The glass is in a metastable state. It is changing constantly. So what will a piece of glass look like in 500 years in room temperature?

  • 3
    $\begingroup$ You can find 2000 year old Roman glass that looks absolutely stunning: metmuseum.org/toah/hd/rgls/hd_rgls.htm. $\endgroup$
    – CuriousOne
    Dec 31, 2015 at 6:01
  • 2
    $\begingroup$ I'm not an expert (maybe there's someone here who is) but I believe it's debatable whether glass is really a "slow liquid" rather than a solid. But even if it is a liquid, I think it will takes a lot more than 500 years for it to flow appreciably. $\endgroup$
    – N. Virgo
    Dec 31, 2015 at 6:41
  • 1
    $\begingroup$ It was once claimed that the fact that Medieval era glass panes are often thicker at the bottom than at the top is proof that the glass has flowed over a period of about 500 years, but that's been shown to be false. ( scientificamerican.com/article/fact-fiction-glass-liquid ) $\endgroup$
    – user93237
    Dec 31, 2015 at 6:49
  • 3
    $\begingroup$ Duplicate? physics.stackexchange.com/q/65740/8521 $\endgroup$ Dec 31, 2015 at 7:23
  • 1
    $\begingroup$ @JohnRennie Actually I believe there is a very clear answer to this question that is grounded on physics observations. $\endgroup$ Nov 23, 2016 at 7:47

1 Answer 1


It's a common misconception that glass flows appreciably over time. So the answer is that your glass will have exactly the same shape as it does now, to within under a wavelength.

You can debunk the myth of flowing glass with simple experiment / observation. There are on Earth several very large refracting telescopes over 100 years old with huge lenses ground to precise optical specifications. Astronomy, as with microscopy or any precision imaging, is an application that is extremely sensitive to optical aberrations, and even sub-wavelength deformations in the refractors of these telescopes over time would degrade the telescopes' optical performances. Yet even these sensitive devices have suffered no plastic deformation that would shift their surfaces more than a significant fraction of a wavelength, which they would do over 100 years if glass were in the state of flow that is commonly claimed.

You can make this simple observation quantitative if you can borrow of the order of a week's time on a decent interferometer. Set up a large, low cost lens (say of the order of 0.1m diameter or more) to focus into the focus of a spherical retroreflector and light the system with the probe beam of a Fizeau interferometer. Cement a heavy piece of steel or lead to the top of the lens (you'll need to do this so that you can still pass an appreciable beam from the interferometer through the unobscured part of the lens) and record the fringe pattern. You'll find that after a week, the fringe pattern has not shifted AT ALL. You can then derive an upper bound to the amount of "flow" that the lens has undergone from this observation. If the top surface hasn't moved 20 nanometers (which shift would be highly observable on the interferometer) in a week (the result I can assure you you will observe), its motion is then less one micron every year, i.e. less than than 1 millimeter deformation every thousand years!

  • $\begingroup$ While I like your argument that no effects have been observed in the time scales you mention, I think that you should specify what you mean by "1 mm deformation every thousand years". 1mm deformation at a thickness of 5mm would be considerable but negligable at a thickness of much more, so I'm a bit confused. $\endgroup$
    – Sanya
    Nov 23, 2016 at 8:35
  • $\begingroup$ @Sanya I'm speaking about the apex of the upper convex surface of a 0.1m or so diameter lens. "Fluid" behavior would see that point sag; 1mm sag in the center would be small compared to the flows needed to make the bottoms of cathedral walls thicken significantly. $\endgroup$ Nov 23, 2016 at 11:21
  • $\begingroup$ @WetSavannaAnimal Will you also say that glass is not in a metastable thermodynamic state and relaxing slowly over time? This article scientificamerican.com/article/fact-fiction-glass-liquid says that "Like liquids, these disorganized solids can flow, albeit very slowly. Over long periods of time, the molecules making up the glass shift themselves to settle into a more stable, crystal-like formation" Of course, it also says that such flow cannot account for why some medieval windows are thicker at the bottom. $\endgroup$ Jul 5, 2019 at 2:55

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.