MOSFET structure and operation

A MOSFET is built as a metal-oxide-semiconductor (MOS) stack: a gate electrode sits on top of a thin oxide layer that insulates it from the semiconductor substrate below. Source and drain regions are diffused into the substrate on either side of the gate. This makes the MOSFET a fundamental electronic device in modern electronics.

The MOS capacitor: gate, oxide, and semiconductor

Because the insulated gate is separated from the channel by the oxide layer, no DC current flows into the gate terminal — the gate, oxide, and semiconductor together act as a MOS capacitor. When a positive gate voltage is applied to an N-channel device built on a p-type semiconductor, it first pushes holes away from the semiconductor surface to form a depletion region, then attracts electrons that create a thin conducting channel just under the oxide.

This is the basic MOS structure that every MOSFET symbol represents with its gate, drain, and source terminals. For a P-channel device on an n-type substrate, the polarities reverse and negative gate voltages form the channel.

How to control a MOSFET

A MOSFET is controlled by voltage, not current. An electric field set up by the gate-to-source voltage (VGS) decides whether the channel between drain and source conducts — and almost no current flows into the gate itself.

In the common enhancement-mode device, the MOSFET is normally OFF: it conducts only once VGS rises above the threshold voltage VGS(th), which forms the channel. A depletion-mode device works the opposite way — it is normally ON at VGS = 0 V and is turned OFF by a reverse-polarity gate voltage.

Enhancement-mode and depletion-mode MOSFETs

 

Because the gate is voltage-controlled and draws no steady current, a MOSFET needs very little control power — a large part of why it is so efficient and easy to drive.

The current flow from drain to source

Current in a MOSFET flows through a channel of majority carriers — electrons or holes, depending on the device — between the drain and source terminals (electrons in an N-channel device, holes in a P-channel device).

In an N-channel MOSFET, when the gate voltage is higher than the source voltage, an electron channel forms between drain and source and current flows. In a P-channel MOSFET, when the gate voltage is lower than the source, a hole channel forms and current flows in the opposite polarity.

When fully ON, a MOSFET can conduct in either direction thanks to its symmetric channel. Even when it is OFF, the intrinsic body diode allows current to flow in one direction — a detail that matters in synchronous-rectifier and bridge circuits.

Efficiency of a MOSFET

A MOSFET owes its efficiency to two things: low conduction loss and low switching loss.

Conduction loss is governed by the on-resistance RDS(on) — the lower it is, the less power the device wastes as heat while carrying current (P = I² × RDS(on)). Switching loss is tied to the gate charge Qg, the charge needed to turn the gate on and off each cycle; a lower Qg means faster, lower-energy switching.

Because MOSFETs are majority-carrier devices, they switch very fast, which keeps switching losses low even at high frequencies. Advanced structures push this further: trench MOSFETs combine very low RDS(on) and low Qg for low-voltage, high-current efficiency, while superjunction MOSFETs achieve very low RDS(on) at high voltage.

In practice, power MOSFETs typically switch from tens of kHz up to several hundred kHz, with low-voltage trench devices capable of more than 1 MHz in high-frequency DC-DC converters. Looking further ahead, wide-bandgap GaN FETs can switch roughly 5–10× faster and enable up to 4× higher power density than silicon, leading to cooler, more compact designs.

Types of MOSFETs

MOSFETs are classified along several axes:

These distinctions influence key parameters such as RDS(on), gate charge, and breakdown voltage, and they guide which device suits a given design.

Driving high-side and low-side MOSFETs

How a MOSFET is driven depends mainly on where the switch sits in the circuit.

A low-side switch is placed between the load and ground. With the source connected to ground, the gate only needs to be driven a few volts above ground to turn the device on, making low-side N-channel MOSFETs among the simplest and most efficient to drive.

A high-side switch sits between the supply and the load, so the source floats near the supply voltage. This makes gate driving more challenging:

  • A P-channel MOSFET simplifies high-side control because it turns on when the gate is pulled below the source — no bootstrap or charge-pump needed.
  • A high-side N-channel MOSFET is more efficient (lower RDS(on)) but needs a dedicated gate driver that lifts the gate above the supply rail, using level shifting and a bootstrap or charge-pump circuit.

In summary:

  • Use N-channel MOSFETs for low-side switching and efficiency-critical high-side designs with a suitable driver
  • Use P-channel MOSFETs when simple high-side control is the priority

Learn more about MOSFETs

Recommended documents

Frequently asked questions about MOSFETs

A metal–oxide–semiconductor field effect transistor (MOSFET) is a voltage-controlled semiconductor device used for switching and amplification. It has three main terminals — gate, drain, and source (plus an internal body diode) — and is controlled by the electric field from the gate rather than by a control current.

Learn more about MOSFET technology.

In most power and digital designs, yes — a MOSFET is used as an electronic switch. A gate voltage above the threshold turns it fully ON (a low-resistance path), and below the threshold it is OFF (an open circuit).

It can also act as an analog amplifier in its linear region, but switching is by far its most common role — see how does a MOSFET work as a switch for using a MOSFET as a switch in real circuits.

A MOSFET is voltage-controlled. The voltage between gate and source (VGS) sets how much current flows from drain to source, while the gate itself draws virtually no current. This is the key difference from a current-controlled BJT.

The gate is separated from the channel by a silicon dioxide insulating layer, so no DC current can pass through it.

A MOSFET turns on when the gate-to-source voltage rises above the gate threshold voltage (VGS(th)), which forms a conducting channel between drain and source. It turns off when VGS falls back below that threshold and the channel disappears.

For an enhancement-mode MOSFET, a positive gate voltage switches an N-channel device on; a depletion-mode device works the other way around.

A MOSFET is one specific kind of transistor — a metal-oxide-semiconductor field effect transistor. It belongs to the field effect transistor (FET) family and behaves differently from bipolar transistors (BJTs): it is controlled by an electric field at an insulated gate rather than by a base current.

For the general principle behind all transistors, see how does a transistor work.

The parameters that govern conduction and switching behavior are:

  • RDS(on): influences conduction losses and thermal rise
  • Gate charge (Qg): determines the drive energy needed per switching cycle
  • Safe operating area (SOA): defines the permitted combinations of voltage, current, and time during transients
  • Body diode: affects reverse-current behavior and synchronous-rectifier design

Datasheet parameters guide MOSFET selection:

  • VDS defines the maximum voltage capability, with margin for transients
  • ID reflects continuous current under specific thermal conditions
  • RDS(on) determines conduction loss
  • Gate charge (Qg) and capacitances influence switching losses and gate-driver sizing

Designers must also review SOA curves, thermal resistance, package ratings, and diode characteristics.

For how to read and apply breakdown-voltage specs in detail, see MOSFET voltage ratings - how to read and apply breakdown voltage specs.

MOSFET selection depends on application requirements, operating voltage, control strategy, and desired switching performance. A few rules of thumb:

  • Motor control: prioritise low RDS(on) for conduction losses, sufficient VDS margin for inductive spikes, and a manageable Qg for efficient switching at the motor's PWM frequency.
  • Switching power supplies: focus on adequate VDS breakdown voltage, minimal RDS(on) for efficiency, and low Qg and reverse-recovery charge (Qrr) for reduced switching losses.

The choice comes down to where the switch sits and how it is driven:

  • N-channel MOSFETs use electrons (higher mobility, lower RDS(on)) and, with the source tied to ground, are ideal for low-side switching. They dominate high-efficiency designs such as synchronous rectifiers and DC-DC converters.
  • P-channel MOSFETs use holes (higher RDS(on) for the same die area) but simplify high-side switching, where the source is referenced to the positive supply.

A MOSFET is driven entirely by the gate-to-source voltage (VGS): no steady current flows into the gate, but the gate capacitance must be charged and discharged every cycle — which is what gate charge (Qg) quantifies.

An enhancement-mode device stays OFF until VGS exceeds VGS(th), so a gate-driver IC or a resistive pull-down is used to keep the OFF state defined. How the gate is driven then depends on high-side vs. low-side placement — see Driving high-side and low-side MOSFETs above.

Enhancement-mode N-channel MOSFETs are the most common, thanks to their low conduction losses, simple drive requirements, and wide availability.

"Normally-on" refers to depletion-mode MOSFETs, which conduct current at zero gate voltage and require an external gate bias to turn them off.

A power MOSFET uses a vertical structure to handle high current and high voltage, with low RDS(on) and fast switching for power conversion. A small-signal MOSFET uses a lateral structure and is optimised for low-power, low-current analog or logic-level tasks.

CMOS (complementary metal-oxide-semiconductor) circuits pair an N-channel MOSFET (NMOS) with a P-channel MOSFET so that one is always off, drawing almost no static power. That efficiency is why CMOS logic is the basis of nearly all modern digital integrated circuits — processors, memory, and logic chips.

Unlike bipolar junction transistors (BJTs), these MOSFETs need almost no drive current, which keeps CMOS circuits cool and dense and lets them work across both digital and analog circuits.