Let's say that we have a gaseous or liquidus compound (I don't know if elements or compounds make a difference, take this as a thought experiment), and we have a tungsten or steel block that's 5cm (or less, you choose) thick. Is there any physical method for that gas or liquid to pass through that thick heavy metal block (not by drilling etc.)? Maybe vibrating or something else, I am asking for I have no information about this. Every quantum mechanical or unorthodox ideas and theories are accepted. Maybe some solid state physicist could help me. Maybe some proposal which works like diffusion, I don't know. I am here to listen and learn.


  • 8
    $\begingroup$ Does this answer your question? Gas permeation through solid metal? $\endgroup$
    – KingLogic
    Oct 15, 2020 at 4:29
  • 1
    $\begingroup$ If you want more.."creative" ideas you should try the worldbuilding stackexchange site. You can also use the hard science tags if you want to stay realistic. $\endgroup$ Oct 15, 2020 at 8:27
  • 1
    $\begingroup$ @GeorgeMenoutis I thought I could get some answers on theoretical physics to enhance this diffusion event too, by crazy I meant that. I think you don't know the difference between "fictional" and "crazy/hard to believe". I asked for ideas that are still on discussion, which some of them might be really hard to believe/crazy, but I wanted to learn about them too. Crazy/hard to believe things can be real too, sir. And I think "you" can try the worldbuilding site, and start to create a world where its people's understanding on physics is restricted with just kinematics. $\endgroup$ Oct 15, 2020 at 9:40
  • $\begingroup$ If something sounds "crazy" to a post-1950ies physicist, you can be pretty sure it's impossible. ;) Surprises are extremely rare. $\endgroup$
    – Karl
    Oct 16, 2020 at 19:02

4 Answers 4


Yes, some gases can diffuse into and through metal. It is the bane of the high-vacuum engineer's life. Hydrogen is the worst because it tends to dissociate into atoms at the surface and the nucleus, a single proton, can then leave its electron behind and wander through the metal lattice until it picks up another electron when it leaves.

For example Mu-metal, favoured for some applications, typically has to be annealed in hydrogen at high temperature. Once that is over, it can take weeks or months for the residual hydrogen to diffuse out of the metal before a high enough vacuum can be achieved and the work proceed.

A "virtual leak" occurs where a small bubble of gas is embedded in the material inside a vacuum chamber. The leak usually happens because a tiny hole exists for the gas to diffuse out through, but sometimes the "hole" is no more than an ultra-thin skin of metal (invisible to the frustrated technician) and the gas diffuses through it. These little horrors can keep going for months or even years and generally mean replacing suspected parts and pumping down over and over again until the dodgy one is finally stumbled on.

Helium is both monatomic and the physically smallest atom. It can diffuse more easily than any other neutral atom or molecule, making certain metal foils unsuitable as say gas-tight liners for airships. As noted in another answer, in quantity it can also affect the bulk properties of the metal.

On a more energetic scale, hydrogen and helium nuclei (protons and alpha particles) can pass through thin metal foils if fired with sufficient energy, and this has been used to establish the crystalline structures of some metals and alloys (where, for whatever reason, electrons were unsuitable).

Other gases have much larger atoms (neon and other noble gases) or molecules (nitrogen and other diatomic molecules, water and other hydrides), but they can still diffuse extremely slowly through some metals. This can limit the lifetime of some microchips. A related phenomenon occurs where there is a defect in the lattice at the surface, such as a grain boundary, and a gas atom attaches to it. Defects are sometimes quite mobile and can migrate through the lattice; the gas atom will stabilise the defect and may be able to hitch a ride.

Quantum processes such as tunnelling are not really relevant, as they work over distances smaller than the atomic wavelength, which in turn is typically far smaller than the thickness of any metal atom or foil. The probability of a gas atom tunnelling across is so infinitesimal as to be effectively zero.

  • 2
    $\begingroup$ Were you thinking of mu-metal, rather than Muntz metal? $\endgroup$
    – Sneftel
    Oct 16, 2020 at 8:35
  • 3
    $\begingroup$ Yes, indeed I was. 46 upvotes and you are the first to spot my blooper. Congratulations. I'll correct it. $\endgroup$ Oct 16, 2020 at 10:49
  • 2
    $\begingroup$ Hydrogen is effectively soluble in some metals, especially palladium, and can diffuse through it relatively quickly. $\endgroup$
    – matt_black
    Oct 16, 2020 at 13:42
  • $\begingroup$ The first sentence should read, "Yes, gases can diffuse into and through metal, some problematically so." I don't know of any evidence that there's a subset of gases that don't diffuse through metal. $\endgroup$ Feb 10, 2021 at 20:44

Not really, but sort of.

Helium atoms do not form molecules, and the atoms are small. They fit between the spaces in iron, and can diffuse around inside.

This is not a fast process. That is to say, it does not create a measurable leak.

I have heard of one case where that actually created a problem, though I have forgotten most of the details. Some facility had pipes with high pressure He. There was a port for an instrument. The port had thin steel bellows on which the instrument we mounted to allow for some movement.

Over the years, He had diffused into the steel. It had stiffened the bellows and made them brittle. They broke, badly injuring someone.

  • 2
    $\begingroup$ Hydrogen also does it. Heating speeds up the process. $\endgroup$ Oct 14, 2020 at 14:46
  • 13
    $\begingroup$ @mmesser314 e.g. palladium becomes so permeable to H₂ at slightly elevated temperatures (300-400 °C), that a heated Pd pipe is used in a commercial Pd hydrogen purifier. Helium permeability through thin-walled mylar windows at room temperature is so high that it causes troubles with cryogenics (a rule: cool with LN₂ first, then let He in). $\endgroup$
    – dominecf
    Oct 14, 2020 at 16:40
  • 3
    $\begingroup$ @mmesser314 another example: en.wikipedia.org/wiki/Hydrogen_embrittlement is a problem caused by hydrogen diffusing into metals in e.g. nuclear reactors $\endgroup$
    – kevinsa5
    Oct 15, 2020 at 19:01
  • 2
    $\begingroup$ @dominecf - Come to think of it, Cold Fusion was supposed to have worked something like this. Hydrogen atoms could fit inside a unit cell of Palladium. Two atoms could fit, but they were squeezed together so closely that there was expected to be a reasonable rate of fusion. $\endgroup$
    – mmesser314
    Oct 15, 2020 at 21:36
  • 1
    $\begingroup$ I have a nice example behind me of He diffusion: 7 months ago we vacated our labs in a hurry leaving a rubber bellows filled with 37mbar of He transfer gas. I came back to find the bellows collapsed as if I'd put a vacuum pump on them - all (or near enough) the He had diffused through the rubber $\endgroup$
    – Chris H
    Oct 16, 2020 at 8:00

Permeation of atomic hydrogen through metal has been performed in a study(ref.1):

[...] atomic hydrogen is supplied to the metal surface by reaction with an acid, by electrolysis, or by ionization, rapid permeation occurs even at room temperature . Hydrogen atoms can easily dissolve in metal, diffuse, and readily leave the exit surface to recombine as molecules. Dissociation of molecules into atoms on the entrance surface is, for this low-temperature example, the process which determines the permeation velocity of molecular hydrogen.

The rate of permeability is expressed in terms of membrane thickness and pressure differential across the membrane according to Fick's law, and to temperature by the Arrhenius rate theory. You can find the calculation in the paper.

Also research on metal membranes to absorb hydrogen gas has also been discussed(Ref.2,3).


  1. Webb, R W. PERMEATION OF HYDROGEN THROUGH METALS. United States: N. p., 1965. Web. doi:10.2172/4583045.
  2. Hydrogen-permeable metal membranes for high-temperature gas separations David Edlund, Dwayne Friesen, Bruce Johnson, William Pledger, Gas Separation & Purification, Volume 8, Issue 3, 1994, Pages 131-136, DOI: 10.1016/0950-4214(94)80023-5
  3. Shigeyuki Uemiya (1999) State-of-the-Art of Supported Metal Membranes for Gas Separation, Separation and Purification Methods, 28:1, 51-85, DOI: 10.1080/03602549909351644

This actually happens when you use acid to etch or treat metal parts. Hydrochloric acid is a common and very effective rust remover, but when acids attack a substrate, hydrogen is liberated. Enough of the hydrogen enters the steel to cause what's known as hydrogen embrittlement; for low strength/non-hardened steel it tends not to matter too much, but for high strength steel it can have a very negative effect on the toughness. The same thing can happen when welding with oxyacetylene, or arc-welding with rods which have absorbed moisture into the flux coating.

So if you ever decide to weld or derust a high strength steel part or tool, make sure you understand the effects of hydrogen embrittlement first. Where practical, controlled baking can be used to diffuse the hydrogen back out again.

(Hydrogen embrittlement can also affect copper and some other less common metals.)



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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