Published: 19 July, 2000

Is glass a liquid disguised as a solid?

Why is there discussion as to whether or not glass is a liquid or a solid?

Because glass has properties of both. When glass is used at room temperature, for example, as a glass brick, it is a rigid material. Therefore, most people consider glass to be a solid. However, if you could look at glass at a couple million times magnification, you would see that the atoms are arranged randomly, which is typical for a liquid.

In solids such as iron, steel, and aluminum, the atoms are arranged very orderly, in a repeating pattern, like tiles on a bathroom wall. Because of the random atomic structure of glass, physicists tend to think of glass as a liquid, or more precisely, as an "undercooled" liquid. This view is not as absurd as it looks, because it accurately explains what happens when you heat glass. Metals and other materials—think of ice as an example—are solids up to their melting point, where they suddenly turn into liquids. When I worked in a foundry, I found out that liquid iron, for example, is a very good liquid, with a surface tension much less then that of water. You shook it the least little bit and it would splash you with droplets.

But glass is different. When you heat glass, there is never a sudden transition from solid to liquid. Rather, glass becomes gradually more deformable. First like asphalt on a hot summer day, then butter, then honey. That is why glass, heated to a few hundred degrees Celsius, can easily be formed into objects such as light bulbs. However, there is no precise temperature at which glass turns "liquid." There can't be such a temperature, since glass' internal structure is already that of a liquid.

It is a very, very interesting question as to why solid glass has the structure of a liquid. After all, when you first melt all the ingredients to make glass, you clearly do have a liquid. When you cool this liquid, you expect it to freeze at some temperature. But unlike iron, water, silicon or sodium, this liquid never "freezes" into a solid. It just forms a stiffer and stiffer liquid! That's why the system is called an undercooled liquid; it should have turned into a regular solid, with orderly placed atoms, but it did not. Only a few systems fail to do this. It is an unnatural thing not to "freeze" because atoms forming a solid like to stick together—and they can do this best by forming regular pattern.

Now, why is there a discussion? Mostly because some scientists think: Since glass has the atomic structure of a liquid, shouldn't it flow—under its own weight—a tiny bit? To find the answer to this question is not simple! You first have to think about the smallest amount of "flow" that you can measure. Clearly, the smaller, the better! Could you measure a millionth of an inch ? Second, you must consider how much time you want to spend on this project. The longer you wait, the better the chances you will find something. Would you be willing to spend a year on this? And finally, would it be OK to speed things up by putting a big weight on top of the glass or raising the temperature? Could we somehow cleverly work our way back from such "accelerated tests," perhaps by working with different weights, to decide if glass would flow under its own weight at room temperature?

Glass researchers at Corning Inc. have very carefully investigated all of these questions and concluded that glass, at room temperature, is a perfectly rigid solid. In fact, it is many times more rigid than a metal such as copper. The bonds between glass atoms are simply so strong that the atoms cannot move relative to each other. In that regard, glass is very different from ordinary liquids, which are held together by much weaker bonds.