Glass Windows Do Not Flow
Feb. 2nd, 2006 11:34 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
lederhosen(X-posted to ![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-community.gif)
sceptics)
This is one of those urban myths that just WILL NOT DIE. Every couple of years somebody claims that old windows are thicker at the bottom because glass is a slow-flowing liquid, somebody demolishes it, and a couple of years later it rears its head again. But it's an interesting subject none the less, because the reasons *why* it's so hard to dispel touch on several different aspects of science.
I was originally going to call this 'Glass Is Not A Liquid', but that was a little harder to defend; over very large timescales the distinction between liquid and solid becomes less clear, as discussed below. But I'm happy to assert that for practical purposes, glass is as 'solid' as many other things we have no hesitation in calling solid.
States of matter
Once upon a time, science recognised three phases of matter: solid, liquid, and gas. (Many more have since been discovered - see Wikipedia article for a starting point - but you're not likely to see most of them in your day-to-day life except for plasma and liquid crystals. I'm going to ignore those other phases for this post.)
Liquids and gases are collectively called 'fluids' because they flow under any provocation whatsoever. Apply the slightest force to them, and they'll move out of the way; honey won't move as quickly as water, but they'll move and won't stop until you remove the force (or they have nowhere else to go). Put a liquid or gas in a container, and it will eventually reshape itself to match the shape of the container exactly.
Solids don't like to flow. If you apply a small force to them, they'll give a little way, and no more. No matter how long a fly stands on a gold brick, his feet are never going to sink into that brick; to make a large change to its shape you need a large force, large enough to overcome its material strength.
The difference between these different phases of matter is in how their component parts hang together. Let's start with a nice simple example: iron.
In a slab of cold iron, you have a whole lot of iron atoms tightly packed together by electrical forces that pull them towards one another. If you were to reach into that slab with a very fine pair of tweezers, you'd find it very hard to move one of those atoms around - they're all nestled together closely, so there is no place for that atom to go that isn't already occupied.
If you heat that slab up a bit, the atoms start to jiggle around. (This is what heat is: random jiggling at a microscopic scale.) They pull away from one another, and so the iron expands slightly, but they're not jiggling hard enough to actually break free from their position in the structure; they're like cranky toddlers squirming in a car harness. If you were able to 'tag' a specific atom and then come back to it an hour later, it'd still be in the same position.
Heat it up a lot, though - to about 1538°C - and things change. Some of the atoms are now jiggling hard enough to break out of position and stay out. They're still packed close together, but that packing no longer holds them in place; they're like people shoving past one another in a crowd. Overall, the forces pulling those atoms towards one another are still strong enough to keep the atoms together, so it doesn't expand much when it melts (one well-known solid actually contracts on melting), but that freedom of movement means the grid is no longer permanent; push against it gently for long enough, and it will move aside. At 1540°C, these breaks from position only happen occasionally, and so the liquid is still very thick; as you heat it up they happen more and more often, and it gets easier to pour.
Heat it up further, to 2861°C, and that jiggling becomes so intense that it now overcomes the attractive forces. The atoms fly apart, and they go bouncing off into infinite space. The iron has become a gas, which expands to fill whatever space there is available.
Viscosity
Viscosity describes the 'thickness' of a liquid. The more viscous a liquid is, the slower it pours. Tip a glass of water over and the water falls out instantly; tip a jar of honey, and it'll take a few seconds. So how viscous can a liquid get, and still be a liquid?
As it turns out, very viscous indeed. The classic example is the University of Queensland's pitch experiment: back in 1930 they put a dollop of pitch in a glass container with a hole in the bottom, and left it to flow. It's still there; seven and a half decades later, just eight drops of pitch have fallen into the cup below. To the casual bystander, it's very easy to mistake pitch for a solid. So it's not unreasonable to wonder if glass might also be a very slow-flowing liquid. How else can we distinguish liquids from solids?
Solid structures
If you could look at that cold slab of iron close-up, you'd see the atoms arranged in a regular lattice like cannonballs stacked in a neat pyramid. On a large scale that lattice has a lot of imperfections, but overall it follows a pretty regular structure. As scientists researched further, they discovered that this was pretty much standard for solids. Iron atoms in that slab, sodium and chloride ions in a crystal of table salt, water molecules in a block of ice from the freezer - they all freeze in regular, tesselated crystal structures.
Looking at a material's crystal structure is a much more appealing prospect than sticking it in a container for a decade and watching to see whether it does anything, so it's not too surprising that scientists were eager to define solids this way; a couple of years ago I ran across a modern dictionary that still defined a solid on the basis of a regular crystal structure.
Unfortunately, it doesn't work that way. It turns out that some materials are solid in the not-flowing sense, but don't have a regular, repeating crystal structure - these are known as amorphous solids. One example of this is polystyrene, which is made of very long molecules tangled together like a microscopic hairball. Another, as you might've guessed, is glass, which is made up mostly of silicon and oxygen jumbled together in no particular structure. Cribbed from this UCR site, an image of that:
It's also possible to make amorphous ice. A water molecule is shaped rather like a boomerang. In the normal course of things, when water freezes, it forms a structure where the 'boomerangs' form a regular hexagonal pattern - each end of the boomerang connects to the middle of another boomerang, or to another boomerang-end. This is a nice low-energy arrangement (it keeps like charges away from one another, and opposite charges close together) but it's quite an open structure - there's a lot of empty space in the middle of those hexagonal rings, which is why regular ice is less dense than water.
But water doesn't automatically convert to ice the moment the temperature drops below zero. It needs some sort of 'seed' to get that structure forming. This can be pre-existing ice, or an impurity in the water; without that to start it off, you can cool water as far as -42°C without freezing it into a solid. This is what we call a 'supercooled liquid' - it's under temperature and pressure conditions where its most stable state is solid ice, but it needs some help to reach that state.
If you take that supercooled water and cool it further, very fast - for instance, by spraying droplets of it into a cold gas - you freeze it into a solid; the molecules no longer have the energy to shove past their neighbours and move around. But it doesn't have time to rearrange itself into that neat crystalline structure. Instead you get a dense, amorphous solid; if you took a microscopic snapshot you'd have great difficulty distinguishing it from liquid water, because the liquid's random jumble of molecules has been frozen into the solid. This is pretty much the same thing that happens in glass, and indeed this particular form of ice is known as 'hyperquenched glassy water'.
(In fact, there are at least a dozen different forms of ice known, including several different amorphous forms, and even at least one alternate form of liquid water; for more info, see Wikipedia articles on ice and amorphous ice. From here on, though, unless I qualify 'ice' I'm talking about the regular kind.)
Because regular glass is of the same sort of nature as HGW, it has been referred to as a 'frozen supercooled liquid'; unfortunately, the 'frozen' is often forgotten when people hear the 'liquid'.
Discontinuous properties
When you pump energy into a block of ice, its temperature goes steadily up... and up... and then, at 0°C, it stops. It stays at 0°C for quite some time, until the ice has entirely melted away into water; then it starts to rise again. If you were to plot temperature on the x-axis, and energy on the y-axis, you'd see an abrupt jump in energy as the temperature goes from -0.1°C to 0.1°C. Many other properties - density, optical properties etc. etc. - change abruptly at this point. Because ice has a regular structure, the energy (& hence, temperature) required to break its bonds at one point is the same as that required to break corresponding bonds anywhere else in that structure; at -0.1°C they hold, at 0.1°C they're broken and you just have a jumble. It's like a workplace where everybody knocks off at 5 pm sharp - one minute it's business as usual, the next there's nobody there.
Not so with glasses. Because of that disorder, every atom in a glass is in its own unique situation, and so it takes a different amount of energy to break free from its position. The result of this is that at around 1000 °C (depending on exact composition), some of the atoms start to get a little freedom... but others are still caught up in their structure. It's more like my workplace, where some people head home at 4 pm and others stay until 6 or 7; the transition is much more gradual. You don't see that sharp jump in material properties against temperature. Cooling it down again, it works the same way - some bonds reform before others, and the freezing point is not well-defined, which leads many people to wonder - if we don't know where it becomes a solid, are we sure it actually did?
Experiment
The scientist's favourite answer, of course, is 'try it and see'. Since the supposed rates of flow are very slow indeed, setting up an experiment from scratch isn't very convenient (though I'm told one engineer at Dow Corning did run a 27-year experiment on whether glass rods subjected to a bending force experienced flow; anybody who can find me a good reference for this wins an imaginary gold sticker). Instead, most people start out by looking at old glassware, leading to the most famous piece of 'evidence' for glass flow:
If you look at old glass windows, they're thicker on the bottom than at the top, because the glass has flowed downwards. (And a couple of variants, e.g. "old glass windows have ripples in them because of glass flow".)
The pitfall with this approach is the assumption that old glass started out like new glass. Modern glass windows are usually made by the 'float process': a layer of liquid glass is poured out on top of a bath of molten tin. This results in a very flat, even sheet of glass, which is then cut into smaller pieces as required. This process was invented by Alistair Pilkington, around 1950.
Before that, many other processes were used. Glass might be blown into a cylindrical mold, cut along one side, and unrolled into a sheet - this technique was commonly used before Pilkington's invention. Before that, molten glass was spun into a disc ('Crown process') before being cut into smaller pieces. Many other techniques were also used, but none of them were much good at making flat glass of an even thickness.
The result was that these old glass panes were already uneven when they were fit into these windows. If this unevenness was obvious, it would've made sense for glaziers to install them thick side down for stability. But not all were; many sceptics who went out and checked old windows for themselves found a fair percentage had the thickest edges on the top, or even on the sides.
This is where confirmation bias plays a big part. Humans are very good at remembering the evidence that fits the theory they're trying to prove, and forgetting or playing down that which contradicts it. When we see an apparent pattern - say, we see three windows in a row where the bottom edge is thicker than the top - we read meaning into that pattern. But one in eight observations will turn out that way, just by chance; the other seven people who don't see this 'flow' pattern are much less likely to talk about their experience than the one who did. So the positive evidence is much more widely heard than the negative, and acquires an unearned weight.
What are some other things we could look at? Well, forget medieval glass - we have Roman glassware from two thousand years ago, which does not show noticeable sagging. We have any number of optical devices that require far more precision in thickness than a glass window; while I've seen one claim for sagging in very large telescopes, it didn't come with any cites that would allow me to check it out. And we have ancient pieces of obsidian, volcanic glass, with air bubbles caught in it at its moment of formation; even when that obsidian has since been buried under huge pressures for millions of years, those bubbles don't show the flattening we'd expect in a slow-flowing liquid.
Those who want to believe usually answer these with variants of "well, those are different sorts of glass", and so they are. But these glasses show that it's certainly possible for an amorphous glassy solid to hold shape for thousands and millions of years; that leaves the burden of proof with those who claim that certain other glasses flow at room temperature. If glass really does flow fast enough for effects to be perceived by the naked eye in a windowpane that's a few centuries old, modern techniques should be more than capable of measuring those effects within days in the lab. So far, I'm yet to see a single sourced claim for measurable flow.
Creep
The distinction between solids and liquids is not quite as clear-cut as I've made out above. Temperature measures the average energy of the atoms within a material, but not all atoms have average energy. Once in a while, chance will give a single atom a lot more energy than the average; in these circumstances, it might have enough to break out of its position and not go back to where it was. Material structures are not completely perfect; there will be the odd gap here, an impurity there, and a transition between two different 'grids' that formed from different seeds somewhere else. This sort of thing offers a very limited capacity for atoms to migrate within a 'solid' material - it's a bit like one of those puzzles with fifteen tiles and one empty square, where you have to rearrange the tiles to form a picture.
Over a very long period of time, all materials will slowly deform under stress. The speed of this deformation (known as 'creep') depends greatly on the temperature; physicist Edgar Zanotto (Am. J. Phys, 66(5):392-5, May 1998) estimated that at room temperature, the characteristic flow time for cathedral window glasses would be in excess of 1032 years. So if you're willing to wait a few billion billion times longer than the current lifetime of the universe, perhaps we can say that glass really is a liquid... but then, so are most other 'solids' too. (By comparison, ice flows fairly quickly, as anybody familiar with glaciers will know; this is also a form of creep.)
Further reading
All of these were helpful in writing this, be it by telling me things I didn't already know or by jogging old and rusty memories:
Glass: Liquid or Solid? - Florin Neumann.
Is Glass Liquid or Solid - Philip Gibbs et al.
Amorphous Ice and Glassy Water - Martin Chaplin.
Physicsforums.com discussion
Wikipedia articles on glass, amorphous solids, ice, amorphous ice, ice Ih, creep, and dislocation
sceptics)This is one of those urban myths that just WILL NOT DIE. Every couple of years somebody claims that old windows are thicker at the bottom because glass is a slow-flowing liquid, somebody demolishes it, and a couple of years later it rears its head again. But it's an interesting subject none the less, because the reasons *why* it's so hard to dispel touch on several different aspects of science.
I was originally going to call this 'Glass Is Not A Liquid', but that was a little harder to defend; over very large timescales the distinction between liquid and solid becomes less clear, as discussed below. But I'm happy to assert that for practical purposes, glass is as 'solid' as many other things we have no hesitation in calling solid.
States of matter
Once upon a time, science recognised three phases of matter: solid, liquid, and gas. (Many more have since been discovered - see Wikipedia article for a starting point - but you're not likely to see most of them in your day-to-day life except for plasma and liquid crystals. I'm going to ignore those other phases for this post.)
Liquids and gases are collectively called 'fluids' because they flow under any provocation whatsoever. Apply the slightest force to them, and they'll move out of the way; honey won't move as quickly as water, but they'll move and won't stop until you remove the force (or they have nowhere else to go). Put a liquid or gas in a container, and it will eventually reshape itself to match the shape of the container exactly.
Solids don't like to flow. If you apply a small force to them, they'll give a little way, and no more. No matter how long a fly stands on a gold brick, his feet are never going to sink into that brick; to make a large change to its shape you need a large force, large enough to overcome its material strength.
The difference between these different phases of matter is in how their component parts hang together. Let's start with a nice simple example: iron.
In a slab of cold iron, you have a whole lot of iron atoms tightly packed together by electrical forces that pull them towards one another. If you were to reach into that slab with a very fine pair of tweezers, you'd find it very hard to move one of those atoms around - they're all nestled together closely, so there is no place for that atom to go that isn't already occupied.
If you heat that slab up a bit, the atoms start to jiggle around. (This is what heat is: random jiggling at a microscopic scale.) They pull away from one another, and so the iron expands slightly, but they're not jiggling hard enough to actually break free from their position in the structure; they're like cranky toddlers squirming in a car harness. If you were able to 'tag' a specific atom and then come back to it an hour later, it'd still be in the same position.
Heat it up a lot, though - to about 1538°C - and things change. Some of the atoms are now jiggling hard enough to break out of position and stay out. They're still packed close together, but that packing no longer holds them in place; they're like people shoving past one another in a crowd. Overall, the forces pulling those atoms towards one another are still strong enough to keep the atoms together, so it doesn't expand much when it melts (one well-known solid actually contracts on melting), but that freedom of movement means the grid is no longer permanent; push against it gently for long enough, and it will move aside. At 1540°C, these breaks from position only happen occasionally, and so the liquid is still very thick; as you heat it up they happen more and more often, and it gets easier to pour.
Heat it up further, to 2861°C, and that jiggling becomes so intense that it now overcomes the attractive forces. The atoms fly apart, and they go bouncing off into infinite space. The iron has become a gas, which expands to fill whatever space there is available.
Viscosity
Viscosity describes the 'thickness' of a liquid. The more viscous a liquid is, the slower it pours. Tip a glass of water over and the water falls out instantly; tip a jar of honey, and it'll take a few seconds. So how viscous can a liquid get, and still be a liquid?
As it turns out, very viscous indeed. The classic example is the University of Queensland's pitch experiment: back in 1930 they put a dollop of pitch in a glass container with a hole in the bottom, and left it to flow. It's still there; seven and a half decades later, just eight drops of pitch have fallen into the cup below. To the casual bystander, it's very easy to mistake pitch for a solid. So it's not unreasonable to wonder if glass might also be a very slow-flowing liquid. How else can we distinguish liquids from solids?
Solid structures
If you could look at that cold slab of iron close-up, you'd see the atoms arranged in a regular lattice like cannonballs stacked in a neat pyramid. On a large scale that lattice has a lot of imperfections, but overall it follows a pretty regular structure. As scientists researched further, they discovered that this was pretty much standard for solids. Iron atoms in that slab, sodium and chloride ions in a crystal of table salt, water molecules in a block of ice from the freezer - they all freeze in regular, tesselated crystal structures.
Looking at a material's crystal structure is a much more appealing prospect than sticking it in a container for a decade and watching to see whether it does anything, so it's not too surprising that scientists were eager to define solids this way; a couple of years ago I ran across a modern dictionary that still defined a solid on the basis of a regular crystal structure.
Unfortunately, it doesn't work that way. It turns out that some materials are solid in the not-flowing sense, but don't have a regular, repeating crystal structure - these are known as amorphous solids. One example of this is polystyrene, which is made of very long molecules tangled together like a microscopic hairball. Another, as you might've guessed, is glass, which is made up mostly of silicon and oxygen jumbled together in no particular structure. Cribbed from this UCR site, an image of that:
It's also possible to make amorphous ice. A water molecule is shaped rather like a boomerang. In the normal course of things, when water freezes, it forms a structure where the 'boomerangs' form a regular hexagonal pattern - each end of the boomerang connects to the middle of another boomerang, or to another boomerang-end. This is a nice low-energy arrangement (it keeps like charges away from one another, and opposite charges close together) but it's quite an open structure - there's a lot of empty space in the middle of those hexagonal rings, which is why regular ice is less dense than water.
But water doesn't automatically convert to ice the moment the temperature drops below zero. It needs some sort of 'seed' to get that structure forming. This can be pre-existing ice, or an impurity in the water; without that to start it off, you can cool water as far as -42°C without freezing it into a solid. This is what we call a 'supercooled liquid' - it's under temperature and pressure conditions where its most stable state is solid ice, but it needs some help to reach that state.
If you take that supercooled water and cool it further, very fast - for instance, by spraying droplets of it into a cold gas - you freeze it into a solid; the molecules no longer have the energy to shove past their neighbours and move around. But it doesn't have time to rearrange itself into that neat crystalline structure. Instead you get a dense, amorphous solid; if you took a microscopic snapshot you'd have great difficulty distinguishing it from liquid water, because the liquid's random jumble of molecules has been frozen into the solid. This is pretty much the same thing that happens in glass, and indeed this particular form of ice is known as 'hyperquenched glassy water'.
(In fact, there are at least a dozen different forms of ice known, including several different amorphous forms, and even at least one alternate form of liquid water; for more info, see Wikipedia articles on ice and amorphous ice. From here on, though, unless I qualify 'ice' I'm talking about the regular kind.)
Because regular glass is of the same sort of nature as HGW, it has been referred to as a 'frozen supercooled liquid'; unfortunately, the 'frozen' is often forgotten when people hear the 'liquid'.
Discontinuous properties
When you pump energy into a block of ice, its temperature goes steadily up... and up... and then, at 0°C, it stops. It stays at 0°C for quite some time, until the ice has entirely melted away into water; then it starts to rise again. If you were to plot temperature on the x-axis, and energy on the y-axis, you'd see an abrupt jump in energy as the temperature goes from -0.1°C to 0.1°C. Many other properties - density, optical properties etc. etc. - change abruptly at this point. Because ice has a regular structure, the energy (& hence, temperature) required to break its bonds at one point is the same as that required to break corresponding bonds anywhere else in that structure; at -0.1°C they hold, at 0.1°C they're broken and you just have a jumble. It's like a workplace where everybody knocks off at 5 pm sharp - one minute it's business as usual, the next there's nobody there.
Not so with glasses. Because of that disorder, every atom in a glass is in its own unique situation, and so it takes a different amount of energy to break free from its position. The result of this is that at around 1000 °C (depending on exact composition), some of the atoms start to get a little freedom... but others are still caught up in their structure. It's more like my workplace, where some people head home at 4 pm and others stay until 6 or 7; the transition is much more gradual. You don't see that sharp jump in material properties against temperature. Cooling it down again, it works the same way - some bonds reform before others, and the freezing point is not well-defined, which leads many people to wonder - if we don't know where it becomes a solid, are we sure it actually did?
Experiment
The scientist's favourite answer, of course, is 'try it and see'. Since the supposed rates of flow are very slow indeed, setting up an experiment from scratch isn't very convenient (though I'm told one engineer at Dow Corning did run a 27-year experiment on whether glass rods subjected to a bending force experienced flow; anybody who can find me a good reference for this wins an imaginary gold sticker). Instead, most people start out by looking at old glassware, leading to the most famous piece of 'evidence' for glass flow:
If you look at old glass windows, they're thicker on the bottom than at the top, because the glass has flowed downwards. (And a couple of variants, e.g. "old glass windows have ripples in them because of glass flow".)
The pitfall with this approach is the assumption that old glass started out like new glass. Modern glass windows are usually made by the 'float process': a layer of liquid glass is poured out on top of a bath of molten tin. This results in a very flat, even sheet of glass, which is then cut into smaller pieces as required. This process was invented by Alistair Pilkington, around 1950.
Before that, many other processes were used. Glass might be blown into a cylindrical mold, cut along one side, and unrolled into a sheet - this technique was commonly used before Pilkington's invention. Before that, molten glass was spun into a disc ('Crown process') before being cut into smaller pieces. Many other techniques were also used, but none of them were much good at making flat glass of an even thickness.
The result was that these old glass panes were already uneven when they were fit into these windows. If this unevenness was obvious, it would've made sense for glaziers to install them thick side down for stability. But not all were; many sceptics who went out and checked old windows for themselves found a fair percentage had the thickest edges on the top, or even on the sides.
This is where confirmation bias plays a big part. Humans are very good at remembering the evidence that fits the theory they're trying to prove, and forgetting or playing down that which contradicts it. When we see an apparent pattern - say, we see three windows in a row where the bottom edge is thicker than the top - we read meaning into that pattern. But one in eight observations will turn out that way, just by chance; the other seven people who don't see this 'flow' pattern are much less likely to talk about their experience than the one who did. So the positive evidence is much more widely heard than the negative, and acquires an unearned weight.
What are some other things we could look at? Well, forget medieval glass - we have Roman glassware from two thousand years ago, which does not show noticeable sagging. We have any number of optical devices that require far more precision in thickness than a glass window; while I've seen one claim for sagging in very large telescopes, it didn't come with any cites that would allow me to check it out. And we have ancient pieces of obsidian, volcanic glass, with air bubbles caught in it at its moment of formation; even when that obsidian has since been buried under huge pressures for millions of years, those bubbles don't show the flattening we'd expect in a slow-flowing liquid.
Those who want to believe usually answer these with variants of "well, those are different sorts of glass", and so they are. But these glasses show that it's certainly possible for an amorphous glassy solid to hold shape for thousands and millions of years; that leaves the burden of proof with those who claim that certain other glasses flow at room temperature. If glass really does flow fast enough for effects to be perceived by the naked eye in a windowpane that's a few centuries old, modern techniques should be more than capable of measuring those effects within days in the lab. So far, I'm yet to see a single sourced claim for measurable flow.
Creep
The distinction between solids and liquids is not quite as clear-cut as I've made out above. Temperature measures the average energy of the atoms within a material, but not all atoms have average energy. Once in a while, chance will give a single atom a lot more energy than the average; in these circumstances, it might have enough to break out of its position and not go back to where it was. Material structures are not completely perfect; there will be the odd gap here, an impurity there, and a transition between two different 'grids' that formed from different seeds somewhere else. This sort of thing offers a very limited capacity for atoms to migrate within a 'solid' material - it's a bit like one of those puzzles with fifteen tiles and one empty square, where you have to rearrange the tiles to form a picture.
Over a very long period of time, all materials will slowly deform under stress. The speed of this deformation (known as 'creep') depends greatly on the temperature; physicist Edgar Zanotto (Am. J. Phys, 66(5):392-5, May 1998) estimated that at room temperature, the characteristic flow time for cathedral window glasses would be in excess of 1032 years. So if you're willing to wait a few billion billion times longer than the current lifetime of the universe, perhaps we can say that glass really is a liquid... but then, so are most other 'solids' too. (By comparison, ice flows fairly quickly, as anybody familiar with glaciers will know; this is also a form of creep.)
Further reading
All of these were helpful in writing this, be it by telling me things I didn't already know or by jogging old and rusty memories:
Glass: Liquid or Solid? - Florin Neumann.
Is Glass Liquid or Solid - Philip Gibbs et al.
Amorphous Ice and Glassy Water - Martin Chaplin.
Physicsforums.com discussion
Wikipedia articles on glass, amorphous solids, ice, amorphous ice, ice Ih, creep, and dislocation
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2006-02-02 06:12 am (UTC)Edie
no subject
Date: 2006-02-02 06:22 am (UTC)no subject
Date: 2006-02-02 01:43 pm (UTC)You also write well. Perhaps a book is in order?
no subject
Date: 2006-02-02 01:53 pm (UTC)no subject
Date: 2006-02-02 06:39 am (UTC)no subject
Date: 2006-02-02 07:12 am (UTC)no subject
Date: 2006-02-02 12:14 pm (UTC)And incidentally, glass does sag a little, which is why all the big telescopes are reflectors (with big mirrors underneath the glass as support) rather than refractors.
no subject
Date: 2006-02-02 01:37 pm (UTC)Yup, and likewise you'd see the glass at the bottom flattening out against the sill, which also doesn't happen.
And incidentally, glass does sag a little
Sag in the 'bend a little under gravity, and stay the same degree of bent' sense, or in the 'bend, and keep on bending further and further over time' sense?
I always thought the main reason why the biggest telescopes were reflectors was chromatic aberration - different wavelengths refract at slightly different angles, but all reflect equally?
no subject
Date: 2006-02-02 01:41 pm (UTC)no subject
Date: 2006-02-02 01:35 pm (UTC)Well done.
no subject
Date: 2006-02-02 01:49 pm (UTC)Give it another 50 years, maybe...
no subject
Date: 2006-02-02 01:52 pm (UTC)no subject
Date: 2006-02-02 03:17 pm (UTC)no subject
Date: 2006-02-02 03:37 pm (UTC)no subject
Date: 2006-02-02 10:04 pm (UTC)no subject
Date: 2006-02-02 04:04 pm (UTC)