Quantum Gas Goes Below Absolute Zero

First time accepted submitter mromanuk writes in with a story about scientists at Ludwig Maximilian University of Munich who have created an atomic gas that goes below absolute zero. "It may sound less likely than hell freezing over, but physicists have created an atomic gas with a sub-absolute-zero temperature for the first time. Their technique opens the door to generating negative-Kelvin materials and new quantum devices, and it could even help to solve a cosmological mystery."

Comments

better explanation

[ssam]

wikipedia has quite a good explanation of negative temperature.
http://en.wikipedia.org/wiki/Negative_temperature

Re:better explanation

[Anonymous Coward]

Temperature isn't defined in physics as anything to do with heat, but the derivative of energy with respect to entropy. Absolute zero is the temperature at which there is no energy left in the system. At normal temperatures, it is positive. At absolute zero, it's zero. If you can create a system with dU/dS as negative, it's technically negative temperature, even though the system still has energy.
It's hard to explain due to how things like temperature and energy are defined, not because physicists are being smug. There isn't a proper name for it because it doesn't happen very much. That's why it's news.

Re:better explanation

[locofungus]

It's a quirk of the way the temperature scale was defined.

One possible definition of temperature:

Put lots of little magnets in a magnetic field. They will line up with the field. At absolute zero there will be no (technically minimal[1]) deviation from them all being perfectly aligned. As you warm them up they will start to be less and less well aligned until at what we call infinite temperature, there is no alignment with the field at all and the alignment is completely random.

But, if instead of warming them up, you flip the magnetic field they will then "cool" through "infinite" temperature.

If we use this definition of temperature then it would make more sense to have absolute zero as negative infinite temperature, infinite as zero and still hotter temperatures as greater than zero.

This makes the unreachability of absolute zero make more sense. "Infinite" temperatures (and greater than infinite) are only unreachable via trying to add more heat.

Lasers utilize population inversion - which is a state that is impossible via naive thermodynamics and also does not have a sensible temperature as a result.

[1] Zero point energy.

Tim.

Re:better explanation

[cgaertner]

It seems this is a very specific quantum mechanical perversion, and no classical systems can reach the state quantum physicists call "negative temperature".

This is by no means a quantum perversion, just a natural consequence of the definition of temperature as 1/T = dS/dE. There's nothing mysterious about negative temperatures from a thermodynamical point of view, it just happens that calssical systems don't exhibit this property because they do not come with an upper limit on energy, whereas there are quantum ones that do.

The common interpretation of temperature as average energy per degree of freedom comes in via the equipartition theorem, but breaks down in various edge cases, eg when the energy levels cannot be approximated by continuity (eg heat capacity of diatomic gases) or for non-ergodic systems (some plasmas, I believe).

As to the problem of infinite temperature: In a sense, thermodynamic \beta = 1/kT is the more natural measure of hotness and coldness and has a pole at T = 0. Coming from T > 0, this corresponds to infinite coldness, whereas coming from T < 0, this corresponds to infinite hotness.

Re:better explanation

[ArsenneLupin]

Temperature isn't a property of individual atoms. It is a property of a huge set of atoms, describing their energy distribution.

In a body with positive temperature, there are more atoms in the low energy states (moving slowly) than in the high energy states.

In a body with negative temperature, there are more atoms in the high energy states than in the low energy states. Normally that can only happen if there is an upper bound to the energy. Kinda like a speed limit, or, more realistically, if the states being talked about are not related to speed at all, but to some other physical property of the atom (such as orientation of spin within an external magnetic field...)