Nanotube are more efficient IR source than LEDs

Physicists have discovered a way for carbon nanotubes to emit IR light (in the optical fiber wavelengths) more efficient than LEDs. Perhaps they can get them to laze.

formatting link
" ...3 milliamp current was able to produce about 105 times more photon flux than a large area LED."

Robert

Reply to
Robert
Loading thread data ...

The article actually says 10^5, i.e. 100,000 times more flux.

I find that a bit hard to believe, if they mean they get that much light for the same current. That suggests the LED is 10^-5 the efficiency, and I thought LED quantum efficiencies were in the tens of percent.

formatting link

Do they actually mean the device gave 10^5 times more light per surface area, rather than solid angle?

Reply to
Kryten

Pretty much the same thing isn't it?

Reply to
Ian Stirling

No, not at all.

Two small devices might put out the same amount of light into a cone shape solid angle.

One might be an LED of about 1 mm^2 of light emitting area.

The second might have 10^-5 mm^2, e.g. a 3.16 micron square.

Reply to
Kryten

I don't understand that sentence either.

If you have the same amount of light power flowing into the same solid angle, you have the same luminous flux.

And I'm still no clearer as to what the web page meant by that 10^5 factor.

They don't make any mention of lenses.

Reply to
Kryten

I should have been clearer - if you've got a size constraint - then for a given lens size, decreasing source size means higher solid angle output for a given power.

(until diffraction comes into play)

Reply to
Ian Stirling

I should have been even clearer.

Take a light source in a 5mm LED conventional package emitting 100mW IR.

Now. If you have a 1mm source, you're going to get (about) a 15 degree beam. Pretty much nothing other than throwing light away can get much better than this. This gets you about 1.6W/steradian.

If you have a .01mm source, you're in a different ballpark. With perhaps an extra reflector, you can now get a .15 degree beam, in the same package. This is about 16Kw/steradian. There is the same amount of light coming out of both - it's just that the spot pattern of the nanotube based one at 1m will be akin to the LED based one at 1cm.

Reply to
Ian Stirling

Okay so width of the light source makes for more off-ideal-axis light.

But I suspect that most applications will need the same sort of beam width as LEDs have now. One has to be able to read equipment at moderate angles.

Reply to
Kryten

Basically, yes. Smaller light source means more flexibility. You can use simple optics, rather than complex ones, in some cases. If you're just doing flood illumination, then there is no benefit. Even for things like conventional LEDs, you can more closely specify the beam pattern. For example, to eliminate the hot-spot in the middle, and make the light output constant to a sharp cutoff.

Reply to
Ian Stirling

I think it's just merely journalistic incompetence. A photon flux of 10^5 of a large area LED would burn the house down.

I was kind of hoping for something like nanotube-sized nanometer-wave klystron lasers, and feedhorns -

Maybe even something like an array of these nanotubes mounted orthogonally on one of those magical graphittine sheets, to put a 100 MW laser in a pistol... But we need better battery technology.

Thansk, Rich

Reply to
Rich, Under the Affluence

Yeah, I thought that too.

When someone proposed using LEDs for traffic lights, someone else pointed out that there wasn't really enough gallium produced to satisfy that market. It is only produced (as a by product of the aluminium refining industry IIRC) in the order of tonnes per year. Enough for the many existing applications needing tiny indicators.

Carbon nanotubes are currently more expensive than gold, but if they are

10^5 times more light per mass used then they can still be more cost effective. Gold itself is sparingly used in electronics packaging.

I have a vision of a Far Side cartoon:

"Yoo-hoo, Mister Froggatt! My science fair project is ready! Yeah, I'm talking to you, squidface!"

In the original cartoon it was a huge killer robot, but a Flash Gordon laser pistol would do just as well...

Reply to
Kryten

EETimes article on the same subject.

formatting link

Robert

Reply to
Robert

Sounds like they just edited the same press release, I'm still not satisfied with the text.

I'll check out the IBM website...

Reply to
Kryten

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.