So, I've been looking over data on permeabilities and there's something that bothers me. Permeability data on the noble gases is somewhat hard to find, but the Parker O-Ring Handbook has them for a number of materials:


(as an example material to choose in the list, say, PTFE)

What I don't get is... why is the permeability rate for krypton usually higher than that of argon for most materials? ? Their chemical properties are basically identical, krypton is just a larger atom. By all standards one would expect it to permeate slower. Xenon permeates slower than both of them, as expected. So what's up with krypton?

  • 1
    $\begingroup$ The tables are a little hard to compare - the temperatures aren't always the same between, say, the argon entries and the krypton entries for the same elastomer. However, for those that are directly comparable it does appear that Kr can be a little faster. I think, sadly, that one might have to see if there is a paper from one group directly comparing measurements on the same system - I could see there being large differences for measurements done in different ways on different systems. Oh, and Kr should of course be stronger since it comes from the planet Krypton just like Superman... $\endgroup$
    – Jon Custer
    Jul 14, 2016 at 16:11
  • 2
    $\begingroup$ Could possibly be a solubility thing and not just about straight diffusion. $\endgroup$
    – matt_black
    Jul 14, 2016 at 19:19
  • $\begingroup$ Related: chemistry.stackexchange.com/questions/54826/… $\endgroup$
    – aventurin
    Jul 14, 2016 at 21:28
  • $\begingroup$ @Jon: But as you note, in the cases that are directly comparable... yeah, krypton seems faster than argon, and significantly faster than xenon... the latter is expected, while the former is not. And it's not like there's just one entry... there's a lot of them, for a variety of materials. And given that the measured temperatures, etc are identical and not some overly generic thing (for example, "149C" for PTFE), it sure looks like they're all from the same testing run. $\endgroup$
    – KarenRei
    Jul 15, 2016 at 2:18
  • 1
    $\begingroup$ @KarenRei - actually Google pulled up an article by Schowalter et al in Nuclear Instruments and methods A - they have data for all the noble gases through 4 polymers. The diffusivity decreases monotonically from He to Xe. $\endgroup$
    – Jon Custer
    Jul 15, 2016 at 2:27

2 Answers 2


As noted in some of the comments above, a paper that directly compares noble gas permeation through several materials is S.J. Schowalter et al., Permeability of noble gases through Kapton, butyl, nylon, and "Silver Shield", in Nucl. Instr. Methods A 615 267-271 (2010) . They note that $K = Db$ where $K$ is the permeability, $D$ is the diffusivity, and $b$ is the solubility of the noble gas in the material. For the noble gases, the diffusivity monotonically decreases from He to Xe in all materials. However, the solubility has an odder behavior, with Ne being the oddball with lower solubilities than He or Ar on either side of it.

The net result is that the permeation can end up not being monotonic in nature across the noble gases. In particular, for butyl, $K$ is highest for He, drops for Ne, increases for Ar, drops a bit for Kr, then continues lower for Xe.

Lots of interesting data in the paper, a nice direct comparison of the noble gases and several materials in the same aparatus under the same conditions .


I'll assume that the data is for the same pressure and temperature and concentration, and other experimental conditions AND that the units of measure are molar rather than mass based. 300 pages exceeds my patience. OK, if your concept of atoms is that they are hard billiard (pool) balls, then it doesn't make much sense. But consider that Xe is more reactive than Kr which is more than Ar which is more than Ne. That implies that those rascally electrons are not as tightly bound (in the valence shell) for Kr than for Ar. Unfortunately, this is just hand-waving. If you were to say that Ar diffuses faster than Kr, then I'd say "Ah, of course! Because it is smaller!" So, all I can say for sure is that very few situations have only one contributing variable - in the real world. Most things are a balance between two (or more) factors. If there were just one "factor" you'd expect straight line behavior. With two factors, you'd expect a change in slope, at least, or in the most severe cases parabolic behavior would be possible. This is a post hoc (after knowing the facts) explanation. In this case, you have three factors, I think: vdW size, deformability and mass. How they interact in the real world isn't easy to predict. That's one reason why compatibility tables such as for O-rings exist. Note that 1st Ionization E is higher for Ar than Kr, supporting the "Kr is more deformable, more reactive" argument.

  • $\begingroup$ This is very speculative, and the question refers to hard data, with specific examples. You should perhaps support your answer with similar references. $\endgroup$
    – Nij
    Jul 14, 2016 at 22:52
  • $\begingroup$ You can just search for "krypton", for example. Starts on page 40 in the file, "3-30" in the page numbering, and I gave an example material to choose ("PTFE"), although you can feel free to choose others. The reactivity of Xe, Kr, and Ar are tiny, I can't imagine that it applies to this situation. They have full outer valence shells and these are not abnormally energetic scenarios.. And Kr is the middle entry in reactivity, yet the highest diffusion rate. And Kr is, again, the middle entry in terms of 1st ionization energy. $\endgroup$
    – KarenRei
    Jul 15, 2016 at 2:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.