Throughout this series, we’ll focus primarily on Silicon, as it is the most widely used semiconductor, although the same basic principles apply to other semiconductors as well.
Silicon is element number 14 on the periodic table. That means it has 14 positively-charged protons in its nucleus, and 14 negatively-charged electrons orbiting the nucleus. Of those fourteen electrons, ten of them are in the inner
shell, while the remaining four orbit farther away in the outer, or valence
, shell. Those four electrons in the valence shell are of critical importance, as they mean that silicon can make four bonds. That is, silicon would ideally like to bond with four other atoms to reach its most stable state.
http://media.hardwareanalysis.com/articles/small/10632.gif" alt="Semiconductor Physics">Fig. 1 - An atomic model of Silicon. There are 14 Protons in the nucleus, and 14 electrons orbiting. The furthest 4 are known as Valence electrons.
And thus we have the silicon crystal structure, or lattice
. A number of pure silicon atoms will bond together as shown below, each atom bonding with four of its neighbors. The actual crystal lattice structure of silicon is a somewhat more complicated, three dimensional tetrahedral pattern, but this simplified, two dimensional model will suffice for the purposes of this article.
http://media.hardwareanalysis.com/articles/small/10633.gif" alt="Semiconductor Physics">Fig. 2 - A simplfied, two dimensional model of a Silicon crystal. Each atom makes a bond with four of its neighbors.
As can be seen, each silicon atom bonds together with four of its neighbors to form a rigid crystal structure. Why did we use a double line to represent each bond? There’s a method to our madness: the silicon atom ‘gives’ one of its valence electrons to each one of the four bonds it makes, as does each of its neighbors. Both of the two silicon atoms involved in the bond in effect ‘share’ the two electrons (one from each of them) that are involved in the bond. The result is that instead of having four valence electrons all to itself, each silicon atom now has eight valence electrons, but they’re all shared with neighboring atoms. It is this ‘sharing’ of valence electrons that holds the atoms together in their crystal structure.
http://media.hardwareanalysis.com/articles/small/10634.gif" alt="Semiconductor Physics">Fig. 3 - A more complicated lattice structure, showing the valence electrons associated with each bond. Notice that each silicon atom now has eight valence electrons, but that they are all shared, two with each of its four neighbors.
It is important to note here that most electrons involved in the bonds are essentially trapped in those bonds, and are not available for conduction. As a result, silicon is a poor conductor (no free electrons to conduct current means poor conductivity). A very small number are able to escape their bonds, however, and thus pure silicon is able to conduct very minimally, making it - you guessed it - a semiconductor.
But why is it that some electrons are allowed to escape their bonds while others are not, and what is it that determines when, and how many are able to escape?