4 ms·

To make the perfect mirror, physicists confront the mystery of glass

77 points by sohkamyung 7y ago

moonliscuits 7y ago
That was pretty interesting -- although a bit short. Happen to know of something that's a bit more in-depth about the two-level motion of glass?
- magicalhippo 7y ago
  Wonder if it's similar to oscillators[1] in Game of Life. I.e. a some molecules have two different arrangements with same energy, so they can swap between them by absorbing a bit of thermal energy or something. [1]: https://www.conwaylife.com/wiki/Toad https://www.conwaylife.com/wiki/Toad
- sohkamyung 7y ago
  There's a linked article about the hunt for ideal glass that includes some details about 'two-level systems'. [1] [1] https://www.quantamagazine.org/ideal-glass-would-explain-why-glass-exists-at-all-20200311/ https://www.quantamagazine.org/ideal-glass-would-explain-why...
- Phaune421 7y ago
  Turns out that this was my PhD thesis subject. To put it simply two-level systems (TLS) are groups of atoms in a disordered solid that present two stable configurations, that is two physical states of similar potential energy separated by a small energy barrier. The barriers need to be small for it to be overcome by thermal fluctuation alone and thus for the TLS to resonate at a frequency close to the frequency of interest. If you are interested by what exactly are these TLSs at the atomic level, have a look at this paper: https://journals.aps.org/prb/pdf/10.1103/PhysRevB.97.014201 https://journals.aps.org/prb/pdf/10.1103/PhysRevB.97.014201
metaphor 7y ago
> ...Ediger and his team dropped molecules one by one onto a surface as if they were Tetris pieces, allowing them to find snug fits. Let's assume a 40kg super mirror was made from the same base silica[1]/tantala[2] substrate as the legacy mirror. Let's also assume 50% distribution by mass. That's a (20kg/60.08g/mol)+(20kg/441.893g/mol) ~= 378.1mol tetris endgame. So if we took this analogy quite literally and wanted to build 1 mirror in 10 years, you'd have to play precision atomic tetris a mere 17 orders of magnitude faster than this guy[3]. Not going to lie...sure does make watching grass grow sound a lot more pragmatic. [1] https://en.wikipedia.org/wiki/Silicon_dioxide https://en.wikipedia.org/wiki/Silicon_dioxide [2] https://en.wikipedia.org/wiki/Tantalum_pentoxide https://en.wikipedia.org/wiki/Tantalum_pentoxide [3] https://www.youtube.com/watch?v=v3TJXSFnVS4 https://www.youtube.com/watch?v=v3TJXSFnVS4
- mrob 7y ago
  Most of the mass is in the supporting base layer, which presumably does not contribute much noise. If this layer also needs to be special glass then why are the LIGO scientists even considering using it? I assume they are not idiots, and only the wavelength-thick layers need to be assembled molecule by molecule.
- ipnon 7y ago
  Quanta Magazine is not great reading for the practical and expeditious.
  - JadeNB 7y ago
    Why not? I've found Quanta Magazine great reading for pretty much everyone. All I can judge as an expert is their articles on mathematics, and they're not technical reports, but a howler in a Quanta math article is rare, whereas in any other popular article I would expect to find at least one major error per statement.
    - craftinator 7y ago
      Quite sensationalistic. They go out of their way to obfuscate titles.
at_a_remove 7y ago
I have often thought that technologically advanced civilizations would not have FTL travel or transporters or the like, rather, they would be masters of specced-out materials engineering, producing items by placing just the right atom here and a different one over there to uniquely suit the purpose of whatever it was.
- ape4 7y ago
  If you could do that... what are some other possible products?
  - izzydata 7y ago
    Replicators, brain backups maybe. Probably plenty of seemingly sci-fi technologies.
- quesera 7y ago
  Popular treatment: Engines of Creation, Drexler 1986. There are newer options, but this is basically the beginning of post-Feynman Nanotechnology. Version 2.0 (2007) of the book is a free PDF: https://web.archive.org/web/20140810022659/http://www1.appstate.edu/dept/physics/nanotech/EnginesofCreation2_8803267.pdf https://web.archive.org/web/20140810022659/http://www1.appst...
- N1H1L 7y ago
  Technically we already do that. Molecular beam epitaxy is layer by layer growth of crystals, some of which like superlattices can only be engineered from the ground up. Source: My PhD is in Materials Science
  - sunstone 7y ago
    I do this in my sleep. But then, so does everyone else.
NikolaeVarius 7y ago
For some cool stuff about mirror development, I recommend reading up on the Hale telescope. https://en.wikipedia.org/wiki/Hale_Telescope https://en.wikipedia.org/wiki/Hale_Telescope Some massive improvements to mirror making techniques were derived from the construction.
newfeatureok 7y ago
Reminds of me the Three Body Problem series. The "droplet" was a perfect mirror because a mystery reflective material was tied together using the strong force. So I guess that means the "perfect" mirror must be bound by strong forces.
ebg13 7y ago
If the ideal state is perfectly flat and the proposed mechanism is to place individual atoms one at a time, why are they still talking about glass and not crystal lattice sheets? > ideal glass are theoretically packed together in the densest possible random arrangement Why would you want it to be random at all?
- VorticesRcool 7y ago
  You would like it to have a lot of entropy. It's not about being random in the sense if being up to chance, it's about making the interactions between the electrons so complicated that any passing photon does not get absorbed or does not produce stimulated emission. When there is enough entropy/complexity/randomness the electrons have no state that they can jump to of a similar range of energy to the frequencies of light that you would like the glass to be transparent to.
mannykannot 7y ago
I was wondering whether a design using total internal reflection within 90-degree prisms at each end of the arms would be feasible for an instrument of this precision. I assume it is not, and I have a few guesses as to why not: 1) the Fabry-Perot resonator arrangement of mirrors in each arm creates a much higher intensity between the mirrors, and this is important for getting a measurable signal out; 2) the resonator rejects a lot of noise; 3) there's no good way to get power into, and a signal out of, an arrangement of prisms; 4) an arrangement of prisms puts too much glass in the path. I would be interested in hearing what the actual answer is. Update: there's a lot more information here (though it is about the Advanced LIGO concept), but I don't think it addresses the failings of a prism arrangement directly: https://iopscience.iop.org/article/10.1088/0264-9381/34/1/015001/ampdf https://iopscience.iop.org/article/10.1088/0264-9381/34/1/01...
- chris_va 7y ago
  AR coatings are not perfect at the interface boundary (kind of the same problem as a perfect mirror in reverse), plus glass has attenuation.
  - mannykannot 7y ago
    I get the point about glass attenuation, but with regard to AR (anti-reflection?) coatings, are you referring to the surface through which light enters the prisms? If those surfaces are normal to the laser beam, would the reflected light not still be contained within the beam path, just as it is in the case of a Fabry-Perot resonator?
    - chris_va 7y ago
      Yes, but at the wrong path length (and offset by some vector, if you had a TIR cube)

77 points by sohkamyung 7y ago

That was pretty interesting -- although a bit short. Happen to know of something that's a bit more in-depth about the two-level motion of glass?

magicalhippo 7y ago
Wonder if it's similar to oscillators[1] in Game of Life. I.e. a some molecules have two different arrangements with same energy, so they can swap between them by absorbing a bit of thermal energy or something. [1]: https://www.conwaylife.com/wiki/Toad https://www.conwaylife.com/wiki/Toad
sohkamyung 7y ago
There's a linked article about the hunt for ideal glass that includes some details about 'two-level systems'. [1] [1] https://www.quantamagazine.org/ideal-glass-would-explain-why-glass-exists-at-all-20200311/ https://www.quantamagazine.org/ideal-glass-would-explain-why...
Phaune421 7y ago
Turns out that this was my PhD thesis subject. To put it simply two-level systems (TLS) are groups of atoms in a disordered solid that present two stable configurations, that is two physical states of similar potential energy separated by a small energy barrier. The barriers need to be small for it to be overcome by thermal fluctuation alone and thus for the TLS to resonate at a frequency close to the frequency of interest. If you are interested by what exactly are these TLSs at the atomic level, have a look at this paper: https://journals.aps.org/prb/pdf/10.1103/PhysRevB.97.014201 https://journals.aps.org/prb/pdf/10.1103/PhysRevB.97.014201

metaphor 7y ago

> ...Ediger and his team dropped molecules one by one onto a surface as if they were Tetris pieces, allowing them to find snug fits. Let's assume a 40kg super mirror was made from the same base silica[1]/tantala[2] substrate as the legacy mirror. Let's also assume 50% distribution by mass. That's a (20kg/60.08g/mol)+(20kg/441.893g/mol) ~= 378.1mol tetris endgame. So if we took this analogy quite literally and wanted to build 1 mirror in 10 years, you'd have to play precision atomic tetris a mere 17 orders of magnitude faster than this guy[3]. Not going to lie...sure does make watching grass grow sound a lot more pragmatic. [1] https://en.wikipedia.org/wiki/Silicon_dioxide https://en.wikipedia.org/wiki/Silicon_dioxide [2] https://en.wikipedia.org/wiki/Tantalum_pentoxide https://en.wikipedia.org/wiki/Tantalum_pentoxide [3] https://www.youtube.com/watch?v=v3TJXSFnVS4 https://www.youtube.com/watch?v=v3TJXSFnVS4

mrob 7y ago
Most of the mass is in the supporting base layer, which presumably does not contribute much noise. If this layer also needs to be special glass then why are the LIGO scientists even considering using it? I assume they are not idiots, and only the wavelength-thick layers need to be assembled molecule by molecule.
ipnon 7y ago
Quanta Magazine is not great reading for the practical and expeditious.
- JadeNB 7y ago
  Why not? I've found Quanta Magazine great reading for pretty much everyone. All I can judge as an expert is their articles on mathematics, and they're not technical reports, but a howler in a Quanta math article is rare, whereas in any other popular article I would expect to find at least one major error per statement.
  - craftinator 7y ago
    Quite sensationalistic. They go out of their way to obfuscate titles.

at_a_remove 7y ago

I have often thought that technologically advanced civilizations would not have FTL travel or transporters or the like, rather, they would be masters of specced-out materials engineering, producing items by placing just the right atom here and a different one over there to uniquely suit the purpose of whatever it was.

ape4 7y ago
If you could do that... what are some other possible products?
- izzydata 7y ago
  Replicators, brain backups maybe. Probably plenty of seemingly sci-fi technologies.
quesera 7y ago
Popular treatment: Engines of Creation, Drexler 1986. There are newer options, but this is basically the beginning of post-Feynman Nanotechnology. Version 2.0 (2007) of the book is a free PDF: https://web.archive.org/web/20140810022659/http://www1.appstate.edu/dept/physics/nanotech/EnginesofCreation2_8803267.pdf https://web.archive.org/web/20140810022659/http://www1.appst...
N1H1L 7y ago
Technically we already do that. Molecular beam epitaxy is layer by layer growth of crystals, some of which like superlattices can only be engineered from the ground up. Source: My PhD is in Materials Science
- sunstone 7y ago
  I do this in my sleep. But then, so does everyone else.

NikolaeVarius 7y ago

For some cool stuff about mirror development, I recommend reading up on the Hale telescope. https://en.wikipedia.org/wiki/Hale_Telescope https://en.wikipedia.org/wiki/Hale_Telescope Some massive improvements to mirror making techniques were derived from the construction.

newfeatureok 7y ago

Reminds of me the Three Body Problem series. The "droplet" was a perfect mirror because a mystery reflective material was tied together using the strong force. So I guess that means the "perfect" mirror must be bound by strong forces.

ebg13 7y ago

If the ideal state is perfectly flat and the proposed mechanism is to place individual atoms one at a time, why are they still talking about glass and not crystal lattice sheets? > ideal glass are theoretically packed together in the densest possible random arrangement Why would you want it to be random at all?

VorticesRcool 7y ago
You would like it to have a lot of entropy. It's not about being random in the sense if being up to chance, it's about making the interactions between the electrons so complicated that any passing photon does not get absorbed or does not produce stimulated emission. When there is enough entropy/complexity/randomness the electrons have no state that they can jump to of a similar range of energy to the frequencies of light that you would like the glass to be transparent to.

mannykannot 7y ago

I was wondering whether a design using total internal reflection within 90-degree prisms at each end of the arms would be feasible for an instrument of this precision. I assume it is not, and I have a few guesses as to why not: 1) the Fabry-Perot resonator arrangement of mirrors in each arm creates a much higher intensity between the mirrors, and this is important for getting a measurable signal out; 2) the resonator rejects a lot of noise; 3) there's no good way to get power into, and a signal out of, an arrangement of prisms; 4) an arrangement of prisms puts too much glass in the path. I would be interested in hearing what the actual answer is. Update: there's a lot more information here (though it is about the Advanced LIGO concept), but I don't think it addresses the failings of a prism arrangement directly: https://iopscience.iop.org/article/10.1088/0264-9381/34/1/015001/ampdf https://iopscience.iop.org/article/10.1088/0264-9381/34/1/01...

chris_va 7y ago
AR coatings are not perfect at the interface boundary (kind of the same problem as a perfect mirror in reverse), plus glass has attenuation.
- mannykannot 7y ago
  I get the point about glass attenuation, but with regard to AR (anti-reflection?) coatings, are you referring to the surface through which light enters the prisms? If those surfaces are normal to the laser beam, would the reflected light not still be contained within the beam path, just as it is in the case of a Fabry-Perot resonator?
  - chris_va 7y ago
    Yes, but at the wrong path length (and offset by some vector, if you had a TIR cube)