Hacker News new | past | comments | ask | show | jobs | submit login
Five glassy mysteries we still can’t explain (physicsworld.com)
103 points by sohkamyung on June 10, 2022 | hide | past | favorite | 51 comments



Well, everything was going great in the article until I got to this part in the first "Flowing Myth" example:

>At this rate, a piece of common soda-lime glass would take aeons to slowly flow and turn into the more energetically favourable crystalline sodium dioxide – otherwise known as quartz.

Quartz is not sodium dioxide - NaO2, it is silicon dioxide - SiO2. Sodium dioxide is probably not stable and after looking it up the most stable form is probably sodium peroxide - Na2O2. This guy forgot all about the sand he used to make the glass. The sodium part of the deal, sodium carbonate or soda ash - Na2CO3, was used to regulate temperature as he described and the lime was produced from local limestone, CaCO3.

This is a Physics World article and I would've expected someone to proof-read it specifically for stuff like this.

In reference to the Flowing Myth, I first heard that myth in a geoscience classroom and it was stated as a fact that old glass is thicker at the bottoms of cathedral windows because it has slowly flowed.

Taking that further, I guess it is a good thing that these cathedral windows were made of many panes of this old flowing glass since the thickening would not dramatically change the graphic display. If they were made of a single pane and it flowed significantly over time, Jesus may have started in medieval times as a finely cut specimen of manhood and ended up in modern times with a noticeable middle-age spread.


That is not all. It repeats the falsehood: The stability of glass is one of its most attractive characteristics, for example in the storage of nuclear waste.

Vitrified nuclear waste, immersed, would become porous and friable long before its radioactivity decayed significantly.

It also claims that we don't have proof glass doesn't flow in historical time-frames, where in fact we have glass goblets from thousands of years ago.

It further attributes tapering in glass in windows as indicating poor quality, where in fact that was just a feature of how flat glass was made. Window makers oriented glass thicker edge down because that was the sturdier way to make things.

Until quite recently, flat glass was made by blowing a big bottle, and centering it on a turntable while still soft. Spinning, the mouth widened until the whole thing flopped out flat, with a characteristic "whomp!", the bottle mouth becoming the rim.


> Vitrified nuclear waste, immersed, would become porous and friable long before its radioactivity decayed significantly.

Not quite. Silicates form a passive gel layer when put in water, they are quite resistant to leaching (we have examples of ancient Egyptian glass items recovered from the sea). Over very long time frames, way longer than millennia, the glass just dissolves. But for this you’d need them to be immersed and constant water circulation; silica is very stable. In this case you’re not worried about radionuclides release too much anyway, as the seas tend to be very large and have already quite a bit of them. Plus, only the low-activity isotopes remain isotopes after that time.


Radionuclide glasses should be expected to be less stable than ordinary silicates.


It depends on what you call “ordinary silicates”. The matrix is mostly sodium borosilicate with a bunch of additives and is as stable as a glass could be.

There are precipitates caused by some of the waste elements (things like ruthenium, rhodium and palladium, for example). They have issues but still quite far from making the glass crumble. As with all glasses, crystallisation is an open question but won’t happen for a very long time either.


Yes. If glass flowed precision optics would stop working after a few years


>we have glass goblets from thousands of years ago

I think you'd also need to verify what it looked like thousands of years ago and whether it was stored in a way that held a consistent flow direction, if you wanted to use that as evidence.


> Vitrified nuclear waste, immersed, would become porous and friable long before its radioactivity decayed significantly.

Eh, that totally depends on what's your radioactive material. They don't all decay at the same rate, you know?


I think you are a bit harsh on the author here. Sodium and silicon are very easy to mix up when writing, especially in an article where you talk about both quite often. It's also easy to overlook in proof reading because it sort of make sense in the sentence.

Regarding the flowing myth, I also heard the myth talking about cathedral glass planes being thicker at the bottom. There is a much more mundane reason for that though, it's much easier to handle a pane that is heavier at the bottom than the top, thus they were made that way. The myth is also debunked by the fact that there exist examples of cathedral glass panes which were made with the thick end up and haven't changed since then.


I wasn't trying to be harsh on the author. I thought about this late last night and decided that the easiest way to explain the error could be that the author was employed by Physics World to compose interesting articles in layman's terms and that the author does not have a technical background and those persons at Physics World who were supposed to proof-read this work dropped the ball.

In checking the byline you can see that the author is a freelancer and if you follow the link to their site you can see that they do a lot of technical writing for multiple publications on a variety of subjects. It is possible that they had a tight deadline for article submission, more articles to write, etc so that this one was dumped to the machine just in time for publication but without enough time for a good review that would've spotted this slight error.


> I guess it is a good thing that these cathedral windows were made of many panes of this old flowing glass

This is a myth. These are not “flowing glasses”. Silicate glasses have relaxation times in the billion years (the euphemistic “eons” in the article). There is no way medieval glass has had enough time to noticeably flow.


I'm actually a geophysicist and I understand that the glass doesn't flow. As I read the article and thought of the potential implications I was reminded of the old cartoon where the middle-aged man with a beer gut or paunch sees an attractive woman approaching so he stiffens his back, stands up straight and puffs out his chest as she passes so he can appear younger and more fit and virile. Once she has passed he lets gravity take control again.

I set up that scenario (or joke as has been noted by another reply) with the idea that detecting a middle-aged spread in an old window like that would be nearly impossible if the stained glass used to create the image, Jesus in this case, was constructed from a series of small pieces of glass constrained by a framework of lead caming as is used in stained glass construction. In this case, all the flow, if it occurred at all would be constrained at multiple points in the framework so that the image would remain nearly identical over the aeons as long as the framework existed in good condition.

However, if glass did indeed flow and if that flow was detectable over the few hundred years that have passed since those cathedrals were constructed and also if each window was made from a single sheet of this glass with an image painted on the glass - then the image itself could be expected to exhibit noticeable changes over significant time periods and it would naturally follow that young Jesus would eventually become middle-aged Jesus with a paunch, like the guy in the cartoon.

It was late at the time and my mind was wandering again, punching at the lines while trying to leak some color to the outside.


He was joking...


Yes. Thanks.


The Feynman Lectures on Physics offers a simple dimensional argument for why friction acts like it does, and then says that the details of the coefficient of friction are beyond our ability to estimate from first principles.

It then offers a simple experiment to illustrate that the properties have to do with surface impurities. If you slide glass over dry glass, it slides easily. Do the same with wet glass and you can feel a "gritty" motion. This is due to the water lifting the impurities off of the glass and allowing a direct glass on glass contact, at which point they fuse and have to be broken. You can see where this happened because it leaves visible scratches.

I found this unbelievable and had to test it. My mother was not thrilled to see that her good crystal got visible scratches on it...


>This is due to the water lifting the impurities off of the glass and allowing a direct glass on glass contact, at which point they fuse and have to be broken.

I don't believe this. I think it is more due to dirt on the glass that gets mobilised by the water. occam's razor...


That was his explanation.

Not sure how your idea causes scratches to be left on the glass.


Some small tidbit as well: The process of bending and shaping glass, especially single-layer safety glass, e.g. used in automotive rear and side glazing, is still an iterative process for every new model of car, relying mostly on the gut feeling of the operators of the bending and tempering ovens. No hard and fast rules exist, and a lot is just passed on by experience.


Wow. I just sort of imagined they were somehow molded or vacuum formed in some sort of automated fashion by now.


My first mystery for glass, is why is it transparent? Surely random as opposed to crystalline ordering would be better at stopping photons?

Turns out it is about electron excitation levels in glass - the gaps between levels are too large to absorb the energy from a photon, so they pass right through!


If you're intrigued by band gaps... get some cheap synthetic Ruby (you can get a 1.5 carets for $40-50) and see what happens when it gets hot/cold.

Ruby is just Sapphire (Aluminum Oxide) with Iron impurities that make it red. It is red because blue light (everything above the band gap) is absorbed!

When you cool the ruby (e.g. in liquid nitrogen) it will turn pink/orange as it shrinks and the absorption band moves to shorter wavelengths. When you heat it (e.g. with a propane torch) it will turn deep red to black as the crystal structure expands and the absorption band moves up in wave length from red into the IR. Remove the heat/cold and watch it slowly come back to the ruby color you expect!


thats a classic matter in materials science and engineering !


SiO2 glass happens to be transparent to visible wavelengths in particular, which is one of the reasons we use it. But it’s not nearly as transparent to UV and also becomes quickly opaque as you go from near IR to longer wavelengths. In "thermal" IR, glass is actually a pretty good mirror.


I enjoy looking at different glasses with my polarized prescription sunglasses. They reveal all kinds of interesting distortions depending on how the glass was formed.


You are not far from the truth.

Interatomic forces tend to create energy levels with differences on the infrared band. And glass is pretty opaque on the infrared.

But yes, the visible light features are usually caused by electron excitation.


I’d love it if someone can direct to an explanation accessible to a non physics college graduate that actually explains why glass (and other things like quartz and even water/air) are transparent, how exactly does light pass through them that also makes them that causes refraction and what exactly makes them reflect light as well.


The oscillating electric field of the light wave induces a sympathetic oscillation of the electric charges bound to atoms or molecular bonds in the material. The degree to which those charges can align themselves to the field is a characteristic of each material. Taking that polarizability into account in Maxwell's Equations yields a wave equation for electromagnetism, just like it does in a vacuum, but with a different propagation velocity, slower than the speed of light. In a vacuum, there's a corresponding polarizability of free space, related to the speed of light. In undergrad electrodynamics, prior to learning about relativity or quantum, we didn't try to explain it any further than that.

It's transparent because the electrons stay attached to their atoms. If they could break free, then conduction would occur, and another solution of Maxwell's equations shows the wave being quenched. You could imagine conduction causing loss of energy via heating, like the way an electric stove works. If the electrons could absorb energy by making a quantum transition, some of energy could be released as heat rather than light.

Refraction and reflections could be thought of as necessary consequences of continuity at an interface between materials of different composition. A number of things related to electric and magnetic field vectors have to remain the same at both sides of the boundary. This is akin to a conservation law, and requires introduction of a third wave at the boundary -- the reflected wave -- to make everything balance.

Another way to look at refraction is via Huygen's Principle, where each point on a wave front is a source of a spherical wave front, and accounting for the different speeds in the two materials (e.g., air and glass) gives the law of refraction. And there's even yet another explanation which can be found by figuring out the path that the light takes, which results in a minimum for the time of travel. It's interesting that the law of refraction was worked out before electromagnetism was understood.

I'm an industrial physicist, and this is hand-wavy, but I'm trying for an explanation that doesn't require equations.


I have seen a few casual explanations of it and I still don’t really understand it.


Things are transparent that don't absorb, scatter or reflect the light. For stuff like quartz the band gap (energy above which photons get absorbed) is higher than the energy of visible light so the light just goes through and its transparent... Are you asking why certain materials don't absorb visible light?


In general, for the substances that do not contain atoms/ions/molecules which absorb the visible light (i.e. most substances which contain neither transitional metals nor organic molecules with multiple bonds, and which are neither metals nor semi-metals, which have free electrons, nor semiconductors with very small bandgap, where the visible light creates free electrons), the condition to be transparent is that the substance must be in one of the following states: gas, liquid, glass (i.e. solid obtained by very fast cooling) or single-crystal (i.e. solid obtained by very slow cooling).

The reason is that in all these states, where the substance is fluid, amorphous or a single crystal, the substance is homogeneous, all the points inside it have the same properties, there are no internal boundaries, so light propagates in straight lines inside the substance, so light can pass through the substance, with only a pair of refractions on the external surface, when the light enters and exits the substance. Thus the substance is transparent.

Most solids, when they are neither cooled so fast to form an amorphous glass nor cooled so slow to form a single crystal, form at cooling a polycrystalline solid, e.g. like any rock or stone, which is made of a huge number of very small crystal grains.

In a polycrystalline solid, there is a huge number of internal boundaries, which refract and reflect the light. Therefore the light cannot propagate in straight lines, but in extremely long and intricate trajectories. If the substance is extremely pure, so nothing absorbs the light, it will be translucent, i.e. the light eventually gets out but you can not see any image of what is behind. Normal substances have some impurities that absorb the light. Due to the extremely long path of the light inside a polycrystalline substance, it will be eventually absorbed, so the substance is opaque.

So the difference between air, water, common glass, and quartz crystals versus most solids (which are neither amorphous nor single crystals) is that the former do not have internal boundaries, so the propagation of light is affected only at the external surface, while the latter have a huge number of internal boundaries, which reflect and refract the light.

The reason why the light is refracted when passing through a boundary between two different media can be understood without any knowledge about the nature of light, except that it behaves like a wave.

The light waves have different speeds on the 2 sides of the boundary. From the condition of continuity in time at a point on the boundary it results that the frequency of the waves must be the same in both sides, therefore the wavelengths must be different. From the condition of the continuity of the wavelength measured along the boundary on its 2 sides, it results that there must be an angle between the wave propagation directions on the 2 sides.

To explain why there is also reflection at a boundary, that is more complex, because the explanation needs some model of how light is absorbed and emitted and that depends on the chemical nature of a substance (some, like metals, reflect better than others).

Some simplified explanation is at

https://en.wikipedia.org/wiki/Reflection_(physics)#Mechanism


The "slumping glass" meme is surprisingly old: my mum passed it on to me, she a trained architectural historian.

Metals as glass, sintered metal objects, glass as a state not a chemistry (recipe) per se, fascinating stuff.

"Transparent aluminium" wasn't so movie crazy after all


Aluminium oxynitride, which is in fact transparent, was dubbed "transparent aluminum" after Star Trek. Fun stuff.

I wonder how many inventions are inspired or named after that show. "The Next Generation" particularly featured touch screens and tablet computers long before those things were common. And original series communicators bear more than a passing resemblance to cellphones.


Star Trek's PADDs are also interesting for what they don't show. You'll see many instances of characters giving PADDs to each other. There's one particular scene in an episode of DS9 where Sisco has a whole stack of PADDs on his desk, and he laments the workload. Star Trek figured out touch screens and personal data devices, but didn't figure out basic networked data communication, i.e. email attachments. At a time when email existed!

I think of this every time I hand my phone to my wife to show her some stupid joke on Twitter or in a text message thread with my friends.


>Star Trek figured out touch screens and personal data devices, but didn't figure out basic networked data communication, i.e. email attachments. At a time when email existed!

I think it's more that they wanted to convey, to the viewer, that information was being passed from place to place, or have stacks of PADDs serve as easy visual metaphors for "piles of paperwork" or what have you. Star Trek isn't actually trying to extrapolate realistic future technology, it's telling dramatic stories In Space, and the needs of the narrative always come first.


"The needs of the narrative outweigh the needs of the factual" sounds like a spock-kirk dialogue in the high rad chamber scene


Yet the GP still passes his phone around when there's something interesting in it.

I imagine that if they were cheaper, we would do exactly what TNG shows.


Jinx!


Try to imagine how you would visually indicate a big workload without the stack. Kinda dumb, but dramatically effective, in its way.



If you could replicate tablet computers essentially for free, I wonder if in fact the concept of multitasking would fall by the wayside. Why bother switching between apps if you can just make another tablet.


It's kinda weird they didn't just use sapphire as "transparent aluminum". AlOxNy is actually somewhat harder (and more difficult to etch) than sapphire, but weird any way.


> "The Next Generation" particularly featured touch screens and tablet computers long before those things were common

And called them PADDs amusingly enough.


"Personal Access Display Device" for the mildly curious.


> Even more surprising, though, is the onset of glass behaviour in certain computer algorithms. For instance, if an algorithm is designed to seek out progressively better solutions to a problem with a large number of variables it can become overwhelmed by complexity and grind to a halt before the optimal solution is found. By borrowing statistical methods designed for the fundamental study of glasses, however, such algorithms can be improved, and better solutions found.

Are they talking specifically about simulated annealing here, or is this a more general observation?


Pretty sure the former


For anyone curious I believe the statistical method referred to in the last segment is annealing[1]

[1] https://en.wikipedia.org/wiki/Simulated_annealing


To explain a mystery is to tell a story. A story about a thing that joins up nicely with our stories about everything else.

The story is important. More important than the actual thing maybe.

I think it's so important because the story can be kept and carried around, like money. And it can be communicated. Translated into the noosphere or whatever. And everybody likes the noosphere.


This sounds like it was generated by GPT-3.


I think that the popular opinion is that the noosphere (the collective body of ideas) is realer than the physical world. And that the idea about a thing is realer than the thing itself.

Opinion to the contrary is offered only by scientists, artists and similar weirdos and social outcastes.


I always get responses like that when I talk like this. Maybe your reply was generated too.


I guess because it seems so entirely content-free. It's riffing entirely on the title of the post — the words "explain" and "mystery" — and not on the actual subject, which is glass. All the other (human, presumably) commentators responded to the subject, and talked about glass.

It's as if I said to you "Guess what? I just landed the job of my dreams!" and you replied "Guessing is an interesting pastime, people have been doing that for thousands of years."

To an AI it might seem like your comment followed mine, but a human would walk away from you.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: