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

How can I quantify/qualify liquid helium boil off rates based on dewar/cryostat size?

  1. Apr 1, 2012 #1
    I've been looking at various papers which describe the mechanisms and provide calculations describing heat flow into helium. I'm trying to minimize helium loss and am in the market for a new dewar flask/cryostat. I plan on asking the supply companies directly what the proper dimensions are to reduce helium loss while still maintaining low temperatures for our needed time period (~10 min), but I would like an idea of the physics behind these dimensions.

    Currently, we have a glass dewar flask containing LHe sitting in a glass dewar flask containing LN2. Purchasing a metallic cryostat is an option, but I don't know much about them.

    Here is one of the papers I found:
    http://jap.aip.org/resource/1/japiau/v22/i12/p1463_s1#tabs_1_113_1274104113_tab3 [Broken] (Evaporation Rate of Liquid Helium. I). It provides good qualitative description, but the dewar flask I was looking at doesn't have a reduced-diameter neck as the one they describe does.

    Any input would be valuable!
    Last edited by a moderator: May 5, 2017
  2. jcsd
  3. Apr 1, 2012 #2
    Re: How can I quantify/qualify liquid helium boil off rates based on dewar/cryostat s

    These days most dewars and low-temperature cryostats tend to be all-metal designs. Stainless steel has a rather low thermal conductivity and can be made much thinner than glass without risk of breaking.

    There are several sources of heating: Radiation, conduction, and convection, especially if you have a wide-neck cryostat. Estimating these quantitatively is not so easy.

    10 min holding time at 4.2K should be trivial to achieve. If you want to go lower, you need to pump. Then there are several options of how to build the cryostat, essentially with a Joule-Thompson stage (continuous operation) )or a secondary, closed bath (single shot operation).
    JT-type cryostats you can get as flow-type without a bath. They are fed directly from the storage dewar and can be very efficient.

    Other things to think about:
    Will your sample be immersed in liquid, or in exchange gas or in vacuum?
    Do you want a single temperature, or do you need variable temperature, to be regulated with a heater. What is the base temperature you want?
    Do you want a cryostat that relies on (external) liquid He supply or a closed-circuit cryocooler that only needs electricity?
  4. Apr 1, 2012 #3


    User Avatar
    Science Advisor
    Gold Member

    Re: How can I quantify/qualify liquid helium boil off rates based on dewar/cryostat s

    A good transport dewar that can hold say 60-120 liters will have a boil-off rate of about 1-2 liters/day.

    However, in many experiments one needs a fairly wide neck-diameter in order to be able to fit a probe (50mm or even wider), in which case the boil-off rate goes up quite a bit (say 5 liters/day); and the boil-off rate will increase while the probe is inserted (and will of course go up even further if you are runnning a 1K pot).

    I wouldn't bother trying to quantify this using formulas etc. , there are so many different factors that come into play that I don't think you can ever get a realistic estimate.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook