[CORRECTIONS ABOUND! It isn't as bad as my first napkin suggests.]
Well done scientists[1]! Cue the engineers…
And it must be focused to an intensity of 10 million watts per square centimeter. Sunlight isn't this intense on its own…
Sunlight is about 1000watt/m^2. They are using 100 billion watts/m^2. That is, you would need to concentrate sunlight to 100 million times to reach this level.[2]
Put another way, if your fiber is about 1mm in diameter, you will need a perfect[3] mirror array 10 meters in diameter to drive it[4]. [CORRECTED from 400 meters, thanks to loup-vaillant]
Wild speculation follows:
• This has got to be used in pulsed mode. 100 kilowatts down a 1mm optical fiber is going to vaporize it if done for more than a tiny fraction of a second. Are there the optical equivalent of super conductors that are suitable for this?
• I went with a 1mm fiber because it "only" required a 10m mirror to drive it, but at what voltage are we going to get 100 kilowatts of power out of a non-conductor with a 1mm^2 cross section? Sounds like 50 zillion is the answer.
• Lets guess 10% efficiency, and the rest comes out as light at the end of the fiber (so our fiber doesn't turn to plasma). Where are you going to point that? I suppose into a light tight chamber contained in the molten salt loop of a steam power plant would be a good start.
• Maybe this all works better at the nano scale. Heat dissipation is easier since essentially all of the optical media can be in contact with heat sink. The mirrors get smaller to achieve the required light density. There is a shape, sort of stretched parabolic looking, that is a non-imaging optical funnel which might get your light all gathered to the same point, albeit going in a variety of direction. The total power would be lower, lowering the voltage required, letting you use sane power converters.
[1] And yet again, you have degraded the value of my undergraduate education by falsifying my hard learned facts.
[2] I may well have lost a zero or more each way. It's early. Don't use these calculations to place orders for mirrors.
[3] From 10+ meters away it is going to have to focus to a 1mm target. [CORRECTED from 200+, it doesn't sound so bad now]
[4] On proofreading I see I have a [4] footnote, in the text but no corresponding note. It was probably vaporized by a poorly aimed fiber.
…but… if we don't rely on spatial compression to achieve the density, but instead use temporal compression…
1) Consider a mirror which rotates through n angular positions.
2) At position zero it reflects off "secondary mirror #0" and onto our target at X microseconds.
3) At position one it reflects off "secondary mirror #1" and onto our target at X-k microseconds.
4) At position n it reflects off "secondary mirror #n" and onto our target at X-k * n microseconds.
5) If we turn the rotating mirror from position 0 to 1 after k microseconds, we have darkness on the target for X microseconds, then light from both secondary mirrors 0 and 1 for k microseconds.
6) If we rotate through all n positions at k microseconds/position we get n times the light at a 1/n duty cycle.
7) As n goes to infinity: PROFIT!
Mathematicians might solve for the shape of the now continuous secondary mirror. I would just write a javascript program to approximate it for segments of length epsilon. Ultimately the CNC equipment that makes the mirror will have an epsilon anyway.
You may wish to use a mirrored polygon to sweep the usable angle of your secondary mirror repeatedly.
Yes! It wasn't until lunch when it dawned on me that viewed backwards it made a time dilation device. I fear the continuous version could not be reconciled with imaging optics, but for non-imaging applications it can slow down an event for more leisurely capture.
I didn't go to University, and wasn't much for science in high-school, but (using your figures) sunlight is 1000watts/m^2/second (or millisecond or minute), isn't it?
If so, aren't your concerns about the size of the mirror, etc. imposing a limitation, kinda like saying we can build a jet that will fly from NYC to San Fran in 30 seconds, but we can't keep the people in one piece during the trip. Therefore, we slow it down to a more humanly acceptable speed. With the volume of sunlight hitting the planet, would we not be able to take only a portion of that sunlight and concentrate it over time to get the same effect??
Of course, like an airplane, you need a minimum speed to create lift, and this magnetic effect may have such a limiting speed/force(?).
Of course, I have no idea what I'm talking about, but you seem quite knowledgable.
watts is a measure of power. The often confused "kilowatt hour" is a measure of energy and would indeed need the time "/second" (or other time unit) added. But I am working in power here.
would we not be able to take only a portion of that sunlight and concentrate it over time to get the same effect
Yes! We went the same way here. I replied to the parent comment with a description of a temporal light concentrator.
I've never heard of such a device, maybe people that know the name of them can order them from a catalog, but it is an intriguing solution because it solves the "massive collector" problem as well as the "holy sh*t! what am I going to do with all this waste heat!" problem.
It's also possible that people more comfortable with light as a wave spit coffee out of their nose and had a good laugh when they read it.
I didn't go to University, and wasn't much for science in high-school, but (using your figures) sunlight is 1000watts/m^2/second (or millisecond or minute), isn't it?
No, sunlight is 1000 watts / m^2, or 1000 joules per square metre per second. The Watt is a unit of power (ie energy over time).
As a footnote, there is a pitfall here in the English language.
The word "over" might mean divided by, or it might mean "integrated across", which are, in the case of constant power completely opposite meanings.
power is energy ÷ time (assuming the power is constant, and thinking of horizontal lines for fractions you might say "power is energy over time")
energy is power × time (assuming the power is constant, but if you are thinking "integration" you might say "energy is power over time")
[3]: Err: 100m² are enough get the 100M concentration on the square millimetre, right? A perfect lens of that surface would only need to be about 11 meters in diameter. Error on your part, or did I miss something?
Here, we expect to have a very low heat load. Instead of the light being absorbed, energy is stored in the magnetic moment. Intense magnetization can be induced by intense light and then it is ultimately capable of providing a capacitive power source.
I work for a living doing research on solar cells.
I downloaded the technical paper and had a quick look. It would have taken me days of hard study to understand what this undoubtedly clever people have done, even though I have a PhD in physics. From how excited they seem to be about their own work, I would guess its probably a very nice piece of theoretical physics. But a few things are clear, from their own words:
- They have not actually fabricated a solar cell, their most optimistic expectation based on this technology is 10% efficiency. With efficiencies bellow 14% you cant compete with other energy sources (you can not reach grid parity), even if your solar cells are so cheap that they are free. That is because the solar cell itself is only a fraction of the costs of solar energy. Think of land, maintenance, support structures, cabling, power electronics, and so on. That means that this concept, as it is, is useless for solar energy production, just like "organic" and other fashionable solar cell technologies.
- The effects they are talking about happen when light has an intensity, in their own words "as low as 10^7 W/cm2" that is a hundred million times more intense than the light from the sun. You need a big, expensive laser to observe the effects they are talking about.
- This is just one more case of people trying to spin their work as a revolution in solar energy, just because it is fashionable and it gives you publicity and funding for your research. They are probably fooling themselves in the first place, but that is no excuse. This type of thing is terribly detrimental for all of us, as it takes funding and attention from the solar cell technologies that actually work. This is a shame for the people involved, and for the University of Michigan.
- If you want to make a bet on solar cells, bet for III-V concentrator solar cells. These are the most efficient ones, by far, (above 40%), and can be used under optical concentration (up to a thousand suns) making the cost of the cell itself almost irrelevant.
From my not-still-incomplete engineering education I came to understand that "solar panels" are huge photodiodes.
Doesn't sound exiting at all. I get to the point of having a PhD like you, I guess nothing will excite me on the mass media when it comes to technology.
> The university is pursuing patent protection for the intellectual property
They make a discovery that may significantly benefit all humanity, then they ensure it won't spread as fast as possible. Besides, I wonder who fund the research so the university can claim property on new knowledge. (I'm not accusing the university specifically here, they may have little choice given the current system. Nevertheless, this sucks.)
From my understanding, they can't patent the physics, and that is the real discovery here.
They have potentially hit on a way to create energy from a previously unknown or ignored phenomenon, but just like you can't patent gravity, you can't patent this.
Now that people are paying attention to this phenomenon, lots more research will go into the best way of capturing it. Sure all those will likely be patented as well, but there is nothing (in theory) stopping you from developing your own way of capturing this energy and making it freely available to the world.
> They have potentially hit on a way to create energy from a previously unknown or ignored phenomenon, but just like you can't patent gravity, you can't patent this.
Not so fast. While you can't patent gravity, you can patent using gravity to do things (subject to novelty and obviousness).
That's how Feynmann ended up with a patent on nuclear submarines. You can't patent radioactivity or heating things with radioactivity, but you can patent using heat from radioactivity to do specific things.
This begs the question, if you can magnetically polarize light, can you bend a beam of light with a strong enough magnetic field? and if so, there is tremendous potential beyond just energy storage.
This discovery doesn't have to do with magnets 'polarizing' light. Rather it's using light's magnetic component to setup a charge separation. You also can't "steer" light by polarizing it. Polarization is just the 'angle' that EM makes with its direction of travel. Polarizing light just means selecting which angle you want your light to have.
I recommend a look at "circular polarization" for anyone that thought they understood polarization as the direction the light vibrates.
Take two pair of Real-3d glasses from the movie theater and try the different orientations (including front-front, front-back permutations). The vertical/horizontal model will not explain your observations. Hence, I know about circular polarization and had a confusing pre-movie experience at Iron Man 2.
I missed the first 3 or 4 minutes of Avatar because I was playing with the glasses and a laser pen pointer :-)
On one level I agree with you that this seems, at least initially, as tidy bit of theoretical physics. Most people 'get' that light is an electromagnetic wave, and of course Maxwell tied the two together quite elegantly.
The clever bits will be these two:
1) Is there a material, either natural or 'meta' in which the structure can convert a fraction of the light passing through it into a magnetic field. If so, and the light is modulated, you can induce a current in a conductor. Could be useful, could be a parlor trick.
2) Can you run it backwards? Which is to say if you generate a magnetic field of the proper type and orientation in the presence of such a material, can you convert the magnetic field into light? If so what frequency? What coherence? Does this paper provide the foundation for a LAAMR (Light Amplification by Amplified Magnetic Resonance) (no, its not a pun on 'lamer' :-)
I expect if you can create the latter you could probably get a Nobel prize (or at least share it).
Am I the only one who enjoys finding out we're wrong about some long held scientific beliefs (in this case the magnetic influence of light)
As a complete aside from most of the conversation happening here..
The ISS uses mainly circular polarisation in their radio communications because horizontal flips to vertical in a range of 15 minutes, relative to position over the Earth (90 minute period).
A horizontal antenna _can_ receive a vertically transmitted signal, with a 22 dBi loss. The reverse is also true. However, if a vertical or horizontal antenna receives a circularly polarised radio signal, the loss is only 3 dBi.*
Spoken as a user and experimenter of amateur radio.
* And I also found a link showing experimental loss when one receives a wrong "handedness" of circular polarisation. There is a clockwise and counter-clockwise polarisation as well. Theoretical maximum loss is also around 22 dBi, the same if you used a vertical to receive a horizontal.
So far, the types are: Horizontal, Vertical, CW circular, CCW circular, CW elliptical, CCW elliptical. Admittedly, I have never done anything with elliptical, but is noted here for completeness.
Meta comment: it's kind of awesome to see more umich.edu links on HN lately. Would there be interest in an Ann Arbor meetup, say early May, after Funded by Night?
Well done scientists[1]! Cue the engineers…
And it must be focused to an intensity of 10 million watts per square centimeter. Sunlight isn't this intense on its own…
Sunlight is about 1000watt/m^2. They are using 100 billion watts/m^2. That is, you would need to concentrate sunlight to 100 million times to reach this level.[2]
Put another way, if your fiber is about 1mm in diameter, you will need a perfect[3] mirror array 10 meters in diameter to drive it[4]. [CORRECTED from 400 meters, thanks to loup-vaillant]
Wild speculation follows:
• This has got to be used in pulsed mode. 100 kilowatts down a 1mm optical fiber is going to vaporize it if done for more than a tiny fraction of a second. Are there the optical equivalent of super conductors that are suitable for this?
• I went with a 1mm fiber because it "only" required a 10m mirror to drive it, but at what voltage are we going to get 100 kilowatts of power out of a non-conductor with a 1mm^2 cross section? Sounds like 50 zillion is the answer.
• Lets guess 10% efficiency, and the rest comes out as light at the end of the fiber (so our fiber doesn't turn to plasma). Where are you going to point that? I suppose into a light tight chamber contained in the molten salt loop of a steam power plant would be a good start.
• Maybe this all works better at the nano scale. Heat dissipation is easier since essentially all of the optical media can be in contact with heat sink. The mirrors get smaller to achieve the required light density. There is a shape, sort of stretched parabolic looking, that is a non-imaging optical funnel which might get your light all gathered to the same point, albeit going in a variety of direction. The total power would be lower, lowering the voltage required, letting you use sane power converters.
[1] And yet again, you have degraded the value of my undergraduate education by falsifying my hard learned facts.
[2] I may well have lost a zero or more each way. It's early. Don't use these calculations to place orders for mirrors.
[3] From 10+ meters away it is going to have to focus to a 1mm target. [CORRECTED from 200+, it doesn't sound so bad now]
[4] On proofreading I see I have a [4] footnote, in the text but no corresponding note. It was probably vaporized by a poorly aimed fiber.