Any stable amorphous substances?

[snorkack]

I´ve seen it asserted at the general physics forum that melting point of amorphous substance is "always" lower than the "melting point" of corresponding crystal.
Also, amorphous solids are often described as metastable against crystallization.

Does it necessarily apply to all amorphous solids?

For the viscosity of liquids is found in a broad range.
Among simple common substances - water conveniently has viscosity of 1 cP at 20 C (less than 1.01). The viscosity rises on cooling - to less than 1,8 cP at 0 C. Water easily freezes. At boiling point of 100 Celsius, viscosity of water is 0,28 cP.
Glycerine has freezing point of 18 Celsius, and freezing point viscosity of around 1700 cP. And glycerin readily supercools. On heating, viscosity of glycerin drops a lot. While heating glycerin is nasty (tends to decay to acrolein), the boiling point can be measured at 290 Celsius. Could not find viscosity measured or estimated at that region, but already at 170 Celsius, it is under 3 cP.
Molten silica comes to true melting equilibrium with crystal - cristobalite - around 1710 Celsius. And at that region, the viscosity of the melt is about 3*10⁹ cP. On heating, it falls. But hot molten silica is too hot to handle - it tends to attack solid vessel materials. Its boiling point is hard to measure, but estimated at 2700...2800 Celsius (cannot be measured better). Its viscosity under those conditions? Still around 50 000 cP.

As you see, freezing point viscosity is found in a wide range. But beyond the range of these three? Any liquids with yet bigger freezing point viscosity?

On cooling liquids, while viscosity changes continuously, in the range of 10¹⁵ cP some other features like thermal expansion and heat capacity undergo a rapid change in a narrow range (how narrow, in terms of viscosity?). It´s called glass transition.

Are there any substances for which true thermodynamic equilibrium between ordered crystal and disordered phase does exist, but at conditions where the viscosity of the disordered phase is on the amorphous solid side of the glass transition? And therefore, a true thermodynamic equilibrium amorphous solid exists?

 

[trurle]

Many common plastics (PE, PP, PET) have crystalline fraction as solid, and that fraction is quite resistant to crystallization conditions change. Therefore, you can talk about equilibrium between amorphous and crystalline phases of plastic, although it is driven by entanglement of molecules, not by free energy.

In general, highly asymmetric molecules like (uncured) SU-8 polymer are impossible to crystallize at all, due very high entropic contribution to Gibbs energy, even at room temperature. SU-8 therefore can be called "truly amorphous", although technically everything would crystallize at near-
absolute zero temperature and very long time.

 

[snorkack]

Many common plastics (PE, PP, PET) have crystalline fraction as solid, and that fraction is quite resistant to crystallization conditions change. Therefore, you can talk about equilibrium between amorphous and crystalline phases of plastic, although it is driven by entanglement of molecules, not by free energy.

You could have a true equilibrium if you could shift the condition to spontaneous transition into amorphous solid.

In general, highly asymmetric molecules like (uncured) SU-8 polymer are impossible to crystallize at all, due very high entropic contribution to Gibbs energy, even at room temperature.

Water is impossible to crystallize at room temperature because of entropic contribution to Gibbs energy (but easy to crystallize below room temperature).

SU-8 therefore can be called "truly amorphous", although technically everything would crystallize at near-
absolute zero temperature and very long time.

Helium won´t, but it has low viscosity even when not superfluid. I think solid He (both 3 and 4) are readily nucleated... how stable is supercooled liquid He?

So what I´m looking for is spontaneous transition to a solid of higher entropy which specifically lacks long range order.

 

[trurle]

Water is impossible to crystallize at room temperature because of entropic contribution to Gibbs energy (but easy to crystallize below room temperature).

I feel you are playing with definitions. I meant entropic contribution at melting point, not above it.

The criteria for the liquid to amorphous transition is to have entropic part of Gibbs energy high enough to have melting point (defined as point then TS~dH(melting) )well below glass transition point (although the glass transition point definition is somewhat arbitrary, been based on human experience timescales). For most materials, the melting point is above glass transition point, but it is not a law. Nothing prohibits the reverse.

Alternative way to define liquid to amorphous transition is "direct transition from state dominated by 2-dimensional defects (slip planes in liquid) to state dominated by 0-dimensional defects (vacancies), skipping entirely the state dominated by 1-dimensional defects (dislocations in crystal)".

Using this definition, you can correctly predict what any substance which do not easily form dislocations (typically due irregular molecular packing happening with large, flexible and asymmetric molecules) will directly transit from liquid to glass.

Polymorphic substances may have even more complicated picture, because molecular structure changes with temperature in non-trivial way. For example, sulphur has small molecules and low viscosity at melting point, but polymerizes and become viscous liquid if heated further.
Cyanogen is even more bizarre - it melts at -28C, vaporizes at -21C, and condense back to glass if heated to 300C.