The bipolar junction transistor (BJT) was invented in the late 1940s at the Bell Telephone Laboratories by Bardeen, Brattain, and Shockley and became the first commercially successful three-terminal solid-state device.
Although the FET has become the dominant device technology in modern integrated circuits, bipolar transistors are still widely used in both discrete and integrated circuit design. In particular, the BJT is still the preferred device in many applications that require high speed and/or high precision such as op-amps, A/D and D/A converters, and wireless communication products.
The basic physical structure of the BJT consists of a three-layer sandwich of alternating p- and n-type semiconductor materials and can be fabricated in either npn or pnp form.
The emitter of the transistor injects carriers into the base. Most of these carriers traverse the base region and are collected by the collector. The carriers that do not completely traverse the base region give rise to a small current in the base terminal.
A mathematical model called the transport model (a simplified Gummel-Poon model) characterizes the i-v characteristics of the bipolar transistor for general terminal voltage and current conditions. The transport model requires three unique parameters to characterize a particular BJT: the saturation current IS and the forward and reverse common-emitter current gains βF and βR.
βF is a relatively large number, ranging from 20 to 500, and characterizes the significant current amplification capability of the BJT. Practical fabrication limitations cause the bipolar transistor structure to be inherently asymmetric, and the value of βR is much smaller than βF, typically between 0 and 2.
The classical Ebers-Moll model can be obtained from a rearrangement of the transport model equations.
SPICE circuit analysis programs contain a comprehensive built-in model for the transistor that is an extension of the transport model.
Four regions of operation -- cutoff, forward-active, reverse-active, and saturation -- were identified for the BJT based on the bias voltages applied to the base-emitter and base-collector junctions. The transport model can be simplified for each individual region of operation.
The cutoff and saturation regions are most often used in switching applications and logic circuits. In cutoff, the transistor approximates an open switch, whereas in saturation, the transistor represents a closed switch. However, in contrast to the "on" MOSFET, the saturated bipolar transistor has a small voltage, the collector-emitter saturation voltage VCESAT, between its collector and emitter terminals, even when operating with zero collector current.
In the forward-active region, the bipolar transistor can provide high voltage and current gain for amplification of analog signals. The reverse-active region finds limited use in some analog- and digital-switching applications.
The i-v characteristics of the bipolar transistor are often presented graphically in the form of the output characteristics, iC versus vCE or vCB, and the transfer characteristics, iC versus vBE or vEB.
The transconductance gm of the bipolar transistor in the forward-active region relates differential changes in collector current and base-emitter voltage and was shown to be directly proportional to the dc collector current IC.
In the forward-active region, the collector current increases slightly as the collector-emitter voltage increases. The origin of this effect is base-width modulation, known as the Early effect, and it can be included in the model for the forward-active region through addition of the parameter called the Early voltage VA.
The collector current of the bipolar transistor is determined by minority-carrier diffusion across the base of the transistor, and expressions were developed that relate the saturation current and base transit time of the transistor to physical device parameters. The base width plays a crucial role in determining the base transit time and the high-frequency operating limits of the transistor.
Minority-carrier charge is stored in the base of the transistor during its operation, and changes in this stored charge with applied voltage result in diffusion capacitances being associated with forward-biased junctions. The value of the diffusion capacitance is proportional to the collector current IC.
The capacitances of the bipolar transistor cause the current gain to be frequency-dependent. At the beta cutoff frequency fβ, the current gain has fallen to 71 percent of its low frequency value, whereas the value of the current gain is only 1 at the unity-gain frequency fT.
A number of biasing circuits were analyzed to determine the Q-point of the transistor. Design of the four-resistor network was investigated in detail. The four-resistor bias circuit provides highly stable control of the operating point and is the most important bias circuit for discrete design.
The current mirror circuit, which is extremely important for biasing integrated circuits, relies on the use of closely matched transistors for proper operation.
Techniques for analyzing the influence of element tolerances on circuit performance include the worst-case analysis and statistical Monte Carlo analysis methods. In worst-case analysis, element values are simultaneously pushed to their extremes, and the resulting predictions of circuit behavior are often overly pessimistic. The Monte Carlo method analyzes a large number of randomly selected versions of a circuit to build up a realistic estimate of the statistical distribution of circuit performance. Random number generators in high-level computer languages, spreadsheets, or MATLAB can be used to randomly select element values for use in the Monte Carlo analysis. Some circuit analysis packages such as PSPICE provide a Monte Carlo analysis option as part of the program.
