1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Glass transition temperature; slowly cooling it

  1. Mar 15, 2015 #1
    One aspect of glasses that has confused me was the fact that the transition temperature Tg drops as you slow the rate of cooling.

    This has confused me because if the glass is slowly becoming more crystalline, then I would of thought that the Tg would start climbing up towards the crystal melt point (Tm is it?) as by cooling it slower, the molecules would have more time to organize themselves.

    One of the ways I was trying to reason this out, was that in the case that you cool glass quicker, the less ordered amorphous result would require you to put in my energy to allow the glass molecules to move around since there are less directions in which the lattice can slide in since very little of it is crystalline.

    Since cooling it slower allows a more crystalline structure to form, the Tg temperature drops since there are more directions in which the molecules can move (easier to form dislocations I guess...?).

    However (if my visualization above at all is at least somewhat true) I would think that there must be a point where the binding energy of the lattice starts to reduce the mobility of the molecules; ie if you cool it slowly enough you do actually create a crystal, so there must be a point where Tg starts to increase for a very slowly cooled glass (which eventually starts looking like a melting point since the viscousity would become so high as to be almost indistinguishible from being solid).

    I guess i'm looking to understand why Tg drops when the binding energy increases in slowly cooled glasses; and the above explanation kind of makes sense to me, but I might be looking at it the wrong way.

    Any one have any good explanations for the drop in Tg for glasses that are cooled slower?

    Thanks!
     
  2. jcsd
  3. Mar 16, 2015 #2
    Weird that this post was moved to physics; after all I learned these concepts in a chemistry class (which is why I posted in chemistry). Still if a physicist can answer this question well, no harm done.
     
  4. Mar 16, 2015 #3
    Getting an answer isn't looking good (many looks, no replies).

    I think the problem is a lot more difficult then I believe it to be. For example Fused quartz is a glass that is made of melting high purity sillica and the result has a melting point that is quite high (1715 degrees celcius). Since from the below phase diagram none of the forms of quartz at standard atmospheric pressure are going to be solid above 1705 degrees celcius; the glass form has a higher melting point then the crystalline form.


    silica_phase_diagram.png

    I should not forget though the transition temperature still might be lower (the point where a certain viscosity is met; wikipedia: 10^13 poise (or 10^12 Pa·s)).

    Ahh!, I found a link:
    http://www.corning.com/WorkArea/showcontent.aspx?id=59293 [Broken]

    Turns out 10^13 poise is usually the Annealing point of glass; and so in the above link High-purity fused silica has a Tg of 1042 degrees celcius (if Tg is taken to be 10^13 poise).

    Of course I now think I might be at the start again of my question; but I did think for a moment that the melting point of the glass was somehow exceeding the melting point of a pure quartz crystal.

    Still they do place the melting point of Fused silica above that of the quartz crystal; does that mean that while the quartz is liquid, the Fused silica is still in a semi-solid state (or just a really viscous liquid)?

    If you could cool fused silica fast enough, would you be able to increase the Tg of fused silica, above that of the melting point of quartz?
     
    Last edited by a moderator: May 7, 2017
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook