💎 Semiconductor Device Physics Basics: Diodes, BJTs, and MOSFETs Explained

Doped Silicon: The Starting Material

Pure (intrinsic) silicon is a poor conductor at room temperature because very few of its electrons have enough thermal energy to break free and carry current. Semiconductor devices become useful once silicon is deliberately doped with small, controlled amounts of impurity atoms. Adding an element with one more valence electron than silicon, such as phosphorus or arsenic, contributes extra free electrons and creates n-type silicon, where electrons are the majority charge carrier. Adding an element with one fewer valence electron, such as boron, creates an absence of an electron at each dopant site — a hole — that behaves like a mobile positive charge carrier, creating p-type silicon, where holes are the majority carrier. Every device covered in this guide — the diode, the BJT, and the MOSFET — is built from carefully arranged regions of p-type and n-type silicon placed next to each other.

The PN Junction and the Depletion Region

Where a p-type region meets an n-type region, a PN junction is formed, and this junction is the fundamental building block underlying every semiconductor device. Right at the interface, the sharp difference in carrier concentration drives electrons to diffuse from the n-side into the p-side and holes to diffuse from the p-side into the n-side, exactly the way any substance diffuses from a region of high concentration toward one of low concentration. As these mobile carriers cross over and recombine with carriers of the opposite type, they leave behind the fixed dopant ions that had originally supplied them — positive ions exposed on the n-side, negative ions exposed on the p-side — which are no longer balanced by a nearby mobile carrier of opposite charge. The result is a thin region straddling the junction that is depleted of mobile charge carriers, called the depletion region, containing only these fixed, exposed ions. Those exposed charges set up a built-in electric field pointing from the n-side toward the p-side, and an associated built-in potential (roughly 0.6-0.7V for a silicon junction at room temperature) that opposes further diffusion. This built-in field and potential are what every diode, BJT, and MOSFET ultimately manipulates to control current flow.

The Diode: Forward Bias, Reverse Bias, and Breakdown

A diode is simply a PN junction with a terminal on each side — the p-side is the anode, the n-side is the cathode. Applying an external voltage across it changes the width and effect of the depletion region.

Under forward bias (positive voltage applied to the anode relative to the cathode), the applied field opposes and narrows the built-in field, shrinking the depletion region. Once the applied voltage approaches the built-in potential, the junction's opposition to carrier diffusion collapses and current flows readily. Under reverse bias (positive voltage applied to the cathode relative to the anode), the applied field reinforces the built-in field, widening the depletion region and strongly suppressing conduction — only a very small reverse leakage current flows, carried by thermally generated minority carriers.

This behavior is conceptually captured by the diode equation: I = Is × (e^(V / (n·VT)) - 1), where Is is a small reverse saturation current characteristic of the specific device, VT is the thermal voltage (about 25.85 mV at room temperature), n is an ideality factor typically between 1 and 2, and V is the voltage across the junction. The exponential term dominates under forward bias, which is why diode current grows so steeply with only a small increase in forward voltage; under reverse bias, the exponential term becomes negligible and the equation collapses to the small, roughly constant leakage current -Is.

If reverse bias is increased far enough, the diode enters reverse breakdown, where current increases sharply at a roughly fixed reverse voltage. Two distinct physical mechanisms can cause this: avalanche breakdown, where carriers accelerated by the strong reverse field gain enough energy to knock loose additional carrier pairs in a multiplying chain reaction (dominant in most rectifier diodes at breakdown voltages above about 6-8V), and Zener breakdown, where the field itself becomes strong enough to directly pull electrons out of their covalent bonds in a very narrow, heavily doped junction (dominant at lower voltages). Zener diodes are deliberately designed to operate reliably in this reverse breakdown region, using the resulting stable voltage as a voltage reference.

Worked Example: Diode Current from the Diode Equation

Take a diode with Is = 1×10⁻¹² A, an ideality factor n = 1, forward voltage V = 0.7V, and room-temperature VT = 0.02585V. The exponent is V/(n·VT) = 0.7/0.02585 ≈ 27.08, and e^27.08 ≈ 5.8×10¹¹. Current is then approximately I ≈ Is × e^27.08 = 1×10⁻¹² × 5.8×10¹¹ ≈ 0.58A. This confirms the earlier FAQ point numerically: a forward voltage of roughly 0.7V on an ordinary small-signal silicon diode corresponds to a current in the hundreds-of-milliamps range, and it also shows just how sensitive the exponential relationship is — a change of only about 60mV in forward voltage would change this current by roughly a factor of ten.

The Bipolar Junction Transistor (BJT)

An NPN BJT is built from three alternating regions: a heavily doped n-type emitter, a thin, lightly doped p-type base, and a moderately doped n-type collector. In normal (active) operation, the base-emitter junction is forward biased and the base-collector junction is reverse biased. The forward-biased emitter-base junction injects a large flow of electrons from the heavily doped emitter into the thin base region. Because the base is deliberately made very thin and lightly doped, the great majority of those injected electrons diffuse straight across it before they can recombine, and are then swept into the collector by the reverse-biased base-collector junction's field. Only a small fraction recombine in the base, and it is that small recombination current that must be continuously supplied as external base current (Ib) to sustain the whole process. The ratio of the resulting collector current to that small base current is the transistor's current gain, beta (β or hFE), typically in the range of 50 to a few hundred for small-signal devices: Ic = β × Ib.

A BJT operates in one of several distinct regions depending on the bias applied to its two junctions. In the active region (base-emitter forward biased, base-collector reverse biased), the transistor behaves as a current amplifier, with Ic proportional to Ib via beta — this is the region used for linear amplification. In saturation (both junctions forward biased), the collector-emitter voltage drops to a small, roughly fixed value and the transistor behaves like a closed switch, no longer obeying the Ic = β × Ib relationship because the collector current is now limited by the external circuit rather than by base drive. In cutoff (both junctions reverse biased, or the base-emitter junction below its forward-conduction threshold), essentially no current flows and the transistor behaves like an open switch. Digital switching applications deliberately drive a BJT between cutoff and saturation to realize a clean off/on switch; analog amplifier applications keep it firmly in the active region.

The MOSFET: Voltage-Controlled Conduction

An enhancement-mode NMOS transistor is built with two heavily doped n-type regions, the source and drain, separated by a p-type body region, with a conductive gate sitting above the gap between them, insulated from the silicon by a thin layer of oxide. With zero gate voltage, no continuous n-type path connects source to drain, so the device is off regardless of drain voltage. Applying a sufficiently positive voltage to the gate relative to the source attracts electrons up toward the oxide interface, and once the gate-source voltage exceeds the device's threshold voltage (Vth), enough electrons accumulate to form a thin conductive channel — an "inverted" n-type layer within what was p-type body material — directly connecting source to drain. This is why it is called enhancement mode: the conductive channel does not exist at all until the gate voltage enhances it into being. Crucially, the gate is electrically insulated from the channel by the oxide layer, so essentially no steady-state current flows into the gate to sustain this channel — a fundamental difference from the BJT's continuously flowing base current.

Once the channel exists, the MOSFET's behavior depends on the drain-source voltage, VDS, relative to the "overdrive" voltage, VGS - Vth. In the triode (or linear) region, when VDS is small (VDS less than VGS - Vth), the channel behaves roughly like a voltage-controlled resistor, and drain current follows approximately ID ≈ k × [(VGS - Vth)·VDS - VDS²/2], where k is a constant set by the device's geometry and process parameters — this is the region used when a MOSFET acts as a low-resistance switch. In the saturation region, once VDS grows large enough that VDS ≥ VGS - Vth, the channel pinches off near the drain end and drain current becomes largely independent of VDS, instead following approximately ID ≈ (k/2) × (VGS - Vth)² — this is the region used when a MOSFET acts as a voltage-controlled current source in analog amplifier applications. Note carefully that MOSFET "saturation" describes a current-source-like region, essentially the opposite of a BJT's saturation, which describes a low-resistance, fully-on switch state — the MOSFET region that behaves like a BJT's saturation is its triode region, a naming collision that trips up many students moving between the two device families.

Why MOSFETs Dominate Modern Digital ICs

Digital logic chips are built almost entirely from MOSFETs, specifically in the complementary pairing known as CMOS, and the reason ties directly back to the gate's insulated, near-zero-current control. In a CMOS logic gate, one transistor of each complementary pair is always fully off in any static (unchanging) logic state, and since a fully off MOSFET, and an idle MOSFET gate, both draw essentially no steady current, a CMOS chip's static power consumption stays extremely low no matter how many billions of transistors sit on the die — power is consumed mainly during switching transitions, not while sitting idle in a fixed state. A chip built the same way from BJTs would require continuous base current flowing through a substantial fraction of its transistors at all times just to hold a static logic state, which becomes thermally and electrically impossible to sustain once transistor counts reach into the millions, let alone the billions found in modern processors (the transistor-scaling trends behind those counts are covered in the companion semiconductor manufacturing article on this site). MOSFETs are also considerably simpler to shrink in size across process generations than BJTs, whose performance depends on precise, thin base-region doping profiles that do not scale down as cleanly. This combination of near-zero static gate current and superior scalability is why the MOSFET, not the BJT, became the transistor of the digital age.

BJT vs MOSFET: Side-by-Side Comparison

Both devices remain essential rather than one simply replacing the other: BJTs are still favored in specific analog and RF applications for their high transconductance and low noise characteristics, and in some high-current power switching roles, while MOSFETs dominate digital logic and the majority of power conversion switch applications (as covered in the companion power electronics article on this site). Understanding both devices at the physics level — what the depletion region, the base current, and the gate-controlled channel are each actually doing — is what makes datasheet parameters like beta, threshold voltage, and switching times meaningful rather than arbitrary numbers to memorize.