Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Why is glass 'see-through'?

  1. Apr 8, 2007 #1
    I'm confused with a simple physical explanation as to why glass is so transparent.

    I would assume crystalline material would be 'more' transparent than glass since the lattice structure is ordered and light can pass through in an ordered manner.

    Normal glass is an amorphous structure (random) and I would've thought light would get reflected at random clusters of the lattice. Yet its more transparent than a crystal.
  2. jcsd
  3. Apr 8, 2007 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    But note, however, that ordinary glass is very opaque over a large portion of the EM spectrum. It is opaque to the UV range, for example. You need special glass such as quartz or fused silica to get transmission which, at best, is about 90%.

    For most material, the optical property depends very much on the phonon structure of the material (see one of the FAQ entry). For glass, it is a bit more complicated than that because the "glassy phase" is quite unique and is a whole area of study in itself. The phonon structure, while it exists, isn't that well-defined, and I think that is the reason why visible range light cannot excite such modes and cause it to be absorbed.

    Someone who is more an expert in the glassy phase study may have a better answer than this.

  4. Apr 8, 2007 #3
    I must admit, i dont know a huge amount on the subject, but i have a few theories.

    Maybe the quantum/wave behaviour of light is a factor, dont take my word on it, but maybe the structure of glass is such that electromagnetic waves with the wavelength of visible light can pass through unaffected.

    Also, reflection isnt the only factor in transparity/opacity, there is also scattering to take into account.

    Im quite interested in this too, i hope somebody with a bit more knowledge than me in the subject will reply

    EDIT: Ah, ZapperZ replied before me with a much better explanation than mine
    Last edited: Apr 8, 2007
  5. Apr 9, 2007 #4
    I think your explanation of the phonon structure not being able to absorb light is an excellent one. It makes sense too.

    Would that mean all/most amorphous structured material be transparent to more wavelengths than crystallines?
  6. Apr 19, 2008 #5
    So far, this is what i have settled on as an explanation for that behaviour. I would love to hear any valid opinions on what i should correct......

    When a photon enters a material, it becomes a phonon for description in all intents and purposes.
    The photon itself should be viewed only as having an electric and magnetic field that is oscillating. The material itself is composed of atoms which also have excited fields which are due to the background energy but all the most certain will have a discrete range of oscillating field arrangements.

    In the case of a crystal, there will be a global and periodic arrangement of these atoms which will cause particular arrangements of fields within the structure.
    The photon field, while passing through the crystal is reacting with the field arrangements within the material.
    This determines its direction of travel as the field arrangement of the photon will most likely fall into the path where the combination of fields has its least effect or where they strike a balance so to speak given its unique momentum. The average exit angle is mostly as a result of this.

    In non-crystalline materials, the arrangments of the atoms are such that the solid is still organised along the lines of its field strengths and will still exhibit some average properties towards photons.

    Many opaque material if cut thin enough will allow transmission of photons.

    If the photon has a frequency equivalent to a vibratory mode within the field arrangment that is unoccupied, then it will be absorped. If there are no modes present at or near the surface of the material that would allow transmission, it is reflected.

    Higher Z material tend to be more tightly packed and exert greater field effects which lend themselves to greater refraction angles.
    High grade optics make use of this phenomenon by adding high Z materials to glass to increase their index of refraction.

    If the photon should fly through the material without having been influenced in anyway, then it should be considered that the photon actually went through a hole in the material and not through the material itself.

    If the photon should hit a atom or experience a near hit, then scattering is the resulting effect.
    We have to realise that the material is held and bonded together through its fields and so the gap that exists in between is actually occupied by the interacting fields of the atoms.

    On the whole, the wavelength isnt changed much unless you have effects like photon pumping wherein more than one photon will interact in a given field area to produce a photon of higher frequency but this is very marked and the light produced is generally in keepin with one of the vibraional modes of the material....ND:YAG lasers make use of this....infact, most lasers do as well.
    Any interactions by the fields in the material on the photon field will be small and should fit a very tight(slim) gaussian profile that can easily go unnoticed by a regular spectrometer unless the accuracy is way above the wavelength being examined and yet is in keeping with the small variations of the discrete levels of the atomic fields involved.

    However, all said and done, the photon while travelling through the material should be considered as a phonon AND could be treated as an acoustic shock pulse through the material. This makes it easy to see how the frequency of the wave would 1. determine its speed through the material and 2. interact with the various vibrational modes.
  7. Apr 20, 2008 #6
    Can you say photons "become" phonons? Not all photons are absorbed by the material creating vibrational atomic resonances known as 'phonons'. (specific conditions need to be met like intensity).

    I assume you are defining how the polarization of the incident light is defined as it propagates through the material? That being said, all light is polarized upon exit of any crystal?

    A random structured material (eg. glass) would then have a higher probability of absorption/scattering than an ordered structure (crystal). Yet, some crystals appear more opaque at white light than ordinary glass.

    ZapperZ was saying that the amorphous structure of glass prevents the excitation of vibrational modes within the material, now this would mean that Laser Microphones (based on the principle of opto-acoustic modulation) would work better on Crystals than standard window glass. And normal glass has a fairly decent SNR for those laser microphones (I've tried it).

    What do you think?
  8. Apr 21, 2008 #7
    Why not? After all, the propagation of the wave through the material is soley done by interaction with the material....hence the change in speed....
    To argue against it being a phonon, we would have to say that it doesnt interact at all....
    if it doesnt interact, then how is it that the material is able to act so efficiently as a filter to some momentum and not others? I'm going on the basis that any wave propagation in a material is a phonon and not a photon....i could be wrong, but as long as the wave exists within the boundry, then it remains a phonon....I'd welcome advice from experts in the area on this.
    I found this link interesting in looking at optical and acoustical modes.
    (web page addy: chembio.uoguelph.ca/educmat/chm729/Phonons/cont.htm)

    I'm not sure on the all light is polarized bit, what i am saying there, or doing rather is giving an example of a mode of travel through a material.....using a crystal as a clearer example.
    While glass isnt your highly organised structure, it doesnt prevent it from being somewhat arranged. The atomic arrangements within the glass are, agreed, more amorphous and so we dont get those lovely channels as we do in crystals.....what i dont know is whether this presents minute surface boundary conditions in the material that the wave has to negotiate...however, the overall field arrangement is such that it allows propagation from one surface to the other unless there is a distinct breakdown within the glass.
    i think its easy to see the photons being transmitted soley by the electrons, and more difficult to see that its the field thats doing all the work. I dont think the major work done in transmission is via absorption and then retransmission by the electron. The delays produced by this would be significant to say the least. This would also set a greater limit on the number of photons allowed to pass at any one time....I just dont see this myself....and i must be careful here not to venture into what i believe rather than what is proven experimentally....but to stick my toes in a bit....
    warning! conjecture!
    when a wave passes an electron, i assume the behaviour is like that of a cork on water wherein it responds to the passing field of the wave and there is some affinity or lack of exerted, as well as a dependence on its previous or natural state of vibration, with respect to the passing wave . It is this interaction that determines 1. it propagtes on, 2. gets absorbed or 3. is resisted.
    it doesnt actually need to touch the charge body, just the fields need to interact.

    does this make sense at all? because in viewing this, i'm imposing the limit that the photon/phono is completely wavelike in nature.

    Yes i agree that it looks upside down, but lets face it, we know looks are decieving...
    but the proof as far as i'm concerned is in the pudding...;)
    A great example is the piezo effect (field effect) of many crystals.....one optical use of this is found in Q-Switched lasers. By applying a voltage field or an acoustic wave, depending on the material, they can influence the light's passage.
    I take it that they have realigned the electronic fields within the cyrstal and so present a resistive/assistive field to the waves.
    What may appear opaque to the visible portion of the spectrum may infact be completely clear to the infra-red or microwave portions or it could be that the fields are arranged such that 1. the majority EM is absorped or 2. reflected.

    In the case of scatter, as is obvious when white light then gets jumbled to become milky...this suggests a completely disorganised field or possibly one that has many different field structures though organised. Sort of like a staircase that has many steps of different sizes and directions.

    I'm not going to disagree with zapperZ at all on the modes issue and I think i have covered pretty much what i think in showing that what may be amorphous can in fact be organised though its atomic arrangements do not lend the idea easily.
    i believe the discontinuity of effects you describe are covered if we view it as fields rather than picture perfectly organised charge bodies.
    I can easily twang any nearly any material with an ultrasound wave and examine it for defects....admittedly, its not a optical phonon but a phonon nevertheless.
    my point being that though photonics require lovely discrete modes on which to build devices geared to the semiconductor industry, it doesnt mean that we cannot make use of the natural field arrangement that materials make use of all the time when conducting all manner of EM radiation....ie heat, light and so on....

    My only other addition is that i believe, not sure i'm right but, that the higher you get in frequency, the more the wave tends towards the particle effect and hence why at energies above xray...ie gamma....we see more billiard ball effects and the scattering becomes more important.....

    opto-acoustic behaviour is best in peizo type materials which aren't neccesarily pure crystal objects.....most piezos are made by heating a mixture of powders and allowing them to cool, much the same way as glass is made....so i expect that both would have similar effects....in this case, i think its radiation pressure thats the pronounced behaviour....have a look at photo-acoustic spectroscopy to see whats happening there and you will see the similarities.
    I would have thought that any material under pressure would distort....see what i am saying?

    we need some optical experts to clear up our thoughts....:)
  9. Apr 22, 2008 #8


    User Avatar
    Science Advisor

    You are rather wrong here. Photons and phonons are particles/quasiparticles, whiche couple to each other in material, but a photon does not become a phonon instantly and automatically. In the case of weak coupling they just scatter for example. In the case of strong coupling (near resonance) however the behaviour of photons inside materials gets quite complicated and is best described within the polariton picture.

    A polariton is another quasiparticle needed to describe the (near resonant) interaction between photons and phonon, where perturbation theory does not lead to sensible results. and explains features like the observed anticrossing between photonic and phononic modes. While this model is usually just used for strong coupling conditions, the results are of course also valid for weak coupling. Iirc there are also some theoretical groups, which claim that a photon in bulk material must always be described by polaritons (Maybe the Koch group, but I am not sure here). So you might want to do a google search for phonon polaritons or polaritons in general.
  10. Apr 22, 2008 #9
    You raise some interesting points deakie. But I think we are confusing ourselves by explaining these situations where light is either a particle or a wave. We need to realise that both pictures are valid and that a complete explanation involves an understanding of classical and quantum optics (as Cthugha gave a quantum view).

    An applied voltage across the crystal does change the electronic fields and thus polarization of the input EM field. However, a polarizer and analyzer before and after the crystal controls the output intensity. It is these polarizers that cause the light to switch 'on' and 'off'. Thats how the Q-switched lasers use them. The crystal does not turn 'on/off' the light simply by a certain voltage.

    I agree, but should the level of distortion be dependent on the materials "electronic" structure or "field" arrangement? If so, what kind of material would give a lower threshold for distortion. You quote that piezos don't necessarily possess the same structure as crystals, yet they distort easier (I don't know if "easier" is the right word because some piezo's require thousands of volts).

    lol, we certainly do have a lot of thoughts :approve:
  11. Apr 22, 2008 #10
    When silica cools into the amorphous phase the electrons in the glass do not absorb the energy of photons in the visible spectrum. Transparent plastics behave the same way.
    Last edited: Apr 22, 2008
  12. Apr 22, 2008 #11
    Typical!....i came out to bat but the umpires came out with a new set of balls to play with...

    i need a couple of days n0_3sc before we continue...need to examine Cthugha point on...
    i'm not quite convinced of the particle only view....so i'm looking at these polariton groupings....
    I haven't seen them before.....:cool:and they look really interesting....
  13. Apr 22, 2008 #12
    No Worries! :smile: I'm not going anywhere - this is purely out of interest.
  14. Apr 25, 2008 #13
    whew!! i'd say about a dozen or so papers later....i have a wee grasp of these blighters...
    Thanks Cthugha!!

    I see what you mean about their groupings and how they have locked down phonons to specifics.
    However, my problem is that if i use polaritons, it would only be as a direct substitute of phonons.....and would that solve my issue for viewing it as electromagnetic interactions? not sure about that as i'm convinced more by wave description than by particle.

    I can see how the particle description is more apt when furnishing a behaviour but it really doesn't take away from the fact that its waves as far as i can see...
    excepting that they utilise the term quasi particle to represent real particles with odd behaviours but confine their definition to the unreal....weird....arrrrgh is more appropriate...
    So in effect, a quaisparticle is an almsot real charged particle in this instance....
    I can certainly see how having a particle of sorts or the nature of one would make it easier to model a behaviour.....thats for sure....

    so i'll desist from refering to the wave like properties that i'm refering to as phonons....as phonons do not cover the range according to the conventions....
  15. Apr 25, 2008 #14
    Of course, i have a few more thoughts on the issue and hopefully it wont be too horrible if i aired them a bit...:shy:

    I noticed that conduction of current in some way gets tied to these quasiparticle interactions wherein i choose to seperate this...
    i'm not saying that it doesnt occur but im choosing to keep them seperate.
    for now, let me keep my wave as a wave and a particle as an electron.

    if i have a wave entering a material and it has some kind of interaction, then one of two things must occur with respect to a particle....it either perturbs the particle, which im saying occurs through interaction through the fields or it acts on the particle as in shifting it out of its current field bonding with the atom.....
    in the latter, that conduction effect is somewhat like scattering isnt it. Surely the wave loses some energy in that process?
    in the former, the interaction is such that the electron would move with respect to the rise and fall of the passing wave's field without much change in energy.....in the first half it either adds or takes away energy from the electron and in the second half, the opposite occurs leaving a net 0 effect.....
    however, this view is blown apart by experiments at low K where narrow spectrum lasers are used to exite cooled material and the resulting spectra indicates strong lines around the energy levels of the atom's electrons and also a production of a spread at other energies with reduced amplitude....so its proven that the interactions can cause a loss of energy that is significant. I find this very interesting.

    as i said earlier....a field can excite the material to provide some shift....this is born out with the use of graphene as a semiconductor...

    Why would it not have an effect on the wave that is transposed during its passage?
    graphene, while somewhat crystalline, isnt quite diamond like with a really rigid structure but more like your average arranged material......

    @ n0_3sc
    not neccesarily....the polarisers are used as alignment....
    the RF field is applied to create an electroacoustic effect which results in either a shear or longitudinal wave....it's the waves that does the switching...
    thats how i understood it......
  16. Apr 25, 2008 #15
    Oh right yes I'm thinking of something different. Yes I do know about the technique your referring to now.

    Deakie, your understanding of this thread is beyond me. But to clear up:
    When low level light enters a material, it interacts with the electrons which 're-emit' the light?
    When high level light enters a material, it interacts with the phonons which 're-emit' the light down-shifted in frequency (Raman scattering)?
  17. Apr 26, 2008 #16

    I take it you mean low and high level to be the energy of the photon right?

    Just saying what conclusions i'm coming to....while trying to leave out any percieved logic which may distort the picture...

    if a photon has a direct hit with an electron mode and gets absorped, then its highly likely that it will be remitted by that same electron.
    if a photon enters a material and interacts by means of the field, then its highly likely that that photon emerging from the material will be the same photon which has either been added to, reduced in or remains the same in energy...in the raman effect, the photons always lose energy....so raman cannot be the only effect....

    Cthugha said that the photons will couple with the phonons already in the material and he gave instances that they would be treated differently.

    Personally, i'm begining to think of the material as a complex, wide (with discrete levels) bandwidth, multiharmonic, mixing oscillator....
    does this make sense?
  18. Apr 27, 2008 #17
    Yeah I guess it does make sense.
    Its weird that in all my years of physics I never new that light is re-emitted from a material upon incident light.

    Would that imply that there exists a delay between the outgoing light and the incoming light - a delay given by the electron's decay lifetime?
  19. Apr 27, 2008 #18


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Glass, being essentially a supercooled liquid, is transparent for about the same reasons that most liquids are transparent ("Eg" > 4eV, and suitable phonon dispersion). In fact, the common method used to make most glasses, plastics and even candy more transparent is to melt them and then rapidly quench them.
  20. Apr 27, 2008 #19
    where T=transmission coefficient,R=reflection coefficient and A=Absorption coeff. and all depend on material and wavelength of radiation you are talking about..

    So for any material, if T is to be high, A and R need to be low...

    Since metal surface has a lot of free electrons, when the radiation strikes the material, the free surface electrons start oscillating and in the process they emit the radiation which falls on them (oscillating charge will emit radiation). This is how metals are supposed to be good reflectors..

    Absorption is something i'm not very sure of but since glass has high T, obviously its A is low for visible radiation..

    p.s: i could not read the previous posts due to lack of time so i'm sorry if i'm repeating something.. :)
  21. Apr 27, 2008 #20


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    You really DO need to read the thread before responding with something like that.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook