IBM Claims Brighter LED

[jollyrgr]

IBM has come up with a way to make light using silicon that is 1000 time brighter than current LEDs.

http://www.informationweek.com/story/sh ... =174301044

[L. Daniel Rosa]

I'm not sure I care for Colin Johnson's choice of words. He said the "infrared light" is 1000 times brighter. Will he next say that the emissions from cellular telephones are trending towards a greener shade of purple microwaves?

Also with the "n-fold increase", is it a multiplier or a doubler? I've heard it used both ways, but prefer the brainier one- so 10 fold increase would be *1024.

Okay, I'm done whining now. I promise.

[jimandy]

Personally, I'm for bringing back the Nernst Lamp.

[paulrevelcet]

Here is some interesting faqs about the nernst lamp http://www.nernst.de/lamp/nernstlamp.htm

[rshayes]

The emitting area on an LED may be close to a square millimeter or 10^-4 meter^2. The area in the press release is described as 2 nanometers in diameter, or an area of of about 3x10^-18 meter^2. A thousand times brighter than an LED would imply a total output power about 10^10 less than an LED when the emitting area is considered.

LED's are not particularly bright light sources. The emission from solid state lasers is much brighter than that of an LED, since the radiation occurs in a much smaller area. This looks like a straw man type of comparison.

The wavelength of the emission is not given. It is described as infrared radiation, which would imply a wavelength longer than 1 micron. The diameter of the source is given as 2 nanometers or about 1/500 of a wavelength.

Since most of the figures given are relative and vague, it is hard to tell exactly what IBM is claiming.

[Chris Smith]

“Photonics Spectra” [trade magazine] reported all of this over two years ago.

You can check the back issues for all the details.

The carbon nano tube structures are doing more than just producing light. Like a diamond, they have many uses including the new carbon fiber structures for structural strength in things like bridges, airplanes, medicines, light emitting devices, and computational devices.

Added with metals like nano-nickel, their usage is even further boosted.

[rshayes]

A better source is probably the web site of IBM's research department (http://www.research.ibm.com/nanoscience/nanotubes.html). Along with the research papers, this site has an article from the December 2000 issue of Scientific American.

In 2000, the basic material cost $1500 per gram. The material is not pure, but consists of a mixture of straight and twisted tubes of different sizes. The straight tubes are conductors and the twisted tubes are semiconductors with differing band gaps, depending on the amount of twist. Individual nanotubes are individually selected and manipulated to make various devices.

Research in this area appears to have been going on for about twenty five years. The one actual product mentioned is the use of a nanotube for the probe in a scanning tunnelling microscope. Until a method of producing uniform nanotubes is found, production of most of the devices mentioned is probably wishful thinking.

It took nearly thirty years to learn how to consistently produce mercury cadmium telluride for infrared detectors. Producing carbon nanotubes with a specified twist (to get semiconductor characteristics with the correct band gap) might be even more difficult than that.