PORTLAND, Ore.—Today's state-of-the-art thermoelectrics are only about 5 percent efficient, but new research indicates that a class of material called skutterudites—plus a new technque for aligning their atoms—could improve thermoelectric efficiencies to as much as 20-percent, enough for commercialization.
Such high-efficiency thermoelectric converters on the exhaust pipe of an automobile, for instance, could convert enough heat into electricity to charge the batteries of a hybrid vehicle.
Sutterudites conduct electricity well, but conduct heat poorly. However, University of Michigan professor Ctirad Uher recently discovered that certain configurations of a barium alloy in the compound could drastically increase the materials' efficiency. The technique effectively lowers the thermal conductivity of skutterudites, thus drastically increasing their conversion efficiency. Uher performed the work with fellow professor Massoud Kaviany.
The researchers claim that automobile manufacturers could use their new material to harvest the heat from the exhaust pipe of an automobile to generate electricity. "That's a big source of heat that you paid for already," said Uher.
The research was funded by the U.S. Department of Energy's Office of Basic Energy Sciences and the University of Michigan's Center for Solar and Thermal Energy Conversion.
The conversion of energy between light and electricity (LEDs/solar cells) and between light and heat (light bulb/greenhouse) is complemented by thermoelectric conversion between electricity and heat.
I agree with @new2coding: from thermodynamics point of view re-charging your batteries using waste heat doesn't make any sense, the math just does not work...perhaps small fraction but I highly doubt whether this gadget will ever pay off in a car, show me the money or ROI...Kris
Interesting! Is there any comparison on the efficiency of those thermal electric generators? I wonder what the efficiency finally means. What we care is how much cost and weight is added in order to get the wanted power. Just for the peltier type device, we can always use matrix of devices to generate as much energy as we want if cost is not a matter.
The act of converting excess heat into electricity consumes that excess heat, so there is a net cooling effect. Putting electricity INTO a TEC just gives you control of the magnitude of the heat differential between sides--heatsinking the hot side in that case gives you a way to more effectively pull heat from the cool side (assuming you can dissipate enough heat on the hot side), but then you use electricity instead of producing it.
To @p_g: it has been shown that solar cooker is more efficient than photovoltaics...the problem is you need industrial scale installation, so you can't do it at home...I believe Spain has the best installations...Kris
Another idea could be to harvest solar heat by consolidating it using solar cooker (glass changer with hi CO2) and converting to electricity. Not sure if it will be more efficient than solar energy conversion or may complement it.
Take a look at Coolchips/Powerchips. These are nanometer gap Peltier-type devices which have been waiting in the wings for years but have a claimed Carnot efficiency of at least 45%, better than this new material, if they can get them to work!
In fact equally as Solar Energy this source of energy is completely left aside and no one has looked much into it. When we drive car excessive amount of heat gets generated and needs to be radiated out using radiator. Same if the case with Refrigerators and Air-conditioners. If we can convert this heat in some other form of energy too much amount of energy can be saved/earned. When food is being cooked out of 270 deg for heat spread is wasted.
I am just wondering whether this technology can be applied to refrigerators. In that case the heat exchanger design will be radically changed and the heat taken out from the fridge compartment could be recycled back to run the fridge itself or may be used to charge the mobile batteries or run some other appliances in the kitchen
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.