Core77

MIT Researchers Produce Micro-Light that Outputs More Energy Than It Takes In

By Rain Noe - June 12, 2012

in Tech

Materials scientists know that things behave differently on a microscale than they do on a larger, more human scale. And now, researchers at MIT have apparently proved that this is true of light as well.

The MIT team, comprised of Parthiban Santhanam, Dodd Joseph Gray, Jr., and Rajeev J. Ram, have created a common-sense-busting ultra-efficient LED with an efficiency of 230 percent. Their light draws 30 picowatts of power, yet produces 69 picowatts of light.

How the heck is that possible? The long answer looks like what Good Will Hunting was scrawling out on the chalkboard between bouts of mopping. The short answer is,

The researchers chose a light-emitting diode with a small band gap, and applied such small voltages that it acted like a normal resistor. With each halving of the voltage, they reduced the electrical power by a factor of 4, even though the number of electrons, and thus the light power emitted, dropped by only a factor of 2. Decreasing the input power to 30 picowatts, the team detected nearly 70 picowatts of emitted light. The extra energy comes from lattice vibrations, so the device should be cooled slightly, as occurs in thermoelectric coolers.

The gist of the "lattice vibrations" is that the device draws in ambient heat from outside of itself and converts that into light. It's an exciting prospect, even if the breakthrough is currently more academic than anything else: "These initial results provide too little light for most applications," the research synopsis admits. A picowatt is just one milion millionth of a watt.

o

Favorite This
Q
16

Comment

Comments

Charles Ivie

9 years ago

Z

Z

Reply

This may be a functioning example of "Max well's Demon". If so it is a major achievement.

!Report as spam
Sean

12 years ago

Z

Z

Reply

So, skip over the light and just use lattice vibrations to create electricity. Thus, convert global warming into an energy supply.
!Report as spam
George Burkhard

12 years ago

Z

Z

Reply

A better way to think of why this doesn't violate thermodynamics is that you actually are conserving photon power (despite what they say in the article). The article doesn't consider the blackbody radiation that the device emits from being at a given temperature. As they mention, because the emitter is taking energy from the thermal bath, it cools the bath, meaning that the blackbody radiative power is decreased. Instead this power goes to emitting photons at or above the bandgap of the material. So the extra photons at the bandgap are at the expense of the photons emitted by blackbody radiation.
!Report as spam
Gerard Vignes

12 years ago

Z

Z

Reply

I think this result is more important than it might first appear---so many of our technologies produce heat as a by-product---we are literally drowning in heat---try operating a modern society without cooling towers, air conditioners, ...---we need some way to convert this waste heat back into useful energy---i realize we are talking about photons (which don't store readily :-)---but perhaps this is the beginning of an exciting new direction
!Report as spam
Bergun

12 years ago

Z

Z

Reply

@ Steve No, this does not disprove anything. The extra energy comes from the thermal excitations of the lattice known as "phonons." The device would not be able to produce "more" than given energy at very low temperatures probably.
!Report as spam
Steve G.

12 years ago

Z

Z

Reply

Steve S.: No, they have not! Did you not read the article??? The lattice vibrations inside the LED make it take up heat from its surrounding, thus the conservation of energy and E=mc^2 stays still true. Heat = energy, you know...
!Report as spam
Luke

12 years ago

Z

Z

Reply

@ Steve
agreed, truly unbelievable

sounds like energy is conserved, but surely it's breaking the second law of thermodynamics?!

!Report as spam
James Wood

12 years ago

Z

Z

Reply

Think it would be more along the lines of contradicting the thermodynamic arrow of time ie. the entropy of the system decreasing. Amazing if it's true.
!Report as spam
JP Pennypincher

12 years ago

Z

Z

Reply

Cool. I'm looking forward to glowing AC in my server closet, Fridge 2.0, and eliminating those infernal CPU fans (in 150 years).

Actually, this would be quite synergistic with a move to optical system buses.
!Report as spam
Tristram H.

12 years ago

Z

Z

Reply

@Steve S

Not quite, the extra energy is drawn from "Lattice vibrations", which are the vibrations of the internal structure of the LED. In effect it's thermal energy. The LED basically turns it's own heat into light.
!Report as spam
Juho

12 years ago

Z

Z

Reply

It is good to remember that this does not violate the thermodynamics as we understand them. In the matter of fact, the LED does not even emit more energy than it consumes. What the paper says is that it does radiate more light than it consumes _electrical_ energy. The rest of the energy consumed comes in the form of heat being converted to radiation. As long as dE(light) < dE(electrical) + dE(heat), this isn't a perpetual motion device.

Compare to some "perpetual motion devices" where the source of the motion was, in fact, a hidden motor - here, the heat source is the motor. Might still be useful if this can be done in larger scale, since usually electrical circuits do produce quite a bit of excess heat, and if this heat can be used, then to total efficiency of the electrical device can be slightly improved.
!Report as spam
Rudolf L

12 years ago

Z

Z

Reply

No, they didn't disprove the 1st law.

"The gist of the "lattice vibrations" is that the device draws in ambient heat from outside of itself and converts that into light."

Ambient heat provides the rest of the energy needed.
!Report as spam
Quite N.

12 years ago

Z

Z

Reply

Steve S: Not quite. Here's the relevant quote from the article:

"The gist of the "lattice vibrations" is that the device draws in ambient heat from outside of itself and converts that into light."

So this disproves E=mc^2 about as much as solar panels do; it's merely converting heat energy in the surrounding atmosphere into light energy. Very cool, nonetheless.
!Report as spam
⬡

12 years ago

Z

Z

Reply

It's not violating conservation of energy. It's drawing in ambient heat, so the total output is still less than or equal to the total input; the input is just coming from more than one source. Still, it's very cool. No pun intended.

I hope to see further development in the "converting ambient heat to electricity" field. Would be a very nice way to cool CPUs...
!Report as spam
Ricrardo R.

12 years ago

Z

Z

Reply

quite hard to belive !
but it is another demostration of why should we rely more on tecnology as a part of an enviromental change !

tumbs up as usual to MIT !
!Report as spam
Steve S.

12 years ago

Z

Z

Reply

did MIT just disprove the 1st law of thermodynamics "conservation of energy" or Einstein's "E=mc^2"?

If so, may be they can hook up a bunch of these and connect them to solar panels to produce free energy, since the bulbs "produce more energy than they take in"....May be we really did find faster than light particles

Did MIT just solve the world's energy problem....truly unbelievable accomplishment.
!Report as spam