Better living through condensed matter physics: Thermoelectric edition!
A new thermoelectric (TE) material developed at Ohio State made modest headlines this week. It doubles the efficiency of TE devices that are best applied to obtaining electrical energy from otherwise wasted thermal energy in, say, internal combustion engines. It's kind of a big breakthrough in a field that is decades old. "Let me 'splain. No, there is too much. Let me sum up."

The TE effect itself is nothing new. In purely phenomenological terms, it is the relation between a temperature gradient across a material and a corresponding voltage drop. It can work either way: apply a voltage to get a temperature difference, or apply a thermal gradient to get a voltage. (Note we're talking about a thermal gradient. The normal heating you associate with current going through a wire is energy loss from scattered electrons, and it is essentially uniformly distributed.) Simple enough, right? How it works is only slightly complicated.
Imagine some container with water in it. If you make one side of the container hot and the other cold, your a priori thermodynamic intuition (which will turn out to be correct) is that water molecules on the hot side will heat up and start flowing to the cold side, and the ones on the cold side will flow to the hot side to fill in the gap left behind. This is oversimplified, of course, because really these suckers are bouncing off each other like crazy, but the basic idea is right: convection.
A metal is pretty much analogous, except instead of water molecules, you have conduction electrons, which behave like a fluid (albeit a fluid that carries electrical charge). So we put two ends of a metal at different temperatures and the electrons will convect. Great. But notice that there's no electric current - not yet at least. Remember the back-filling: there are just as many hot electrons going one way as cold electrons going the other.
So what's a condensed matter physicist to do? One trick is to employ a diode (see p-n junction), which is an everyday electrical circuit component that only allows electrons to flow in one direction. (OK, technically you want to split the diode in half and sandwich it with some metal, but the idea's still the same.) Just like that, you have a one-way valve for hot electrons. Want to pump heat across the device? Connect the TE device to a battery. Want to convert thermal energy to electricity? Replace that battery with your waffle maker, and heat one end of the TE while cooling the other.

So now that you have a sense of how this stuff works, what happened at Ohio State to bring about this breakthrough? Well, that's a long and technical explanation (band structure, density of states, Fermi energy, etc.) way beyond the scope of this already too-long post! Let it suffice to say that quantum mechanics works.
TE devices have found a lot of uses in the last couple decades, particularly for small cooling applications. They're great - cheap, low power consumption, no acoustical noise or vibrations, extremely long operating lifetime, and no coolants required (water, freon, etc.). High-sensitivity CCD cameras in astronomy and other fields use TE cooling to reduce camera noise due to thermal "dark" current (not at all related to dark matter/energy, BTW). Some high-power lasers use TE's for cooling as well. DIY computer builders, gamers, and home theater enthusiasts are beginning to catch on. I've even spotted TE camping/tailgate coolers that plug into your car's cigarette lighter.
Hopefully the future will see TE's in our homes in the form of refrigerators, A/C and heating units, and computer coolers. Unfortunately, it's my opinion that the success of the technology will rely heavily upon our willingness to restructure our electrical infrastructure. These are low-voltage, DC devices, while your home is mostly wired with 120 V at 60 Hz. Sure, you can step down the voltage and rectify it, but so far that's surprisingly inefficient - just put your hand on your AC adapter next time you use your laptop. That baby's putting out some heat, and as a corollary, wasting a lot of energy before any of it ever even gets to your processor. This is one of the reasons why a related technology, LED lighting (also in the press this week), is not in widespread use in buildings but is instead mostly limited to applications that use batteries, like the lights on your bike.
That's all nice Back to the Future II-type stuff, but what should really excite us, and what makes the Ohio State discovery an important milestone, is the potential to use the TE effect to help offset the inefficiencies in other devices. Anything - and I do mean anything - we ever build is inherently inefficient (see Laws of Thermodynamics, 2nd). And almost always, a significant portion of the wasted energy is lost in the form of heat. TE devices give us a way to easily convert some of this energy back into electricity. Considering that even the best combustion engines can't break about 20% efficiency (and even some small electric motors have room for improvement), the potential energy (and carbon) savings are enormous. Moreover, consider that a standard hybrid vehicle, in order to charge its batteries, uses a generator to actually steal energy that would otherwise go to the wheels. (Caveat: regenerative braking doesn't have this drawback.) So with the Prius, for example, you're actually making the combustion engine more inefficient in the short term in order to save gas over the long haul. TE devices, by contrast, aren't so parasitic - they get their energy from throw-away heat that would have been lost anyway, and, on top of that, they help cool the engine block, thereby allowing the engine run even more efficiently!








All my life I've intuitively understood that humanity is a disease, but now there's definitive proof of our ignoble origins! Geneticists recently announced that 8% of the human genetic code originated from













Recent Comments