Slashdot Mirror
Researchers Pave Way For Compressor-Free Refrigeration
Hugh Pickens brings news that scientists from Penn State have developed a new method for heat-transfer that may replace the common compressor-based system used in household appliances. Quoting:
"Zhang's approach uses the change from disorganized to organized that occurs in some polarpolymers when placed in an electric field. The natural state of these materials is disorganized with the various molecules randomly positioned. When electricity is applied, the molecules become highly ordered and the material gives off heat and becomes colder. When the electricity is turned off, the material reverts to its disordered state and absorbs heat. The researchers report a change in temperature for the material of about 22.6 degrees Fahrenheit... Repeated randomizing and ordering of the material combined with an appropriate heat exchanger could provide a wide range of heating and cooling temperatures."
Not at all. Microwave works by emitting electromagnatic waves that exite water molecules, thus making them and (indirectly) whatever they are a part of warmer.
It's still based on compression (and out of Penn State, licensed to Ben and Jerry's, of course), but it's a much *faster* compression, at the frequency of the sound waves used, and it takes advantage of air's intrinsic nonlinearity at high acoustic amplitudes, rather than the much slower effects inherent in traditional refrigeration techniques.
http://www.acs.psu.edu/thermoacoustics/refrigeration/benandjerrys.htm
...the second law of thermodynamics that states that the entropy of all isolated systems always increases.
There. Fix'd it for you.
When external energy is applied to the system (like, say, electricity), then the system isn't isolated.
TFA is written very poorly and describes a phenomena involving polymers that is already widely known. There are many examples. Here is one you can try using something far less exotic than the polymers mentioned in the article.
For this example, take a rubber band. Stretch it out. Touch the stretched rubber band to your lips. It will feel warm. Hold it in the stretched position for a few seconds to let it cool down to room temperature. Now let the rubber band relax, and once again touch it to your lips. You should now notice that it will feel cool.
The above process uses exactly the same principles described in TFA. Stretching the rubber band causes reduction of disorder by aligning the polymer chains. It also warms the rubber band because of the work applied. As you hold the rubber band in the stretched state it will cool to room temperature releasing some of the energy needed to heat it. This is equivalent to the step where the electrical field is applied.
Now release the rubber band. The polymer chains now revert back to a disordered state, cooling the rubber. Since the rubber band started in a stretched room temperature state the relaxed rubber band will now be below room temperature. this is equivalent to turning off the electric field as mentioned in the article.
Voila. This is a wonderful new refrigeration system that will replace all existing known cooling systems. NOT.
There are so many issues with practical application of this it is not funny. If these issues didn't exist we would have been using rubber band refrigerators for many decades already.
Also, please note that from a thermodynamics point of view this is essentially how a conventional refrigeration system works (albeit fat far more efficiently).
Going by the rough description in TFA, it sounds like electricity's effect on the ferropolymer causes its bonds to strengthen, or perhaps to magnetically align, increasing rigidity, reducing the material's potential for containing kinetic energy.
If the material's new state caps the amount of kinetic energy it can store, it has to move on - first law of thermodynamics and all.
This may be the next interesting bit in applying their discovery - finding a compatible heat conductor, and also learning the optimal frequency, voltage, current etc. at which to apply voltage.
O lord, bless this thy holy hand grenade, that with it thou mayest blow thine enemies to tiny bits, in thy mercy.
I did a lot of HVAC systems in the past, and many were large scale water source heat pump systems. This is, as expected, where the air handler cools the air in the space to, say, 55 F at the coil.
The fluids go to a condensing unit (compressor) which, instead of going to a coil with a fan directly to outside air, goes instead to a heat exchanger with water. The water runs throughout the building taking heat away from all of the water source heat pumps.
Typically, what I remember, the water will gain 10 degrees from the loop and dump 10 degrees at the cooler. The cooler will either be an evaporative cooling tower or a "fluid cooler" but it is basically always dumping 10 degrees of heat multiplied by however many gallons per minute of flow.
Yeah, the individual space gets a larger than 10 degree F temperature difference, and the SYSTEM gets "just" 10 degrees, but it's apples and oranges, different flow rates of fluids, CFM, BTU's, etc. Energy is energy, the temperature difference is only one part of the equation.
So, my point, and I do have one, is that ~20 degrees C or ~20 degrees F or whatever, it's enough if applied correctly.
Flappinbooger isn't my real name
I work in a group researching magnetocaloric refrigeration at room temperature. I read the Science paper, and this is about the same, except with electrical polarization instead of magnetic. It's promising in some ways, but have some potentially fatal problems.
1. 12 deg C is a really large temperature change, we have to do with 1-3C. My group would kill for a material like that, $EVIL_GENIUS_LAUGHTER. (With a design like this, it's possible to have a much greater cumulative change of temperature than what any single piece of material does, so that's how to cool from +25 to -18 C).
2. The hysteresis is not too high, look at fig. 1 in the paper. This is important, because hysteresis means you're converting electricity to heat inside your fridge. Many materials have great change in entropy and temperature when you put an electric or magnetic field on them, but it's killed for practical purposes by hysteresis.
3. You need a really high electric field. The curves in the paper are done at 100-200 MegaV/m, meaning that you need 100-200 kV to polarize a layer of 1 mm thickness. A CRT uses voltages of around 20 kV, and so it's plausible to use thin layers, or just live with the fact that you'll only get 1-2 C temperature change. (Which means it has to compete with magnetic refrigeration on an even footing).
4. It's hard to polarize and depolarize the material without electric losses. (This is a problem for ferroelectric cooling in general). You're basically charging and discharging a huge capacitor, and you'll lose the charge on the capacitor every round. This could be fixed by putting it as the "C" in an oscillating (LCR) circuit with some inductance, but it's not easy to get an inductance (L) high enough, unless you run at high frequency. This material looks to work at high frequency (the hysteresis curves are taken at 1kHz), but how do you transport the heat into/out of it? If you run at 1kHz, you'll have less than half a ms to transfer heat to the cooling fluid, which means you'll need to use a very thin layer indeed. (Incidentally this will make it easier to get a strong field gradient). Then there's the problem of moving the cooling fluid back and forth over many layers of sub-mm thickness polymer. I'm not saying it can't be done, and there might very well be smart solutions I haven't thought of, but it's not trivial. (And btw, magnetic cooling doesn't have this problem, because we can use a permanent magnet with a several cm gap, and balance material moving into the gap with material moving out.)
Any sufficiently advanced libertarian utopia is indistinguishable from government.