MIT physicists have managed to create a Light emitting Diode that operates with greater than 100% efficiency!


The LED produces 69 picowatts of light using 30 picowatts of power, giving it an efficiency of 230 percent. That means it operates above "unity efficiency" -- putting it into a category normally occupied by perpetual motion machines.

However, while MIT's diode puts out more than twice as much energy in photons as it's fed in electrons, it doesn't violate the conservation of energy because it appears to draw in heat energy from its surroundings instead. When it gets more than 100 percent electrically-efficient, it begins to cool down, stealing energy from its environment to convert into more photons.

In slightly more detail, the researchers chose an LED with a small band gap, and applied smaller and smaller voltages. Every time the voltage was halved, the electrical power was reduced by a factor of four, but the light power emitted only dropped by a factor of two. The extra energy came instead from lattice vibrations.
http://www.wired.co.uk/news/archive/2012-03/09/230-percent-efficient-leds

That sounds pretty awesome :)

Conceptually, this sounds like a potential paradigm shift.

The thing that gets me, in a larger context, is that when people discuss "energy efficiency" in terms of large-scale solutions, the discussion often becomes too compartmentalized. That is, sure, one particular system might only be able to achieve a certain efficiency, but one really ought to be able to distribute inefficiency in such a way as to make multiphasic energy production and distribution systems considerably more efficient than the marketplace has thus far found profitable.

I'm just sayin' ....

That is, when you start throwing around ideas that could include the phrase "cooling lamp", it's a whole new discussion.

There's something here that the casual reader might easily overlook. Granted, 69 picowatts is VERY small amount of power (it's 69 trillionths of one watt if i have all my zeros right), but it's still energy.

And here's the implication that's easily missed: This device presents a mechanism for reversing entropy locally. Small scale, certainly, but a reversal nonetheless. Residual ambient heat is the final product of all energy-consuming activities - everything from a decaying tomato to all the stars in the cosmos which are eventually doomed to die someday (leading to what's called "heat death" of the universe when there is no longer a difference in temperature anywhere).

And apparently this thing can take that low-level (waste) heat energy and ramp it up to a considerably higher level. AND if the device can be scaled up or produced en mass and tandemed together, it could produce USABLE energy.

A pico refrigerator with an interior light to boot?

I can't wait to see the gigawatt version.

I am curious to know to what value the temperature drops, into fully insulated enclosure, inside which there is such a LED ?

All that is needed now is a solar power generator operating at over 50% efficiency, and Bingo!
Limitless free power.
Take that, perpetual motion machine deniers!

I also liked the adjoining article that popped up: http://www.wired.co.uk/news/archive/2011-06/10/hoverbike-chris-alloy

Anyone want to start a thread on this?

As to the LED, that appears to open a whole new vista of extracting ambient temperature energy and excreting it as photons, cooling down the device. Excellent way to make an air-conditioner that produces energy in the process. This might be the answer to the energy problem in the distant future, if this can be scaled -up and produced in large quantity.

I'm going to need to see the paper describing the measurement. If P = 39 picowatts and T = 300 K

\frac{P}{ \sigma T^4} = \frac{15 h^3 c^2 P}{2 \pi^5 k_B^4 T^4} \approx 8.5 \times 10^{\tiny -14} \, \textrm{m}^{\tiny 2} = 0.085 \, \textrm{\mu m}^{\tiny 2}

Likely that the reverse entropy begins to break down at large scales.

My first thought was embedding these LEDs in processors, essentially using them to dissipate some of the heat as light. Everyone seems to want these interior lighted cases anyway, might as well make it useful as well.

Likely that the reverse entropy begins to break down at large scales.

My first thought was embedding these LEDs in processors, essentially using them to dissipate some of the heat as light. Everyone seems to want these interior lighted cases anyway, might as well make it useful as well.

Heh-heh - except that you would gain practically nothing that way. ;) Only the small amount of light that escaped the case would carry heat away; the rest of the light would be absorbed inside and return the heat right back to where it came from. :)

Fiber optics then. You'd need the same thing with a refrigerator or air conditioning. I'd suspect the big issue would be, is the power consumption to run the LEDs worth the heat they absorb. If it's only a fraction of what can be accomplished with conventional means, then it's pointless.

But from a LED point of view, it's very practical, because there's usually extra heat to use somewhere.

What is meant by "The LED produces 69 picowatts of light..." ?
What is meant 1 watt of light? What kind of unit of measure is that?

It seems to be a genuine measure, although it is not one I have ever come across until now: http://www.answers.com/topic/light-watt

Measuring Light (http://www.mediacollege.com/lighting/measurement/):
"There are many different units for measuring light and it can get very complicated. Here are a few common measurement terms:
Candela (cd):
Unit of luminous intensity of a light source in a specific direction. Also called candle.
Technically, the radiation intensity in a perpendicular direction of a surface of 1/600000 square metre of a black body at the temperature of solidification platinum under a pressure of 101,325 newtons per square metre.
Footcandle (fc or ftc):
Unit of light intensity, measured in lumens per square foot. The brightness of one candle at a distance of one foot. Approximately 10.7639 lux.
Lumen (lm):
Unit of light flow or luminous flux. The output of artificial lights can be measured in lumens.
Lux (lx):
Unit of illumination equal to one lumen per square metre. The metric equivalent of foot-candles (one lux equals 0.0929 footcandles). Also called metre-candle."

It seems to be a genuine measure, although it is not one I have ever come across until now: http://www.answers.com/topic/light-watt
From the link:
"(optics) A unit of luminous power equal to the luminous power of light of a single wavelength λ whose radiant power is 1/Vλ watts, where Vλ is the value of the luminosity function at λ."

"photics The nominal unit in the reciprocal ratio of the efficacy of a source of light to the maximum efficacy of 680 lm·W-1, which occurs at a wavelength of 555 nm (for the light-adapted eye). (The reciprocal of this maximum, i.e. 1.47~ mW·lm-1, is the mechanical equivalent of light.)"

But:
radiant power (http://www.thefreedictionary.com/Luminous+power):
"luminous flux: A measure of the radiant power of light emitted from a source without regard for the direction in which it is emitted. It is measured in lumens."

I can not find a correlation between power consumption (electrical) and power of light.
It is only a reference to the maximum efficiency (so far) of 680 lm / W

What is meant by "The LED produces 69 picowatts of light..."

A watt is a measure of power, which is energy per unit time - and you can express light as energy.

To get light power in watts, take all the photons the thing produces a second, multiply by h/2*pi, then multiply by frequency.

A watt is a measure of power, which is energy per unit time - and you can express light as energy.

To get light power in watts, take all the photons the thing produces a second, multiply by h/2*pi, then multiply by frequency.
lol....:D

Wow, a lightbulb that will help my A/C cool the place!

Groovy.

Might be useful for optical computing, fast operation, low power consumption and helps reduce heating. Win-win.

I'm wondering if it would be possible to make a huge array of tiny LED systems. Scale each individual down in size to where you could fit billions on a sheet. They by linking millions of sheets, you could scale-up the process. I can just see such a system immersed into warm air, cooling it down, while at the same time emitting a beam of light targeted towards a photovoltaic grid to produce electricity. What an amazing thing that would be, to simply convert ambient temperature into utilizable electricity.

The principle is similar to OTEC, I believe. But instead of a colder 'heat sink', the quantum principle of converting ambient heat into a photon allows for the photon to carry away that energy at above-unity efficiency. It in essence radiates photons into space, making it the equivalent of a heat-sink.

While it took decades to miniaturize transistors, it might take decades to make this into a practicable device. But it does appear theoretically possible. Comments anyone? RP? Prom? AN?

I'm wondering if it would be possible to make a huge array of tiny LED systems. Scale each individual down in size to where you could fit billions on a sheet. They by linking millions of sheets, you could scale-up the process. I can just see such a system immersed into warm air, cooling it down, while at the same time emitting a beam of light targeted towards a photovoltaic grid to produce electricity. What an amazing thing that would be, to simply convert ambient temperature into utilizable electricity.

The principle is similar to OTEC, I believe. But instead of a colder 'heat sink', the quantum principle of converting ambient heat into a photon allows for the photon to carry away that energy at above-unity efficiency. It in essence radiates photons into space, making it the equivalent of a heat-sink.

While it took decades to miniaturize transistors, it might take decades to make this into a practicable device. But it does appear theoretically possible. Comments anyone? RP? Prom? AN?
Even at 230% efficiency for the LED, you'd end up losing energy because of the low efficiency of solar cells. I wonder if this "drawing of energy from lattice vibrations" could be used to produce electricity directly, or is it a phenomena specific to the emission of photons?

I'm wondering if it would be possible to make a huge array of tiny LED systems. Scale each individual down in size to where you could fit billions on a sheet. They by linking millions of sheets, you could scale-up the process. I can just see such a system immersed into warm air, cooling it down, while at the same time emitting a beam of light targeted towards a photovoltaic grid to produce electricity. What an amazing thing that would be, to simply convert ambient temperature into utilizable electricity.

Unfortunately the very best PV cells out there are around 42% efficient, which means you would still lose energy (and gradually increase temperature) even if you captured every single photon (which is next to impossible.)


The principle is similar to OTEC, I believe. But instead of a colder 'heat sink', the quantum principle of converting ambient heat into a photon allows for the photon to carry away that energy at above-unity efficiency. It in essence radiates photons into space, making it the equivalent of a heat-sink.

OTEC (and other Stirling engine based systems) are heat engines that work on a heat differential. In other words they work by transferring heat from one place to another and extracting work in the process. This is fundamentally very different, if repeatable.

OTEC (and other Stirling engine based systems) are heat engines that work on a heat differential. In other words they work by transferring heat from one place to another and extracting work in the process. This is fundamentally very different, if repeatable.

Different, yes, because there is not an actual heat differential, which is what allows for extraction of the energy from the higher-temperature body (by, for example, using colder water to cool a low-vapor-pressure gas used in a gas-turbine at the OTEC facility; comparable to a conventional boiler turbine that then uses cold water plus radiator to chill the steam back into a liquid).

However, this is theoretically on a similar basis. But instead of a colder temperature heat differential, there is the equivalent of a colder temperature in the form of a radiator - when it radiates, it cools. That is, it dumps the energy into a 'heat-sink' that is actually an ability to remove a photon, i.e. the 'heat-sink' is at a lower energy potential, even if at the same temperature, and utilizes a quantum phenomenon to radiate away a photon.

Or so it appears from the article and the comments.

However, this is theoretically on a similar basis.

To me it's completely different. Heat engines obey the laws of thermodynamics, and their energy comes from temperature differentials. This device violates the second law of thermodynamics. That, to me, makes it a different beast completely. Definitely interesting but many of the rules don't apply any more.

It's like claiming you have created a new magnet and claiming "the flux density created by this magnet doesn't fall off with distance." That wouldn't be similar to a magnet in terms of theory; that would require a whole new way of looking at Maxwell's Equations.

This device violates the second law of thermodynamics.

It's not violating anything if you measure the total input and total output. The 230% is not factoring in the other source of energy. What's important is how outside sources are being tapped to produce more photons, using the initial energy as a catalyst.

It's using the room power. I do that when I breath. :D

It's not violating anything if you measure the total input and total output. The 230% is not factoring in the other source of energy.

Well then it's definitely violating the second law.


What's important is how outside sources are being tapped to produce more photons, using the initial energy as a catalyst.

Which outside sources?

There have been any number of devices that attempt to tap Brownian motion to do useful work. They have all failed due to the second law of thermodynamics. If this gets around that, that's a very big deal - and is way different than a heat engine.

Well then it's definitely violating the second law.

Yes, that was my initial take on it. I discussed this (via an email exchange) with a retired NG engineer, and he said as such also, that it violates the second law, concluding that the article will be disproven because of that.

But I looked at it from a slightly different perspective. It is essentially extraction of energy based on a heat differential. But in an unusual sense. The ambient room temperature is the one heat level. The other heat level is outer-space (quite cold). The hot-room is linked to the outer-space heat-sink by the photons, which convey the heat-energy to the heat-sink. This is allowable apparently because of the quantum effects 'forcing' emission of photons.

If you read the comments section for that article, you will read a few similar comments (though I did not read them until after I had theorized as such above.)

It is essentially extraction of energy based on a heat differential. But in an unusual sense. The ambient room temperature is the one heat level. The other heat level is outer-space (quite cold).
For this I asked:
"I am curious to know to what value the temperature drops, into fully insulated enclosure, inside which there is such a LED ?"
This would imply the existence of an absolute temperature of reference (not theoretical but practical).
Otherwise it violate, as others have said, the second law of thermodynamics.

They are explicitly heating the LED, so the electrical input is not the total input energy.
http://prl.aps.org/abstract/PRL/v108/i9/e097403

http://physics.aps.org/synopsis-for/10.1103/PhysRevLett.108.097403

http://www.physorg.com/news/2012-03-efficiency.html (Check out the graph)

They are explicitly heating the LED, so the electrical input is not the total input energy.

That was given in the opening post, that the electrical input is not the total input energy, and that it is not violating the First Law in that it is drawing energy from the ambient temperature:

"However, while MIT's diode puts out more than twice as much energy in photons as it's fed in electrons, it doesn't violate the conservation of energy because it appears to draw in heat energy from its surroundings instead. When it gets more than 100 percent electrically-efficient, it begins to cool down, stealing energy from its environment to convert into more photons." http://www.wired.co.uk/news/archive/2012-03/09/230-percent-efficient-leds

Elsewhere they discuss that a higher ambient temperature, by heating, makes for an easier demonstration:

"These initial results provide too little light for most applications. However, heating the light emitters increases their output power and efficiency, meaning they are like thermodynamic heat engines, except they come with the fast electrical control of modern semiconductor devices." http://physics.aps.org/synopsis-for/10.1103/PhysRevLett.108.097403

What the posters here have been concerned about is not the violation of the First Law (energy conservation), but the Second Law. At first glance, it appears to do as such.

However, as I noted, it actually doesn't because the higher ambient temperature of the device is higher than the temperature of "outer-space" into which the energy is deposited (or very close to outer-space; i.e. the photons escape to a heat-sink equivalent). Unless someone else has a better take on how it doesn't violate the Second Law.



http://physics.aps.org/synopsis-image/10.1103/PhysRevLett.108.097403

That was given in the opening post, that the electrical input is not the total input energy, and that it is not violating the First Law in that it is drawing energy from the ambient temperature:

"However, while MIT's diode puts out more than twice as much energy in photons as it's fed in electrons, it doesn't violate the conservation of energy because it appears to draw in heat energy from its surroundings instead. When it gets more than 100 percent electrically-efficient, it begins to cool down, stealing energy from its environment to convert into more photons." http://www.wired.co.uk/news/archive/2012-03/09/230-percent-efficient-leds

Elsewhere they discuss that a higher ambient temperature, by heating, makes for an easier demonstration:

"These initial results provide too little light for most applications. However, heating the light emitters increases their output power and efficiency, meaning they are like thermodynamic heat engines, except they come with the fast electrical control of modern semiconductor devices." http://physics.aps.org/synopsis-for/10.1103/PhysRevLett.108.097403

What the posters here have been concerned about is not the violation of the First Law (energy conservation), but the Second Law. At first glance, it appears to do as such.

However, as I noted, it actually doesn't because the higher ambient temperature of the device is higher than the temperature of "outer-space" into which the energy is deposited (or very close to outer-space; i.e. the photons escape to a heat-sink equivalent). Unless someone else has a better take on how it doesn't violate the Second Law.



http://physics.aps.org/synopsis-image/10.1103/PhysRevLett.108.097403

Yes, but not in this thread.

Yes, but not in this thread.

See post #24: "This device violates the second law of thermodynamics."

See post #27: "Well then it's definitely violating the second law."

This has been a common theme by persons versed in physics.

But I looked at it from a slightly different perspective. It is essentially extraction of energy based on a heat differential. But in an unusual sense. The ambient room temperature is the one heat level. The other heat level is outer-space (quite cold). The hot-room is linked to the outer-space heat-sink by the photons, which convey the heat-energy to the heat-sink. This is allowable apparently because of the quantum effects 'forcing' emission of photons.

If that is the case, and you're effectively generating energy via thermophotovoltaics, then I agree - it does not violate the second law. However, it's also not 230% efficient, since you are exploiting the difference between a heated emitter and a cooler photovoltaic cell. In a real world application, for example, you could not run this thing at 70F and have it cool things down while the energy was absorbed by another 70F receiver. You'd have to keep the emitter warm and the receiver cool, and that would require work to be added to the system to maintain the temperature differential.


"However, while MIT's diode puts out more than twice as much energy in photons as it's fed in electrons, it doesn't violate the conservation of energy because it appears to draw in heat energy from its surroundings instead. When it gets more than 100 percent electrically-efficient, it begins to cool down, stealing energy from its environment to convert into more photons."

This is similar to blackbody radiation in a thermophotovoltaic system. An object (say the filament of a bulb) is heated and begins to emit thermal photons. This cools it down. Unfortunately it cannot be used to "pump" heat from one place to another, since it relies on being hotter than its surroundings. (and it is certainly not more than 100% efficient.)

See post #24: "This device violates the second law of thermodynamics."

See post #27: "Well then it's definitely violating the second law."

This has been a common theme by persons versed in physics.

Versed in scientific models is more accurate. Which is why I'm not allowed to answer.

Fallacy in logic. No such thing is physically possible in the universe we have come to know and love. Reading the study reveals it is simple journalistic incompetence.

Nothing beats the second law-the law of entropy in the long term.

doesn't this violate first law of thermodynamics?

doesn't this violate first law of thermodynamics?

No. It is not 230% efficient overall, it merely puts out more light that the electricity put into it can account for. The extra energy it needs is taken from the environmental heat. If you count the total energy input (not just the electricity) and the total energy output, I'm sure it's less than 100% efficient.

As we typically consider ambient heat to be "wasted" energy anyway, though, it is an exciting find.