"Light recycling" tech could save incandescent bulbs from obsolescence

"Light recycling" tech could save incandescent bulbs from obsolescence
The prototype two-stage incandescent light bulb
The prototype two-stage incandescent light bulb
View 1 Image
The prototype two-stage incandescent light bulb
The prototype two-stage incandescent light bulb

Incandescent light bulbs may put out a warmer-looking, more familiar type of light than LEDs or compact fluorescents, but they're far less efficient – the majority of the energy they use is wasted, mainly in the form of heat. Technology may save them yet, however. Scientists at MIT and Purdue University have developed an ultra-efficient new incandescent bulb that reuses the heat it gives off, converting that heat into more light.

With traditional incandescent bulbs, both visible and infrared light are created by heating a tungsten filament, causing it to glow. Both wavelengths flow unimpeded out into the room, with the infrared doing nothing other than dissipating as heat.

In the case of the new two-stage incandescent, however, the filament is surrounded by structures known as photonic crystals.

Made from abundant elements and manufactured using conventional material-deposition technology, these crystals allow visible light to pass through, but reflect the infrared back onto the filament. This helps keep the filament heated, glowing and emitting more visible light, while using much less electricity than it would otherwise.

The bulb could conceivably score very high when it comes to luminous efficiency – this is a measure of how well a light source produces visible light. While regular incandescents have a luminous efficiency of 2-3 percent, with compact fluorescents coming in at 7-15 percent and LEDs at 5-15, the two-stage incandescent could reportedly manage up to 40 percent once developed further.

The current proof-of-concept model sits at around 6.6 percent, although even that figure is in line with some LEDs and fluorescents, and is three times better than conventional incandescents.

A paper on the research was recently published in the journal Nature Nanotechnology.

Source: MIT

If the efficiency gains ultimately realised place this on a par with the better fluorescent and LED sources, this will be a good thing for photographers who prefer to use continuous lighting.
I'm not used to seeing luminous efficiency expressed as a percentage but rather in terms of lumens per watt.
I googled and the max luminous efficiency (100%) is 683 lm/W meaning 6.6% is 45 lumens/watt and 40% would be 273 lumens/watt.
Popular CFL's are > 60 lumens/watt and even inexpensive LED's are now over 80 lm/watt. The good LED's are passing 100 lumens/watt and Cree broke 300 lumens/watt with LED in the lab in March 2014 which puts it ahead of theoretical maximum of 40% efficiency listed above.
So basically even ignoring costs/complexity the 6.6 efficiency they were able to achieve in a lab environment is already still behind disposable priced CFL and LED's on the market today.
My Nanoleaf Bloom from last year was expensive but it's 2.4 times their lab tested efficiency. It's an improvement but incandescent bulbs are unlikely to make a comeback. In the time they improve LED technology isn't going to stand still either.
Great insight Daishi! The only bonus of this I can see is that it's probably AC powered vs DC in LEDs - meaning less complex/costly. However electricity microproduction is accelerating quickly and that is mostly DC.
I think this new type of lamp will not be able to beat the price of LED lighting. By the way, I wish people would stop referring to the harsh white light of LEDs. Yes this is one of the colours available which some people actually want. Otherwise you can have warm white.
There are literally hundreds of types of LED lamps. Some of the latest designs replicate vintage light bulbs, as the LED elements now have the appearance of a glowing tungsten filament. They can also be dimmed. The wattage is low, by comparison even with the ghastly fluorescent lights, and of course they are promoted as lasting for many thousands of hours. A 10 W LED bulb is roughly equivalent to a 60 W incandescent lamp. See the savings in electricity. My daughter's kitchen has 10 50 W halogen spotlights. I am trying to persuade her to fit LED replacements! I did in fact buy her some but the bulbs were a bit longer than the space she had in the ceiling!I am trying to find a suitable lamp for the job. By the way the replacement lamps would probably be 6 to 7 W
The prices are now very affordable. I do not work for the company, but have bought a number of different bulbs from banggood.com I have had a few failures, but the company has sent a free replacement when necessary. Another company is myled.com
Bob Stuart
The more efficient lights can look warm and familiar. Their problem is not the average hue, but the usual cheap mix of phosphors that produces the desired average, but not all the wavelengths, making colors look off and eyes feel tired. Health is affected by a spiky spectrum.
Just move on already. The days of incandescent lights are over. Yes, it 'could' reach 40%, but it's at 6% while LED's are on average over 10% (100 lumens per watt is about what you get in the shops now - my bike light from 3 years ago was already making 100 lumens per watt, lab tested).
Why compare what it could reach with what LEDs actually reach? Why not compare it to what LED's could reach? LEDs will reach 200 lumens per watt in less than a decade. At 100 lumens per watt it already is way ahead of incandescent lights.
@swaan People are generally just using LED's to retrofit standard lightbulbs but longer term it probably makes sense to use something like PoE instead that will supply DC voltage and allow simple connection of smart bulbs and sensors.
Homes of the future should be wired to have PoE wall outlets. Things like controllable bulbs, security cameras, temperature sensors, smoke/CO2 detectors, smart TVs, and even wireless access points become incredibly simple to connect from that point forward.
Hopefully this is something more businesses embrace to help lead the way for the consumer market.
I read about this technology in Science News magazine more than TEN YEARS AGO. I have wondered what was going on with it.
amazed W1
The luminous efficiency as usually defined is a bit too crude. Where maximum efficiency is needed the lumens per watt should be defined and tested over the range of wavelengths at which the (human) eye is most sensitive. Basically this means in the yellow part of the "visible" spectrum, though there is a justifiable argument as to how far either side of the peak sensitivity the band should be.
Why do this? Well humans have evolved to be active during daylight hours so our eyes are most efficient at the wavelengths of the solar spectrum with the highest energy per wavelength, at sea level and after the considerable absorbtion at longer and shorter wavelengths by the water vapour and comparatively small net total from CO2. (This is why sodium lighting is effective and No, I am not a climate change denier, though the significance of this is rarely treated honestly by the global warming gurus.)
There are some applications that Led lighting is not suited. Inside an oven, incandescent bulbs work well and solid state led devices to fail and melt, additionally arctic temperatures in Antarctica and Alaska require led devices with heating elements making them far less efficient, and possibly less desirable than this incandescent technology. Fridges regularly ship with led lighting, but most still use bulbs in the freezer. This technology can be used to increase tube efficiency in tubes used in rf applications and in audio amplifiers. Tubes sound more pleasant due to the harmonic nature of the distortion the produce, causing the to sound warm and smooth instead of harsh when they distort. They are also a current switched technology instead of a voltage switch technology. This causes a lot of people to prefer them and their sonic character. I have thought about mimicking this sonic character in DSP as a filter to adaptively smooth the distortion in a class a transistor amp with variable bias headroom maintained with predictive algorithms in DSP before the DAC.With the DAC output directly part of the amp. Darpa has a program developing new tube technology for various applications. With the development of metallic hydrogen and the possibility of room temp superconductors, the effect on generators, alternators, and electric motors is known as is the possibility of high-speed levitating trains and a superconducting power grid. What I think has not been widely talked about is how much this could potentially affect tubes including rf, incandescent lighting, led lighting, lasers weapons, maser weapons, rail guns, plasma weapons, and superconducting rf antennas that remain cool in the megawatt range. superconducting tube audio fully class A, in synthetic sapphire or diamond tube for durability. Not to mention single stage to orbit rockets that are reusable and 60% lighter and at least 40% cheaper to make and around 70% cheaper to orbit. The recent development of metallic hydrogen will revolutionize civilization as we know it. making spreading out into the star system and mining and colonizing the planets and moons and asteroids possible. It will also allow hydrogen fuel cells without the hassles of cryogenic fuel storage. This will allow it to be used in combustion engines, fuel cells, boats, cars, jet and fuel cell electric planes, and fuel cell electric military and research subs. Allowing the quietest military subs ever, with nuclear like performance. An interesting idea would be a device that is both a tube and transistor that is superconducting and can simultaneously switch and modulate current, voltage, frequency and amplitude all at the same time! cell towers, tv, and radio, as well as satellite broadcast; all use klystron tubes. This technology could make them all far more efficient, light it seems to me may have some limited uses where led lighting is not as desirable but tubes have never and will never go away. They have applications that are simply not met by transistors in performance or cost and in some instances at all in fact.