In an intrinsic semiconductor, why don't electrons go out from both valence and conduction bands?

Question

My textbook ("Electronic Principles" by Malvino and Bates) seems to suggest the free electrons in conduction band move left and reach the positive terminal of the battery. Does this mean electrons in valence band are not allowed to move left(hole moving right) and enter the positive terminal of the battery?

Similarly, it says the electrons from the negative terminal of the battery enter from the right directly into the valence band holes. Does this mean the electrons from the battery are not allowed to enter into the conduction band?

In summary: Electrons in an intrinsic semiconductor always leave from the conduction band and always enter from the valence band. This doesn't feel right. What am I missing?

EDIT: Saying electrons in the valence band are not free, so the battery cannot pull them doesn't make sense because of the following scenario: If a p-type semiconductor is connected across a battery, the battery's positive terminal has no problem attracting the electrons from valence band.

Answer 1

If it's intrinsic, there's as many electrons in the conduction band as holes in the valence band by definition, because each electron leaves a hole. The electrons in the conduction band are bound to their atoms by (relatively weak) electrostatic forces, and can move fairly freely.

Electrons in the valence band are bound more tightly, and largely move only from one atom to an adjacent one that contains a hole.

Bottom line is, all the electrons are moving to the left--some in long unfettered paths, and some from one atom to the next. Also, in the valence band, it's kind of tag-team...an individual electron moves to an adjacent atom leaving a hole, then another electron from the other side fills that hole. It's just as easy to model it as a hole propagating than a tag-team of electrons.

Answer 2

Saying electrons in the valence band are not free, so the battery cannot pull them doesn't make sense

The valence band is localized around an individual atom or molecule, so nothing in the valence band can be moved without first being removed from the valence band (which takes energy from the battery, thus contributing to resistance).

Answer 3

In an intrinsic semiconductor like pure silicon crystal, the valence band is entirely occupied at 0deg K, thus can't contribute to current flow, just as an entirely full bottle of water can't slosh about inside. At room temp, there are a relative few free electrons in the conduction band and holes in the valence band, though not enough to be a good conductor.

I think Cristobol's comment "Electrons in the valence band are bound more tightly, and largely move only from one atom to an adjacent one that contains a hole." is just wrong. It's my understanding that existing holes move as freely within the valence band as existing electrons do in the conduction band, which just means electrons are moving freely in the opposite direction. And both are technically electron current, although hole current is more like a line of people (electrons) moving forward in a queue in the direction opposite to the hole current. There just aren't many free electrons and holes in pure silicon. The addition of P and N type dopant atoms greatly increase the population of free electrons in the conduction band and free holes in the valence band to serve as carriers of current.

But beccaboo, doesn't a high concentration of holes in the valence band imply electrons being free to move in the opposite direction of hole current (see above).