Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How are materials brought to (near) absolute zero?

  1. Apr 16, 2007 #1
    Hey, an article on the new Discover Mag talks about underground labs trying to detect dark matter. It mentions that the sensors have to be brought down to almost near absolute zero... I then realized that, even though I keep reading about things being put in extremely low temperatures, I have no clue how they do this.

    How do they get materials to such low temperatures in labs? ... Any cooling agent I can think of (liquid N, etc.) is nowhere near absolute zero. My best guess is they get Dick Chaney to blow on the material in question.
     
  2. jcsd
  3. Apr 17, 2007 #2
  4. Apr 17, 2007 #3
    wow, that's pretty clever.
     
  5. Apr 17, 2007 #4
    There are several methods.

    Liquid nitrogen gets you sort of cold.

    Liquid helium gets you down to 4.2K.

    You can use a dilution refrigerator to get to tens of milli-Kelvin. It works by immersing your sample in a mixture of liquid He3 and He4. You then pump really hard on the vacuum above the mixture of liquids. The helium4 will evaporate, and when the atoms leave, they take energy with them, thereby cooling the rest of the mixture down.

    I believe this is the best you can do without going with the laser cooling method already described.
     
  6. Apr 17, 2007 #5

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    That technique, btw, isn't suitable to be used to cool "materials" if we are talking about macroscopic solid. Atomic gas and molecules, sure. But bulk solid, no.

    The latest technique in cooling bulk solid was recently published. Read here for a review.

    Zz.
     
  7. Apr 17, 2007 #6

    neu

    User Avatar

    mutistage adiabatic demagnetisation refridgerators are a method which can potentialy operate completely mechanically.

    If you take a paramagnetic material and magnetise it (Apply a strong magnetic field to allign spin domains) then you reduce it's magnetic entropy. Of course a reduction in magnetic entropy is compensated by an increase in thermal entropy(energy) which you can remove by placing in contact with a heat bath of equal temperature (initial environment ~10-20K).

    If you then isolate the material and lower the magnetic field then the spin domains start to loose their alignment and become disorderd, so the magnetic entropy increases. As the material is in isolation the energy required to fuel this motion can only come from the materials thermal energy (you can also consider the magnetic field as splitting and separating the energy levels which redistribute the spins to a distribution corresponding to a lower temperature)

    The result therefore of demagnetising a magnetised paramagnetic material in isolation-adiabaticaly-is that the thermal energy lowers. current 2 stage ADR's can maintain a few mK environment for over 24 hours.

    a Multistage stage ADR operating from ~20K which is the limit of current mechanical cryocoolers, would provide an entirely mechanical process from room temperature to near 0K.

    This process is also proposed for use on the ESA's XEUS space mission in 2020 to cool cosmic X-ray detectors such that the resolution of single photons are possible, and is being developed by a British team at UCL. The adcantage is that it's small, does not require lots of liquid He which evaporates and is expensive, the disadvantage is that you need massive B-fields.
     
    Last edited: Apr 17, 2007
  8. Apr 17, 2007 #7

    neu

    User Avatar


    Or place it in thermal equilibrium with Maggie Thatcher's heart.
     
  9. Apr 17, 2007 #8
    hahaha


    Thanks guys, these methods are really interesting.
     
  10. Apr 17, 2007 #9

    Gokul43201

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Just for the sake of completeness, there's two important stages of cooling between the above stage (4.2K) and the dilution refrigerqator (~ few mK).

    #1. He-4 cryostat - you can typically go down to nearly 1K by simply pumping on the vapor above liquid He4 (the equilibrium temperature of the liquid is a function of the vapor pressure). It's very hard to get below about 1K this way (at least using standard, commercial pumps).

    #2. He-3 cryostat - additional cooling below 1K can be had by using He-3 as the refrigerant and pumping on its vapor. Because of the lower mass of the He-3 atoms, the equilibrium temperature for a given vapor pressure is lower than for He-4. So, by pumping on the He-3 vapor, you can get down to a little below 300mK.

    The next stage of cooling, in conventional helium cryostats is the dilution fridge described below.

    The important part of the cooling in the dilution fridge comes from the mixing chamber, where cooling is produced (not by evaporation, but) by the migration of He-3 across a phase boundary (between a He-3 rich phase and a He-4 rich phase).
     
  11. Apr 20, 2007 #10
    if you reach 0 kevlin,what would happen?have they reached it?
     
  12. Apr 20, 2007 #11

    neu

    User Avatar


    It is impossible to reach absolute zero by a finite number of steps. That is: an infinite number of adiabatic and isothermal steps are required to get to T=0.

    as to what would happen if someone did reach T=0?
    my guess is the universe would collapse into a nipple
     
    Last edited: Apr 20, 2007
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: How are materials brought to (near) absolute zero?
Loading...