Is glass an amorphous solid or supercooled liquid?

I have been informed that glass is a super-cooled liquid and is also considered to be an amorphous solid.

Can it be both and, if not, what category does it fall into?

• Window glass does not flow under it's own weight. That's a myth. – Karl Jun 26 '16 at 18:33
• By (rheological) definition, the storage modulus of a solid, under linear deformation, does not vanish for low frequencies. That's also true for all glasses. Under high stress, also solids can flow, and many of them do, before ultimately breaking. – Karl Jun 26 '16 at 18:38
• And why can't a super-cooled liquid be an amorphous solid? Just note that not all amorphous solids are super-cooled liquids, but many, including glass, are. – Jon Custer Jun 26 '16 at 19:18
• @Jon Custer, a liquid is considered to be super-cooled if it passes its freezing point WITHOUT becoming a solid, if it isn't a solid, it shouldn't be an amorphous solid as well – Prabhdeep Singh Jun 26 '16 at 21:14
• There are a lot of good things written about this actually. I think the best thing I've read basically says the easiest way to do this is to say it's kind of undefined. There is no latent heat transitions that a glass goes through when melting, but there is also a dramatic change in measurable properties when glass can flow. So, it doesn't change phases, but it changes properties. Beyond that, the definition is arbitrary. – jheindel Jun 26 '16 at 21:55

I'll focus on the phase of the glass, since it seems there is some confusion in the comments (and the other answer). In glass forming systems, as you drop the temperature below the crystal phase melting temperature, one would expect to nucleate the crystal phase. For various reasons this does not happen. Instead, the liquid continues to cool, and the viscosity continues to increase as one might expect upon cooling. Below a critical point, the glass transition temperature, the viscosity rapidly increases by several orders of magnitude. Atomic mobilities are now so low that crystallization cannot occur. Furthermore, the viscosity is so high that the material is a 'solid' in common usage - all the atoms are frozen in place.

So, what is this 'solid'? Well, it isn't the stable crystal phase - you can tell that by x-ray diffraction. True, it is not 'liquid' in the common sense of flowing like water. But, two things point to it actually still being in the liquid phase.

First, in a disordered material if you do x-ray diffraction you measure a radial distribution function (rdf). This gives you the average spacings between atoms in the material. If you do x-ray diffraction from above the melt down to the glass formation, you will note that the rdf is not changing (well, the average spacing will change a bit with thermal contraction).

Second, the other indication of a phase change would be, of course, a signature of latent heat in calorimetry. Again, as you cool the liquid down to the glass formation temperature (and beyond), there is no latent heat measured. There is no first order phase transition. There is only the glass transition where the visocity increases.

Taken together, a glass is a configurationally (kinetically) frozen liquid. It has the structure of the liquid, and it does not have a first order phase transition to the liquid. I do not see how one could consider it anything but a super-cooled liquid.

To further emphasize - a super-cooled liquid is one below the melting point of the crystal phase that has not nucleated the stable crystal phase. Nothing says that the liquid viscosity can not get so high as to be considered 'solid'.

• en.wikipedia.org/wiki/Supercooling and it's source webcitation.org/5mjJauR7S you're against definition of supercooling? Supercooling is before turning into glass. – Mithoron Jun 27 '16 at 18:59
• @Mithoron - supercooling is required to turn it into a glass. And the glass is frozen in a liquid configuration. And there has been no first order phase change. At best one argues about where you are on a viscosity versus temperature curve. It is likely also a sub-area definition. For silicon, we repeatedly attempted to make glassy silicon and failed. Amorphous silicon is not a configurationally frozen liquid, and exhibits a first order phase change to either diamond cubic silicon or liquid silicon. – Jon Custer Jun 27 '16 at 19:43
• Also helpful: The $\ce{^{29}Si}$ NMR of glass (visible in $\ce{^{29}Si}$ spectra because of the NMR tube) is a poorly defined ‘mountain’, showing the strong disorder present in glass. – Jan Jun 30 '16 at 12:48
• What I have been able to gather from all the comments and answers it that, supercooled liquid is an intermediate state in formation of glass, silicate and other substances are 'supercooled and then frozen. So glass is an amorphous solid but in its formation, comes a time where it is a supercooled liquid – Prabhdeep Singh Jul 5 '16 at 12:17

You are correct: it's not both.

Why? What you've heard is nonsense. Very old window panes are thicker towards one side because of how they were made at the time. It became common practise to put the thicker edge at the bottom, because it was less likely to exceed the acceptable stress on the thinner area.

Contemporary windows have been found with the thicker edge to one side, or even the top. Gravity on Earth has been consistently towards the core since at least the Middle Ages, so the "flowing glass" hypothesis is false.

Glass is an amorphous solid by definition (per Karl's comment) and is not a super-cooled liquid (per Prabhdeep Singh's comment).

• For further reading: m.cmog.org/article/does-glass-flow – Nij Jun 27 '16 at 0:11
• Gravity on Earth has been towards the core since the Earth formed. Suggesting anything else is ludicrous. – bon Jun 27 '16 at 17:28
• Glass is not a solid by definition. There is a definition for what is a solid (vanishing loss angle for low frequencies), and glasses fit that description. – Karl Jun 27 '16 at 19:35
• @bon "since at least" because this is the relevant time period; it was meant to be a bit of dry humour given that someone might wonder of gravity could shift enough to make liquids flow horizontally (yes, there are some people who aren't sure). It doesn't suggest anything about the time prior to that because it doesn't refer to any time prior to that. – Nij Jun 30 '16 at 3:09

I have been informed that glass is a super-cooled liquid and is also considered to be an amorphous solid.

Can it be both and, if not, what category does it fall into?

This is dependent on temperature, but before I go into that, we need to understand what a solid and liquid are as that is where mush of the misinformation comes from. A liquid is defined as:

a nearly incompressible fluid that conforms to the shape of its container but retains a (nearly) constant volume independent of pressure.

Note a liquid is NOT defined as the lack of a crystal structure, but as being a fluid (Fluid: a substance that has no fixed shape and yields easily to external pressure; a gas or (especially) a liquid.) . A solid is defined by:

structural rigidity and resistance to changes of shape or volume. Unlike liquid, a solid object does not flow to take on the shape of its container, nor does it expand to fill the entire volume available to it like a gas does.

This should make the question rather simple -- at ambient temperatures, its a solid, but then some people try to proliferate the myth that glass can flow over extended time scales and thus technically a liquid. This is just that, a myth, and is patently false. If you were to measure any flow of glass over as much time as the universe has existed, there would not be a change large enough to detect. Contrast that to something that is indisputably solid and certainly not fluid but still "flows" -- metals. Yes metals exhibit deformation due to creep when heated above a certain temperature and subjected to stress. Despite this deformation, and the fact that metals soften at high temperature, no one is arguing that metals are crystalline liquids nor does anyone argue that metal powders are liquids. Thus again glass is a solid at ambient temperature.

So the only question left is when is glass a liquid as it does become fluid at a certain temperature, and when is it a solid as it resists deformation below a certain temperature?

Given glass does not have an abrupt transition from liquid to solid such as with metals, but rather a continuously thickening viscosity, there are a few values glass scientists use to characterize a glass and that one might consider the glass to finally be a liquid. For our stringent definition of a liquid, the strain point where no more strain can be introduced into the hot glass making is a good demarcation as to below what temperature glass is definitively solid.

I think the thicker bottoms could be due to a phenomenon known creep (the internal flow of molecules under stress and over time). Amorphous materials below their glass transition temperatures show no molecular chain mobility, yet under a constant stress (gravity) and a very very long time (700,1000) years some very, very slow creep motion increases the volume at the bottom of the glass and thins it towards the rest of the body and in direction to the top.

• Yes, some supercooled viscous things flow: bitumen, for example. But not glass. Glass shows no flow over any measurable timescale, but it does sometimes crystallise. Glass windows made before the float glass process was invented (mid 20th century) were thicker in some parts than others as a byproduct of the production process not because they flow. We know this isn't due to creep. – matt_black Aug 2 '17 at 23:36