Ultracold temperatures
How should we think about temperature?
Temperature is a measure of the average energy in a system. For gases of atoms, we think of temperature as the average speed of the particles.
What kinds of temperatures are we talking about? The scale shown here is a good way of thinking about temperature. Unlike a conventional thermometer, this shows temperature on a logarithmic scale: each tick represents a 1000x change in temperature. The thing about a scale like this is that you can never reach zero, just like we can never get to Absolute Zero -- we may only get to smaller and smaller fractions
Typical temperatures
The kinds of temperatures that are familiar to us fall on a fairly narrow range in a logarithmic temperature scale. From the surface of the sun, at 6000 K, to the temperature of outer space, as measured by the cosmic microwave background at 3 K, occupy only a factor of 2000 in temperature change.
To reach temperatures much colder than this, artificial means must be employed: pumped helium systems at a few K, dilution refrigeration reaches a few mK, and laser cooling reaches the 10s of µK scale and Bose-Einstein condensates of rubidium atoms are on the order of a few 10s of nK: 10 billionths of a degree above Absolute Zero.
Typical speeds
If temperature is just a measure of average speed, let's consider what kinds of atomic speeds these temperatures correspond to. For particles of a mass of the rubidium atoms we use in our experiments, the speeds range from about 300 m/s at room temperature, to a few mm/s at the temperatures of ultracold gases.
While thinking of extreme temperatures can stretch the imagination, it is less difficult to imagine a particle moving at 1 cm per second than it is to think of a 200 nK object. At this kind of speed, we experimentalists realize that we can have a chance to control our systems, and that's just what we do: slow our atoms down.