The AFU and Urban Legend Archive Science Glass Flow amorphous solid Select a topic - - - - - - - - Home Searches AFU FAQ - - - - - - - - AFU Animals Books Celebrities Classic Collegiate Death Disney Drugs Food GIF Language Legal Medical Misc Movies Politics Products Religion Science Sex Songs TV - - - - - - - - Other sites

Newsgroups: alt.folklore.urban
From: ctf2m@kelvin.seas.Virginia.EDU (Chris Fishel)
Subject: Re: Glass flows. Angus is wrong.
Date: Fri, 16 Jun 1995 18:47:34 GMT

dave@mercury.utb.edu writes:
>
> What is a solid? Something with a definate crystalline structure. Glass
> has no crystalline structure, hence it is NOT a solid. I'm sorry, guys.

Well, since I don't see any of the official troll clues, I'll assume this was meant seriously. Please explain the following comments from _Physics of Amorphous Materials_ by S. R. Elliott (London: Longman Group Ltd, 1983; ISBN 0-582-44636-8), both from the definitions section, p. 5:

1. "A glass is an amorphous solid which exhibits a glass transition" (which contradicts both your 'glass ain't a solid' and 'solids must be crystalline' statements)
2. "A solid is a material whose shear viscosity exceeds 10^14.6 poise" (Again contradicting the idea that it is crystallinity that defines a solid)

Although 10^14.6 poise is a semi-arbitrary cutoff point for the solidliquid transition, it perhaps should be noted that common silicalite glasses have this viscosity at temperatures greater than 1100 C (p. 32). At room temp., the viscosity is even larger.

Chris "Of course if you have an actual source to back up your

        claims about glass *really* being a liquid, by all means bring 
        'em out." Fishel

From: s-sehlhorst@ds.mc.ti.com (Scott Sehlhorst)
Newsgroups: alt.folklore.urban,alt.folklore.science
Subject: Glass Blows! was re:wit, pith, tornado(e)s, straw and 2x4's
Date: Mon, 19 Jun 1995 17:11:36

In article <waterrd.26.000DD1E2@mins2.msfc.nasa.gov> waterrd@mins2.msfc.nasa.gov (Ronald D. Waters) writes: [...battling snip quips continue...]

>>If the glass did flow, then it maintained a constant volume. Also, it would
>>flow 'towards' becoming first a cone, and then a pool of molten glass.

> The shape it flowed toward would depend heavily on the manner in which it
>was supported. *If* it was only supported at the bottom (i.e., sitting
>loosely in a window pane), your suggestion would be accurate. If it was
>supported *around* its perimeter (via caulking or some other method), I would
>expect it to flow towards another shape (in particular, the thickness at
>corners would be expected to be greater than that in the center, in this case).

I agree with your assessment, _if_ there is chemical bonding between the glass and the supporting members. _If_ that premise were true, not only would the shape be as you described, but, more importantly, the glass would be slightly thicker at the very top, in the immediate vicinity of the supporting member, than it would slightly less close (a little below). This would also follow in your toothpaste analogy.

According to J.A. Pask, in
Chemical Bonding at Glass-To-Metal Interfaces, presented at the Winter Annual Meeting of the American Society of Mechanical Engineers, Boston, MA, Dec 13-18, 1987.,
regarding glass to metal chemical bonding, "Compatibility of electronic structures of the two phases is required but not easily acquired because of the incompatibility of the metallic and ioniccovalent bonding."

He basically says that a chemical bond, although achievable is unlikely without the presence of an oxide layer (today generated by a redox reaction with air on the metal, prior to attachment to the glass), which is probably beyond the technology of 'old' glass manufacturers.

> [...remainder deleted...]

The remainder you deleted was the point I was trying to obfuscate. If a time period sufficient to allow glass to flow (if it could flow) were to pass, then mechanical bonding would have long since broken down, unless the glass were loaded under a compressive stress. Now some science references...

First, one that I do not have, but should serve as a cathouse-caliber-answer... Shay, R.M. Jr., Ph.D. dissertation, Purdue University, West Lafayette, IN (1989) : An ideal glass analysis, referenced in... The Effect of Thermal History on the Mechanical Behavior of Amorphous Polymers, R.M. Shay, Jr., and J.M. Caruthers, Purdue University, School of Chemical Engineering, presented at the Winter Annual Meeting of the American Society of Mechanical Engineers, San Francisco, CA Dec 10-15, 1989.

In all of the references (6 or 7 that I just checked), although not at all conclusive, the researchers treat glass as a brittle amorphous structure (at least below about 600C) with no dissenting arguments in favor of other models.

Perhaps if we identify the glass in question (soda-lime-silica glass, used in window glass, for those 'old factory on the Allegheny' arguments, or a pressed glass, for the '400 year old monastary' argument), we can get some specific data on the bonding to support structures.

***
this is probably the important paragraph for those interested only in conclusions
***
Clearly, we've reached an academic point regarding the potential mechanism of flow (if it exists), as we concur that any flow will result in specifically measureable non-prismatic characteristics, not just "duh...thicker at the bottom...", through analysis of edge effects in the glass, whether surface tension held them to the supports, or an unsupported existance resulted.

> Regards,
> R.W.

I'm afraid I don't have the familiarity with the Bingham model needed to show that it is inapropriate (or validate your hypothesis). I hope I have supplied some fodder for actual analysis, and the end of an undying thread.

Scott "maybe some garlic and a stake driven through it (by a tornado)" Sehlhorst

From: s-sehlhorst@ds.mc.ti.com (Scott Sehlhorst)
Newsgroups: alt.folklore.urban,alt.folklore.science
Subject: Re: Glass flows... [was: Tornado shoves straw through 2x4]
Date: Tue, 20 Jun 1995 07:54:13

In article <3s66un$2n5@nnrp2.primenet.com> btrosko@primenet.com (Brian Trosko) writes: >J.Young (ccjy@sun.cse.bris.ac.uk) wrote:

[...coments of pinhead dancing on head of angel deleted...]

>And here we go yet again, dammit. Once more, for the folks who have
>missed it.

>From the archives at Cathouse:

> >And indeed I have. Some interesting points arise in the article "Antique
> >windowpanes and the flow of supercooled liquids" by Robert C. Plumb
> >(Worcester Polytech. Inst.)in the Journal of Chemical Education, 66(12),
> >994-6, 1989.

[...familiar summary deleted...]

>Brian "And wood is also a liquid, right?" Trosko

More ammo Brian! The Mark's Standard Handbook for Mechanical Engineering also supports the cathouse data, although 594C is Tg for silica lime glass.

Also,

Regarding discussions of bonding keeping glass from 'gaping' at the top...

According to J.A. Pask, in
Chemical Bonding at Glass-To-Metal Interfaces, presented at the Winter Annual Meeting of the American Society of Mechanical Engineers, Boston, MA, Dec 13-18,1987.,
regarding glass to metal chemical bonding, "Compatibility of electronic structures of the two phases is required but not easily acquired because of the incompatibility of the metallic and ioniccovalent bonding."

He basically says that a chemical bond, although achievable is unlikely without the presence of an oxide layer (today generated by a redox reaction with air on the metal, prior to attachment to the glass), which is probably beyond the technology of 'old' glass manufacturers.

Shay, R.M. Jr., Ph.D. dissertation, Purdue University, West Lafayette, IN (1989) : An ideal glass analysis, referenced in... The Effect of Thermal History on the Mechanical Behavior of Amorphous Polymers, R.M. Shay, Jr., and J.M. Caruthers, Purdue University, School of Chemical Engineering, presented at the Winter Annual Meeting of the American Society of Mechanical Engineers, San Francisco, CA Dec 10-15, 1989.

Shay, btw, treats glass as a solid.

Scott "but if a tree flows in the forest, will anyone hear it?" Sehlhorst

From: spindler@alcoa.com (Doug Spindler)
Newsgroups: alt.folklore.urban
Subject: Re: Glass does not flow. David is wrong.
Date: 20 Jun 1995 13:37:11 GMT

Brian Trosko (btrosko@Primenet.Com) wrote: > Greg Stevens (stevens@prodigal.psych.rochester.edu) wrote:
> : Uh, actually, it's liquid because it has no crystal structure.
> >... Some interesting points arise in the article "Antique
> >windowpanes and the flow of supercooled liquids" by Robert C. Plumb
> >(Worcester Polytech. Inst.)in the Journal of Chemical Education, 66(12),
> >994-6, 1989.
> > [standard good glass/liquidity points]

Although we're all a bit glazed over on this topic, I thought I would pile on with a semi-pertinent aggressively pro-rigidity statement I found in the literature a year or so ago.

If I am desciphering my notes correctly this was quoted in a book by Reiner, _Deformation, Strain and Flow_, from an article on glass in J. Appl. Phys. (13) pp623-654, 1942 by Preston(?).

[Describing a use of glass in a particular telescope] We use glass in this case, not because it is transparent, but because its rigidity and permanence of shape are better than steel or concrete.

Doug "rigid assets" Spindler

Newsgroups: alt.folklore.urban,alt.folklore.science
From: nathan@pact.srf.ac.uk (Nathan Sidwell)
Subject: Re: Glass flows... [was: Tornado shoves straw through 2x4]
Date: Tue, 20 Jun 1995 14:09:49 GMT

J.Young (ccjy@sun.cse.bris.ac.uk) wrote:

: Confirmed information.

: Glass IS a supercooled LIQUID.
A super cooled liquid is a substance which remains in a liquid state below its freezing point. Given a nucleation center it will immediately freeze.

Glass does not show this phenonmenon. (doesn't undergo a phase transition when scratched)

As others have pointed out it is an amorphous solid. The atoms have the same ordering as the molecules of a liquid (ie no long range order), so in certain ways glass resembles a liquid. *but* the bonds between the atoms are covalent (ie strong) not weak inter-molecular bonds as in acetone, or the hydrogen bonding which occurs in water.

In order to flow, these bonds must be broken (and then reform). The energy required to do so is much higher than the thermal energy available at room temperature -- an additional source of energy must be provided (such as stress).