How Low Can We Go?

It's unlikely we'll ever get all thermal motion to stop in an object. But we can get close enough in many experiments that "it’s basically absolute zero for all practical purposes."
  • Transcript

TRANSCRIPT

Question: Will it ever be possible to get a temperature down to absolute zero?

William Phillips: Well, that’s an interesting question.  And sadly, the answer isn’t simple.  The simple answer is, no.  But now I’ve got to explain why I’m saying that the answer is no.  And answer is that every process for cooling either also introduces the possibility that you can introduce some extra energy into the system.  You see, cooling means taking energy out, and heating means putting energy in.  But in order to take energy out, then it turns out that you open the door for energy to go in. 

Take laser cooling.  Laser cooling takes energy out by having an atom coming along and then a photon hits the atoms and slows the atom down, but then that photon has to go someplace.  And when that photon is shot out by the atom, the atom recoils and more energy goes in.  so, there’s a balance between the cooling and the heating and you can try to make that balance work more and more in your favor, but you can never make it work 100% cooling and no heating. 

So that’s one of the reasons why you don’t expect to ever get to absolute zero.  On the other hand, what does it mean to be at absolute zero?  It means that all of the thermal motion stops.  Well, I can take one atom and I can take as much energy out of it as possible so that it’s in what would call the ground state, the lowest possible state of energy.  Is it absolute zero?  Not really because in order to be at absolutely zero, I really have to have a whole bundle of things.  I can’t really talk easily about the temperature of a single object.  I should really talk about a whole ensemble.  And if I do that with a whole bunch of atoms, what’s going to happen is, maybe if I’m lucky, maybe 99 percent of them are going to be in the ground state and then one percent isn’t.  So, it’s not absolute zero. 

I can’t come up with any procedure that is going to say 100 percent of the time this atom’s going to end up in the ground state.  And that’s what I would need to be able to claim that I really had gotten down to absolute zero.  But on the other hand, I can get so close to absolute zero that for many experiments, it’s basically absolute zero for all practical purposes.  But not for all experiments and we are constantly working on making things colder because for some experiments, it really matters that were not quite there.

Recorded June 4, 2010
Interviewed by Jessica Liebman