Semiconductors and doping

A pure, undoped semiconductor such as silicon conducts electricity only weakly at room temperature, since its small but nonzero band gap allows only a modest number of electrons to be thermally excited from the filled valence band into the empty conduction band. Practical semiconductor devices rarely rely on pure material; instead they exploit doping — deliberately introducing a small, carefully controlled concentration of impurity atoms into the crystal lattice — to engineer the material's conductivity and charge-carrier type on purpose.

Silicon has four valence electrons, each forming a covalent bond with a neighbouring silicon atom in the crystal lattice. Replacing an occasional silicon atom with a donor impurity from group V of the periodic table, such as phosphorus (five valence electrons), leaves four electrons participating in bonds just like silicon's, with one extra electron left over — loosely bound to the donor atom and easily freed into the conduction band by thermal energy at room temperature. This produces an n-type semiconductor, named for the extra negative charge carriers (free electrons) it contributes; the donor impurities themselves stay fixed and become positively charged once they have donated their extra electron, but it is the freed electrons, not the fixed ions, that carry current.

Replacing a silicon atom instead with an acceptor impurity from group III, such as boron (three valence electrons), leaves one bond short an electron — a vacancy called a hole. A hole is not a literal particle, but it behaves like one: a nearby valence electron can hop into the vacancy, filling it but creating a new vacancy at the electron's former position, so the hole effectively migrates through the crystal exactly like a mobile positive charge carrier, in a direction opposite the electrons that are actually moving. This produces a p-type semiconductor, named for its extra positive charge carriers.

Neither doping type by itself is especially remarkable; the real power of semiconductor doping emerges when n-type and p-type regions are combined within a single crystal. A junction between them — a p-n junction — allows current to flow easily in one direction and blocks it in the other, forming a diode, and more elaborate combinations of doped regions form transistors, the switching elements at the heart of essentially all modern digital electronics.