MIT develops 200% efficient LED

[argleargle]

Griswold Christmas roof, I am now a power station and a landmark for aircraft.

 

[TEEJ]

OMG!

We can stop global warming!

:santa:

Install these where the heat from the sun will drive them at 200% efficiency, so for every watt the sun hits them with, they cool the earth by 2 watts.

Install them in the deserts and we can return them to jungle!


Install them on people, and they can both be brighter, and, cooler!

 

[awenta]

LMAO. Teej just solved all the world's problems.

 

[jtr1962]

I'm actually more intrigued at the possibility that these can be developed into a more efficient solid-state heat pump than existing bismuth telluride thermoelectric modules. If they happen to emit a little light also, wonderful-you can light your refrigerator and cool it with the same device. If not, it doesn't really matter. I love the possibility of getting to cryogenic temperatures. That's well out of the range of standard thermoelectric modules which are hard pressed to get much under about 200K. I'd love to build a little insulated box which could get close to liquid nitrogen temps, or at least as cold as dry ice.

 

[arek98]

Interesting. Imagine multiemitter light. Mixed regular LEDs for heating and new once. Circuit that controls it. Regular LEDs are cooled boosting their efficiency and new ones convert otherwise wate heat into light. I would sayy combinet efficeincy may be pretty good and no problem cooling environment to much (where it is not intended)

 

[awenta]

I'm actually more intrigued at the possibility that these can be developed into a more efficient solid-state heat pump than existing bismuth telluride thermoelectric modules. If they happen to emit a little light also, wonderful-you can light your refrigerator and cool it with the same device. If not, it doesn't really matter. I love the possibility of getting to cryogenic temperatures. That's well out of the range of standard thermoelectric modules which are hard pressed to get much under about 200K. I'd love to build a little insulated box which could get close to liquid nitrogen temps, or at least as cold as dry ice.

You can buy a box of liquid nitrogen.

 

[Esko]

Some older threads with some links:

http://www.candlepowerforums.com/vb/showthread.php?334523-230-Efficient-LEDs
http://www.candlepowerforums.com/vb...D-converts-heat-into-light-efficiency-gt-200-!

 

[Anders Hoveland]

I call shenanigans on 200% efficiency. This implies that the device is within a closed system; it is not. It requires ambient heat, correct?

It is still essentially 200% efficiency. The heat being consumed is ambient heat, without any temperature gradient. It is essentially free energy.

Thermodynamic law states that in a closed system, entropy must always increase (the lack of the ability of energy to do work.)

Except this "law" of thermodynamics is not actually a real law. It could potentially be broken, and there is nothing in fundamental physics which says this cannot happen. But so far it has never actually been clearly demonstrated. It is very challenging to try to turn the random order of tiny molecules into useful work energy, without some sort of temperature difference on a larger scale.

Also, "free energy devices," aka "perpetual motion machines" tend to get debunked in 5 minutes or less once you get your hands on one. Past history has shown that they have all been scams so far.

I would definitely lose my composure and giggle like a schoolgirl if someone managed to actually disprove conservation of matter and energy, or even F=ma. It'd be *GRAND!!!* Science fiction becomes reality, infinite batteries become possible, on and on...

A "free energy" machine is not necessarily the same thing as creating energy out of nothing. Some types of free energy machines attempt to convert some form of ambient energy, such as heat or "zero-point". Of course, a successful demonstration has never be done yet. That does not mean it is impossible, it may be beyond current technological reach. This LED phenomena offers a chance that such energy could be harnassed, although it is still not clear yet whether it is just a dead end.

 

[argleargle]

Except this "law" of thermodynamics is not actually a real law. It could potentially be broken, and there is nothing in fundamental physics which says this cannot happen. But so far it has never actually been clearly demonstrated.

From:
http://en.wikipedia.org/wiki/Second_law_of_thermodynamics
"The second law is an empirically validated postulate of thermodynamics"

Basically, "they have actual proof."

It is very challenging to try to turn the random order of tiny molecules into useful work energy, without some sort of temperature difference on a larger scale.

You speak of Maxwell's Demon.
http://en.wikipedia.org/wiki/Maxwell's_demon
Scroll down to "criticism and development." 2nd thermo isn't actually violated if you're taking energy from outside the system, hence an "open system." A closed system cannot do this. It's just a matter of definition of the scope of your system. Example: Your light bulb is powered by the electrical grid. By stating that the bulb itself is a closed system, you can draw a conclusion that isn't true.

As far as using a temperature difference on a larger scale to extract energy, this is how Stirling engines work. There's another thread open about those here at the moment.

A "free energy" machine is not necessarily the same thing as creating energy out of nothing. Some types of free energy machines attempt to convert some form of ambient energy, such as heat or "zero-point". Of course, a successful demonstration has never be done yet. That does not mean it is impossible, it may be beyond current technological reach. This LED phenomena offers a chance that such energy could be harnassed, although it is still not clear yet whether it is just a dead end.

I was referring to "Perpetual Motion of the Second Kind," basically "zero-point" energy. Harvesting energy from the environment is not "free." The energy comes from somewhere.

2nd Thermo Law basically states that your batteries will run out and won't magically recharge themselves. They'll need to be charged from outside the system of the LED circuit, so to speak. The Sun will run out of hydrogen and approach a state where the energy of its system cannot do work. It's more like that.

Good post, Anders! I like your thinking and I like a good physics discussion, regardless of the outcome.

Disclaimer: I never said I was perfect at physics. I'm sure there is still room for discussion. I am not opposed to new breakthroughs permitting the harvest of energy from the environment. I'm not against what you said at all. A solar panel is an example of a non-science fiction device that harvests energy from the environment; they don't violate Thermo 2nd.

 

[2xTrinity]

It is still essentially 200% efficiency. The heat being consumed is ambient heat, without any temperature gradient. It is essentially free energy.

I read the paper. The device was only operating above-unity electrical efficiency when the substrate was heated to well above ambient. It was at 200% at 135C. When operated at room temperature, the substrate was cooling down but it was a net consumer of electrical power like you'd expect from a thermoelectric cooler.

What's interseting however is that unlike simply creating a temperature gradient, or emitting photons due to blackbody radiation, this is emitting wavelengths that are much much shorter wavelength than would be expected from the temperature. At 135C a passive device is going to radiate mostly at ~10microns, but this device peaks at more like ~1-2 microns.

This type of technology might be useful to extract useful energy from heat with small tempreature gradients. A huge heat source with small temp gradient might not be hot enough to drive a traditional engine, but a device like this might be able to create a small amount of high energy photons that can be absorbed efficiently with a photodetector and extract some net energy that way.

 

[Anders Hoveland]

I read the paper. The device was only operating above-unity electrical efficiency when the substrate was heated to well above ambient. It was at 200% at 135C. When operated at room temperature, the substrate was cooling down but it was a net consumer of electrical power like you'd expect from a thermoelectric cooler.

True, but is does demonstrate the principle that ambient heat could potentially be harnessed. Yes, 135C is higher than room temperature, but still no temperature gradient is needed.

When operated at room temperature, the substrate was cooling down but it was a net consumer of electrical power like you'd expect from a thermoelectric cooler.

Except that the energy it radiated could be a more useful form than the heat energy it absorbed. If you can turn ambient heat into light, that is free energy. It doesn't matter if it consumes 1 watt of power, if 2 watts of power can be generated from converting the light back into electrical current.

However, this has not been verified, and it could very well be that there is some type of entropy consideration that may prevent more electrical energy from being extracted than what is being put in, when the maximum efficiency of the photovoltaic cell is factored in. Right now we just do not know. Since it is ambient heat there might not be any increase in entropy.

What's interseting however is that unlike simply creating a temperature gradient, or emitting photons due to blackbody radiation, this is emitting wavelengths that are much much shorter wavelength than would be expected from the temperature. At 135C a passive device is going to radiate mostly at ~10microns, but this device peaks at more like ~1-2 microns.

Nothing real surprising there. Incandescent photonic crystals have been found not to follow the law of blackbody radiation either.

This type of technology might be useful to extract useful energy from heat with small tempreature gradients.

But that's the thing: no temperature gradient is needed, only the presence of heat. This is what makes it so different from any other heat engine.

"The second law is an empirically validated postulate of thermodynamics"

Basically, "they have actual proof."

But it's still not a fundamental law of physics, meaning it could potentially be violated. True, so far this has never been demonstrated, but being that it is really a statistical law, unlike many other simple phenomena, validating a violation would actually require the building of a complex device and quantitative measurements.

Yes, the second law is a proven observable phenomena, but does not necessarily imply that it will always hold under any conditions.

You speak of Maxwell's Demon.

Scroll down to "criticism and development." 2nd thermo isn't actually violated if you're taking energy from outside the system, hence an "open system." A closed system cannot do this. It's just a matter of definition of the scope of your system. Example: Your light bulb is powered by the electrical grid. By stating that the bulb itself is a closed system, you can draw a conclusion that isn't true.

As far as using a temperature difference on a larger scale to extract energy, this is how Stirling engines work. There's another thread open about those here at the moment.

Yes, it would still violate the second law. Entropy could be decreasing. It would be like transferring more heat from a cold area to a hot area, without venting out heat anywhere else.
Energy can be taken from outside the system, but is being concentrated here. Concentrated such that it would be an easy matter to return the energy to its original state.

Unfortunately, the concept of entropy is not really a mathematically precise one, so it can be difficult to quantify its existence it changes from one form of energy to another. Entropy is more of a statistical generalization.

A better analogy would be a heating unit which used cold air to heat the inside the house without consuming any energy itself. That is what we are talking about here. If you want to call it a heat pump, fine, but it is a "free energy heat" pump. Obviously there are huge implications if this is actually possible.

Look, entropy or not, it would be an amazing discovery if we could just set a device down and it would give out a continuous supply of electric power, without being constrained by the availability of sunlight. Any generator that can absorb ambient heat could effectively "suck" it all in from its surroundings. This is just not the case with harvesting sunlight by solar panel.

 

[AnAppleSnail]

Well, that's what a heat pump does. It uses energy (Electricity) to move heat 'uphill.' Your house aircon probably does this, Anders. If you design your heat pump right, the 'heat' from its input energy also warms the side you'd like it to. Either way, you have a Coefficient of Performance, not an Efficiency.