To understand semiconductors, we need to first understand conductors themselves. On a superficial level, we can say that conductors are substances that permit the flow of electrons. (For the purposes of this discussion, we'll call this flow of electrons a "current".) The conductors that we will be comparing semiconductors to will be metals; in this sidebar, we will be ignoring ionic solutions.

If we recall back to the days of VSEPR, we remember that when there is a bond made between two atoms, there are various molecular orbitals formed. As we learned, for each pair of atomic orbitals that overlap, there is one such molecular orbital formed. Also as we recall, metals have a lattice-like structure; in other words, all of the atoms are closely packed together. We would imagine that there are a lot of overlapping atomic orbitals in this case if they're all so close by, and in this case, we're right.

So what does this have to do with metals conducting? Well, when we have all these molecular orbitals, they each have their own slightly different energy levels, for reasons that are too complicated to go into here. In the case of the many orbitals that we have in any reasonable amount of metal, the energy levels are so close together that we don't even distinguish them as energy levels, and we just call them an energy band. With this 'band', it takes very little energy to push electrons into other levels -- little enough energy that a difference of electric potential (i.e., differing voltages) can do it. This works because the band is only partially filled with electrons, and hence the current can push electrons into the band, and pull electrons out. This pushing and pulling allows for the flow of electrons that we talked about earlier, also known as a current.