Color-Tunable LEDs Open Up Possibilities Of Configurable Semiconductors

The invention of the blue LED was groundbreaking enough to warrant a Nobel prize. For the last decade, researchers have been trying to take the technology to the next level by controlling the color of emission while the device is in operation. In a new research paper, by the guys over Osaka University, Lehigh University, the University of Amsterdam and West Chester University have presented a GaN LEDs that can be tuned to emit different colors from the same substrate.

GaN or Gallium nitride is a wide band-gap semiconductor that has been employed in the manufacturing of FETs that are known to have higher power density due to its high thermal capacity while increasing efficiency. In the the case of the tunable LED, the key has been the doping with Europium for creating energy bands. When an electron jumps from a higher band to a lower band, it emits energy in the form of light and the wavelength or color depends on the gap of energy jumped as per Plank-Einstein equation.

By controlling the current density and duty cycle, the energy jumps can be controller thereby controlling the color being emitted. This is important since it opens up the possibility of control of LEDs post production. External controllers could be used with the same substrates i.e. same LEDs to make a lamp of different intensity as well as color without needing different doping for R,G and B emissions. The reduction in cost as well as size could be phenomenal and could pave the way for similar semiconductor research.

We have covered the details of the LED in the past along with some fundamentals on the control techniques. We are hoping for some high speed color accurate displays in the future that don’t break the bank on our next gaming build.

Thanks for the tip [Qes]

26 thoughts on “Color-Tunable LEDs Open Up Possibilities Of Configurable Semiconductors

  1. Displays based on tunable LEDs seem to have one giant drawback IMHO. They are impossible to calibrate efficiently. Unlike CRT displays or LCD, for which it is possible to assume colours are uniform across the whole display, each tunable LED pixel can be different.

    1. What ? Why ?
      It can be done efficiently. For instance you could tool the pick and place machine to calibrate the LED and remember the calibration constant of every pixel. It could even discard outliers and replace them immediately before they are even assembled on the display.

      1. Except when making a screen, one builds the LED directly on the screen’s surface.

        So the variation in performance between the individual LEDs would matter, and not really something we could do much about other then scanning over the screen with some optical solution to calibrate each pixel.

        To say the least, time consuming, and likely expensive. And potentially irrelevant if aging/ambient-temperature creates noticeable drift.

        Though, it is still likely that neighboring pixels would have somewhat similar behavior. (though, dependent on what the LED characteristics gets effected by in production.)

  2. So does this mean plasma screen will not longer need the three R-G-Bs anymore? And please, whomever responds, don’t get sarcastic or rude with me because you know what I mean. Thanks, Jim

      1. If you read the article you will see that these LEDs are capable of emitting multiple wavelengths at once. I wouldn’t expect a pixel with one single LED but I can imagine the addition of this type of LED to be able to increase the amount of control or the color fidelity significantly

    1. This is still experimental and it might be years until it is ready to hang on a wall, but if they succeed, yes. Unlike a plasma (or LCD) screen, this would require no R-G-B subpixels. Each pixel would be able to emit light in a range of color and intensity.

      1. Perhaps if they use POV (persistance of vision) to mix different colors both output by the same LED. It could not be done in steady state. For instance there is no brown in the spectrum. You only get that by mixing colors.

        1. I don’t have access to the full article, but the abstract states:

          “we show that it is possible to attain all three primary colors due to an emission originating from two different excited states of the same Eu3+ ion mixed with near band edge emission from GaN centered at ∼430 nm. The intensity ratios of these transitions can be controlled by choosing the current injection conditions”

          Note the phrases “mixed with” and “intensity ratios”; my understanding is that the device emits multiple wavelengths simultaneously–“mixing colors” is actually what it does.

  3. Old silicon carbide blue LEDs can do this… (kind of).

    If you put them on a variable current supply, and slowly turn up the current.

    They will shift from blue to yellow, and then let out the magic smoke.

    I was 10 or 11 years old when I discovered this (’89 – ’90)…
    My dad wad not amused, since I used up about half of his expensive blue LEDs. ;)

    1. Was just going to comment on this….
      I had one of those spring-board X-in-one electronics education kits… well, I hooked up the red LED to the variable resistor and slowly cranked the resistor to “see how bright” I could get the LED… instead the LED went from red to orange and after a few seconds, the resistor had smoke coming off it.

      Thinking back…. It may of been a green LED… or something… however, I managed to change the LED colour by over driving it. (Silicon LEDs back then)

  4. As one of the authors of the paper, I should point out that critical parts of the research have been performed by female scientists. I would recommend to not use terms like “guys” and suggest that such work can only be done by male.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.