Novel LED Design May Boost Efficiencies (10-16x!)

[OKIsItJustMe]

http://www.osa.org/About_Osa/Newsroom/News_Releases/Releases/10.2011/Renewable-Energy-the-Environment-Congress-Presentation-Highlights.aspx#5

…

Novel LED Design May Boost Efficiencies
Light-emitting diodes (LEDs) are vastly more efficient than incandescent light bulbs and even compact fluorescent bulbs, but they still have design constraints that limit their potential. Current systems rely on scattering and diffraction of light from within an LED device to boost the total light emitted. This design enables extraction efficiencies of approximately 50 percent in some red-yellow LEDs and approximately 80 percent in some blue-green LEDs. This technique of producing multiple reflections of light within the device, however, may have reached its limits. A new technique proposed by Xue-Lun Wang and his colleagues at the National Institute of Advanced Industrial Science and Technology (AIST) in Japan takes a more direct approach and may pave the way to even more efficient LEDs. The researchers designed a flat-top, ridge-shaped semiconductor surface that channels two waves of energy toward the top of the semiconductor. Known as evanescent waves, these waves travel along the surface/air interface of the semiconductor. When they intersect and combine at the ridge top, they couple with each other and are transformed efficiently into light. By adding an additional layer of silicon dioxide to the semiconductor, the light extraction efficiencies can be enhanced even further. Their initial results indicate that the ridge-shaped design is 10 to 16 times more efficient than a flat design. By using this technique, the researchers were able to extract light without involving the multiple reflection processes in conventional techniques. Their next step is to combine the system with a metal mirror in hope of creating efficiencies much higher than those in conventional devices.

Presentation SDThB5, "A Novel LED Light Extraction Technique Based on Evanescent Wave Coupling." Xue-Lun Wang, Guo-Dong Hao, and Tokio Takahashi, Nanosystem Research Institute, National Institute of Advanced Industrial Science and Technology, Tsukuba, Japan, takes place Thursday, Nov. 3 at 12 p.m.

…

[kestrel91316]

1. More lumens! More lumens!!!

:rofl:

I really want to get on board with LED lighting. I am installing LED nightlight/emergency lighting in the clinic and have one at home also. Handy little things, have proved their worth during our 4 transient blackouts so far the past month (LA is like a 3rd world country anymore with the power failures).

But for general lighting they are woefully inadequate.

[OKIsItJustMe]

2. “…for general lighting they (LED’s) are woefully inadequate.”

I disagree.

I recently replaced some spotlights in a high ceiling with these:
http://www.amazon.com/Dimmable-Replaces-Incandescent-Halogen-Recessed/dp/B004ENC68S/


I couldn’t be happier!

[HillbillyBob]

5. That brand looks like the 'Lights of America' that Sams sells

they are made in China of course. I m looking for and forward to getting some Cree led equiped lights they are supposed to be brightest per watt. I have eaglelight.com commercial 55k color temp 4ft fluro tubes as replacements for the kitchen..I can actually see out there now. The 22 watt puts out 2000 lumens or about 2x what a regular 4 ft fluorescent does and you take out the ballast. There is a video of how to on the page for the tube lights. I have also converted my lights for wintering over my tropical plants..Im still a bit short on lumens, but instead of 1000 watts im now using about 200 220 watts and my plants do well under them.

Aside from changing all of the most used general lighting to LED and we still have maybe 30% CFLs as those die off I will LED them.
We bought a front loader washer and HE dishwasher, both paid for themselves in savings in 2 1/2 yrs.

When we moved in to this place it had all old appliances, many leaks, no weather striping, and all incandescent lights.
Power bills were about 300 a month and 450 500$ a month for AC or heat. Our highest bill this summer(using 2 HE window units 746 watts and 2200 watts for a total of 22,000btu, since I have sealed up the leaks those 2 units kept the house about 75 degrees and power bill was 250(highest and most 100 degree days) other bills were 143 and 150 that is a big savings over the old bills.
If you have a dark colored roof change it or paint it with henrys kool seal it made a huge difference.

In winter for now at least we are using a 23,000 btu kero heater and take advantage of winter time solar gain. Eventually we plan to go with solar powered under floor water heat.
We are not sitting in the cold and dark either, I look for lower power use stuff as I need to replace old stuff, but not till the old dies and i can recycle it.

[OKIsItJustMe]

6. They are Lights of America

I believe the boxes said they were made in the USofA. (I’ll check though…)

[OKIsItJustMe]

12. Assembled in the USofA

Which (I assume) means that at least some of the components (e.g. the LED’s themselves) were made elsewhere.

[kestrel91316]

9. I'll stick with fluorescent lighting for my office. Can't afford to spend thousands

putting in a new ceiling that would accomodate the necessary new fixtures, lol.

[Tesha]

3. A "10 to 16x" increase in LED efficiency is impossible.

For any given wavelength of light, a watt of input
electrical energy can only be converted into an
equivalent watt of optical energy output.

For broad spectrum light that we would call "White",
a watt of electrical input energy can only produce about
250 Lumens of light even if the conversion is done at
100% efficiency.

Right now, the best "white" LEDs manage about 105
Lumens/Watt so there's only a factor of about 2.5 to
go before those "white" LEDs reach perfect efficiency.

(I don't know the numbers off-hand for other, pure-
color LEDs but I assume they're better, given that
the "white" LED wastes a lot of energy down-converting
blue photons to yellow photons; the pure-color LEDs
don't have to do that.)

Tesha

[Geoff R. Casavant]

7. I wonder if they were approaching efficiency from the opposite direction.

So when they said efficiency was 10 to 16 time greater, what they really meant was that wasted energy was reduced by a factor of 10 to 16. In other words, from losing 145 lumens/watt down to losing 10 or 15 lumens/watt.

[Tesha]

8. Well, they might be, but I'd consider that "lying with statistics".

More likely, they were discussing the efficiency of
some small part of the entire LED "system" and that
wasn't conveyed, either by the article or by the excerpt.

Tesha

[kristopher]

10. Perhaps you should reserve your charges of "lying"...

They are not addressing energy conversion but the amount of light that crosses the semiconductor to air interface without refracting.
This looks like it is reporting at a conference work published about a year and a half ago. The paper is clear and "light extraction efficiency" of their shape over a "flat design" was specified. Since the "flat design" is a baseline by which other approaches are measured (ie flat design times X) it is hard to see how that could be interpreted as "lying".

A more detailed explanation of their work is here:
http://www.nanowerk.com/news/newsid=16229.php

You'll note that in this write up they directly compare their method with the "multiple refraction" method that is in the first para of the OP and report a 1.5X improvement in extraction efficiency over that method.

[Tesha]

4. (Duplicate post removed by author)

[txlibdem]

11. +1