I have a Klean Kanteen flip-top growler, allegedly leakproof at up to 35psi of pressure, that I filled with some freshly-carbonated water 4 days ago — just to see how long it would last. I used a Soda Stream for carbonation. The bottle was filled to the brim.

When fresh, the carbonated water fizzed and sparkled. You could see the bubbles constantly rising to the surface. It felt like pop rocks when you put it in your mouth.

Now, 4 days later, the water is still clearly carbonated, but it's a lot more subdued. You can taste the fizz when you drink it, but there's no more surface bubbling. When you pour it into a cup, you see bubbles for about a second before the water clears up.

The obvious theory is that some CO2 escaped over the course of the week, but I'm fairly certain there were no liquid leaks. I also couldn't hear any hissing. The bottle was stored on its side in the fridge.

Is it possible for a container to be liquid leak-proof, but not gas leak-proof?

If not, does the nature of carbonated water change over time? Does the water, perhaps, absorb more of the CO2 the longer it sits? If so, why does this not happen with soda cans and such?

Update 2015-11-3:

I repeated the experiment with a 20oz screw-top Klean Kanteen I had lying around, ensuring that the water was as close to the cap as possible and that the lid was screwed on as tightly as possible. The results after 4 days were more carbonated than my initial trial with the flip-top Kanteen, though still not as much as fresh. Currently repeating the experiment once again with the flip-top, this time making 100% sure that the water is touching the lid.

  • $\begingroup$ What temperature was the water when you carbonated it? What temperature was the water when you opened the container? $\endgroup$
    – ringo
    Commented Oct 30, 2015 at 20:41
  • $\begingroup$ Going in, it was chilly, but not ice-cold. The bottle was stored in the fridge and was fridge-temperature when opened. I can measure the exact temperature if necessary. $\endgroup$
    – Archagon
    Commented Oct 30, 2015 at 20:49
  • $\begingroup$ No, I don't think that will be necessary. You say the container is filled the the brim, but was there any air space between the brim and the cap? $\endgroup$
    – ringo
    Commented Oct 30, 2015 at 20:56
  • $\begingroup$ It's likely there was a small amount (half a centimeter?), but I tried to make it tight. $\endgroup$
    – Archagon
    Commented Oct 30, 2015 at 21:02

1 Answer 1


The difference is in what is leaking. Gas leaks out a lot easier than fluid. The container is rated not to leak liquid, not that it won't leak air. So if you turned it upside down with 35 psi inside (a carbonated drink) the fluid won't leak out.

So turn the Klean Kanteen flip-top growler upside down. ;-)

  • $\begingroup$ Hey, now that's an experiment! I agree that gas will leak much easier than the liquid, but I suppose if it has to also diffuse out of solution, that might make the difference. It's rather a trick to get a good leak tight seal for small gas molecules. You could also try to find an analytical balance and weigh before and after. Carefully. Several times. Just for kicks. $\endgroup$
    – repurposer
    Commented Nov 2, 2015 at 5:02

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.