Bands

scott_alexsk

I do not know if you saw this in my other thread, but I suppose this makes my question clearer. Are the conductive and valence bands defined in an unbonded atom? I think that it would be defined either way, but it seemed to be suggested that it is not, by other posts.
-Scott

Gokul43201

There are no bands when you have only a single atom. Bands are nothing but a macroscopic number of closely-spaced, discrete energy levels formed from having a macroscopic number of atoms in a solid.

scott_alexsk

So what changes within the bonded atoms to change the postion of the bands whereas the colvelant band be overlapping or not overlapping the valenece band? What dictates in both atoms these band characteristics?
-Scott

ZapperZ

The concept of "valence bands", "conduction bands", etc. in solids is a many-body concept. Such concepts are meaningless when you have, let's say, 12 atoms group together. That's why you don't have a "conduction band" in a molecule, even large ones such as bucky balls.

A many-body physics requires the collective effect where the individual no longer dictates the properties. It is the interaction with A LOT of other individuals that produces such collective behavior. So the "overlapping" of interaction is not just from one, two, three, etc particles, but from a gazillion.

Zz.

scott_alexsk

Even at the marcoscale, there is some effect of the constituent atoms on the postion of the bands. What is this variation with certain atoms that causes a variation of postion of bands at the marcoscale whereas the conductive band overlapping the valence band or vice versa? Is the presence of bands only a characteritic of metal compounds?
-Scott

ZapperZ

Come again?

What are these "effects"? And what do you mean by "position" of the bands?

Ordinary metals, semiconductors, and insulators. It is why the energy gap in a semiconductor is called a "band gap".

Zz.

Gokul43201

If I understand the question correctly, there are many factors : the size of the atoms involved and the crystal geometry; the number of valence electrons per atom; the density of atoms; and the atomic number, to name some big ones.

scott_alexsk

Just to clarify one point at a time, do only metalic compounds have bands or can all marcomolocuels have bands?
-Scott

ZapperZ

This is getting quite confusing. What is a "metalic compound"? Macromolecules? A compound doesn't become metallic. It is only when it has a gazillion other partners can it be called a metal. You need to understand the significance of many-body physics here.

And did you miss my remark about semiconductors and insulators?

Zz.

scott_alexsk

Yes I missed that sorry . How does the band gap vary with insulaters? The valence band is over the conductive band, right? In metals I suppose as a result of the higher energy conductive band overlapping the valence band the electrons are delocalized, because they do not want to occupy that higher energy state, right?
-Scott

Gokul43201

Please read this :

http://hyperphysics.phy-astr.gsu.edu...lids/band.html

scott_alexsk

Thanks for the link Gokul, it helped very much. One more question, where is the valence band located? I mean is the conduction band essentially everywhere on the substance and the valence band localized to the individual atoms?
-Scott

Gokul43201

The conduction electrons are delocalized and the valence electrons are localized to each atom. However, you can't speak of "valence bands being localized" to atoms, because a band is nothing more than a graph of energy vs. crystal momentum.