[thedoc (OP)]

Stefan Johansson  asked the Naked Scientists:
   I understand that it is a complex endeavor to reach so low temperatures, but my question is:
 
   - How is the temperature measured? I imagine that some sort of thermo couple measurement is not possible at such low temperatures, but maybe it is a indirect method being able in some way to the measure particle velocities in the material.

Measuring extremely low temperaturesConsidering the following facts:

1) According to the podcast 365 days of astronomy, http://cosmoquest.org/x/365daysofastronomy/2016/02/14/feb-14th-baby-pictures-of-a-solar-system/:
"...Where is the coldest place we know of in the Universe? Well it's right here on Earth! The coldest temperature ever recorded was in a laboratory on Earth where the temperature dipped to a frigid -273  °C (less than 1 ° C above absolute zero!). That's colder than the empty space between galaxies!"

2) NASA JPL's Cold Atom Laboratory, http://coldatomlab.jpl.nasa.gov/mission/:
"... CAL will be a facility for the study of ultra-cold quantum gases in the microgravity environment of the International Space Station (ISS). It will enable research in a temperature regime and force free environment that is inaccessible to terrestrial laboratories. In the microgravity environment, up to 20 second long interaction times and as low as 1 picokelvin temperatures are achievable, unlocking the potential to observe new quantum phenomena. The CAL facility is designed for use by multiple scientific investigators and to be upgradable/maintainable on orbit. CAL will also be a pathfinder experiment for future quantum sensors based on laser cooled atoms."


Please help me with this answer, because readers to my blog, Medieborgaren.se, are asking me and I do not want to disappoint them.

Thanks in advance,

Stefan



 
What do you think?

[SeanB]

A few methods are commonly used. Thermocouples are very common, but need compensation as they are slightly non linear in measurement. The most common method is to use a Pt100 or Pt1000 sensor, as this is very linear and has a well defined temperature coefficient. Another method is to use a semiconductor diode, as this also has a very well known and easy to measure temperature coefficient, or to use a bandgap multiplier where you get a voltage proportional to absolute temperature out of it.

Of these the thermocouple is the most robust, as the others need to be cooled slowly so as not to stress them.

As well there are well defined states of superconductivity that are easy to measure at very low absolute temperatures, though this generally only is a way to know if the temperature is below a certain point.

[evan_au]

Reaching extremely cold temperatures (millikelvins and below) often uses some form of laser cooling.

In this case, the behavior of the atoms in the laser beam tells you about the temperature - are they absorbing energy from the laser beam or not?

[chris]

Reaching extremely cold temperatures (millikelvins and below) often uses some form of laser cooling.

Evan - using your unique skills in clarity and brevity, can you please explain what that impenetrable page on Wikipedia is actually trying to say about laser cooling please? Maybe I was being impatient, but I didn't find it at all accessible.

[agyejy]

Evan - using your unique skills in clarity and brevity, can you please explain what that impenetrable page on Wikipedia is actually trying to say about laser cooling please? Maybe I was being impatient, but I didn't find it at all accessible.

I can give it a try if you don't mind. So basically as we should all know photons carry linear momentum as well as angular momentum. The angular momentum is conserved in the absorption process through the selection rules. The linear momentum is conserved in the absorption process through the atom recoiling in a direction opposite the direction of the incoming photon. When the atom then decays to the ground state and emits a photon spontaneously that emission happens in a nearly completely random direction. So if you can figure out a way to make sure only atoms that are traveling toward your photon source can absorb photons they will lose a bit of momentum in a specific direction through the absorption recoil and will gain the same amount of momentum in a random direction through the emission recoil. The differences in direction are important and over many interactions the overall momentum of the atom decreases (i.e. the emission recoil doesn't always cancel the absorption recoil because of the differences in direction) especially in the direction toward the light source.

So if you use photons with an energy slightly less than an absorption transition the doppler effect means that only atoms moving toward the source can absorb a photon (because the atom moving at the source sees the photons has having slightly higher energy) and thus you get the situation described above. That is to say only atoms moving in a specific direction absorb the photons. Now in general you want atoms to lose momentum in more than one direction so you fire a couple of lasers from many directions.

[chris]

Dear agyejy - very nice; thank you for doing an excellent job of explaining it to me so clearly.