Semiconductor Devices

P-N junction in semiconductors

This post is covers the topic of pn junction definition. A semiconductor itself does not give any special properties to a semiconductor device. However, when sections of p-type and n-type semiconductors are brought together, a semiconductor device obtains some interesting features, which are described below. Nowadays, a p-n junction is the base for a lots of diodes- and transistor-type semiconductor devices. Figure 1 shows the classical p-n junction.

On the p-side of the contact positive charges are prevailing, on the n-side of the contact negative charges are prevailing. There is a thin layer in place of the contact of two semiconductors, which is called the depletion layer, where the charges come into contact and recombine. Then the depletion region of the p-type ionises negatively, and the depletion region of the n-type semiconductor ionises positively.

Usually charge carriers in semiconductors are compensated by the lattice, and the semiconductor is neutral. Charge separation at the depletion region causes an electric field and contact potential in the adjacent region. Typically, contact potential is less then 1eV for most semiconductors. If these two semiconductor sections are closed in a circuit, electric currents will flow through the circuit. The first current is saturation current, related to the electric field and contact potential. The second current is opposite to saturation current, and is called diffusion current. This current is caused by the holes and electrons, having sufficient energy to cross the p-n junction. Diffusion current dominates above the saturation current.

Figure 2 depicts forward- and reverse-biased connection for p-n-junction. Contacts between the battery and semiconductor wires are assumed to be ohmic. Forward-biased  connection lowers the contact potential. Battery voltage ${V}_{battery}$  is opposed to the contact potential ${V}_{depletion}$ in this case. In the external voltage source the battery increases the diffusion of charge carriers, and: .

This equation is a diode equation. For saturation current ${I}_{0}$t  is very small for most semiconductors $\left({10}^{–15}–{10}^{–9}\right)A$, and the diode equation can be approximated to the exponential form: ${I}_{diode}={I}_{0}e\frac{q{v}_{D}}{kT}$.

Reverse-biased connection increases the potential contact between p- and n-semiconductors. It means that the charger carriers have to overcome a greater barrier and a wide depletion layer. Diffusion current is nearly equal to zero, only small reverse  still exists through the junction. ${I}_{diode}=–{I}_{0}$

The p-n junction is characterised by the ability to conduct current in one direction – in the forward-biased state. This feature makes it useful in many electronic devices and circuits. The single p-n-junction with ohmic contacts on its terminals is called a semiconductor diode, or diode.

Figure 3 shows the i-v characteristics for the semiconductor diode. This curve is important for device characterisation and shows the working regimes of the device. i-v characteristics of the diode consist of three regions. Forward-biased, reverse-biased and reverse breakdown region. Forward-and reverse-biased regimes were discussed above.

When the reverse voltage reaches a certain level, (with high magnitude, reverse voltage), the diode starts to conduct current, but in the reverse direction. This phenomena is related to the avalanche breakdown (the nature of avalanche breakdown was explained in the previous chapter about semiconductor materials). In this case large reverse bias delivers enough energy to ionise charge carriers and causes the reverse current.

This reverse current is bigger than reverse saturation current. The electric field energises the electrons of the high energies. These energised electrons collide with the other charge carriers and delivers an excess of energy. This process is called impact ionisation, and it progresses like an avalanche. An important factor is the Zenner breakdown voltage, ${V}_{Z}$. The Zenner breakdown usually occurs in the doped regions on the junction of a semiconductor and metal (ohmic contact in this case). The high density of charge carriers causes intensive reverse current. This phenomena is also useful in some applications which we will discuss later on.

$I–V$ characteristics are very important during the designing of circuit models. Let’s discuss large-and small-signal modelling.

Read more educational posts at our Reddit community r/ElectronicsEasy.