In the previous subsections, we have discussed the pn and metal-semiconductor junctions. A pn junction is a bipolar junction because both electrons and holes act as charge carriers whereas a metal-semiconductor junction is a unipolar junction (also called a monopolar junction) because either electrons or holes act as charge carriers.
To create a pn junction, a lightly doped n-type (or p-type) semiconductor substrate is heavily doped with boron (B) or other p-type dopant using a diffusion, epitaxial growth, or ion implantation process. Therefore, the lightly doped n-type (or p-type) substrate acts as a series resistor. Note that the conductivity modulation of the pn junction causes its series resistance to decrease.
Typical pn junction diodes consist of heavily doped p-type (p+) and n-type (n+) regions on either side of a lightly doped n (n–) region. Diodes with an extremely lightly doped n– region are called PIN diodes. Typical pn junction diodes have a structure similar to that of PIN diodes although I layers of pn junction diodes are more heavily doped.
Electrons from a power supply flow into the n– region via the n+ region. To maintain electrical neutrality, holes are injected into the n– region from the p+ region. These electrons and holes recombine and disappear ultimately. Carrier lifetime is defined as the average time required for this recombination. Because both electrons and holes exist in the n– region during this period, it exhibits low resistance as if it were a heavily doped region.
The longer the carrier lifetime, the stronger the conductivity modulation effect, yet at the expense of an increase in reverse recovery time (i.e., the time required for a diode to stop conducting). Toshiba provides a type of diodes with a reduced reverse recovery time (i.e., a reduced carrier lifetime) called fast-recovery diodes (FRDs). The FRD has a shallower forward voltage-vs-forward current curve (i.e., higher forward resistance) than the typical pn junction diode as shown in Figure 2-10.