The Nature of Transparency

[Iconoclast]

Although I feel I have a pretty good grasp of mathematics and physics (not that weird Quantum Mechanics stuff, regular physics), I'm completely hopeless when it comes to anything to do with chemistry. My electrical engineering degree course contained very information on chemistry except for a subject or two on materials science.

So anyway, every so often I start wondering about transparency of solids, and why some solids are opaque and some are transparent.

I stumbled across a discussion about this very subject on the ABC (Aust) science forums a few months ago. Someone asked why it is that glass is transparent. Someone else replied that glass is actually a liquid, not a solid, so it's transparent because water is also transparent. After that the thread degenerated into a debate about whether or not glass is a solid or a liquid, so I didn't manage to learn anything new there. You see, even if glass is a liquid, that doesn't necessarily explain why we can see through it.

So, let's forget about glass and avoid that wrinkle altogether, let's use diamond as our example. Now, diamond is undoubtably a solid, it has a regular lattice structure, and if it's cut into a rectangular shape (i.e. not faceted), we can all agree that it's perfectly transparent.

So, what is it that lets light pass through diamond, but not through (say) steel? I guess the real question here is not "why are some solids transparent?" but rather "why are some solids NOT transparent?", since a solid is mostly empty space anyway.

Is anyone here able to shed some light on this subject?

 

[Guest]

Whether or not light gets absorbed by a solid depends on the energy state of the electrons in the solid and the frequency of the light.

edited :

The reason glass and diamond (and many other crystals) are transparent is because there is only one main sort of atomic structure. Things made of different substances have lots of different energy-state electrons and therefore absorb all the frequencies. Impurities in diamonds cause coloration.

 

[Iconoclast]

Re: Re: The Nature of Transparency

Whether or not light gets absorbed by a solid depends on the energy state of the electrons in the solid and the frequency of the light.

Could you elaborate on this point?

The reason glass and diamond (and many other crystals) are transparent is because there is only one main sort of atomic structure. Things made of different substances have lots of different energy-state electrons and therefore absorb all the frequencies. Impurities in diamonds cause coloration.

OK cool, the part about impurities being visible because they are impurities makes sense. However, my understanding is that the turbine blades in super duper jet engines are "grown" as a single, faultless crystal, but (I assume) these turbine blades are opaque. Am I missing something else here?

 

[Guest]

Re: Re: Re: The Nature of Transparency

Could you elaborate on this point?

The energy state of the electron means which electron shell it is in, how many other electrons are in that shell, and its tendency to be able to flip to another energy state when hit by a photon. When a photon passes through an atom one of several things can happen - it can pass through unaffected, it can be reflected/deflected, it can be absorbed (only to then be released as luminescence) or it can be absorbed and converted into heat. It is the last of these cases that causes colouration. That is of course also why dark substances get hotter faster when exposed to light.

OK cool, the part about impurities being visible because they are impurities makes sense. However, my understanding is that the turbine blades in super duper jet engines are "grown" as a single, faultless crystal, but (I assume) these turbine blades are opaque. Am I missing something else here?

I'm guessing now....

Well, those are metal crystals. Most metals are opaque because they contain 'metallic bonds' with delocalised electrons in many different states. How metal crystals compare to normal metals in terms of their electron states I do not know.

Anyone else?

 

[Agammamon]

"Simply stated, a solid material will appear transparent if there are no processes that compete with transmission, either by absorbing the light or by scattering it in other directions.

In pure silicon, there is a very strong absorptive process at work: the incident visible light is absorbed by electrons that then move from one electron energy state to another (an occurrence technically known as a band-to-band transition).

Glass, being silicon dioxide--not pure silicon--does not have this band structure, so it cannot absorb light as pure silicon does.

Sand, on the other hand, is also silicon dioxide, but it is so filled with impurities that light simply scatters outward incoherently and does not pass through to a noticeable extent.

Scientific American

 

[Agammamon]

The arguments about what form of matter glass belongs to revolves around two main issues: temperature and time. Generally speaking, as a substance is heated or cooled, it takes on different properties. The temperature at which a solid becomes a liquid for example, is called the melting point.
Most polymers (a molecular chain of repeating smaller units) including glass, have what is called a glass transition temperature (Tg). This is the temperature range at which a substance moves from a tough and rubbery state, to that of a stiff, brittle (or glass like) state. An everyday example of this is when we want to remove chewing gum from clothing. We can put ice on the gum to lower the temperature below Tg. (4) The gum then takes on different properties to become brittle and break apart.

It is generally agreed upon by both groups that glass is an amorphous substance, which means that the molecules have no long range order and lack a well-defined arrangement. Conversely there exists crystalline substances, which are defined as having long range order and the molecules occupy specific positions rather then being random. Common examples include ice and salt.

The characteristics that define a supercooled liquid are: a) it is only slightly stable liquid and b) it has been cooled below its melting point but not below the glass transition temperature. (3)

Below Tg, glass can undergo structural rearrangement only extremely slowly and at temperatures well below Tg. As it cools below the Tg it becomes more difficult for the molecules to move around and they eventually reach a point where they can no longer rearrange themselves, and appear rigid. Then and only then is it correct to call this material glass for a material is considered a solid when observable long-range order is present, during the duration of the experiment. (1)

As far as the whole window pane phenomenon goes, some people point to the now outdated process of crown glassmaking as the cause for their uneveness. It was made by blowing a glass bubble and then spinning it flat, so the more of the glass was spun, the thicker the edges would be. Whether or not this is the cause for old windows being thick on the bottom, i can't say... but it would make sense that if the glass was uneven, one would want to put the thick end towards the bottom for structural reasons.

I think glass best fits under the category of a solid. It is a amorphous substance with no long range disorder when restricted to a laboratory experiment. One study suggests that the rate at which glass actually "flows" is impossible to see since the resultant calculation is beyond the supposed age of earth! (2) Heres a final thought taken from a paper written by R.Jeanloz, and Q.Williams,

"As kinetically frozen forms of liquids, glasses are
characterized by a complete lack of long-range crystalline order and are the
most structurally disordered types of solids known! (1) Yep! That's a good way to sum things up.

References:
1) Science vs Urban Legend
2) Glass Myth Shattered
3) Supercooled Liquids
4) Chemistry for the Changing Times, written by John W. Hill and Doris K. Kolb

 

[Stimpson J. Cat]

One important thing to consider when considering optics is the question "Why isn't everything transparent?"

When light enters a substance, the electrons in that substance begin oscillating at the same frequency as the light. These oscillating electrons produce a propagating field of their own. This is called stimulated emission.

It is the interference of this new wave with the original one which causes absorption and reflection. In the case of reflection, the wave that is generated at the interface of the substance destructively interferes with part of the incoming wave, thus reducing the amount of light that penetrates. The new wave also extends out from the interface, as a reflected wave, carrying away the energy that was lost due to destructive interference. For the light that is not reflected, the energy lost to destructive interference is converted into thermal heat. That is absorption. The two waves will never interfere constructively within the substance*.

The destructive interference within the substance causes an exponential attenuation of the wave amplitude as you move into the substance. But in order for this to happen, the electrons in the substance have to react to the wave. This is where what UCE was talking about comes into play. Pure crystals have very sharply defined, and specific, modes of vibration. Only frequencies which are near these modes will be absorbed or reflected. Such materials react very weakly to the incident light, and thus have very low absorption rates. They are thus very transparent.

Substances with lots of different modes will tend to interact very strongly with incident light. They therefore tend to be opaque. This can occur when there are impurities in the crystal. It can also be the case for crystals with very complex molecular structures.

Metals are a special case. The free-moving electrons do not have specific modes of vibration, but can instead react to a continuous range of frequencies. This is also why metal is good at blocking X-Rays and Gamma rays. Those frequencies are usually well above the resonating frequencies of molecular electrons, but are still able to effect free electrons in a conductor, because the frequency response function of free charges is a gradually decreasing function of frequency.

Note that ionic conductors (like salt water) do not have this property, because the more massive ions cannot react to high frequencies. The rate at which the response function drops off depends on the mass of the charged particles, and an Na+ or Cl- ion is more than 10000 times as massive as an electron.

*Actually, constructive interference can occur in a substance. This is how lasers work. But it only happens under a very strange (and highly artificial) condition, known as population inversion.

Dr. Stupid

 

[Iconoclast]

UCE, Agammamon, and Stimpy, thank you all for your responses. All of your replies have helped me get a handle on this mechanism.

 

[Andonyx]

Glass is NOT Thicker at the bottom because it flowed. That's a pervasive old myth. In many places in Europe it's thicker at the TOP. It depends on how it was installed.

This has been debunked ages ago, and I have several articles in books by Ira Flatow and others about it.

For the immediate future here is one link:

http://www.robinsonglass.com/analysis.htm

Here's some more:

http://math.ucr.edu/home/baez/physics/General/Glass/glass.html

 

[garys_2k]

Glass is NOT Thicker at the bottom because it flowed. That's a pervasive old myth. In many places in Europe it's thicker at the TOP. It depends on how it was installed.

This has been debunked ages ago, and I have several articles in books by Ira Flatow and others about it.

For the immediate future here is one link:

http://www.robinsonglass.com/analysis.htm

Here's some more:

http://math.ucr.edu/home/baez/physics/General/Glass/glass.html

Sigh... If only we could get high school science teachers to learn this! They vector more urban legends than the students.

 

[Tez]

Iconoclast I dont have the time to go through and read peoples replies, but I'm sure they contain the right ideas.

A beautiful, short and simple book that discusses this and a myriad of similar questions is "Why things are the way they are" by Chandrasekar. I highly recommend it.

 

[Iconoclast]

Iconoclast I dont have the time to go through and read peoples replies, but I'm sure they contain the right ideas.

A beautiful, short and simple book that discusses this and a myriad of similar questions is "Why things are the way they are" by Chandrasekar. I highly recommend it.

Thanks Tez, I'll check it out.

 

[LucyR]

When light enters a substance, the electrons in that substance begin oscillating at the same frequency as the light. These oscillating electrons produce a propagating field of their own.


In the approximation of a driven harmonic oscillator, yes. However, second and higher-order harmonic generation, for example, can also occur under the right circumstances.


It is the interference of this new wave with the original one which causes absorption and reflection. In the case of reflection, the wave that is generated at the interface of the substance destructively interferes with part of the incoming wave, thus reducing the amount of light that penetrates. The new wave also extends out from the interface, as a reflected wave, carrying away the energy that was lost due to destructive interference. For the light that is not reflected, the energy lost to destructive interference is converted into thermal heat. That is absorption. The two waves will never interfere constructively within the substance*.


Right. This is basically the statement of the (classical) Ewald-Oseen extinction theorem. Macroscopically, the kinematics of the scattering are summarized by Snell's law.

In a real material, however, this is once again an approximation. Consider, for example, random thermal motion. Such motion results in an excess polarization upon the application of an incident field. The latter then behaves as a source term in the Maxwell equations. Derivation of the actual scattering cross-section is highly non-trivial in most cases. Basically, an additional electric (and magnetic) field is radiated over 4pi, and has a frequency slightly different to that of the incident field. Under most circumstances, the effect is pretty small. It can nevertheless be easily observed given a light source of sufficient intensity and a spectrometer of adequate resolution and contrast. It is the basis of experimental techniques such as Brillouin (inelastic interaction with acoustic phonons, i.e. omega tends to zero, as k tends to zero) and Raman scattering (optical phonons). I should note that the process occurs in metals too.

Hope you found this interesting.