Ionisation of semiconductor atoms

[mendes]

When we provide thermally or optically en energy higher than the semiconductor bandgap we create elctron-hole pairs, that is electrons are extracted to the valance band and they become free to move inside the crystal. My question is: are the semiconductor atoms then ionized, which means for example under the form Si+ ions ? Or that those electrons are just delocalized and that the silicon atoms are at anytime with all the electrons since at anytime an electron is in the neighborhood of the atom making it "neutral" and not an ion ?

 

[ZapperZ]

When we provide thermally or optically en energy higher than the semiconductor bandgap we create elctron-hole pairs, that is electrons are extracted to the valance band and they become free to move inside the crystal. My question is: are the semiconductor atoms then ionized, which means for example under the form Si+ ions ? Or that those electrons are just delocalized and that the silicon atoms are at anytime with all the electrons since at anytime an electron is in the neighborhood of the atom making it "neutral" and not an ion ?

A VERY important message that needs to get across here is that, the presence of the conduction band, valence band, and band gap are the result of a collective properties of atoms that make up the material. The valence shell of these atoms have combined (hybridize, etc.) to form these continuous band. So in many cases, it is no longer valid to talk about the individual atoms of the material. Instead, one has to talk about the collective behavior of the material.

Now, let's do a bit of a quick correction. When an electron is excited from the valence band into the conduction band, the whole material is still neutral. Nothing has changed. The electron in the conduction band does not belong to any atom, the same way an electron in the conduction band of a conductor does not belong to any particular atom. The fact that they form these bands means that they have contributed to some "mutual funds" whereby none of these belong to any individual atoms.

Zz.

 

[chrisbaird]

Yes, the atoms that loose electrons are ionized. That's what ionization means. The atom now has more positive charges in its nucleus than electrons bound to to it locally. We call this net positive charge a hole and treat it like a particle because it can move around. The electron that enters the conduction band is delocalized and forms a state bound to the entire crystal, but the atom it leaves behind stays localized.

Because the conduction electrons are essentially free, they move and bunch up in places different then where the holes are. While the entire solid may stay globally net neutral, it is not locally neutral. The non-zero space charge leads to an electrostatic potential known the built-in voltage which effects charge transport. These effects are very important in semiconductor heterojunctions such as in diode lasers, transistors, quantum well devices, etc.

The picture gets more interesting if you dope a portion of the semiconductor with donor or acceptor atoms. With a donor atom, when the electron is freed to the conduction band, a hole is not left behind because the donor atom already had one extra electron compared to its neighbors. But the atom left behind is still net positively charged. So you get a fixed positive ion (not a mobile positive charge, i.e. a hole), which also contributes to the space charge. In the most general form, the space charge is the sum of the conduction electron distribution, the hole distribution, the positive ion distribution (ionized donor atoms), and the negative ion distribution (ionized acceptor atom). Not every atom gets ionized. It depends on the temperature.