Naked Science Forum
General Science => General Science => Topic started by: chris on 13/04/2008 21:44:59
-
Why are some surfaces the colour they are? What is happening at the level of the atoms and electrons and the light photons to make a surface look white, or indeed any colour?
Chris
-
Why are some surfaces the colour they are? What is happening at the level of the atoms and electrons and the light photons to make a surface look white, or indeed any colour?
Chris
To answer your question would need an indefinite amount of time...
-
Let's make a start, though.
For a start, you have to take into account the fact that the eye/brain try really hard to eliminate the effects of the colour tint / temperature of the illumination, they do this by 'integrating to grey' the whole of a scene. This works over a large range of conventional lighting and we assess and remember many colours well under all daylight conditions and even with tungsten lighing but it falls down with fluorescent and LED lights. To get 'white' you need to reflect all wavelengths well - you can't approximate to white by mixing - this would give you 'grey'.
The colour white implies that the object is reflecting pretty well all the light incident on it but not in a coherent way.
A metal surface will reflect all wavelengths quite well because of the high mobility of the surface electrons but, if it is polished, it will give specular reflections which will give you virtual images of things but a 'black' image of black areas. If you make the surface irregular enough on a small scale it should reflect a mixture of light from all objects, in all directions from any part of its surface. This would 'look white'. I can't think of a brushed / matt finish on a metal that looks actually white, though.
Scattering by Total Internal Reflection in multiple small crystals (e.g. snow) is a good way to get a white surface. TIR is easiest looked at as a bulk effect but can be modeled in terms of electron mobility (Ionspheric Reflection - TIR- of HF radio can be explained in terms of changing ionisation levels and the density of free electrons at different heights).
Pigments would be much harder to explain - as I mentioned earlier, because colour mixing is subtractive, you can only aim at a 'grey' but not a good reflective white. For a non-conducting surface to reflect all wavelengths you need a lot of electrons which are bound tightly enough to their molecules to prevent dc conduction but can move freely about, locally, in some sort of band structure with a large number of possible optical transitions so that photons of all wavelengths are absorbed and then re radiated, without significant opportunity for energy loss into other wavelengths.
I think that a good white colour is only produced in living things by the 'snow' / TIR method - as in bird's wings - rather than with pigments.
Naturally occurring chalk looks pretty white - it is amorphous but does it let light into its grains and then TIR sends most light back out again? Perhaps that's what paint pigments use.
Is that all right for a kick-off?
-
Titanium IV oxide is the whitest substance known. So what's special about the crystal structure of this oxide that makes it reflect all wavelengths of light so well?
-
Here comes an essentially classical argument - the link between QM and refractive index is there if you want to find it but, in the same way that people discuss electrical circuit theory without constantly referring to QM, it must be ok to bypass QM for a satisfactory explanation of this.
TiO seems to have a very high refractive index - higher than diamond. This means that light entering the crystals at most angles will hit a face, be Totally Internally Reflected (no loss of energy at all) and emerge back in a similar direction from which it arrived. The light will emerge, more or less intact.
To get a good 'white' you need a lot of small reflecting faces, at a good range of angles. TiO seems to form lots of small crystals so it fills the bill. This link discusses some of the practicalities of pigments. I imagine that diamond powder would be quite good as a white pigment, too.
http://www.azom.com/details.asp?ArticleID=2809#_The_Importance_of (http://www.azom.com/details.asp?ArticleID=2809#_The_Importance_of)