LED produces 2.3x more light energy than electrical energy consumed

in LEDs by Arup | 9 comments

MIT physicists developed an LED that emits more light energy than the electrical energy consumed. These LEDs supplement electrical energy with heat from the surrounding environment. Unfortunately this has only been observed at very low power levels.

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 pW, the team detected nearly 70 pW of emitted light. The extra energy comes from lattice vibrations, so the device should be cooled slightly, as occurs in thermoelectric coolers.

Via: Adafruit

This entry was posted in LEDs and tagged , .

Comments

  1. kevin says:

    now that is cool!

  2. J. Peterson says:

    So can you shine it on a solar cell and get infinite electricity?

    • Brian says:

      No… Electron transport in materials is also related to thermal transport, this coupling with the particular band structure allows a small use of thermal energy in the room to help emit photons. (The light emission process induces a temperature gradient on the junction, which then helps assist the emission).

      You can also cool things via stimulated emission effects IE by shining light on them. People I know have done this to cool mechanical beams to near the quantum ground state.

  3. Fractal says:

    Caveats: Only proven to work for near infrared (~2.3um), the LED needs to be heated to 135 degrees C, and necessarily only works for very low input power. The electronic effect (as in actual electrons) can be imagined as similar to the peltier effect. The paper is interesting, and worth a read.

  4. JBeale says:

    Of course, if you get an object (any object) hot, it emits some infrared photons even without any LEDs involved…

  5. Brian says:

    As Fractal notes the outcome is interesting on a purely scientific basis. Essentially what the device is, is a pn heterojunction that allows for Seebeck effect to assist with photon generation. There are some nice band diagrams to explain how.

    The black body radiation produced by heating the junction greatly, greatly exceeds the AC intensity they have measured with a lock-in, and the effect is quenched if the junction is biased to emit a reasonable amount of light.

    These things taken together mean that result is not that practical. It remains to be seen if co-design for Seebeck like effects and emission is practically cost effective. (IE given how small the effect is a conventional thermoelectric could be added perhaps to greater effect at lower cost than a highly optimized LED that does both).

  6. hardcore says:

    yep…. but you have to put in the heat energy as well as the electricity.

    Then we get the issue if how to prove this near Infra-red is not actually the ‘heat’ we put in being re-radiated at a slightly different wavelength.

    It’s like saying…. i can reduce down my petrol consumption WHEN travel down hill in my car, and I get 2.3 times more ‘travel’ energy out than petrol in. (but we will overlook the gravity because that wont make such a good headline)

  7. Chris says:

    Seebeck? Thermocouple? Both require a temperature differential across the junction as a whole to produce anything, which I don’t see referenced here. This seems to be more of a Maxwell’s Demon sort of thing, producing order from entropy. Sure it’s a minor effect, but still interesting that such a thing is even possible.

  8. unkewl says:

    First law of thermodynamics >.> At least there on the right path for creating new things. Need to aim higher though.

Leave a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Notify me of followup comments via e-mail. You can also subscribe without commenting.