LED Lighting Update
Latest developments in producing low-cost LED bulbsPosted Apr 29, 2009
LED light bulbs are extremely energy-efficient, using only 2 – 10 watts of electricity. They are more efficient than CFL bulbs and do not contain mercury. Built using durable solid-state technology, these bulbs are now available with standard bases which fit common household light fixtures. Among lighting developers, LEDs are considered the next generation in home lighting.
The high cost of producing LEDs, however, has been a roadblock to widespread use. Several months ago, researchers at Purdue University achieved a LED production breakthrough which clears the way for low-cost, high-efficiency lighting. This development has stimulated interest in the availability of LEDs using the new silicon technology, and homeowners and businesses are eager to learn when low-cost LEDs will be available on the market.
Eartheasy has been in contact with Tim Sands, the Director of the Birck Nanotechnology Center at Purdue, to get a better idea of when these new LEDs may become available. Here is the text of our conversation:
Hi Tim –
We are following up on your research with anticipation about the availability of low-cost LEDs. Can you bring us up to date on developments in the new LEDs? Do you have a “best guess” as to their eventual market availability?
Our research on metallized silicon as a substrate for GaN LEDs certainly struck a nerve! Of course, that is why we have pursued this direction.
The work was funded by the Dept of Energy’s Solid-State Lighting program, with help from the Next Generation Lighting Industry Alliance (NGLIA). We have received considerable interest from VCs and companies who might be interested in helping to commercialize the technology, but there are constraints associated with the licensing that must be worked out before we can move forward. While we are working on overcoming these obstacles, students from Purdue (engineering students from my group teamed up with MBA students from Purdue’s Krannert School of Management) are competing in business plan competitions. They are doing quite well, having recently won the Global Venture Challenge at Oak Ridge National Laboratory. They are competing at Rice this week, and have a competition at MIT in the near future.
Regarding the technology itself, one has to put it in the proper perspective. Our invention describes a new way to marry GaN-based LED heterostructures (emitting UV, blue and green light) with low-cost silicon substrates. This process has the potential of greatly simplifying the manufacture of the LED itself. The invention does not directly address other challenges that are keeping the cost of high-performance white LED lighting from dropping as fast as everyone would like.
The invention does not directly address other challenges that are keeping the cost of high-performance white LED lighting from dropping as fast as everyone would like.
Among these challenges at the chip level are the “green gap” and “droop.” The green gap refers to the fact that there is no efficient LED that emits light near 555 nm (the yellow-green part of the spectrum where our eyes are most sensitive). The fact that our eyes are sensitive here means that we can see what appears to be bright light from relatively inefficient LEDs in the green to yellow part of the spectrum (e.g., traffic lights), but the low efficiency (still much better than incandescents for single-color applications!) means that we cannot use these LEDs in general lighting applications. Instead, we take blue or UV light from an (In,Ga)N LED and down-convert it to green, yellow, orange and red using phosphors, an inherently inefficient process, but still better than the direct emission from an LED at these colors. Our group at Purdue, as well as other research groups around the world, are working hard on the green gap problem.
The droop problem refers to the observation that the efficiency of green LEDs drops as the electrical power input increases. This problem is being tackled by several groups (Philips Lumileds lighting and the university group at RPI in NY are perhaps the most prominent), and I have some confidence that this problem is on its way to resolution, at least in the green. Once we figure out how to make efficient LEDs in the yellow-green, and yellow parts of the spectrum, I think we will see similar droop problems that need attention.
Another chip-level problem is light extraction (even if LEDs generate light efficiently, we still have to get the light out of the material). Much progress has been made on this in recent years, and my personal opinion is that this will not become a limiting factor.
Beyond the chip itself, we have issues of heat extraction, driver circuitry, form factor, luminaire design (how do you get the angular distribution of the light that you desire?), etc. Engineers are working hard on all of these issues.
From a broader perspective, a big concern is the proliferation of cheap, low-quality, white LED lamps that have the potential to disappoint eager consumers.
From a broader perspective, a big concern is the proliferation of cheap, low-quality, white LED lamps that have the potential to disappoint eager consumers. We all have experience with disappointments over the quality of cheap CFLs. As an LED consumer myself, I have been purchasing replacement bulbs that run from $60-$120 (still cheaper from a lifecycle perspective than incandescents). I stick to “warm white” and I look to see what manufacturer provided the LEDs themselves. Even with these high-end LED lamps, I have encountered problems (not related to the chips, but to the packages and wiring). I think the experiences in commercial and public lighting (warehouses, street lamps, garages, etc) have been more favorable. Hopefully, the quality of the lamps for residential market will continue to improve, and the costs will come down steadily.
I would go all-LED if I were building a house today
Our technology is just one piece of the puzzle. Improving quality and performance, and reducing the initial cost will require continued innovations (and scale-up) on several fronts simultaneously. Based on recent history, there is no reason to doubt that this progress will continue for the foreseeable future. As for my personal view as a technology-literate consumer of LED lighting, I would go all-LED if I were building a house today, and I expect to have replaced half of the incandescent and fluorescent bulbs and fixtures in my house with LED lamps by 2015.
Timothy D. Sands
Mary Jo and Robert L. Kirk Director, Birck Nanotechnology Center
Discovery Park, Purdue University
Basil S. Turner
Professor of Engineering, School of Materials Engineering & School of Electrical and Computer Engineering