I am trying to reconcile the optics explanations (from “Why glass is transparent?”) with the bits I learnt about band gaps and fermi levels. It is hard to write down a single question for now, will have to edit this question to suite the StackExchange format later. I welcome all forms of comments and answers.

My current understanding is that, glass being transparent (to visible light) is because it is not absorbing any photons at all. Its structure is so rigid ( meaning the energy band will be narrow and far apart ) the visible light just didn’t happen to contain the right amount of energy to make any electrons in any fully filled band to jump to a partially filled band. If there is no jumping then there is no absorption. [NOTE, This understanding is proven to be correct and confirmed]

I also think, the propagation delay in glasses is caused by impurities and crystalline imperfections so that some absorption occur.

I can not explain why a pure silicon crystal is not transparent. Visible light contains just the right amount energy to make electrons in silicon crystal to jump band? It is a semi conductor so it is probably less draconian about its electrons freedom. [NOTE, this question is solved, as the visible light's energy is 1.65~3.1ev, and minimum required energy for silicon electron jump is 1.11ev, which is its band gap energy]

Visible light is a very very narrow band, I would expect many solids in this world to be transparent, I would expect lots of transparent insulators. But that's not the case in real world. There are way more insulators than transparent objects.

I do not know how “elastic” absorption works. Inter-molecular bond or inner-molecular bond, shouldn’t all of the electrons in whatever bonds be in the same fermi sea and be subjected to the same rule?

I am also aware of the uncertainty principle at this scale. light can behave strangely just by itself. I learnt how to calculated diffraction but never learnt the true cause. I am not sure if that has anything to do with this topic.

  • $\begingroup$ Glasses (i.e., purely amorphous) do neither have a band structure nor a crystalline structure - you might consider this difference in your explanations. But I'd also advise you to reconsider what exactly you want to be asking :) $\endgroup$
    – Sanya
    Dec 12, 2016 at 23:23
  • $\begingroup$ You may some helpful observations in why is diamond transparent when graphite is not? (not an exact duplicate, but a lot of the same physics) $\endgroup$
    – Floris
    Dec 12, 2016 at 23:41
  • $\begingroup$ @Sanya I know it as a fact (not why) Glasse is amorphous, so it doesn't have crystalline structure. But when you say it doesn't have a band structure... I can not think thru the reason way. Because ... there are lots of gaps between small infinitesimal pockets of crystalline structures? so those gaps would isolate the electrons into many infinitesimal fermi seas...? $\endgroup$
    – eliu
    Dec 13, 2016 at 0:49
  • $\begingroup$ @Floris Thank you, that link confirms my original thoughts, but the post does not make the dots connect with "elastic absorption" $\endgroup$
    – eliu
    Dec 13, 2016 at 0:55

1 Answer 1


Silicon has a band gap of 1.2 eV. It is transparent to infrared with lower photon energy, there are no transitions in which those energies can be absorbed. This is the same reason why air and glass are transparent in the visible.

Silicon strongly absorbs (or reflects) photons in the visible part of the spectrum, as those induce electronic transitions of electrons from the valence band to the conduction band.

Transparent substances have a bandgap larger than the photon energy of blue light, larger than about 4 eV. Not all wide-gap insulators are transparent - there are other processes that can absorb light, local transitions.

  • $\begingroup$ That confirms my own line of thought regarding band gap. Would you care to comment on my other assertions? thanks $\endgroup$
    – eliu
    Dec 13, 2016 at 0:46
  • $\begingroup$ I cannot guess what you mean by "elastic absorption". $\endgroup$
    – user137289
    Dec 13, 2016 at 7:59
  • $\begingroup$ I read about "elastic absorption" on this forum somewhere... the key idea was that: electrons in some material will absorb a photon, jump the band, jump back , then release an identical photon. That's elastic. electrons in other material will absorb a photon, jump to a very high band, jump back a bit, release a smaller photon, wait a bit, jump back a bit again and release a smaller photon again. Or just convert the original photon into vibration without ever releasing a photon. That is inelastic. $\endgroup$
    – eliu
    Dec 13, 2016 at 16:32
  • $\begingroup$ @eliu The term would be elastic scattering. Descriptions in terms of photons are problematic, I do not find them helpful at all. There is no measurable delay, for example. $\endgroup$
    – user137289
    Dec 13, 2016 at 17:10
  • $\begingroup$ "phonons" is what other people are talking about. Almost like band structure view of matters but with classical mechanics mixed in. I m glad band structure can explain transparency, but it can not explain refraction and optical phenomenons. $\endgroup$
    – eliu
    Dec 14, 2016 at 21:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.