Q6Microwave Engineering 2

(a) Numerical: Pinch-off Voltage of GaAs MESFET

Given data:

• Channel height, a = 0.1 micrometre = 0.1 x 10^-6 m = 1 x 10^-7 m
• Electron (doping) concentration, Nd = 8 x 10^17 cm^-3 = 8 x 10^23 m^-3 (converting cm^-3 to m^-3 by multiplying by 10^6)
• Relative dielectric constant, epsilon_r = 13.10

The pinch-off voltage of a MESFET (the gate-to-channel reverse voltage at which the depletion region extends completely through the channel height, cutting off conduction) is given by the standard parallel-plate-capacitor-derived formula:

where q=1.602x10^-19 C is the electron charge, and epsilon_0=8.854x10^-12 F/m is the permittivity of free space. Substituting the given values:

Calculating the numerator: (1.602x10^-19) x (8x10^23) x (1x10^-14) = (1.602x8)x10^(-19+23-14) = 12.816 x 10^-10 = 1.2816x10^-9. Calculating the denominator: 2 x 8.854x10^-12 x 13.10 = 2.32 x 10^-10 (approximately).

Result: the pinch-off voltage of the given GaAs MESFET is approximately 5.52 V. This calculation directly follows from treating the depleted region beneath the gate as a parallel-plate structure whose total depletion charge (q times Nd times the depleted volume per unit area, i.e., q Nd a) must be supported by the electric field integrated across the channel height a, giving a total pinch-off voltage proportional to the doping concentration and to the square of the channel height, and inversely proportional to the material's permittivity — this relationship shows that a MESFET with a thinner channel (smaller a) or lower channel doping (smaller Nd) would have a correspondingly lower pinch-off voltage, an important design trade-off since channel thickness and doping also directly affect the device's maximum current-carrying capability and transconductance.

(b) Structure, Operation, and Layout of Microwave BJT

A microwave bipolar junction transistor (BJT) is a specialized BJT design specifically optimized to provide useful current gain and power output at microwave frequencies (up to several GHz), requiring substantial departures from the simple planar geometry of a low-frequency BJT to overcome the frequency-limiting effects of base transit time, junction capacitance, and parasitic base resistance.

Structure and layout: rather than the simple single, large-area emitter region used in a low-frequency BJT, a microwave BJT uses an interdigitated (comb-like, finger-shaped) emitter-base geometry, in which many long, narrow emitter 'fingers' alternate with corresponding narrow base contact fingers across the die surface, all sharing a common underlying collector region — this layout maximizes the total emitter periphery (edge length) relative to the total emitter area, which is important because high-frequency current tends to crowd toward the emitter edge (emitter current crowding effect), and a design with a high periphery-to-area ratio makes more effective use of the emitter for useful current conduction at high frequency. The base region between the finger-shaped emitter and its adjacent base contact is also made as physically narrow as possible (minimizing the lateral base resistance the base current must flow through to reach each point along the emitter edge), and the base layer itself is made very thin (minimizing the vertical minority-carrier transit time across the base, one of the fundamental frequency-limiting delays in any BJT).

Operation: the basic operating principle of a microwave BJT remains the same as that of a conventional low-frequency BJT — a forward-biased base-emitter junction injects minority carriers into the base region, most of which diffuse across the thin base and are collected by the reverse-biased base-collector junction, giving current gain (Beta = collector current / base current) — but achieving useful gain at microwave frequencies specifically requires minimizing every frequency-limiting delay and parasitic element in this basic process: the base transit time (via an extremely thin, heavily-doped base layer, often using a graded or heterojunction base-doping profile to build in an additional accelerating electric field that speeds minority carriers across the base), the base resistance (via the narrow-finger interdigitated layout described above, and heavy base doping), the collector-base junction capacitance and collector depletion-region transit time (via careful collector doping and thickness optimization, balancing capacitance against breakdown voltage and power-handling requirements), and the emitter-base junction capacitance (via minimizing emitter junction area for a given current-carrying requirement, again favoring the narrow-finger layout).

Practical application range: with these specialized structural optimizations, microwave BJTs (including modern heterojunction bipolar transistor, HBT, variants using compound semiconductor materials such as GaAs or InP for further improved high-frequency performance) can provide useful power gain and reasonably efficient power output up to several GHz, making them suitable for microwave power amplifier and oscillator applications at the lower end of the microwave frequency spectrum, though at the very highest microwave and millimeter-wave frequencies, field-effect devices (such as the MESFET and HEMT discussed elsewhere in this paper) generally offer superior high-frequency performance and have become the dominant active device technology for those applications.

Sensitivity of the pinch-off result to the given parameters: it is worth noting how strongly the calculated pinch-off voltage depends on each of the three given parameters in this problem, since Vp is proportional to Nd and to the square of channel height a, but inversely proportional to the relative dielectric constant epsilon_r. This means that halving the channel height a would reduce the calculated pinch-off voltage to one quarter of its original value (5.52V would fall to approximately 1.38V), while halving the doping concentration Nd would simply halve the pinch-off voltage (to approximately 2.76V) — this strong quadratic sensitivity to channel thickness is precisely why MESFET channel thickness must be controlled to very tight tolerances during epitaxial growth, since even a small unintended variation in as-grown channel thickness across a wafer can produce a significant, and generally undesirable, spread in pinch-off voltage (and hence threshold/bias-point behavior) among nominally identical devices fabricated on the same wafer.

Comparison of microwave BJT with the MESFET discussed elsewhere in this paper: it is useful to briefly contrast the microwave BJT structure described in part (b) with the MESFET structure analyzed numerically in part (a) of this question, since both are commonly used active devices in microwave amplifier and oscillator circuits but rely on quite different physical operating principles. The BJT is fundamentally a minority-carrier (bipolar) device, relying on the diffusion of injected minority carriers across a thin base region, whereas the MESFET is a majority-carrier (unipolar) device, relying on modulating a majority-carrier channel's conducting cross-section via a reverse-biased Schottky gate junction depletion region — this majority-carrier-only operation generally gives the MESFET (and, at even higher performance, the HEMT) inherently faster switching and higher-frequency capability than a comparable BJT, since majority-carrier devices avoid the minority-carrier storage/recombination delay that fundamentally limits BJT high-frequency response, which is precisely why field-effect devices have become the dominant active-device technology at the very highest microwave and millimeter-wave frequencies, while microwave BJTs (and HBTs) remain competitive and widely used at the lower end of the microwave spectrum, particularly where their generally lower 1/f noise and good linearity characteristics are specifically valued.