The Naked Scientists

The Naked Scientists Forum

Author Topic: Glass isn't a supercooled liquid  (Read 13631 times)

Kevin Davey

  • Guest
Glass isn't a supercooled liquid
« on: 31/10/2008 23:13:33 »
Kevin Davey  asked the Naked Scientists:

Hi Chris,
 
I just listened to the podcast of the 12th Oct 2008 show and heard your footnote to the the answer given by Dave regarding magnifying glasses losing power.  In this you said that glass is a supercooled fluid and flows over time.
 
I am currently studying for a Master of Science in Astronomy and the question of glass lenses changing shape came up in a group discussion board.  Research I did suggests that glass is not a supercooled liquid and does not flow over time.  My response to the discussion was:

It seems to be incorrect to refer to glass a supercooled liquid, instead it is an amorphous solid [1] - it does not contain crystals.
 
Some conclusions have been wrongly drawn from old window panes being thicker at the bottom than at the top.  This is not due to glass flowing over time, but because that is the way the glass, which could not be made with uniform thickness, was installed [2].  Placing the thicker and therefore heavier end down would make the glass panels more stable in their fittings.
 
The lens of the 40" Yerkes refractor has the largest telescope lens as larger ones would deform due to their weight.  This deformation is bending and flexing, it is not due to fluid flow.  Glass rods can be twisted and bent and will resume their original shapes when the deforming forces are stopped - assuming the glass has not broken first, of course!

Correcting for this bending deformation in large reflecting telescopes is the key component of active optics, but is impossible for lenses.
 
The borosilicate glass used in modern mirrors (first used in the 200" Hale reflector [3]) is stiffer and has a higher melting point than the glasses used in both the Yerkes refractor and the 100" Mount Wilson telescope.  None of these older telescopes exhibit any flow in their glass and have not required reconfiguring to remove any flow effects.  Such flows would surely show in the precise optics of these instruments.

[1] Neumann, F. 1996,  "Glass: Liquid or Solid -- Science vs. an Urban Legend",
http://dwb.unl.edu/Teacher/NSF/C01/C01Links/www.ualberta.ca/~bderksen/florin.html
 
[2] Plumb, R. C. (undated)  Antique Windowpanes and the Flow of Supercooled Liquids, http://dwb.unl.edu/Teacher/NSF/C01/C01Links/www.ualberta.ca/~bderksen/windowpane.html
 
[3] http://www.astro.caltech.edu/palomar/history/
 
Regards to you and your team and congratulations for making such a brilliant show!
 
Kevin Davey
Adelaide, South Australia

What do you think?


 

Offline chris

  • Neilep Level Member
  • ******
  • Posts: 5339
  • Thanked: 65 times
  • The Naked Scientist
    • View Profile
    • The Naked Scientists
Glass isn't a supercooled liquid
« Reply #1 on: 31/10/2008 23:20:54 »
Thanks Kevin, this is superb information.

Chris
 

Offline syhprum

  • Neilep Level Member
  • ******
  • Posts: 3820
  • Thanked: 19 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #2 on: 01/11/2008 16:30:32 »
I have always believed that medieval glass was produced by blowing a bubble which was then punctured and spun into a disk as it cooled.
The technique is also used in the production of large telescope mirrors that cool on a spinning bed to produce a parabolic shape.
 

lyner

  • Guest
Glass isn't a supercooled liquid
« Reply #3 on: 02/11/2008 21:11:37 »
Which way would the curve go?
It sounds counter-intuitive to me because the force would increase with distance from the centre. Perhaps a diagram would help me understand.
 

Offline daveshorts

  • Moderator
  • Neilep Level Member
  • *****
  • Posts: 2583
  • Physics, Experiments
    • View Profile
    • http://www.chaosscience.org.uk
Glass isn't a supercooled liquid
« Reply #4 on: 03/11/2008 11:40:48 »
If you are spinning a piece of molten glass then the glass will end up thinner in the middle than the outside. Although the 'centrifugal force' on each gram of glass will increase with radius, the glass in the centre will be under much more tension than the glass at the outside. This is because the glass at the centre has to support all of the glass outside it - the sum of the centrifugal forces on everything outside, so the glass will be thinner at the centre if you start off with a flat piece of glass and spin it.

Although you wouldn't be, you would be starting off with a blob in the middle so it could all get more complex...
 

lyner

  • Guest
Glass isn't a supercooled liquid
« Reply #5 on: 03/11/2008 17:43:22 »
I see - The horizontal force will be proportional to the distance from the centre. This has to work against (balance) the hydrostatic pressure of the glass piled up on top - proportional to depth.
Ummm  I suppose you just might get a parabola.
Actually, the details of the profile may not be too critical. You could always correct for it with the sub-reflector.
The method is more likely to give a known shape than the old method of grinding two pieces of glass over each other and throwing the convex half away.
A cool bit of technology.
 

Offline yor_on

  • Naked Science Forum GOD!
  • *******
  • Posts: 11993
  • Thanked: 4 times
  • (Ah, yes:) *a table is always good to hide under*
    • View Profile
Glass isn't a supercooled liquid
« Reply #6 on: 07/11/2008 00:26:49 »
Kevin, if it's not a fluid, what allows a coin to work its way through a window over time if you put one upon it.
 

Offline daveshorts

  • Moderator
  • Neilep Level Member
  • *****
  • Posts: 2583
  • Physics, Experiments
    • View Profile
    • http://www.chaosscience.org.uk
Glass isn't a supercooled liquid
« Reply #7 on: 07/11/2008 14:42:07 »
Sorry I was still on spinning windows... The telescope mirror thing does come out to exactly a parabola.

If you spin a liquid the centrifugal force per unit mass will be proportional to the distance from the centre.
F = kx    where k = ω2

You can therefore think of a centrifugal potential energy. Potentials are the integral of  which is the integral of the force per unit mass over a distance
so

Vc = -kx2

Liquid will flow so that the total potential (graviational and centrifugal) of the surface is a constant so

Vc + Vg = const

this constant is arbitraty so we can measure it so that 0 is where we want it


Vc + Vg = 0


-Vc = Vg

--kx2 = gh

h = k/g x2

a parabola.
 

lyner

  • Guest
Glass isn't a supercooled liquid
« Reply #8 on: 07/11/2008 16:31:48 »
I love it!
My hero.
Mwah!

Has anyone ever found time to push a penny through glass? My coffee table must be a healthy and safety risk.
 

Offline yor_on

  • Naked Science Forum GOD!
  • *******
  • Posts: 11993
  • Thanked: 4 times
  • (Ah, yes:) *a table is always good to hide under*
    • View Profile
Glass isn't a supercooled liquid
« Reply #9 on: 07/11/2008 21:45:19 »
Well you gave to put some force to it:)
It's over an extended time period, we're talking decades if i remember right.
It might be a urban legend though but I heard it in a BBC program? i think??
Anyway, there might be another explanation allowing for it too?
but I'm still curious toward if it is true.
 

Offline RD

  • Neilep Level Member
  • ******
  • Posts: 8128
  • Thanked: 53 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #10 on: 07/11/2008 22:00:18 »
The telescope mirror thing does come out to exactly a parabola.

Quote
A LMT [Liquid Mirror Telescope] is a telescope which uses a highly-reflecting liquid as the primary mirror. Currently, the most commonly used liquid is mercury. The surface of a spinning liquid takes the shape of a paraboloid that can be used as the primary mirror of a telescope.
http://www.aeos.ulg.ac.be/LMT/index.php

Rotating mercury mirrors are much cheaper than machined glass ones, but of course a mercury mirror can only point straight up.
The mercury vapour is potentially a health hazard too...



« Last Edit: 07/11/2008 22:06:38 by RD »
 

Offline Bored chemist

  • Neilep Level Member
  • ******
  • Posts: 8665
  • Thanked: 42 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #11 on: 08/11/2008 00:10:14 »
Kevin, if it's not a fluid, what allows a coin to work its way through a window over time if you put one upon it.


and your evidence is..,.
 

Offline yor_on

  • Naked Science Forum GOD!
  • *******
  • Posts: 11993
  • Thanked: 4 times
  • (Ah, yes:) *a table is always good to hide under*
    • View Profile
Glass isn't a supercooled liquid
« Reply #12 on: 08/11/2008 01:09:19 »
Are we in court?

I just asked a question coming from a program I saw some years ago.
As for evidence? I could look around I guess.
Goggle if you like, but to what purpose?

I heard it as I remember and so I'm asking.
I got the impression that he was an 'authority' on the subject?

I don't need no evidence for asking it.
Or are you presuming me to be lying here?

-------------------------------------------------

"By definition as an amorphous solid, the atomic structure of a glass lacks any long range translational periodicity."
Which might explain the coins slow descent through the a windowpane (decades), if that statement now is correct.
And that I don't know, if I did, why would I bother to ask?






« Last Edit: 08/11/2008 01:27:38 by yor_on »
 

Offline Bored chemist

  • Neilep Level Member
  • ******
  • Posts: 8665
  • Thanked: 42 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #13 on: 08/11/2008 17:22:44 »
Strictly, rotating mercury mirrors are only perfect parabolae at the poles. Otherwise Coriolis forces disturb them. This is only important for really big mirrors.

If a coin's weight were enough to push it through a window in any reasonable time- say, a century, then the weight of the glass would have a noticeable effect in that time. However we know this isn't true because there are glass artifacts much older than that.
 

Offline Make it Lady

  • Neilep Level Member
  • ******
  • Posts: 4050
  • Hands-on fun for everyone!
    • View Profile
Glass isn't a supercooled liquid
« Reply #14 on: 08/11/2008 18:31:34 »
I lived in a Georgian house with hand spun glass windows. They were, well, bumpy. Each pane had the thicker section in a different place. There was no exacting place where the glass was thin, not the centre or the top. The glass was rather beautiful though especially with the light shining through it. Great patterns on the wall.
Hope this is of some help. Sorry no equations would fit.
 

Offline syhprum

  • Neilep Level Member
  • ******
  • Posts: 3820
  • Thanked: 19 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #15 on: 08/11/2008 20:42:43 »
Dave

Your spun mirror is not a parabola but an ellipse, I know the difference is small but it would only be a parabola on a flat Earth of infinite extent
 

lyner

  • Guest
Glass isn't a supercooled liquid
« Reply #16 on: 09/11/2008 08:20:26 »
how big is this mirror going to be?
The phrase"near enough for Jazz" springs to mind.
For a 10m mirror, what would be the departure from a parabola?
 

lyner

  • Guest
Glass isn't a supercooled liquid
« Reply #17 on: 09/11/2008 08:27:55 »
I thought that window glass used to be made by lifting it up from a trough in a frame with a T bar. Like honey on a knife or making soap film with your hands.  No simple sums would apply and it would have 'runs' in it.
 

Offline Bored chemist

  • Neilep Level Member
  • ******
  • Posts: 8665
  • Thanked: 42 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #18 on: 09/11/2008 09:55:57 »
Dave

Your spun mirror is not a parabola but an ellipse, I know the difference is small but it would only be a parabola on a flat Earth of infinite extent
That seems odd, what makes it bend in on itself and form a closed loop?
 

Offline syhprum

  • Neilep Level Member
  • ******
  • Posts: 3820
  • Thanked: 19 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #19 on: 09/11/2008 10:35:45 »
It is of course only a very small section of an ellipse the reason for the deviation from a parabolic curve is because the lines of gravitational influence converge in the centre of the earth.
here is an interesting article on liquid mirrors

http://technology.newscientist.com/article/dn15154-morphing-mirror-could-clear-the-skies-for-astronomers.html?DCMP=ILC-hmts&nsref=news9_head_dn15154

PS could this strange idea of a coin passing thru a sheet of glass be due to some chemical effect
 

lyner

  • Guest
Glass isn't a supercooled liquid
« Reply #20 on: 09/11/2008 11:05:12 »
I think we'd need a bit more evidence that is actually happens before we spend too much time on speculation about the coin idea.
It could easily have been put there when the glass was made. You could tell by looking for small bubbles around the coin - like you get in paperweights with embedded items.
 

Offline yor_on

  • Naked Science Forum GOD!
  • *******
  • Posts: 11993
  • Thanked: 4 times
  • (Ah, yes:) *a table is always good to hide under*
    • View Profile
Glass isn't a supercooled liquid
« Reply #21 on: 09/11/2008 12:35:38 »
Well I've got to admit that I've been searching for collaboration to my story on the Internet.
So far I've only seen about coin tricks though:)
The strange thing is that I think it was a 'science program' presenting it.
Never mind, perhaps I'll find if I stop looking.
 

lyner

  • Guest
Glass isn't a supercooled liquid
« Reply #22 on: 09/11/2008 15:25:28 »
Quote
he strange thing is that I think it was a 'science program' presenting it.
Bet it wasn't TNS.
 

Offline Bored chemist

  • Neilep Level Member
  • ******
  • Posts: 8665
  • Thanked: 42 times
    • View Profile
Glass isn't a supercooled liquid
« Reply #23 on: 09/11/2008 15:43:54 »
"PS could this strange idea of a coin passing thru a sheet of glass be due to some chemical effect "
I doubt it.
Coins are made from metals that are relatively unreactive. Glass is used for a lot of the equipmnt in chemistry labs because it too, is relatively unreactive.

Of course it's possible that the effect of some chemicals on the human brain could make people think that the coin had gone through the glass- then turned into a huge multi coloured rabbit.
 

The Naked Scientists Forum

Glass isn't a supercooled liquid
« Reply #23 on: 09/11/2008 15:43:54 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums
 
Login
Login with username, password and session length