Basics of LEDs

Every solid-state diode is a light-emitting diode (LED). The difference is that conventional solid-state diodes are not configured to produce sufficient light of the proper frequency, directed properly so it is useful as a source of visible light.

A solid-state diode is composed of two semiconducting layers, frequently made of crystalline silicon, that have been joined in a special way to make a junction. The junction is where the action takes place.

The usual way of making an LED is to start with N-type crystalline silicon, silicon doped with minute impurities to make it more electrically negative with a surplus of electrons. On top of this material is grown P-type silicon with a surplus of holes. (It’s possible to grow N-type silicon on P-type but this approach is less common.) The terminal attached to the P-type layer is the anode and the terminal attached to the N-type layer is the cathode. The electrons and holes are the charge carriers that constitute electrical current. When positive voltage is applied to the anode and negative voltage is applied to the cathode, the charge carriers are repelled away from both poles. They crowd in toward the junction, and conduction takes place. The diode is said to be forward biased.

When negative voltage is applied to the anode and positive voltage is applied to the cathode, the charge carriers are attracted to the terminals, migrating away from the junction, which in this state is known as the depletion layer.

It happens that when an electron meets a hole, it drops into a lower energy level and emits a photon, the basic unit of light. Because of this property, an LED emits light when it is forward biased. When it is reverse biased, holes and electrons are not combining on an ongoing basis and there is no production of light.

Considered as a wave, the wavelength (and thus frequency) of the emitted light depends upon the energy band gap of the materials comprising the semiconductor junction. Silicon and germanium, conventional semiconducting materials, are not good for most LED applications. They are indirect bandgap materials, the holes and electrons combining by non-radiative transition. The semiconducting material of choice for LEDs is gallium arsenide, although with recent advances new semiconducting materials have entered the LED picture.

The first working models of LEDs were built in 1961, initially operating in the non-visible infrared range, then emitting red, blue and finally white light. The first LEDs were only about 1% efficient, converting 1% of their dissipated power to light. (Back then the lumen-per-watt rating now used as a measure of output efficiency had yet to be invented.) Eventually, much higher power LEDs, suitable for task lighting, automobile headlights and traffic lights, became practical. As with their close relative the PV solar cell, the price has been in free fall. In view of their long service life and low energy consumption, it appears likely that LEDs will be the wave of the future.