Optics.org
daily coverage of the optics & photonics industry and the markets that it serves
Featured Showcases
Photonics West Showcase
Menu
Historical Archive

Cost hinders LED uptake

28 Apr 2003

A high price-tag has held back the uptake of white LEDs for many lighting applications. Jacqueline Hewett asks Frank Steranka, vice-president of research and development at Lumileds, about the company's strategy for creating cheaper, brighter devices.

From Opto & Laser Europe May 2003

White LEDs are now entering the market and finding applications in devices ranging from compact, efficient torches to bright surgical light-sources. Other potential uses lie just over the horizon, such as the integration of white LEDs into car headlights. Boasting a lifetime of up to 50,000 h and an output rivalling those of conventional sources, the LED has truly come of age as a "fit-and-forget" solution. So does its arrival mark the end for incandescent and fluorescent lamps in general lighting?

So far, the answer still seems to be no. Despite their impressive performance, one major factor is holding the gallium nitride (GaN)-based light sources back: price. LEDs cost up to 100 times more than halogen lamps in terms of the price per lumen emitted. As a result, customers are still opting for traditional, established technologies rather than paying a premium for LEDs.

Lumileds, a joint venture between Agilent Technologies and Philips Lighting, is just one of a host of companies in this field trying to make LEDs cheaper and brighter. "In reasonable volumes, LEDs cost a little over 10 cents per lumen," Frank Steranka, vice-president of research and development at Lumileds, told Opto & Laser Europe. "It's our goal to get the LED cost down to one or two cents per lumen over the next few years."

To reduce the cost per lumen of white LEDs to effectively one-tenth of the current price, Lumileds's strategy involves two key factors: increasing the device's efficiency and moving to higher drive currents.

"If we keep working on the crystal growth process and light extraction from the crystal chip, we believe we can double the efficiency over the next five years," said Steranka. "So that gets you a factor of two for the same LED chip area and packaging costs. And you get double the lumens because it's twice as efficient."

But for costs to drop far enough, even more improvement will be needed. Steranka says that part of this will come as volumes increase, which will naturally push manufacturing costs down. Moving to higher drive currents will also help.

The latter is in fact particularly important for reducing costs. An LED light-source needs an array of several individual LED emitters to generate a useful light output. If the drive current is increased, each individual LED will be brighter, fewer will be needed per source, and money will be saved.

Technological challenges "We have a solid plan to consistently improve the drive current per device over the next couple of years," said Steranka. "Today, our 1 W [electrical power consumption] devices are driven at 350 mA. This summer we will be releasing the same device type, [but] driven at 700 mA. And then maybe within the next year we will go up to 1 A."

All these developments mean that the price of an LED may soon be only 10 times more, rather than 100 times more, than an incandescent lamp. At this point Steranka is confident that more customers will start to seriously consider switching to LEDs, due to the long-term savings available.

Lumileds may seem to have the future mapped out for its LED business, but there is a snag. There are several different ways to make a white light LED and Lumileds has had to decide between them before it has become clear which of them will become the method of choice in the marketplace.

One way to make a white light is to combine the output of red, green and blue LEDs. But as the eye is particularly sensitive to green light, sources based on this concept often tend to look green. Manufacturers also have to package three discrete devices into one source, which is troublesome.

A second approach involves coating the packaging surrounding an ultraviolet LED (UV LED) with a red, green and blue-emitting phosphor. The UV light activates the phosphor and the eye sees white light as a result. This has its drawbacks too: today's UV LEDs tend to have short lifetimes and emit only a fraction of the power of their visible counterparts. The amount of white light emitted is therefore also low.

The third alternative, and the one that Lumileds is working hardest to develop, is to use a blue LED and a yellow-emitting phosphor. Here, the blue light activates the phosphor. As well as seeing the yellow phosphor emission, the eye sees the blue light from the LED and mixes them to perceive white.

"Some people reckon the best solution is a UV LED with three phosphors, and they may be right," said Steranka. "But as of today, we still believe that the blue source devices will be the best-performing."

Lumileds is currently working on optimizing this technology. One of its first tasks is to ensure that the phosphor is applied uniformly. "We are focusing on a phosphor-coated dye approach where we conformally coat each individual chip with a uniform-thickness phosphor layer," explained Steranka. "At the moment, we are the only people using this approach."

According to Steranka, this technique has given Lumileds more control over the colour-temperature of its devices, and therefore over the whiteness of the emitted light. A low colour-temperature implies warmer (more yellow/red) light, whereas a high colour-temperature means that a colder (more blue) light is emitted.

"Our current devices have an average colour temperature of 5500 K," said Steranka. "Warmer white would be in the region 3000 to 3500 K. We will have a warmer white LED on the market before the end of the year." To achieve this warmer white, Lumileds has also developed a yellow-red-emitting phosphor to work in combination with a blue LED.

The end of a bitter battle over intellectual property (IP) rights for GaN LEDs may help to accelerate LED development. Lumileds signed a cross-licensing agreement with its Japanese rival Nichia in October 2002. Under the terms of the agreement, the two companies will share IP rights to their respective LED technologies. "This removes the concern that there will be lawsuits," said Steranka. "There is going to be an acceleration of the technology development as we share the IP. It will help market adoption and prices will come down."

Emerging applications With brighter, cheaper devices on the horizon, where are white LEDs likely to have the most impact? Steranka thinks the market will pick up rapidly in flashlamps (torches), while handheld torches and bicycle lamps that make use of white LEDs are already available and are likely to grow in popularity as performance improves and prices fall.

Another promising sector is medicine. Surgeons could soon be wearing white LED spotlights on their heads while performing operations. "The light can then shine where the surgeon is looking without casting shadows," explained Steranka. Lumileds has teamed up with RIMSA, an Italian medical-lighting specialist, to develop the world's first surgical lamp based on white LEDs. The new PentaLED lamp, which will be available this autumn, uses five Lumileds Luxeon V LEDs to give a longer-lasting and energy-efficient alternative to conventional halogen-based lamps. The 25 W lamp emits 600 lm and offers a lifespan of 50,000 h.

There are also signs that companies and department stores are starting to consider white LEDs. "Places that do lifetime calculations and understand the energy savings might be willing to pay the upfront price premium for an LED," said Steranka.

In many cases, the adoption of white LEDs will be driven by market acceptance of coloured LEDs in niche applications. Steranka says there is already rapid replacement of incandescent lamps in traffic lights in the US. In Europe, coloured LEDs are becoming the source of choice in centre high-mounted stop-lights in cars.

Last year, Luxeon V LEDs made their debut in the headlights of a concept car showcased at the Geneva Motor Show. "LED headlights are going to happen," said Steranka. "I think in 2006 we will see the first headlights and then by 2010 there will be a reasonable number."

But just how long will it take for LEDs to break into the general lighting market? "I don't see this as a case where there will be a specific threshold where all of a sudden people will stop buying incandescent lamps and replace them with LEDs. It will be a gradual replacement," said Steranka.

He added: "I believe you will see LED illumination applications in the 10-year timeframe, but you will still see incandescents out there. Getting into the home is going to be the most challenging, where upfront cost is vitally important."

But while it is likely that white LEDs will ultimately start to steal market share from incandescent lamps, they face much stiffer competition from fluorescent tube lighting.

The low price and incredible efficiency (70-80 lm/W) of fluorescent tubes will be very hard to beat, meaning that displacing the technology could prove impossible. "I can't say that I see a clear path for that ever happening," concluded Steranka.

LaCroix Precision OpticsSPECTROGON ABECOPTIKTRIOPTICS GmbHOptikos Corporation Berkeley Nucleonics CorporationOmicron-Laserage Laserprodukte GmbH
© 2024 SPIE Europe
Top of Page