Wavelength tunable LEDs?

[Power_of_the_Sun]

http://www.wipo.int/pctdb/en/wo.jsp?wo=1996041138&IA=US1996008631&DISPLAY=DESC

I came across this article online. Does anyone here have the time, money and intelligence to figure out the range of tuning possible with a single LED? (in other words, could an LED be made to go from infrared to visible light, to UV?)

If so, I think it would be a breakthrough in technology. Imagine, in the future you could have a single light source capable of detecting viruses and bacteria, killing them, and then viewing the area you just sanitized with visible light. My immediate hope is to have a flashlight with these three wavelengths (separate LEDS could probably make this a reality - anyone want to try to make one of these?)
LED 1 - UVC light or 253.7nm (if tuning to that specitivity is possible) - kills 99.9% of bacteria and viruses
LED 2 - Black light or 385nm - detects bacteria and viruses
LED 3 - Visible white light or 650nm

This could make quite a tool for housekeepers and maids. Imagine the money you could make if you produced these commercially! Just a thought...

 

[Steve K]

I've worked a bit with solid state lasers, and know that wavelength would vary with temperature. Don't know if this is because of the change of dimension of the cavity, or a separate property of the semiconductor.

Usually, the wavelength in a LED is a fairly direct function of the bandgap energy of the semiconductor material. I imagine that there is still some variation with temperature, but can't guess why... I don't think that the bandgap energy itself can change.

In general, the variation in laser wavelength was pretty small. Maybe a couple of percent? Certainly not huge changes.

This is probably a good time to consult a semiconductor physics book. Mine isn't handy right now.... this is the book I've got:
http://www.amazon.com/dp/0471056618/?tag=cpf0b6-20
A little pricey, but full of cryptic info.

Steve K.

 

[2xTrinity]

I've worked a bit with solid state lasers, and know that wavelength would vary with temperature. Don't know if this is because of the change of dimension of the cavity, or a separate property of the semiconductor.

Usually, the wavelength in a LED is a fairly direct function of the bandgap energy of the semiconductor material. I imagine that there is still some variation with temperature, but can't guess why... I don't think that the bandgap energy itself can change.

In general, the variation in laser wavelength was pretty small. Maybe a couple of percent? Certainly not huge changes.

This is probably a good time to consult a semiconductor physics book. Mine isn't handy right now.... this is the book I've got:
http://www.amazon.com/dp/0471056618/?tag=cpf0b6-20
A little pricey, but full of cryptic info.

Steve K.

A change in laser frequency is a totally different phenomenon than a change in bandgap. In the case of the laser, there is ususally a gain medium (eg a luminescent gas, in the case of a gas laser, or a semiconductor material as in a semiconductor laser). The bandgap or spectral output basically sets a limit on the possible wavelenths that might be emitted. Say a red LED emits light everywhere frmo 610nm - 660nm in some significant quantity. Lasers are highly monochromatic, so based on dimensions of the cavity, they will basically "choose" a wavelenth from within that range at which to emit their energy. A change in cavity due to thermal expansion miht cause a 650nm diode laser to jump a few nm to like 645nm. But that energy is still within the range of possible wavelengths determined by the semiconductor material itself.

In the case of an LED, you will get SOME radiation form all the wavelengths in that range, in differering amounts, simultaneously. Due to heating, or current flow etc. you might change the domoinant or peak wavelength, but the color won't change significantly. The only way to do that is to sandwich multiple LED dice that have different bandgaps, and control the color by varying the current through them. In other words, separately addressable dice are necessary to have varaible wavelength/color in an LED. there is no way to "tune" it as wavelength is an inherent funciont of the crystal material composition itself, which can't be "modulated" in real time by say, moving a mirror, or changing a bias voltage.

This particular article is talking about very fine tuning of the wavelength, to the naked human eye, such a small change in wavelength output would likely not even register as being a different color. There's no way something like this could be applied to making a red LED turn green, for example.

 

[Baron Von Kline]

Seems like a more practicable approach would be an array of several LED's to "Splash" the desired spectrum range, and then a series of filters to narrow down or pinpoint desired frequencies to be passed.

BVK

Long quote removed and unneccessary.

 

[Bullzeyebill]

I removed one post which was 99.99% a quote. I also edited out a quote of the first post When a post to be quoted is only a few posts away it is not necessary to do the quote. Just respond to the first post, in this case.

Bill

 

[Anders Hoveland]

I came across this article online. Does anyone here have the time, money and intelligence to figure out the range of tuning possible with a single LED?

I am fairly sure the LEDs in that article can only be tuned over a narrow range, less than 5-10nm, perhaps even less.
Some LEDs have a wider bandwidth emission than others, and these could obviously be tuned over a wider range than LEDs with more discrete emissions. The widest bandwidth I have seen for any direct LED emission is still only about 20nm.

That would not be enough for any real color change, not unless the emission was situated in the blue-green part of the spectrum where there is a steep change in human color perception over a short span of wavelengths.

I think these "tunable" LEDs would be intended more for research purposes, when a more exact wavelength is required.