What is negative temperature?

[The Skilled Thief]

Can someone please explain what is negative temperature?
(....5, 4, 3, 2, 1, 0, -1, -2, -3, -4, -5,....)
If 0 is absolute, meaning no heat at all then why is there negatives?
Obviously if there is heat it`s because particles are moving and are creating friction, hence creates heat. However if the temperature reaches zero, I suppose that means, no friction and no heat. If the temperature reaches negatives what is going on the particles? I know this may sound stupid and it`s probably an erroneous question, but can cold actually be colder? sweatdrop

[Darkphantom]

Zero in degrees kelvin is quantitatively "no energy" or absolute zero And no you can't have negative degrees kelvin.

however most temperature scales don't count up from no energy Centigrade for exsample is the freezing point of water in standard conditions. 0 degrees Kelvin in Centigrade is -273'C whitch is the coldest anything can be.

Farenheight I actually don't know what that system counts up from. but absolute zero on that scale is −459.67'F

[SmallTownGuy]

Something like negative temperature on the absolute (Kelvin) scale can happen in an atomic population inversion, when more electrons are in high energy states than in lower states. This happens inside a laser, where electrons are pumped up from a low energy state to a quasi-stable high energy state.

This weird "negative temperature" situation should not be described as "going below zero", though. It's more like "going above infinite temperature and down the other side". For all normal purposes, Darkphantom's answer above is the correct one.

[VorpalNeko]

Temperature is inversely proportional to the rate of change of entropy with respect to energy. That's the definition. Being proportional to average particle energies is only applicable in some special cases, and definitely not in general, because that property lacks the most important feature of temperature: that things in thermal equilibrium are at the same temperature.

So negative temperature happens whenever putting in greater energy makes the entropy go down.

Things are more complicated in practice, but here's an illustration of the basic idea. Suppose you have a physical system that's a giant collection of subsystems that only ever come in two states: low-energy and high-energy. The lowest possible energy for the entire system is that everything is in the low-energy state. There's only one way to do that, so that state has no entropy.

As you put in some heat, some of those subsystems are put in the high-energy state, but most are still in the low-energy state. But now there are many ways to distribute that the energy you put among the subsystems, so that state has somewhat higher entropy. And hence positive temperature. But something stranger happens if most of the subsystems are already in the high-energy state: then putting more energy decreases the number of ways of distributing the total energy among all the subsystems. Entropy decreases as more states are excited. That's negative temperature. If all of them are in the high-energy state, then once again, there's only one way to do that, so there is once again zero entropy.

[Suicidesoldier#1]

Why Kelvin was invented.

[Layra-chan]

Our temperature scale is really dumb. We have 0 on one end, positive temperature, infinite temperature, and then what's above infinite temperature? Negative temperature, starting at negative infinity and heading toward 0 from below.

[Vannak]

Temperature has a long and strange history. Before we had a really good idea of what temperature means, we knew it has something to do with energy. The Fahrenheit scale, like many measurements of the time, deal with things intrinsic to human life and the body, rather than more universal scales.

Fahrenheit sets 0 degrees as the coldest temperature avaliable at the time (frozen sea water, which freezes below pure water) and has 100 set to be the human body temperature (however, it has since been revised as 98.6)

Celsius moves in the direction of a more physics based rather than human biology based scale. 0 is the freezing temperature of water and 100 is boiling. However this is all at on-earth conditions at sea level. Higher elevations let water boil at lower temperatures because its' easier for water molecules to escape the liquid surface.

Finally Kelvin sets zero at the minimum temperature, but when we talk about kelvin we can't really talk about temperature the same way. When we talk about the minimum temperature we're not really talking about "energy" in the same way we do at every day temperatures. We talk about temperature as how energy and entropy interact. VorporalNeko has a good post on entropy.

But in general if you turn on the TV and see a negative temperature, it just means the air has less average kinetic energy than frozen salt water in the US, and less than frozen pure water every where else.

If you hear of negative temperatures in Kelvin, then you are most certainly not talking about weather or kinetic energy.

[Vannak]

One should note that heat, as you've talked about it, is refered to as, "The average kinetic energy" in a certain mass.

If you want to understand entropy in a qualatative way, here's my take.

Imagine a very low temperature gas. It's made up of many millions of atoms, lets say 10 for simplification. As a simplification, we'll say each atom has only a few energy states. They can be at rest (moving very slowly as to have practically no energy) or excited (moving very quickly) or doubly excited (moving even faster) and so on.

Now imagine the gas is initially very cold. All of the atoms aren't really doing anything, they're all at rest. Now lets say we give it enough energy to make one single atom go from rest to an excited state. While we can pick any of the 10 atoms to excite, there's only one state of the gas in total: 9 atoms at rest and 1 excited. One unit of energy can create one state. we take the ratio and get a low temperature.

Now we add this same amount of energy again. We now have two states: 8 at rest and 2 excited OR 9 at rest and 1 doubly excited. Now we have 2 possible ways for the gas to change, but with the same energy. This ratio is 2:1, and has a higher temperature.

Add the same energy again and we start to get a lot more states. We can have:
3 excited 7 at rest OR
1 doubly excited 1 excited and 8 at rest OR
1 triple excited and 9 at rest

We now have a 3:1 ratio and it keeps going up like that. Entropy is related to the number of ways a system can change. A system like we initially described has very low entropy because it can only change in one way. Hence a low entropy state. Our final state with 3 energy units added has a higher entropy. Imagine if you had 10,000 atoms and 10,000 units of energy. The numbers get pretty big pretty quickly.

once you get to a certain point, it becomes pointless to even try to count atoms and energy states with any precision. We use kinetic energy instead because at high temperatures, heat and entropy work in really predictable ways for nearly anything you see around you.

[The20]

Suicidesoldier#1

Why Kelvin was invented.

Why was Fahrenheit invented? Or all the other temperature scales? I'll tell you:
http://xkcd.com/927/

[Kiumaru]

The20

Suicidesoldier#1

Why Kelvin was invented.

Why was Fahrenheit invented? Or all the other temperature scales? I'll tell you:
http://xkcd.com/927/

Well, I would suggest that this isn't wholly accurate. There are various advantages in trying to place Celsius on an absolute 0 scale like Kelvin, or Fahrenheit on the Rankine scale. Of course, if someone was trying to make a scale which would encompass all of them, then the xkcd comic would apply.