Chris Phoenix is providing live blog coverage for us on all the presentations from an important conference on Productive Nanosystems: Launching the Technology Roadmap...
Next talk: Molecular Design of Solid State Lighting for Energy Efficiency
Paul E. Burrows, Laboratory Fellow, Pacific Northwest National Laboratory (and over 100 patents)
Artificial lighting was perhaps the most important/irreplaceable result of our first invention, fire.
Candle: 0.05 lumens/watt
Gaslamp: 0.5 lumens/W
Incandescent lightbulb: 15 lm/W (5% efficient)
Artificial light uses vast amounts of electricity. Conventional light, even fluorescent, can never be more than ~20% efficient. 22% of electricity in US, 8% of total energy consumption, $50B per year, 150 MTon of CO2 per year. Most of this is 19th century technology!
Typical lightbulb has a 2800K blackbody spectrum; 95% of the energy is in the infrared. Most is over a micron; the eye responds to ~400-750 nanometers.
Electrons in a semiconductor can only occupy certain levels; thus, you don't get a blackbody spectrum. The photon energy is defined by the bandgap of the semiconductor.
Commercial LEDs can be expected to reach 50% efficiency and possibly more.
Hot off the press: 1,050 lumens in cool white @ 72 lm/W; 4 amps in a mm^2 die, at 150 degrees C. Very impressive!
"Nanotech isn't a length scale, it's a state of mind; how you think about making materials." Design a molecule for a particular function.
Molecular structure determines color; even in films, the molecules interact weakly, so the photophysics is determined by the bonding inside the molecule. Some molecules can hit 130 lm/W. "All you have to do is convince customers that they want green light bulbs in your living room."
Using phosphorescence, not fluorescence, which means you want high triplet exciton energies. This is because of electron injection--he didn't have time to explain this.
So there are some small molecules (three rings) that have ~3 eV energy, but they're too small to have nice material properties. But you can put phosphine oxide to provide a point of saturation, no conjugation past that point, so it isolates the optical part. And it makes the outer wings of the molecule negative - very high band gap (which means you can inject electrons at the right energy, if I understand correctly).
So you can make films of this stuff, thin-film structures just a few hundred nm thick. Konica Minolta has achieved 60 lm/W. With further development, it may achieve 200 lm/W. And it's diffuse light (because you can't put a lot of power in a small area of organic molecules, you'll fry it) (on the other hand, you can easily print the thin film by the square meter) so you don't need lampshades so you save efficiency there too.
Green fluorescent protein is 80% efficient, but the artificial version of the fluorescent part is 1% efficient and has a broader spectrum of light. Why? It flops around; in natural GFP, it's held in the right position by the rest of the protein. Could this be done artificially? Let's hope...
Fluorescent efficiency can be enhanced by a nanoparticle that creates plasmons to couple the energy out of the molecule. But the spacing must be exactly right. Can this be done by engineered molecule? Let's hope...
Electron transfer rate between organic molecules: very small changes in spacing affect the electroincs of the molecule.
Can we design optimized device components that make an efficient light? Perhaps Schafmeister molecules could be a way to make the right spacing.
Great quote: Report from the Select Committee on Lighting by Electricity", London, House of Commons, 1879: Electric lighting has inherent problems and can never replace gas.
We're not hitting the bleaching lifetime of the molecules; moisture kills it; that's a packaging problem.
Clever idea: using inorganic LEDs, you could flicker them fast enough to transmit lots of data (too fast to see) so you piggyback networking (at least half-duplex) on your lights.
Chris Phoenix
Tags: nanotechnology nanotech nano science technology ethics weblog blog
Comments