The MOSFET

Nearly every transistor inside a modern processor — all tens of billions of them — is a MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor). Unlike the bipolar transistor, which needs a continuous trickle of base current to stay switched on, a MOSFET is controlled almost entirely by voltage, drawing essentially zero steady-state current at its control terminal. That one difference is why the MOSFET, not the BJT, became the building block of digital computing.

The four terminals

A MOSFET has four terminals, though the fourth is often connected internally and left off simple diagrams:

Source (S) — where charge carriers enter the device.
Drain (D) — where charge carriers exit.
Gate (G) — sits above the region between source and drain, separated from the silicon by a thin insulating layer, and controls whether current can flow.
Body / Substrate (B) — the bulk silicon the device is built into.

In an N-channel MOSFET (the type used in the examples below), the source and drain are heavily doped N-type regions embedded in a P-type body.

Cross-section of an N-channel MOSFET

The gate sits on top of a very thin insulating layer of silicon dioxide, so no current ever flows directly into the gate under normal DC operation — the gate simply creates an electric field that reaches down into the silicon beneath it.

How the field-effect switch works

With zero (or low) voltage on the gate, the region between source and drain is just P-type body material — the same polarity as the source and drain are not, meaning no continuous N-type path connects them. Current can’t flow between source and drain because it would have to cross two back-to-back PN junctions in series, and one of them is always reverse biased.

Raise the gate voltage relative to the source ( $V_{GS}$ ) high enough, and the electric field from the gate pushes away holes and pulls free electrons up to the silicon surface directly beneath the gate oxide, forming a thin, continuous layer of N-type-like material called the channel or inversion layer. That channel electrically connects source to drain, letting current flow. The gate voltage at which this channel first fully forms is called the threshold voltage, $V_{th}$ .

🧩 Think of it like… a drawbridge controlled by a magnet, with no physical lever touching the bridge itself. The gate is the magnet sitting above the bridge — energize it enough (raise the gate voltage past $V_{th}$) and the bridge deck (the channel) snaps into place, connecting the two riverbanks (source and drain) so traffic can cross. No current flows through the magnet itself to hold the bridge up — just the *presence* of its field does the job. That's exactly why a MOSFET, unlike a BJT, draws essentially no steady current at its control terminal.

The saturation current equation

Once the channel is formed and the drain voltage is high enough to pull the channel into “saturation” (the regime used for most digital switching), the drain current is approximately:

I_D = \frac{1}{2}\, \mu_n C_{ox} \frac{W}{L} (V_{GS} - V_{th})^2

Where:

$I_D$ is the drain current — the current flowing from drain to source through the channel.
$\mu_n$ is the electron mobility in the channel (how easily electrons move through it).
$C_{ox}$ is the gate oxide capacitance per unit area — a thinner oxide gives a larger $C_{ox}$ and stronger field coupling.
$W$ and $L$ are the channel’s width and length — the physical dimensions of the transistor, which chip designers choose deliberately.
$V_{GS}$ is the gate-to-source voltage, and $V_{th}$ is the threshold voltage.

The key takeaway from this formula, even without doing the math: current turns on only once $V_{GS}$ exceeds $V_{th}$ , and grows with the square of how far past threshold you push it. Below threshold, the device is (ideally) off.

Real MOSFETs aren’t perfectly off below

V_{th}

— a small “subthreshold leakage” current still flows, and controlling this leakage as transistors shrink is one of the biggest challenges in modern chip design, discussed further in FinFET and Gate-All-Around.

From physical device to digital switch

The property that matters most for digital chips is binary: below $V_{th}$ , the MOSFET is effectively an open circuit (off, logic 0); above $V_{th}$ , it conducts (on, logic 1). That clean on/off behavior, controlled purely by voltage, is what the next page builds on directly: The Transistor as a Switch.

Key takeaways

A MOSFET has four terminals: source, drain, gate, and body, with the gate insulated from the silicon by a thin oxide layer.
Applying gate voltage above the threshold voltage $V_{th}$ forms a conductive channel connecting source and drain.
Because the gate is insulated, a MOSFET draws essentially no steady-state current at its control terminal — unlike a BJT.
Saturation drain current follows $I_D = \frac{1}{2}\mu_n C_{ox}\frac{W}{L}(V_{GS}-V_{th})^2$ , growing with the square of the gate overdrive.
This clean voltage-controlled on/off behavior is the direct basis for digital switching, covered in The Transistor as a Switch.

The Bipolar Transistor The Transistor as a Switch