A MESFET is a field-effect transistor that uses a metal-semiconductor (Schottky barrier) junction as its gate to control current flow through a semiconductor channel. Three main components make up the device: the Schottky barrier gate, the semiconductor channel, and the ohmic source and drain contacts.

Schottky barrier gate

The gate is formed by depositing a metal layer directly onto the semiconductor surface, creating a metal-semiconductor junction with rectifying properties. Current flows easily from the semiconductor to the metal but is blocked in the reverse direction.

The Schottky barrier height depends on the work function difference between the metal and the semiconductor. This height directly affects the device's threshold voltage and I-V characteristics.
Common gate metals include titanium (Ti), platinum (Pt), and nickel (Ni).
Unlike a MOSFET, there's no insulating oxide layer between the gate metal and the channel. This is what gives MESFETs their speed advantage but also means the gate draws some leakage current.

Semiconductor channel

The channel is the region between source and drain where current actually flows. It's typically made of n-type gallium arsenide (GaAs) or indium gallium arsenide (InGaAs), chosen for their high electron mobility.

Channel dimensions (length and width) directly influence current-carrying capacity and high-frequency performance. Shorter channels mean faster transit times.
The doping concentration of the channel sets the threshold voltage and determines how the depletion region forms under the gate. Higher doping means more carriers available for conduction but also changes the pinch-off behavior.

Ohmic source and drain contacts

The source and drain contacts are ohmic contacts, meaning they allow current to flow in and out of the channel with minimal resistance and no rectifying behavior.

Ohmic contacts are formed by depositing metal onto heavily doped regions of the semiconductor, creating a low-resistance interface.
The metal choice and doping level near the contacts are critical for keeping contact resistance low. Common metallization schemes include gold-germanium (AuGe) alloys and nickel-germanium-gold (NiGeAu) multilayers.
High contact resistance at these terminals degrades both gain and noise performance, so minimizing it is a major fabrication concern.

Operation principles

MESFETs are voltage-controlled devices. The gate voltage modulates the width of the depletion region under the Schottky barrier, which in turn changes the effective cross-section of the conducting channel and controls the drain current.

Depletion region control

The depletion region is the zone beneath the Schottky gate where mobile charge carriers have been swept away, leaving behind fixed ionized dopants.

The depletion width depends on the applied gate voltage and the built-in potential of the Schottky barrier.
Applying a more negative gate voltage widens the depletion region, squeezing the conducting channel and increasing its resistance.
Applying a less negative (or slightly positive) gate voltage narrows the depletion region, opening up the channel and decreasing its resistance.

Note that for an n-channel MESFET, the gate is typically operated at zero or reverse bias. Driving the gate too far positive would forward-bias the Schottky junction and cause significant gate current.

Voltage-controlled resistance

Because the gate voltage controls the depletion width, it effectively controls the channel resistance. This relationship is non-linear, which gives MESFETs their characteristic I-V curves.

As $V_{GS}$ becomes more negative, channel resistance rises and drain current drops.
This voltage-controlled resistance property is what makes MESFETs useful as voltage-controlled attenuators and RF switches, in addition to amplifiers.

Saturation and pinch-off

As the drain-source voltage $V_{DS}$ increases, the depletion region near the drain side grows wider (because the reverse bias across the gate-drain junction increases).

Pinch-off occurs when the depletion region at the drain end extends across the entire channel thickness. At this point, further increases in $V_{DS}$ don't significantly increase the drain current.
In the saturation region, the drain current stays roughly constant with increasing $V_{DS}$ . The device acts like a voltage-controlled current source.
The pinch-off voltage $V_P$ is the gate voltage at which the channel is completely depleted with $V_{DS} = 0$ . This is a key design parameter.
Saturation behavior is essential for amplifier and oscillator operation, where you need a stable current source controlled by the gate.

Current-voltage characteristics

The I-V characteristics describe how drain current $I_D$ depends on gate-source voltage $V_{GS}$ and drain-source voltage $V_{DS}$ . Two distinct operating regions define the behavior.

Linear region

At low $V_{DS}$ , the MESFET behaves like a voltage-controlled resistor.

Drain current increases approximately linearly with $V_{DS}$ .
The slope of the I-V curve is set by the channel resistance, which is modulated by $V_{GS}$ .
The device's transconductance is highest in this region, which matters for certain low-noise amplifier designs.

Schottky barrier gate, Schottky Diode - Electronics-Lab.com

Saturation region

When $V_{DS}$ exceeds the value needed to pinch off the channel near the drain, the device enters saturation.

Drain current becomes nearly independent of $V_{DS}$ and is primarily controlled by $V_{GS}$ .
This is the normal operating region for amplifiers, oscillators, and power amplifiers, where a constant-current-source behavior is needed.
The boundary between linear and saturation regions occurs roughly at $V_{DS} = V_{GS} - V_P$ .

Transconductance and output resistance

These two parameters largely determine the MESFET's gain and frequency response.

Transconductance $g_m$ measures how much $I_D$ changes per unit change in $V_{GS}$ :

$g_m = \frac{\partial I_D}{\partial V_{GS}}\bigg|_{V_{DS}=\text{const}}$

A higher $g_m$ means more gain. It's largest in the linear region and decreases as the device moves deeper into saturation.

Output resistance $r_o$ measures how much $I_D$ changes per unit change in $V_{DS}$ in saturation:

$r_o = \left(\frac{\partial I_D}{\partial V_{DS}}\bigg|_{V_{GS}=\text{const}}\right)^{-1}$

A high $r_o$ is desirable because it means the current source is more ideal, giving better voltage gain and improved isolation between input and output.

Small-signal model

The small-signal model is an equivalent circuit that captures the MESFET's behavior for AC signals superimposed on the DC operating point. It's essential for designing amplifiers, oscillators, and mixers.

Equivalent circuit elements

The model includes these key elements:

Transconductance $g_m$ : a voltage-controlled current source representing the gain mechanism
Output resistance $r_o$ : models the finite slope of the I-V curves in saturation
Gate-source capacitance $C_{gs}$ : arises from the depletion region under the gate; this is the dominant input capacitance
Gate-drain capacitance $C_{gd}$ : due to fringing fields and the depletion region extending toward the drain; smaller than $C_{gs}$ but important for feedback and stability
Source resistance $R_s$ : resistance of the source ohmic contact plus the semiconductor between source and gate
Drain resistance $R_d$ : resistance of the drain ohmic contact plus the semiconductor between gate and drain

These elements are arranged with $g_m$ and $r_o$ between the intrinsic drain and source nodes, $C_{gs}$ across the intrinsic gate-source, $C_{gd}$ across the intrinsic gate-drain, and $R_s$ and $R_d$ in series with the external source and drain terminals.

Capacitances and resistances

$C_{gs}$ and $C_{gd}$ are both voltage-dependent, meaning they change with the bias point. This nonlinearity matters for large-signal analysis but is treated as constant at a given operating point in small-signal analysis.
$C_{gs}$ dominates the input impedance and largely determines the frequency at which gain starts to roll off.
$C_{gd}$ , though smaller, creates a feedback path from output to input. At high frequencies, this feedback can cause instability if not properly managed.
$R_s$ and $R_d$ add noise and reduce the effective transconductance seen at the external terminals. Minimizing them through good ohmic contact fabrication and material selection is critical for performance.

Frequency response and cutoff frequency

The cutoff frequency $f_T$ is the frequency at which the short-circuit current gain drops to unity. It's the single most important figure of merit for high-frequency capability:

$f_T = \frac{g_m}{2\pi(C_{gs} + C_{gd})}$

To push $f_T$ higher, you need to:

Increase $g_m$ (higher channel doping, shorter gate length, higher-mobility materials)
Decrease $C_{gs}$ and $C_{gd}$ (shorter gate length helps here too, along with optimized device geometry)

GaAs MESFETs routinely achieve $f_T$ values in the tens of GHz range, which is why they're staples in microwave circuit design.

Noise performance

Noise performance determines how well a MESFET can detect and amplify weak signals. Three noise mechanisms dominate.

Thermal noise

Thermal noise (Johnson-Nyquist noise) comes from the random thermal motion of charge carriers in the channel. Its power spectral density is:

$S_{\text{thermal}} = 4k_BTR$

where $k_B$ is Boltzmann's constant, $T$ is absolute temperature, and $R$ is the channel resistance.

This noise is flat across all frequencies ("white noise") and sets a fundamental floor on the minimum noise the device can achieve.
Reducing channel resistance through optimized geometry and high-mobility materials (like GaAs) helps minimize thermal noise.

Shot noise

Shot noise arises from the discrete, random nature of charge carriers crossing potential barriers. Its power spectral density is:

$S_{\text{shot}} = 2qI$

where $q$ is the elementary charge and $I$ is the average current.

In MESFETs, shot noise is most significant in the gate leakage current flowing through the Schottky barrier. The channel current itself doesn't produce shot noise in the same way because it doesn't cross a potential barrier.
Minimizing gate leakage by optimizing the Schottky barrier height and using high-quality semiconductor material reduces shot noise.

Flicker noise

Flicker noise (1/f noise) dominates at low frequencies. Its power spectral density is:

$S_{\text{flicker}} = \frac{K}{f^{\alpha}}$

where $K$ is a device-specific constant, $f$ is frequency, and $\alpha$ is typically close to 1.

Sources of flicker noise include traps and defects in the semiconductor, surface states at the Schottky interface, and carrier mobility fluctuations.
Flicker noise matters most for oscillator phase noise and for systems operating at lower frequencies. Improving material quality, optimizing fabrication, and applying surface passivation techniques all help reduce it.

High-frequency performance

MESFETs are built for speed. Their high-frequency performance depends on transit time, parasitic elements, and device geometry.

Transit time effects

The transit time is how long it takes carriers to travel from source to drain through the channel. The device can't respond to signals faster than this.

Transit time depends on channel length and carrier velocity. Carrier velocity, in turn, depends on the semiconductor material and the electric field in the channel.
GaAs has a peak electron velocity roughly twice that of silicon, which is a major reason GaAs MESFETs outperform Si devices at microwave frequencies.
Shortening the gate length is the most direct way to reduce transit time. Sub-micron gate lengths are standard in modern MESFET designs.

Parasitic elements

Parasitic resistances and capacitances arise from the physical device geometry and interconnections. They aren't part of the intended device operation but they degrade performance.

The main parasitic elements are $R_s$ , $R_d$ , and the pad capacitances at each terminal.
These parasitics introduce extra losses, reduce gain, and can create unwanted feedback paths that cause instability.
Designers minimize parasitics through optimized device layout, advanced fabrication techniques (like T-shaped gate profiles to reduce gate resistance), and careful microwave packaging.

Microwave and RF applications

MESFETs are widely used in microwave and RF circuits, especially in monolithic microwave integrated circuits (MMICs). Key applications include:

Low-noise amplifiers (LNAs): MESFETs serve as the input stage, providing high gain with a low noise figure. This is critical for receiving weak signals in communication and radar systems.
Power amplifiers (PAs): MESFETs in the output stage deliver high output power and efficiency for signal transmission.
Oscillators: MESFETs act as the active element to generate stable, high-frequency signals for radar and wireless systems.
Mixers: MESFETs perform frequency conversion, translating signals between different frequency bands.

Optimizing performance in these applications requires careful selection of the DC operating point, proper impedance matching networks, and appropriate biasing conditions tailored to the specific circuit requirements.

Comparison with other FETs

Choosing the right FET for a given application means understanding the trade-offs between MESFETs and their alternatives.

Junction FETs (JFETs)

JFETs use a p-n junction instead of a Schottky barrier to control channel current.
They typically exhibit lower noise than MESFETs at lower frequencies, making them good choices for audio and low-frequency analog circuits.
However, JFETs have lower maximum operating frequencies and lower transconductance, so they can't compete with MESFETs in microwave applications.

MOSFETs

MOSFETs use an insulated gate (typically $SiO_2$ ) to control the channel, giving them extremely high input impedance and negligible gate current.
Compared to MESFETs, MOSFETs offer lower power consumption, better scaling to small dimensions, and dominance in digital and mixed-signal ICs.
However, silicon MOSFETs have lower electron mobility than GaAs MESFETs, which limits their high-frequency performance. (Advanced Si MOSFETs with very short gate lengths have closed this gap significantly in recent years.)
MOSFETs also can't be easily fabricated on GaAs because GaAs lacks a stable native oxide, which is exactly why MESFETs were developed for III-V semiconductors.

Why MESFETs exist: GaAs and other III-V semiconductors offer superior electron mobility and higher saturation velocity than silicon, but they don't form stable native oxides. The Schottky barrier gate sidesteps this problem entirely, making MESFETs the natural FET structure for III-V high-frequency circuits.

HEMTs

High electron mobility transistors (HEMTs) use a heterojunction to confine electrons in a two-dimensional electron gas (2DEG) at the interface between two different semiconductor materials (e.g., AlGaAs/GaAs).
HEMTs achieve even higher electron mobility and higher $f_T$ than MESFETs because the 2DEG separates the carriers from the ionized dopants, reducing scattering.
For the most demanding microwave and millimeter-wave applications, HEMTs have largely replaced MESFETs. But MESFETs remain relevant where simpler fabrication and lower cost are priorities.

2,589 studying →