Paging Maxwell! Mawell to the white courtesy phone!
We've found your demon.
NIST scientists have demonstrated a solid state refrigerator that sucks energy out of objects using a trick of quantum physics in which hot electrons tunnel through a one-way junction, carrying heat with them. It's been a long project: NIST first demonstrated the use of NIS (normal metal / insulator / superconductor) junctions …
I mean seriously...
The mind boggles.
There is so much clever shit going on, and there is such profound .............
The selective filtering of electrons... and pulling them and "the heat" with them.... out of a junction....
Accurately determining the size and surface spheroidal accuracy of electrons......
The Boson Higgs..... the HUGE nutrino detectors...
The Hobbit Telesccope
And on and on and and on it goes....
Am I right in assuming that this assembly is supposed to operate in a bath of liquid Helium, or similar arrangement? The figure of 700pW of (local?) cooling power is amazing when you consider the amount of heat that must flow into the copper block, even if it were immersed in very cold liquid Helium. Have I got this right?
It is likely to be operating in an more complicated/ even cooler system than liquid helium. The article says it cooled the copper block from 290 mK, which means they probably used a dilution refrigerator to get there, since liquid helium boils at around 4 K (-269 C).
http://en.wikipedia.org/wiki/Dilution_refrigerator.
I would have though that this would be in a very high vacuum, which is perhaps cooled by a liquid helium device: I know the one that I've played with in the past gets quite easily to ~2.7 K with no trouble, and can be lower if you're both lucky and careful.
As it turns out, reading the paper itself, it is an Adiabatic Demagnetisation Refrigerator (ADR) that gets it from ~300K to ~300 mK. This is actually quite a cool method (excuse the pun), especially for one who has a bit of a background with cryogenic stuff. I won't bore you with the details, but NASA have a basic primer here:
http://cryo.gsfc.nasa.gov/introduction/ADR_intro/ADR_intro.html
and something a bit more technical here:
http://cryo.gsfc.nasa.gov/ADR/ADR_primer/ADR_primer.html
which, though a bit space orientated, show the general idea nicely.
The more interesting part I found was, that to disconnect the ADR once it got to 300 mK, they used what is essentially a piece of brass on a rope. I'd have though it would be far more complicated than that.
And to answer the question in the sub-head, yes, the device probably could cool your beer. It might cheaper and more convenient, however, to buy yourself a beer fridge.
Aww, come on. Where's the fun in that? The very genesis of hacking is to see what you can get away with for no other reason than that you (possibly) can (well, OK, there is actually a very good argument for it).
Maybe not "instant", but if you spin a beer container longitudinally in ice without agitating it, it'll cool down from 75F to drinking temperature in a hurry (a couple minutes, in glass bottles). Faster if you salt the ice (under a minute in alumin(i)um cans). Note that you want the water to drain, actual ice contact is important.
Or, if you're not in a hurry & enjoy sillyness: http://www.asciimation.co.nz/beer/
Come now folks, as a physicsy person I can state without fear of contradiction that it's obvious this thing will not cool your beer!
Well technically it could but I'd challenge anyone to detect a 40mK temperature difference between fractions of average pints in a pub ... for tis all the machine achieved ...
In fact the differential between bottom of the pint and the top is probably much more than a degree or so ... until you put your clammy mitts around it at which point it's warming massively and the convection currents increase.
the hottest electrons tunnel across the insulator into the superconductor. As they do make the journey, the electrons carry heat and vibrational energy with them.
So where's the quantum physics in that, you might ask.»
Naive as I am, I have always regarded «tunneling» itself as a quantum phenomenon. Are Richard and the Reg claiming that it can now be explained by classical, pre-20th century physics ?...
Henri