Semiconductors

Between conductors and insulators

In a conductor like copper, electrons roam freely from atom to atom. In an insulator like glass or plastic, charge barely moves at all.^[4] Semiconductors are different: NIST describes them as materials such as silicon whose electrical conductivity can be tuned. They are the base for most electronics.^[8]

Why: the band gap

The electrons in a semiconductor fill a “valence” band of energies. Above it is an empty “conduction” band, and between the two there is a relatively small energy gap.^[1] Electrons lifted across the gap into the conduction band are free to carry current. In a doped semiconductor, the heat of room temperature is enough to lift the extra electrons from the impurity atoms.^[1]

That is also why temperature matters. In most metals, resistance increases as they get hotter, but semiconductors show the opposite behaviour.^[3]

Doping: tuning a semiconductor

Doping means replacing a few of its atoms with impurity atoms of a slightly different valence, which changes its electrical properties dramatically.^[1]

• n-type: silicon has four valence electrons. Add arsenic, which has five, and each arsenic atom brings a spare electron. The main charge carriers are negative electrons.^[1]
• p-type: add aluminium, which has three valence electrons, and each one leaves a gap where an electron is missing, called a hole. The main charge carriers are these positive holes.^[1]

A current in a semiconductor can be seen either as many negative electrons moving one way, or as positive holes moving the other.^[1]

The p-n junction and the diode

Join a piece of p-type semiconductor to a piece of n-type and you get a p-n junction. This is one of the simplest semiconductor devices: a diode, which allows current to flow in only one direction, like a one-way valve.^[2]

• Forward bias: connect the battery’s positive side to the p-type material. The junction’s depletion layer narrows, and current flows easily.^[2]
• Reverse bias: connect it the other way round. The depletion layer widens, and the current is greatly reduced.^[2]

The p-n junction became the most common form of rectifier in the electronics industry, and a fundamental building block of semiconductor devices.^[5]

Light and semiconductors

On 23 February 1940, Russell Ohl at Bell Telephone Laboratories noticed that when a silicon slab was exposed to bright light, the current through it jumped. The slab contained an n-type region and a p-type region, separated by a junction.^[5] Light striking the junction made electrons flow from the n-side to the p-side: the photovoltaic effect, which is the basis of today’s solar cells.^[5]

The US Department of Energy explains how a solar cell works today: its semiconductor absorbs the energy of sunlight and passes it to electrons, and that extra energy lets them flow through the material as an electric current, collected by thin metal contacts. Silicon is by far the most common semiconductor used in solar cells.^[9]

From crystal radios to chips

Semiconductors were used long before anyone understood them. In 1874 Ferdinand Braun noticed that current flowed freely in only one direction through a galena (lead sulfide) crystal probed with a thin metal wire. It found no use until early 1900s radio, where it became the “cat’s whisker” detector in crystal radio sets.^[6]

Understanding the physics made it possible to design devices on purpose. The p-n junction led to the junction transistor,^[5] and the MOS transistor, first made successfully in 1959, is now used in more than 99 percent of microchips.^[7]

Going further

• Transistors: how a semiconductor sandwich lets a small current control a big one.
• OpenStax University Physics Volume 3, listed below, covers band theory and semiconductor devices in depth.