At low and mid-frequencies, transistor behavior is primarily characterized by resistive and transconductance elements in their small-signal models. However, as the operating frequency of the signal increases, the inherent parasitic capacitances within the physical structure of these transistors can no longer be ignored. These internal capacitances provide low-impedance paths for the signal at high frequencies, effectively diverting current away from the amplifying terminals and leading to a significant reduction in amplifier gain. Understanding these capacitive effects is paramount for accurately predicting and designing high-frequency amplifier circuits.

BJTs, due to their P-N junction construction, possess several internal capacitances that profoundly impact their high-frequency performance. These capacitances can be broadly categorized as junction capacitances and diffusion capacitance.

● Junction Capacitances (Depletion Region Capacitances): These arise from the charge storage effects in the reverse-biased depletion regions of the P-N junctions.
As the voltage across a reverse-biased junction changes, the width of the depletion region changes, leading to a capacitance effect.
○ Collector-Base Junction Capacitance (Cµ or Ccb): This capacitance exists across the reverse-biased collector-base junction. In the hybrid-pi model, it is denoted as Cµ. This capacitance is particularly critical because it connects the input (base) to the output (collector) of the transistor. Due to the Miller Effect (discussed below), Cµ can be effectively multiplied at the input of common-emitter amplifiers, drastically increasing the input impedance seen by the signal source and significantly limiting the amplifier's upper frequency response. This is often the dominant factor in determining the high-frequency cutoff for common-emitter configurations.

○ Emitter-Base Junction Capacitance (Cje): This capacitance exists across the forward-biased emitter-base junction. While it has a depletion component, its dominant part in forward bias is the diffusion capacitance.

● Diffusion Capacitance (Cd or Cπ): This capacitance is associated with the storage of minority charge carriers (electrons injected into the p-type base from the emitter, and holes injected into the n-type emitter from the base) in the neutral regions of a forward-biased P-N junction. When the emitter-base junction is forward-biased and conducting, a significant amount of charge is injected and stored in the base region. If the input signal frequency changes rapidly, this stored charge needs to be quickly added or removed, which takes time and presents a capacitive impedance. The Cπ parameter in the hybrid-pi model primarily represents this diffusion capacitance along with the smaller depletion capacitance of the emitter-base junction.

To analyze the high-frequency behavior of a BJT, the small-signal hybrid-pi model is augmented with these parasitic capacitances. The key components include:
● rπ: The input resistance from base to emitter, representing the dynamic resistance of the forward-biased base-emitter junction. It is calculated as rπ = β / gm, where β is the common-emitter current gain and gm is the transconductance.
● gm: The transconductance, which relates the change in collector current to the change in base-emitter voltage (gm = Ic / VT, where VT is the thermal voltage).
● ro: The output resistance from collector to emitter, representing the Early effect.
● Cπ: The total capacitance between base and emitter, primarily diffusion capacitance.
● Cµ: The capacitance between collector and base, primarily junction capacitance.
At high frequencies, the reactances of Cπ and Cµ become small enough to shunt current away from rπ and the collector, respectively, leading to a decrease in gain.

FETs (JFETs and MOSFETs) also exhibit internal capacitances that limit their high-frequency performance. These are primarily derived from the gate electrode's proximity to the channel and source/drain regions.

● Gate-Source Capacitance (Cgs): This capacitance exists between the gate and the source terminals. In MOSFETs, it is primarily due to the gate oxide acting as a dielectric separating the gate electrode from the channel and source region. In JFETs, it is the junction capacitance of the reverse-biased gate-source junction. It is generally the largest of the FET capacitances.
● Gate-Drain Capacitance (Cgd): This capacitance exists between the gate and the drain terminals. Similar to Cµ in BJTs, this capacitance is subject to the Miller Effect and is often the most significant factor limiting the bandwidth of common-source FET amplifiers. It acts as a feedback path from output to input.