hypothetical scenario. You are given spectra for two sources, A and B, below. it is given that the each line in A peaks at 1 W. You are asked to figure the maximum power available from each source, without any consideration as to how you might go about collecting that power. Any thoughts? Please Register or Log in to view the hidden image!A Please Register or Log in to view the hidden image! B
OOPS Hmm? I've always thought that the typical efficiences were in the low teens and the best lab PVs were in the 20% though.....
When I look at solar energy, the problem is inefficiency. It is not competitive in the free market and we have to stack the deck with force of government. This is due to the industry taking the wrong approach. We have been working on solar since at least the first energy crisis of the later 1970's, with the industry still stuck at noncompetitive. Let me outline a better approach. Theoretically, to get the most energy out of sunlight or starlight, you need materials that are very unstable so the electrons can reach maximum absorption energy levels. This approach is only good for one shot absorption, since although such unstable materials would allow maximum energy per cycle, being unstable means it will break down quickly. If we want the absorption materials to last for years, we need to do the opposite and design the material with stability in mind, which then limits the amount of energy it can absorb per cycle to 1-2%. Theoretically, unstable can absorb 100% , once, but is not stable in the long term. With that in mind, my approach would start with the unstable and work backwards toward stability until we get a balance that is competitive with oil. If you look at life, it uses chlorophyl which is about 85% efficient. This is a metal-organic resonance molecule. The solar energy is distributed over a wide range of atoms so even the unstable is stable. Instead of silicon, chlorophyl makes use of magnesium cations at +2, as the central atom.
But you can't use solar energy unless you convert it to something. Placing an unstable chemical in the sun is only one level of the problem. You need the chemical to do something, and give you back the energy in usable form. You haven't suggested how to harness energy from an unstable material. I mean: I could take nitroglycerin and place it in the sun, and odds are, it will eventually explode. I could even devise a (low life) piston engine that runs off the stuff. But the sun really doesn't do more than trigger it, which I can do with a spark or a flame. I'm just burning fuel in this case. Yet I can take a stable chemical, like water, place it in the path of a large solar reflector, and at least develop steam. But you seem to have some sort of reaction in mind? Photovoltaics operate on a principle that exploits the two energy characteristics of doped vs undoped semiconductor material, giving the photon a way to bump an electron to a higher energy level, where it crosses the junction and falls back down, producing current. They may not be very efficient, but at least they harness the light and convert it to electricity. As for harnessing starlight.... we just demonstrated that it would take a collector the size of a city to collect 1 W.. so... Please Register or Log in to view the hidden image!
Yeah, I think this is just an experiment to determine how to improve efficiencies in cheap TCO cells.
Ordinarily you'd figure out the photon flux, multiply each photon by its energy (1/wavelength) and integrate over the entire spectrum to get power. But the case you have presented is a little easier. Since in your first spectrum you have stated that each line represents 1 watt, and the second graph gives total power at each wavelength, the calculation of photon energy has already been done. For the first graph it is the summation of the power in each band, and in the second graph it is the integral of the power over the frequency.
Stable materials (semiconductors) exhibit efficiencies of around 20% for blackbody radiation sources (like sunlight.) Chlorophyll as used in plants is around 3-6% efficient.
Does every atom possess this ability? Or do some just refract a photon that can aid this conduction. Sure. Say it is a photon not absorbed by a voltage producing element. If we passed it through a certain material could it transform into an acceptable voltage producing photon?
