Microrefrigeration... could this be used in flashlights?

cy · Dec 14, 2006

Microrefrigeration... could this be used in flashlights?

One set of candidates are known as paraelectric materials. These act like batteries when they undergo a temperature change: attach electrodes to them and they generate a current. This effect is used in infra-red cameras. An array of tiny pieces of paraelectric material can sense the heat radiated by, for example, a person, and the pattern of the array's electrical outputs can then be used to construct an image. But until recently no one had bothered much with the inverse of this process. That inverse exists, however. Apply an appropriate current to a paraelectric material and it will cool down.

http://www.economist.com/science/displaystory.cfm?story_id=8407182

TorchMan · Dec 14, 2006

I wonder if anyone ever thought of that application? Good thinking!

First reaction is that it might benefit LEDs most of all. As long as it doesn't cool down batteries too much, might be an application there. Maybe the days of a hot incan are numbered done? Especially a hotwire?

Mike Painter · Dec 14, 2006

The device might cool down but the heat from that device has to go someplace and as with other such devices the net will be more heat out than cooling in.
Mass, surface area, heat pipes and "liquid cooling" by holding it firmly in our hot little hands should solve most problems with LED lights except for the group that will in teh near future say "You think your hot wire is bright and throws?"

elgarak · Dec 14, 2006

Something not right here. Paraelectricity is based on electric dipoles within a crystalline material, not on movable electrons. Therefore, they can generate an electric field, i.e. VOLTAGE, but no current.

EDIT: Usual stupid journalism speak. Here's an abstract from Dr. Alex Mischenko himself: "By applying and removing an electric field to a thin film, Alex Mischenko has generated a giant electrocaloric effect that is able to cool the film by 12 degrees centigrade."

See. No current. Huh. This sounds interesting. Though I doubt that it would help much with flashlights; nice big metallic heatsinks and large surfaces seem much easier and still way more effective.

cy · Dec 19, 2006

if I understand correctly.... difference is heat is not getting transfered like conventional cooling. refrigeration theory states one is always removing heat, not creating cold.

Mike Painter said:
The device might cool down but the heat from that device has to go someplace and as with other such devices the net will be more heat out than cooling in.
Mass, surface area, heat pipes and "liquid cooling" by holding it firmly in our hot little hands should solve most problems with LED lights except for the group that will in teh near future say "You think your hot wire is bright and throws?"

jtr1962 · Dec 20, 2006

cy said:
is this case, cold is actually being generated much like an electric heater in reverse. (not an accurate comparison)

That's thermodynamically impossible. When something gets colder you're removing heat from it. The heat has to go somewhere. It can be pumped up to a higher temperature via a heat pump (either compressor-based or solid state). It can also be absorbed via an endothermic chemical reaction. In the latter case the process cannot be continuous unless you continually add the chemical needed for the endothermic reaction and remove the byproducts. For a continuous, closed-loop system such as would be used to cool a microprocessor you need some form of heat pump. There is just no such thing as a "heater in reverse". Although the article's description leaves much to be desired, I interpret the paraelectric effect as simply a thermoelectric effect identical to that which occurs in common bismuth telluride Peltier coolers. Yes, these devices can also generate electricity if one side is kept hotter than the other. They are woefully inefficient at either cooling or generating power. If these scientists have discovered a better thermoelectric material then it is indeed news, but the thermoelectric effect itself has been known about for over 200 years. It's only in the last 50 or so that we've had materials able to exploit it.

cy · Dec 20, 2006

that's exactly how I've always understood things to work.

but reading this articles says differently. maybe I'm interpreting it wrong?

jtr1962 said:
That's thermodynamically impossible. When something gets colder you're removing heat from it. The heat has to go somewhere. It can be pumped up to a higher temperature via a heat pump (either compressor-based or solid state). It can also be absorbed via an endothermic chemical reaction. In the latter case the process cannot be continuous unless you continually add the chemical needed for the endothermic reaction and remove the byproducts. For a continuous, closed-loop system such as would be used to cool a microprocessor you need some form of heat pump. There is just no such thing as a "heater in reverse". Although the article's description leaves much to be desired, I interpret the paraelectric effect as simply a thermoelectric effect identical to that which occurs in common bismuth telluride Peltier coolers. Yes, these devices can also generate electricity if one side is kept hotter than the other. They are woefully inefficient at either cooling or generating power. If these scientists have discovered a better thermoelectric material then it is indeed news, but the thermoelectric effect itself has been known about for over 200 years. It's only in the last 50 or so that we've had materials able to exploit it.

greg_in_canada · Dec 20, 2006

Coolchips (http://www.coolchips.com/technology/index.shtml) seems to be working on pretty interesting cooling technology. (The PDF on that page has the most info.)

They don't seem to have progressed very far towards production in the couple of years I've been watching their web page.

Greg

jtr1962 · Dec 20, 2006

cy said:
that's exactly how I've always understood things to work.

but reading this articles says differently. maybe I'm interpreting it wrong?

It's not you-it's the article. The writer obviously isn't very scientifically literate so the article is full of errors and misused terminology. And so as not to single out this poor guy I've noticed that most so-called "science and technology" reporters really don't have a clue what they're talking about.

This paragraph in the article basically describes the way paraelectric materials function:

"One set of candidates are known as paraelectric materials. These act like batteries when they undergo a temperature change: attach electrodes to them and they generate a current. ..... But until recently no one had bothered much with the inverse of this process. That inverse exists, however. Apply an appropriate current to a paraelectric material and it will cool down."

In essence, this is an exact description of the way commercial thermoelectric or peltier devices work. The only way this "discovery" is worth anything is if the paraelectric materials are more efficient as refrigerators than present day bismuth telluride and if they can be mass produced inexpensively.

Regardless of whether it's paraelectric materials or something else, I feel solid-state refrigeration/power generation is about to undergo a Moores law efficiency improvement within the next decade or two much as LEDs are now doing.

Microrefrigeration... could this be used in flashlights?

cy

Flashaholic

TorchMan

Enlightened

Mike Painter

Flashlight Enthusiast

elgarak

Flashlight Enthusiast

cy

Flashaholic

jtr1962

Flashaholic

cy

Flashaholic

greg_in_canada

Flashlight Enthusiast

jtr1962

Flashaholic

Similar threads

Latest posts