Cambridge University creates low-cost production method for GaN LEDs

Winchester, UK - Cambridge University's Centre for Gallium Nitride has developed a new way of making GaN which could produce LEDs for a tenth of current prices and may see household lighting bills reduced by up to 75 percent within five years.

Currently GaN LEDs can only be grown on 2-inch sapphire wafers. Following the Cambridge innovation nine times as many LEDs can be grown on a 6-inch silicon wafer compared with a 2-inch sapphire wafer. In addition, edge effects are less, so the number of good LEDs is about 10 times higher. The processing costs for a 2-inch wafer are essentially the same as for a 6-inch wafer. A 6-inch silicon wafer is much cheaper to produce than a 2-inch sapphire wafer. Together these factors result in a cost reduction of about a factor of 10.

Unfortunately, as best I understand it, LED's give off less light per watt.

They are built with miniature internal reflectors, so their light is more directional, unlike incandescent or fluorescent lights, which tend to give off light in almost all directions.

So when tested with a light meter, if the LED light is directed at the meter, they seem to give off more light per watt than an incandescent, which they do, in the direction of the meter. But for room lighting, where one wants dispersed light in many directions, LEDs are less efficient than those darn fluorescent lights.

This makes LEDs useful in spotlights, flashlights and recessed lighting, for which strongly directional lighting is appropriate. But it makes them a poor fit for room lamps.

Heat is also a problem with LEDs. They aren't as good at heating a room as halogens or incandescents, but they do tend to have a hot spot right at the LED. That heat has to be dispersed, or else the LED will not last long. This is why the higher power LED flashlights have a big metal heat sink around the head of the light.

The combination of excellent shock resistance and a big fat metal head and long life does make them a good choice for a flashlight that doubles as a baton weapon.

The screw in CFLs are not so great - they are wasteful if nothing else because the transformer circuit is built in and so is predictably cheap and not too reliable. Still I buy them when I can get them for a Dollar or less and they usually last a lot longer than tungsten lamps and they do save some power. But I never leave them on unattended near or over flammable material, because like others have said, sometimes they blow up. And there is the mercury issue that concerns a lot of folks, though I'm not sure it's such a big deal myself.

The fixtures that are built for the plug in "U" shaped compact tubes are much better because they contain a more expensive permanent transformer which is safer and more efficient. I have some of those outside with the built in light sensor switches and five years on they just work. They come on and burn bright even when it's 15 degrees outside (try that with a screw in CFL).

LEDs have potential to turn the lighting industry upside down. They are already the most efficient and portable (in terms of power source) lighting made. The raw material to make them is abundant and to reduce production cost by a factor of 10 or even 5 could be a revolution.

I view LED lighting as a step toward independence, both on a national and individual level. It represents an incremental but measurable reduction in power demand - and a step toward a critical mass that will ultimately make the self powered home a viable option.

http://www.dailytech.com/New+NearProduction+Super+Lightbulb+Lasts+60+Years+Costs+285/article14094.htm

New Near-Production “Super” Lightbulb Lasts 60 Years, Costs $2.85
Jason Mick (Blog) - January 29, 2009 4:00 PM

Cambridge University professor Colin Humphreys shows off his team’s new LED which is near to production. It will cost around $2.85 to produce, and will last 60 years. (Source: Matthew Power MASONS)

The new lightbulbs are bright and more efficient than traditional designs. And unlike other green bulbs they contain no toxic mercury, they turn on instantly, and they do not flicker. (Source: Matthew Power MASONS)
Future of lighting looks bright with new invention

The race towards better, more affordable solid state lighting is heating up quickly. The U.S. government has sponsored a $20M USD prize for the first team of researchers to come up with solid state lighting that meets a strict set of standards. New research has finally helped to eliminate the LED droop typically associated with the higher currents needed to provide greater efficiencies.

Now a team at Cambridge University may be close to having a winning design on their hands, perhaps for the L Prize, if they’re eligible, and for the consumer market. The university has produced a new design which costs a mere $2.85 USD and despite being the size of a penny, produces similar light to a fluorescent bulb while lasting over four times as long with a lifetime of 60 years.

The new design triples fluorescent bulb efficiency and is 12 times more efficient than incandescent designs. Also, it’s capable of instantaneous illumination, so the light lag associated with fluorescent bulbs may soon be a thing of the past.

If installed across all of Britain, the researchers estimate that it could cut the country’s lighting portion of the energy budget from 20 percent to 5 percent a year. The U.S. could muster a similar 10 percent drop with the design, according to recent DOE estimates. The new bulbs last 100,000 hours and unlike other “eco” bulbs, they contain no mercury, a substance that can cause brain damage in humans. They also don’t flicker, while other green designs do, something that’s been blamed for triggering epileptic fits.

Officials say the new design could cut 40 million tons of carbon emissions in Britain alone. Britain recently stopped restocking certain incandescent bulbs in stores. The new design relies on a specially formulated gallium nitride semiconductor, which builds on previous LED work. It is brighter than traditional designs and relatively cheap from a chemical perspective, compared to more exotic chemistries.

The British researchers managed to make the LEDs even more affordable by growing them on silicon wafers instead of on sapphire wafers, the traditional method of production. This makes them at last cheap enough for the consumer market. Growing the LEDs on silicon was assisted by a number of advances at other U.S. and European research institutions.

While some designs take decades to reach the market, Cambridge’s design is already being prototyped and readied for production. RFMD in County Durham, England is the first manufacturer to jump at the opportunity to mass produce and ship the high-performing LED bulbs.

Professor Colin Humphreys, head of the team at Cambridge states, “This could well be the holy grail in terms of providing our lighting needs for the future. We are very close to achieving highly efficient, low-cost white LEDs. That won’t just be good news for the environment. It will also benefit consumers by cutting their electricity bills. It is our belief they will render current energy-efficiency bulbs redundant.”

I had one fail on me a couple of years ago and I looked it up on Google.

A detailed description of the failure modes of fluorescent lamps is given in the fluorescent lamp article. Additionally, the electronic ballast may fail since it has a number of component parts; such failures may be accompanied by discoloration or distortion of the ballast enclosure, odors, or smoke. The lamps are internally protected and are meant to fail safely at the end of their lives. Industry associations are working toward advising consumers of the different failure mode of CFLs compared to incandescent lamps, and to develop lamps with inoffensive failure modes.[56]

Electronic ballasts should be designed to shut down the tube when the emission mix runs out as described above. In the case of integral electronic ballasts, since they never have to work again, this is sometimes done by having them deliberately burn out some component to permanently cease operation.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.