|
|
Posted by D Sherman [72.47.9.228] on Friday, July 15, 2011 at 12:24:55 :
In Reply to: "Peltier effect " is not new. posted by ChrisCase [76.212.215.228] on Friday, July 15, 2011 at 11:10:58 :
It's hard to make out what's going on from that article but it's not the Peltier (thermocouple) effect. There's something about magnetism and a phase change. The "applications" they refer to all involved converting low-grade heat to electricity. If it really worked, the first place to apply it wouldn't be cars and home computers, but large thermal generating plants. Even in the most modern of boilers and turbines, most of the energy released when coal is burned goes up the smokestack and out the cooling tower. I seem to recall something like 35% efficiency is considered good. An old-fashioned single-cycle steam engine is about 10% efficient. The waste heat from a large power plant is at a higher temperature than that of a PC or a car, and more importantly, size and weight are no object when the thing just has to sit on the ground in one place.
You're right that the article hinted at there being no need to cool one side, but that is impossible. The basic laws of thermodynamics require that any "heat engine" (doesn't matter if it's solid state and has no pistons) operates strictly on a temperature difference between a source and a sink. Without a hot side and a cold side, there's no flow of heat, and it's only the flow of heat that can be converted to mechanical motion or a flow of electricity. No matter what the physics of this new material, any heat engine made from it will require a source of cold just as much as a source of heat. As the feeble performance of those wood stove powered Pelier-effect fans shows, keeping the cold side cold is sometimes harder than making the hot side hot.
I read this article as a typical university press release that over-hypes the potential applications of something that is really just an interesting scientific discovery. It may find a niche somewhere, possibly in test and measurement like the Peltier effect and the Hall effect, but the applications they suggest are purely imaginary. It's already possible to build a conventional boiling-liquid heat engine that works on any particular hot and cold temperatures that you have available simply by picking exotic working fluids and picking the right pressures at both ends. "Boilers" using mercury or liquid sodium have been used for higher temperatures than water is used at, usually as a "first stage" in a large thermal power plant where the waste heat from the mercury or sodium turbine is used to boil water for a conventional second stage. At the low end, boilers have been built using ammonia, butane, freons, and probably other low boiling point liquids. All of them are limited to never being able to achieve greater efficiency than a Carnot cycle would at the given source and sink temperatures. The Carnot cycle is the ideal thermodynamic cycle for a piston or turbine engine. The Otto (spark-ignition), Diesel (compression-ignition), and Rankine (steam) cycles are all practical attempts to get as close as possible to the impossible Carnot cycle. Even that "perfect" cycle can only extract power from a temperature difference, not an absolute temperature, though.