Microwave Devices and Semiconductor Technologies

Introduction and Context

Microwave engineering depends on devices that can generate, amplify, control, detect, and convert electromagnetic energy at radio and millimetre-wave frequencies. At these frequencies, device dimensions become comparable with the wavelength, carrier transit times become significant, parasitic reactances dominate, and conventional low-frequency circuit assumptions often fail. Consequently, microwave device engineering is simultaneously a semiconductor-physics problem, a high-frequency circuit problem, and a thermal-packaging problem.

The principal active technologies are semiconductor devices based on silicon, gallium arsenide (GaAs), indium phosphide (InP), gallium nitride (GaN), silicon carbide (SiC), and silicon-germanium (SiGe). Device selection depends on frequency, output power, noise figure, linearity, efficiency, breakdown voltage, temperature, cost, and integration requirements.

A microwave device is useful only when its intrinsic semiconductor properties, package, bias network, transmission lines, and thermal environment are considered together. This chapter therefore treats devices as components of complete microwave systems rather than as isolated circuit symbols.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Explain the physical limitations that distinguish microwave semiconductor devices from low-frequency devices.
2. Compare microwave diodes, bipolar transistors, field-effect transistors, and compound-semiconductor power devices.
3. Derive and apply important expressions for gain, noise, transit frequency, and diode switching behaviour.
4. Select suitable device technologies for low-noise, high-power, switching, oscillation, and frequency-conversion applications.
5. Analyse packaging, thermal, reliability, linearity, and biasing constraints.
6. Evaluate contemporary developments such as GaN HEMTs, SiGe BiCMOS, monolithic microwave integrated circuits, and wide-bandgap technologies.

1. Microwave Device Physics

1.1 Wavelength, transit time, and distributed behaviour

The free-space wavelength is

where λ₀ is wavelength in metres, c ≈ 3 × 10⁸ m/s is the speed of light in vacuum, and f is frequency in hertz. At 10 GHz, λ₀ is approximately 30 mm. A device package or interconnect only a few millimetres long can therefore introduce appreciable phase shift.

The electrical length of a physical length l in a medium of effective relative permittivity ε_eff is approximately

where β is phase constant in radians per metre. This relation assumes a quasi-TEM or TEM-like propagation mode and a sufficiently uniform transmission medium.

A lumped-element approximation is valid only when the largest physical dimension is much smaller than the wavelength in the relevant medium. Otherwise, inductance, capacitance, and resistance must be represented as distributed quantities.

1.2 Carrier transit time

The transit time τ_t is the time required for carriers to cross the active region. A simplified drift estimate is

where L is active-region length and v_d is carrier drift velocity. At high electric fields, v_d approaches a saturation velocity rather than increasing indefinitely with field.

For a simple common-emitter bipolar transistor, the current-gain transition frequency is approximately

where τ_F is the effective forward transit time, including base transit, emitter charging, collector depletion, and other internal delay contributions. This is an approximation; real devices also include parasitic capacitances and resistances.

A transistor intended to operate at frequency f generally requires f_T substantially greater than f. The ratio f_T/f is not itself a guaranteed gain, because available gain also depends on bias, feedback, stability, matching, and parasitics.

1.3 Small-signal and large-signal operation

Small-signal models linearize a nonlinear device about a bias point. They are suitable for gain, input impedance, output impedance, noise, and stability calculations when signal excursions are sufficiently small. Large-signal models are necessary for power compression, harmonic generation, switching, efficiency, and nonlinear distortion.

The hybrid-π model of a bipolar transistor contains transconductance g_m, base-emitter capacitance C_π, base-collector capacitance C_μ, and resistive elements. At temperature T, the transconductance under forward-active operation is

where I_C is collector current and V_T = kT/q is thermal voltage. Here k is Boltzmann’s constant and q is elementary charge. At approximately 300 K, V_T ≈ 25.9 mV. The relation assumes an exponential forward-active junction and neglects high-injection and saturation effects.

For a field-effect transistor, the small-signal drain current is commonly written

where g_m is transconductance, g_ds is drain-source output conductance, v_gs is small-signal gate-source voltage, and v_ds is small-signal drain-source voltage. A more complete high-frequency model includes gate resistance, source inductance, drain inductance, and gate-drain feedback capacitance.

2. Microwave Diodes

2.1 p–n junction diode

A p–n junction diode conducts primarily in one direction. Its ideal current-voltage relationship is

where I_S is reverse saturation current, V is applied junction voltage, n is ideality factor, q is electronic charge, k is Boltzmann’s constant, and T is absolute temperature. The equation applies to an idealized junction away from high-level injection, series-resistance limitation, breakdown, and strong high-frequency effects.

At microwave frequencies, the diode is not characterized only by its static I–V curve. Junction capacitance, package inductance, series resistance, carrier storage, and substrate coupling determine its impedance and switching speed.

2.2 Schottky diode

A Schottky diode uses a metal–semiconductor junction. It is a majority-carrier device and therefore has negligible minority-carrier storage. Its advantages include low forward voltage, fast switching, and low junction charge. Limitations include relatively high reverse leakage, lower reverse breakdown than many p–n diodes, and sensitivity to parasitic inductance.

Schottky diodes are widely used as microwave detectors, mixers, frequency multipliers, samplers, and fast switching elements. In a square-law detector operated at sufficiently small voltage, the output is approximately proportional to the mean-square input voltage:

The proportionality is valid only in the small-signal region; at larger signals the detector becomes linear or compresses depending on its design.

2.3 Varactor diode

A reverse-biased p–n junction can act as a voltage-controlled capacitor. An ideal abrupt-junction approximation gives

where C_j is junction capacitance, C_j0 is zero-bias capacitance, V_R is reverse voltage, and V_bi is built-in potential. The expression assumes an abrupt junction and neglects series resistance, parasitic capacitance, grading, and breakdown.

Varactors are used in voltage-controlled oscillators, electronic tuning, parametric amplifiers, frequency multipliers, and phase shifters. Their practical performance is governed by the capacitance ratio, quality factor, tuning range, power handling, and bias isolation.

2.4 PIN diode

A PIN diode contains p-type, intrinsic, and n-type regions. Under forward bias, stored charge in the intrinsic region reduces RF resistance; under reverse bias, the diode presents a high resistance with junction capacitance. It can therefore function as a microwave switch or variable attenuator.

A simplified forward-biased RF resistance is

where W is intrinsic-layer width, μ_n is an effective carrier mobility, τ is carrier lifetime, and I_F is forward current. This engineering approximation assumes uniform stored charge and neglects high-level effects, recombination nonuniformity, and contact resistance.

PIN switches offer high power handling and good isolation, but their switching speed is limited by charge storage. They are used in radar protection, T/R modules, attenuators, and microwave instrumentation.

2.5 IMPATT, TRAPATT, and tunnel diodes

IMPATT diodes generate negative resistance through avalanche multiplication and carrier transit-time delay. They can produce substantial microwave power but exhibit high noise because avalanche multiplication is stochastic. TRAPATT devices use avalanche-domain dynamics and can provide high efficiency at lower microwave frequencies, although spectral purity is poorer.

Tunnel diodes exploit heavily doped junctions and quantum tunnelling. Their negative differential resistance supports oscillation and amplification over selected frequency ranges. They are historically important and remain useful in specialised high-speed and low-voltage applications, but modern transistor technologies dominate most integrated systems.

3. Microwave Transistors

3.1 Bipolar junction transistors

Microwave BJTs use a thin base and carefully engineered emitter and collector regions. Heterojunction bipolar transistors (HBTs) use different semiconductor materials in the emitter and base to improve injection efficiency and permit a narrow, highly doped base. GaAs and InP HBTs provide high speed, while SiGe HBTs provide excellent integration with silicon circuitry.

The maximum oscillation frequency f_max is influenced by f_T, base resistance, collector-base capacitance, and other parasitics. A useful approximate relationship is

where R_B is base resistance and C_μ is base-collector capacitance. This is an approximate unilateral high-frequency estimate, not a universal device law. It indicates why reducing base resistance and feedback capacitance is important.

Advantages of HBTs include high transconductance per unit current, good linearity, and strong performance in integrated oscillators, mixers, and power amplifiers. Their disadvantages include bias sensitivity, thermal feedback, and potentially lower breakdown voltage than some power FET technologies.

3.2 MESFETs

A metal–semiconductor field-effect transistor controls a conducting channel using a gate depletion region. GaAs MESFETs were important in early microwave integrated circuits because GaAs offers high electron mobility and semi-insulating substrates.

The gate is commonly a Schottky junction. Advantages include low gate capacitance and high speed; disadvantages include gate leakage, limited voltage swing, and lower integration density than modern silicon technologies in many applications.

3.3 HEMTs and pseudomorphic HEMTs

A high-electron-mobility transistor forms a two-dimensional electron gas (2DEG) at a heterointerface. The carriers are spatially separated from ionized donors, reducing impurity scattering and producing high mobility. In a pseudomorphic HEMT, a thin strained layer enhances band engineering and carrier confinement.

GaAs pHEMTs are widely used in low-noise amplifiers and microwave front ends. InP HEMTs offer exceptional high-frequency performance and are used in very high-speed receivers, instrumentation, and photonic-electronic systems. GaN HEMTs combine high electron velocity with high breakdown field and high thermal capability, making them attractive for power amplifiers and radar systems.

A simplified field-effect current model in the gradual-channel, non-saturated region is

where μ is carrier mobility, C_ox is gate-oxide capacitance per unit area, W and L are channel width and length, V_GS is gate-source voltage, V_DS is drain-source voltage, and V_TH is threshold voltage. The model assumes a long channel, uniform mobility, negligible channel-length modulation, and an ideal gate. It is not sufficiently accurate for short-channel microwave devices, where velocity saturation and parasitic effects are important.

3.4 GaN HEMTs

GaN has a wide bandgap, high critical electric field, and high thermal stability. AlGaN/GaN heterostructures generate a high-density 2DEG without conventional intentional doping. The resulting devices support high voltage, high current density, and high power density.

GaN HEMTs are used in satellite communications, electronic warfare, 5G infrastructure, automotive radar, and high-power radar. Important limitations include trapping and current collapse, dynamic on-resistance, surface states, thermal resistance, gate reliability, and nonlinear memory effects. Device passivation, field plates, substrate choice, and pulsed characterization are therefore essential.

4. Microwave Performance Measures

4.1 Scattering parameters

At microwave frequencies, voltage and current are difficult to define uniquely at device ports because travelling waves are more readily measured. Scattering parameters relate incident and reflected waves:

Here a_i and b_i are normalized incident and outgoing wave amplitudes at port i. For a two-port network, S_11 is input reflection with port 2 matched, S_21 is forward transmission, S_12 is reverse transmission, and S_22 is output reflection. Magnitudes are dimensionless; decibels are calculated as 20 log₁₀|S_ij| for voltage-wave ratios.

The transducer power gain is

where Γ_S and Γ_L are source and load reflection coefficients. This expression assumes consistent reference impedances, usually 50 Ω, and linear steady-state operation.

4.2 Stability

A potentially unstable two-port may oscillate for some source or load terminations. The Rollett stability factor is

where

For unconditional stability with passive source and load terminations, a commonly used sufficient condition is K > 1 and |∆| < 1. If these conditions are not met, stability circles and source/load reflection-coefficient regions must be analysed.

4.3 Noise

Thermal noise power available from a resistor at temperature T over bandwidth B is

where k is Boltzmann’s constant. A real amplifier is described by noise figure

where SNR denotes signal-to-noise ratio. Noise figure in decibels is F_dB = 10 log₁₀F.

For cascaded stages, Friis’ formula is

where F_i are linear noise factors and G_i are preceding-stage available power gains, also linear. The formula assumes matched or appropriately defined cascaded stages and shows why the first low-noise stage dominates receiver noise performance.

4.4 Power, efficiency, and linearity

Power-added efficiency is

where P_out is RF output power, P_in is RF input power, and P_DC is supplied direct-current power. Drain efficiency is P_out/P_DC, excluding input RF power. PAE is especially useful for comparing power amplifiers with different drive levels.

Important linearity measures include 1 dB compression point, third-order intercept point, adjacent-channel power ratio, error-vector magnitude, and intermodulation distortion. No single measure is sufficient: a device may have high saturated power but poor spectral regrowth, or excellent small-signal linearity but low efficiency.

5. Semiconductor Material Technologies

Material selection is a system-level decision. High mobility alone does not guarantee superior performance; thermal conductivity, breakdown field, substrate loss, manufacturability, packaging, and reliability may be more decisive.

6. Packaging, Biasing, and Thermal Design

6.1 Packaging as part of the circuit

At microwave frequencies, package leads behave as inductors, bond pads as capacitors, and ground connections as distributed impedances. A package model should include lead inductance, pad capacitance, mutual coupling, substrate modes, and thermal paths. Transition discontinuities can cause reflection and reduce gain even when the intrinsic transistor is excellent.

Best practice includes electromagnetic simulation of the package and launch, calibration structures on the test substrate, short and low-inductance ground returns, controlled-impedance interconnects, and de-embedding of fixtures.

6.2 Bias networks

A bias network must deliver DC while presenting a suitable RF impedance. RF chokes, quarter-wave lines, resistive damping, bypass capacitors, and bias tees are commonly used. The network must remain stable over frequency and bias variation; a bias circuit that is ideal at the operating frequency may resonate at a harmonic or low-frequency envelope frequency.

Gate and base protection are particularly important. Excessive gate voltage can damage a HEMT, while bipolar devices may suffer thermal runaway if collector current increases with temperature and the bias network does not provide negative feedback.

6.3 Thermal management

The junction temperature can be approximated by

where T_J is junction temperature, T_A is ambient temperature, P_D is dissipated power, and θ_JA is junction-to-ambient thermal resistance in K/W. For a mounted device, a more useful network is

where θ_JC, θ_CS, and θ_SA are junction-to-case, case-to-sink, and sink-to-ambient thermal resistances. The expression assumes steady-state heat flow and approximately linear thermal resistance. Transient operation requires thermal impedance rather than a single resistance.

7. Worked Example: Cascaded Receiver Noise

A receiver has a first-stage LNA with gain 15 dB and noise figure 1.5 dB, followed by a mixer with conversion gain −6 dB and noise figure 8 dB. Find the cascaded noise figure.

Convert quantities to linear form:

G₁ = 10^(15/10) ≈ 31.62,

F₁ = 10^(1.5/10) ≈ 1.413,

F₂ = 10^(8/10) ≈ 6.310.

Using Friis’ formula for two stages,

Therefore,

F_total ≈ 1.413 + (6.310 − 1)/31.62 ≈ 1.581.

In decibels,

Thus, the receiver noise figure is approximately 2.0 dB. The mixer has a high noise figure, but the high gain of the LNA suppresses its contribution.

8. Applied Case: GaN Power Amplifier for a Radar Transmitter

Background

Consider an illustrative design team developing an X-band solid-state radar transmitter. The required output is 50 W pulsed RF power near 10 GHz, with high reliability under elevated ambient temperature and a limited cooling volume.

Problem

A silicon transistor offers low cost and easy control integration, while a GaAs device offers mature microwave performance. Neither provides the desired combination of voltage swing, power density, and thermal margin. The team must determine whether GaN HEMT technology is appropriate.

Analysis

The radar waveform has high peak-to-average ratio, so saturated power alone is insufficient. The design must consider pulse droop, trapping, drain efficiency, gain compression, harmonic filtering, load mismatch, gate reliability, and junction temperature. Pulsed I–V and RF load-pull measurements are required because static data may not reveal dynamic current collapse.

The matching network must transform the transistor’s optimum load impedance to the system impedance while limiting voltage and current stress. Electromagnetic simulation is needed for the package, transistor cell, and combining network. Thermal analysis must include duty cycle and transient thermal impedance.

Approach and solution

The team selects a packaged GaN HEMT and designs a two-stage architecture: a driver stage for gain and a power stage for output power. The input and output matching networks use low-loss microstrip structures with harmonic-aware bias networks. A directional coupler monitors forward and reflected power, and protection logic reduces drive under severe mismatch. The heat spreader is connected to a low-resistance thermal path, and the device is qualified using temperature cycling, power cycling, and load-mismatch tests.

Outcome

The resulting design meets the peak-power requirement with fewer combining stages than a comparable GaAs design. However, the final solution requires careful trap management, gate protection, thermal design, and calibration. The lower transistor count reduces assembly complexity but does not eliminate system-level reliability engineering.

Learning points

1. Material choice must follow system requirements rather than frequency alone.
2. Pulsed and dynamic measurements are essential for GaN devices.
3. Matching, packaging, biasing, and thermal design determine usable—not merely intrinsic—performance.
4. Reliability margins should be designed in from the beginning.

9. Contemporary Developments and Research Directions

9.1 Monolithic microwave integrated circuits

MMICs integrate active devices, passive elements, transmission lines, and sometimes control circuits on one substrate. Integration reduces assembly parasitics and improves repeatability, but passive losses, heat removal, process variation, and limited tuning range remain important challenges.

9.2 Silicon at millimetre-wave frequencies

Advanced CMOS and SiGe technologies support radar, imaging, sensing, and 5G/6G transceivers. Their strengths are low cost, dense digital integration, calibration, and beamforming. Challenges include reduced voltage headroom, passive loss, device variability, and power-combining complexity.

9.3 Wide-bandgap and ultra-wide-bandgap devices

GaN and SiC already support high-power microwave and high-temperature applications. Research on diamond heat spreaders, improved GaN substrates, vertical GaN structures, gallium oxide, and diamond-based devices seeks higher breakdown fields and better thermal performance. The engineering challenge is not only material performance but also defect density, contacts, packaging, and manufacturability.

9.4 AI-assisted device and circuit design

Machine learning can accelerate compact-model extraction, electromagnetic surrogate modelling, yield prediction, and nonlinear predistortion. It does not replace physical validation. Training data may omit rare failure modes, and opaque models can conceal extrapolation errors. Best practice combines physics-based constraints, uncertainty estimation, independent test data, and human review.

10. Limitations, Reliability, and Responsible Engineering

Microwave semiconductor designs face coupled limitations: gain decreases with frequency, noise depends on bias and matching, power density raises junction temperature, and nonlinear operation produces unwanted radiation. Device data sheets often specify typical rather than guaranteed performance, so production tolerances and ageing must be included in design verification.

Reliability evaluation should address electromigration, hot-carrier degradation, gate dielectric or Schottky-gate failure, thermal cycling, moisture, vibration, cosmic radiation where relevant, and load mismatch. Safety and regulatory requirements may include electromagnetic compatibility, spectrum occupancy, exposure limits, and export controls for advanced microwave hardware.

Responsible engineering requires accurate performance claims, traceable calibration, protection against unsafe RF exposure, secure handling of radar and communications technology, and consideration of energy consumption and end-of-life material recovery.

Design Best Practices

• Select technology using a weighted system-level trade-off among gain, noise, power, efficiency, linearity, cost, and reliability.
• Use S-parameters only within their specified bias, frequency, power, and temperature conditions.
• Convert decibel quantities to linear quantities before using cascade or power equations.
• Simulate package, bias network, thermal path, and electromagnetic transitions together.
• Verify both small-signal and large-signal behaviour.
• Characterize dynamic trapping, memory effects, and mismatch tolerance for power devices.
• Provide bias sequencing and current limiting for sensitive gates and junctions.
• Design for manufacturing variation and calibrate where appropriate.
• Validate models with independent measurements and document uncertainty.

Chapter Summary

Microwave semiconductor devices operate under conditions in which carrier transit time, parasitic reactance, distributed interconnects, thermal effects, and nonlinear behaviour are central design variables. Schottky, varactor, PIN, IMPATT, and tunnel diodes provide detection, tuning, switching, multiplication, and negative-resistance functions. BJTs and HBTs provide high transconductance and strong integration, while MESFETs, pHEMTs, HEMTs, and GaN HEMTs support high-frequency and high-power operation.

Scattering parameters, stability factors, noise figure, transducer gain, power-added efficiency, and compression measures provide the principal analytical language of microwave devices. Semiconductor choice involves a trade-off among mobility, breakdown field, thermal conductivity, substrate properties, integration, cost, and reliability. Packaging, biasing, electromagnetic transitions, and thermal management are part of the device design, not secondary implementation details. Modern research combines wide-bandgap materials, MMIC integration, advanced silicon technologies, improved thermal solutions, and physics-informed computational design.

Key Terms

Microwave device: A component designed to generate, control, amplify, detect, or convert microwave signals.

Transit time: Time required for carriers to cross an active semiconductor region.

Schottky diode: A majority-carrier metal–semiconductor diode used for fast switching and detection.

Varactor: A reverse-biased junction used as a voltage-controlled capacitor.

PIN diode: A p–i–n structure used as a microwave switch or attenuator.

IMPATT diode: An avalanche transit-time device exhibiting microwave negative resistance.

HBT: Heterojunction bipolar transistor with improved injection and high-frequency performance.

MESFET: Metal–semiconductor field-effect transistor.

HEMT: Heterostructure field-effect transistor using a high-mobility channel.

2DEG: Two-dimensional electron gas formed at a semiconductor heterointerface.

GaN HEMT: High-power HEMT based on gallium nitride heterostructures.

f_T: Current-gain transition frequency of a transistor.

f_max: Maximum oscillation frequency under specified high-frequency conditions.

S-parameter: Ratio describing incident and reflected travelling-wave amplitudes.

Noise figure: Degradation of signal-to-noise ratio caused by a device.

Stability factor: A parameter used to assess unconditional stability of a two-port.

PAE: Power-added efficiency of an amplifier.

Load-pull: Measurement or simulation method for determining performance versus load impedance.

Dynamic trapping: Time-dependent carrier trapping that alters device current and RF behaviour.

MMIC: Monolithic microwave integrated circuit.

Review Questions

Short Answer Questions

1. Why do parasitic capacitances and inductances become significant in microwave devices?
2. Distinguish a Schottky diode from a p–n junction diode in terms of carrier storage.
3. What is the physical significance of transistor transition frequency f_T?
4. Define noise figure and explain why it is usually expressed in decibels.
5. State two important advantages and two limitations of GaN HEMTs.

Long Answer Questions

1. Explain the structure, operation, and microwave applications of varactor and PIN diodes.
2. Compare HBT, GaAs pHEMT, InP HEMT, and GaN HEMT technologies for microwave systems.
3. Derive the role of transit time and parasitic capacitance in limiting transistor high-frequency performance.
4. Explain S-parameters, the Rollett stability factor, and their use in microwave amplifier design.
5. Discuss packaging, biasing, and thermal management as integrated parts of microwave device engineering.

Analytical / Application Questions

1. A two-stage receiver has first-stage gain 12 dB and noise figure 2 dB, followed by a second stage with noise figure 7 dB. Calculate the cascaded noise figure, assuming the second-stage gain is not needed.
2. A power amplifier delivers 20 W RF output, receives 2 W RF input, and consumes 40 W DC power. Calculate its power-added efficiency and explain whether the result is consistent with a physically reasonable amplifier.
3. A designer must choose between a GaAs pHEMT LNA and a GaN HEMT LNA for a low-noise satellite receiver. Develop a justified selection framework considering noise, power, linearity, thermal environment, reliability, and cost.

Multiple Choice Questions with Answers

Question 1

At 10 GHz, the free-space wavelength is approximately:

A. 3 mm

B. 30 mm

C. 300 mm

D. 3 m

Correct Answer: B. 30 mm

Question 2

Which device is primarily a majority-carrier device and therefore has negligible minority-carrier storage?

A. Schottky diode

B. PIN diode

C. Bipolar junction diode

D. IMPATT diode

Correct Answer: A. Schottky diode

Question 3

For unconditional stability using the common Rollett criterion, which condition is required together with |∆| < 1?

A. K < 0

B. K = 0

C. K > 1

D. K = −1

Correct Answer: C. K > 1

Question 4

In Friis’ noise formula, why is high gain in the first stage valuable?

A. It increases the second-stage noise contribution

B. It suppresses the contribution of subsequent stages

C. It eliminates thermal noise in the source

D. It makes all stages noiseless

Correct Answer: B. It suppresses the contribution of subsequent stages

Question 5

Which material is especially associated with high breakdown field and high microwave power density?

A. Silicon dioxide

B. Germanium

C. Gallium nitride

D. Aluminium

Correct Answer: C. Gallium nitride

Question 6

A transistor has collector current 10 mA at 300 K. Using V_T = 25.9 mV, its approximate transconductance is:

A. 0.039 S

B. 0.386 S

C. 2.59 S

D. 25.9 S

Correct Answer: B. 0.386 S

Question 7

A varactor diode is normally operated under:

A. Forward bias only

B. Reverse bias

C. Zero bias with avalanche conduction

D. Thermal breakdown

Correct Answer: B. Reverse bias

Question 8

For a two-port network, S₂₁ primarily represents:

A. Reverse reflection at port 1

B. Forward transmission from port 1 to port 2

C. Output reflection with port 1 matched

D. Input impedance in ohms

Correct Answer: B. Forward transmission from port 1 to port 2

Question 9

If an amplifier provides 20 W output, 2 W input, and consumes 40 W DC, its PAE is:

A. 22.5%

B. 45%

C. 50%

D. 90%

Correct Answer: B. 45%

Question 10

The two-dimensional electron gas in a HEMT is important because it provides:

A. A high-mobility conduction channel

B. Permanent reverse breakdown

C. Zero gate capacitance

D. Perfect thermal isolation

Correct Answer: A. A high-mobility conduction channel