Doped semiconductors, p-n junctions and solar cells

Pure semiconductors are rare, so we dope them on purpose

Intuition

A pure semiconductor has almost no carriers at room temperature, like a hall with a few people. Adding a pinch of the right impurity is like handing out tickets: either extra people walk in (electrons) or empty seats appear that others can shuffle into (holes).

An intrinsic semiconductor is a pure, undoped semiconducting material. An example is a rock-salt type binary compound with no d electrons. The anion s and p orbitals form a fully occupied valence band, and the cation s and p orbitals form an empty conduction band. The gap can be small enough for intrinsic conduction in some cases, especially III-V compounds. The slide lists Cl⁻, S²⁻, P³⁻ as 1s²2s²2p⁶3s²3p⁶ and Na⁺, Mg²⁺, Al³⁺ as 1s²2s²2p⁶3s⁰3p⁰.

Intrinsic semiconductors with gaps of about $kT$ are very rare. Most semiconductors in electronics are extrinsic: they are doped with an impurity atom.

n-type adds electrons, p-type adds holes

Compare

n-type

Carriers are negative electrons. The dopant has occupied levels close to an empty conduction band. Example: As ([Ar]4s²4p³) in Si ([Ne]3s²3p²).

p-type

Carriers are positive holes. The dopant has unoccupied levels close to a filled valence band. Example: Ga ([Ar]4s²4p¹) in Si ([Ne]3s²3p²).

Plain words: As has one more outer electron than Si, so the extra electron sits just below the empty conduction band and is easily lifted into it. Ga has one fewer, so it leaves a hole just above the full valence band that valence electrons can hop into.

Doping can also be done by site vacancies. The lecture example is an oxygen vacancy in ZnO₁₋ₓ, drawn as an n-type donor level just below the conduction band. Mn₁₋ₓO is the p-type example, with an acceptor level just above the filled valence band.

Exam detail

The learning outcomes say: differentiate between intrinsic, p-doped and n-doped semiconductors. Know the carrier (electron or hole), where the dopant level sits, and one example of each.

A p-n junction is a one-way gate that sunlight can drive

Join an n-type semiconductor to a p-type one. Electrons are thermally promoted into the conduction band and diffuse from the n-type to the p-type region. That leaves the n-type region with a net positive charge and the p-type region with a net negative charge. This sets up an electric field across the junction.

An applied current can normally flow in one direction only, so the junction acts as a rectifier. If the regions get a constant supply of new carriers from incoming photons, a current flows. That is a photovoltaic (PV) diode, a solar cell. With the right resistance in the circuit, current and voltage can be optimised.

The solar spectrum ranges from IR (1.1 eV) to UV (3.5 eV). The band gap of Si is 1.1 eV and of GaAs is 1.4 eV. The slide shows P-doped Si as the n-type side and B-doped Si as the p-type side.

Pitfall

An n-type crystal is not charged overall. The charge appears at the junction only after electrons diffuse across, and that charge is what sets up the electric field.

Recall

Plain words: what does a p-n junction in a solar cell do with light?

Light creates electron and hole pairs. The junction’s built-in field separates them, so electrons flow through the outside circuit and do work.