Why do high-melt-point crystals melt so low in molten glass?

[glassyeyed]

I'm looking at a glass "recipe" that heats the raw ingredients to melt at 1550C. This seems to present a problem in that the raw ingredients include:

Al2O3: melting point = 2072 C
MgO: melting point = 2852 C

And even SiO2 (the primary ingredient) has a melting point (range) of 1600-1725 C .


So I'm puzzling over how this can possibly work. Yet apparently it does work, as this is a real recipe.
I can blindly go ahead and do it, but would really like to understand how & why this can be.

I understand that impure crystals have lower melting points. But in searching these forums (and the web in general) I can find no discussion of deltas that big -- (MgO melting a full 1300 C lower than its pure MP?). It seems... surprising.
[Furthermore, the recipe seems to assume no such requirement of impure ingredients.]

So I figure I must be missing something. Probably something very fundamental.

My only idea is that maybe those higher-melting-point ingredients are not really melting. Maybe they are just dissolving in the lower-melting-point SiO2 liquid? (But even then, 1550C doesn't seem quite high enough to melt the SiO2 -- though with impurities, I guess it could.)

Is that it? Or am I indeed missing something?

Thanks much!

 

[fleebell]

I would like to know the answer to that one too.
I cast aluminum at home (hobby) and it's melting temp is only about 660 C and mix copper 1084 C into it to make aluminum bronze. I use old recycled house type copper wire cut up into small bits and just dump it in and stir for a minute or two. It dissolves right into the molten aluminum without a problem.

 

[chill_factor]

Could be a eutectic point, which is a common phenomena in solid solutions to have melting point depression.

 

[Antiphon]

My chemistry is rusty but there is a principle that says when you alloy metals the melting point is lower than for any of the constituent metals.

Any chemists here?

 

[chill_factor]

My chemistry is rusty but there is a principle that says when you alloy metals the melting point is lower than for any of the constituent metals.

Any chemists here?

yep that's what I said about eutectic points. However these aren't metals but refractory ceramics that are getting dissolved in glass. There's something special about metals that make them very different from ceramics - they are made of atoms, instead of ions, and that totally changes their properties.

 

[DrDu]

To glasseyed question: The 1550 C is sufficient for quartz crystals to melt superficially due to the higher energy of the surface. The melt formet will dissolve slowly some of the aluminium and magnesium oxide. As solutions have a lower melting point than the pure solvent, this will lead to complete melting of SiO2 which then dissolves the other oxides.

 

[M Quack]

Even at ~2/3 the melting temperature you already get a lot of diffusion, and the closer you get to the bulk melting point the more surface effects you get (as DrDu pointed out).

If one of the constituents is molten or close to melting then it can probably penetrate into grain boundaries of the others. This greatly enhances the active surface area and will speed up the mixing process.

 

[chill_factor]

To glasseyed question: The 1550 C is sufficient for quartz crystals to melt superficially due to the higher energy of the surface. The melt formet will dissolve slowly some of the aluminium and magnesium oxide. As solutions have a lower melting point than the pure solvent, this will lead to complete melting of SiO2 which then dissolves the other oxides.

if there was a way to thank people in this forum I would thank you.

 

[chemisttree]

I'm looking at a glass "recipe" that heats the raw ingredients to melt at 1550C. This seems to present a problem in that the raw ingredients include:

Al2O3: melting point = 2072 C
MgO: melting point = 2852 C

And even SiO2 (the primary ingredient) has a melting point (range) of 1600-1725 C .


So I'm puzzling over how this can possibly work. Yet apparently it does work, as this is a real recipe.
I can blindly go ahead and do it, but would really like to understand how & why this can be.

I understand that impure crystals have lower melting points. But in searching these forums (and the web in general) I can find no discussion of deltas that big -- (MgO melting a full 1300 C lower than its pure MP?). It seems... surprising.
[Furthermore, the recipe seems to assume no such requirement of impure ingredients.]

So I figure I must be missing something. Probably something very fundamental.

My only idea is that maybe those higher-melting-point ingredients are not really melting. Maybe they are just dissolving in the lower-melting-point SiO2 liquid? (But even then, 1550C doesn't seem quite high enough to melt the SiO2 -- though with impurities, I guess it could.)

Is that it? Or am I indeed missing something?

Thanks much!

Does your glassmaking recipe use pure ingredients? In glassmaking, alumina is usually added as feldspar (alkali and lime feldspars) which may contain sodium, potassium, calcium and quartz. If your recipe also calls for CaO, Na2O or K2O and SiO2, you need to account for the amounts added in your feldspar. Aluminum trihydrate (MP=300C) is also used as are nepheline syenite (mixture of nepheline and alkali feldspar), kaolin (aluminosilicate clay that may have alkalis and iron), aplite (mixture of various feldspars, alkali and/or lime) and slags from metallurgical operations.

The mechanism of melting is described as:

The glass pot, particularly the closed pot, is not a full heat when it receives its charge of raw material and cullet. But in the tank, whatever the method of filling, the charge arrives directly upon the molten bath, exposed to the full heat of the furnace. However great the resulting difference in speed of melting, the essential phenomena are the same.

The alkali members of the batch begin almost immediately to fuse. There ensues reaction with sand, and the alkali silicates, forming eutectic liquids as low as 800^oC. (1472^o F.) are produced. Lime and other bases begin to find their complement of silica and enter solution or perhaps form double silicates with the alkali silicates. The excess of silica now begins to dissolve, as the melt becomes hotter and viscosity lessens. Meanwhile, gases are liberated from carbonates and from hydrates, nitrates and sulfates. The mass is violently agitated by these escaping gases, and this is an aid to complete mixing. The cullet entering with the charge has also played its part (a quiet one) in helping to dissolve the less fusible ingredients.

from Modern Glass Practice, S. R. Scholes (1952).

 

[glassyeyed]

Wow, thanks very much to all who responded! All the responses have been most enlightening.

So it seems that I was thinking generally along the right lines (which is always encouraging!) -- but with much more complexity than I had imagined. But essentially, the answer is that those crystalline structures with higher melting point are indeed dissolving. That quote from the glass book -- though densely packed -- was esp. helpful in tying together the other feedback.

In answer to your question, chemisttree, the recipe does call for pure ingredients -- no doubt because it is on a small scale. It sounds like large-scale production would be notably cheaper by using feldspar, and those other materials that are less demanding of purity. (I'll have to look into that, in the future. I do have alternate recipes that do include Na, K, and others. But in the main recipe, it has only CaCO3.)

I should probably also explain that my motivation for asking (in addition to the benefit of understanding!) is that I need to get a melt furnace, and as it happens, furnaces become notably more expensive when you need to reach temps as high as 1500+ C. So on the one hand, I don't want to needlessly buy a furnace for temperatures that I don't need to hit, and on the other I don't want to throw away money buying a furnace that can't get hot enough.

Thanks again to everyone. Very enlightening and helpful!

 

[Darwin123]

I'm looking at a glass "recipe" that heats the raw ingredients to melt at 1550C. This seems to present a problem in that the raw ingredients include:

Al2O3: melting point = 2072 C
MgO: melting point = 2852 C

And even SiO2 (the primary ingredient) has a melting point (range) of 1600-1725 C .


So I'm puzzling over how this can possibly work. Yet apparently it does work, as this is a real recipe.
I can blindly go ahead and do it, but would really like to understand how & why this can be.

I understand that impure crystals have lower melting points. But in searching these forums (and the web in general) I can find no discussion of deltas that big -- (MgO melting a full 1300 C lower than its pure MP?). It seems... surprising.
[Furthermore, the recipe seems to assume no such requirement of impure ingredients.]

So I figure I must be missing something. Probably something very fundamental.

My only idea is that maybe those higher-melting-point ingredients are not really melting. Maybe they are just dissolving in the lower-melting-point SiO2 liquid? (But even then, 1550C doesn't seem quite high enough to melt the SiO2 -- though with impurities, I guess it could.)

Is that it? Or am I indeed missing something?

Thanks much!

I think that your only idea is very probably correct. "Melting" is not "dissolving". "MgO" dissolves in "SiO2". "MgO" doesn't melt in "SiO2".
The high melting point ingredients don't melt in liquid SiO2. The high melting point ingredients dissolve in SiO2. This example isn't much different from solids that dissolve in room temperature liquids.
Sodium chloride (NaCl) crystals dissolve in liquid water (H2O). At room temperature, NaCl is a solid and H2O is a liquid. In general, ionic solids dissolve in water. However, most ionic solids (i.e., salts) have a much higher melting point temperature than water. You don't even have to heat the water to dissolve the salt.
Glucose (C6H12O6) also dissolves in water. In fact, most of the simple sugars dissolve in water. Yet, most sugars are solid at room temperature.
Benzene is a nonpolar liquid that dissolves nonpolar solids like hydrocarbon polymers (i.e., plastics). Benzene is liquid at room temperature. Many hydrocarbon polymers are solids at room temperature.