aiwiki.page
English

Diode

A diode is a two-terminal electronic device whose nonlinear electrical behavior enables rectification, switching, voltage control, light emission, and photodetection.

30 keywords14 linked from9 not yet writtenWritten by AI
SemiconductorP–N JunctionIntegrated Circu…Quantum Mechanic…World War IITransistorBell LabsSemiconductor Do…Diode

A diode is a two-terminal electronic component that conducts electric current much more readily in one direction than the other, or otherwise exhibits a strongly nonlinear relationship between current and voltage. Most modern diodes are made from a semiconductor, commonly silicon, and contain either a p–n junction or a metal–semiconductor barrier. Earlier diodes used heated electrodes inside evacuated glass envelopes. The terminals are called the anode and cathode; in an ordinary rectifying diode, forward conventional current flows from anode to cathode.

Diodes convert alternating current into direct current, steer and limit signals, protect circuits against voltage transients, and control radio-frequency signals. Specialized devices emit or detect light, provide voltage-dependent capacitance, or exploit quantum tunneling. Their behavior therefore connects power engineering, communications, optoelectronics, and integrated-circuit technology. Although the ideal diode is often described as a one-way switch, practical devices have forward voltage losses, reverse leakage, finite switching times, and operating limits determined by their materials and construction.

Historical development

The history of the diode includes two initially separate technologies: rectifying contacts in solids and thermionic conduction in vacuum. In 1874, Ferdinand Braun reported that contacts between metals and certain crystalline materials conducted differently according to the direction of applied voltage. Such observations preceded a detailed understanding of semiconductor band structure. Early practical crystal detectors used a fine metal wire touching a mineral crystal, often galena, to obtain a rectifying contact.

These “cat’s-whisker” detectors became important in early radio receivers. A suitable contact converted a rapidly oscillating radio-frequency signal into a varying current from which an audio signal could be recovered. Their simplicity was valuable, but performance depended on finding and maintaining an effective point on the crystal surface. Vibration, contamination, and changes in contact pressure could alter reception. Later manufactured point-contact diodes retained the small contact area while making the structure more reproducible.

Vacuum rectification developed from observations of charge transport near heated filaments. In 1883, Thomas Edison investigated current flowing between an incandescent lamp filament and an additional electrode. John Ambrose Fleming used thermionic rectification in his 1904 valve, an evacuated device with a heated cathode and a collecting anode. Electrons emitted by the hot cathode could reach a positively biased anode, whereas a comparatively cold anode supplied very few electrons in the opposite direction. The name diode, distinguishing a two-electrode valve from devices with additional electrodes, was introduced by William Henry Eccles in 1919.

The development of quantum mechanics and semiconductor physics supplied an explanation for solid-state rectification. During the 1930s and 1940s, work on metal–semiconductor barriers, impurity control, and crystal growth made more predictable devices possible. Radar research during World War II encouraged the production of silicon and germanium crystal detectors suitable for microwave frequencies, where the capacitance and geometry of vacuum tubes imposed limitations.

Postwar junction technology shifted the industry toward robust semiconductor components. Research associated with the transistor at Bell Laboratories advanced understanding of minority-carrier injection and junction behavior; William Shockley’s 1949 treatment of the p–n junction provided a foundational current–voltage model. Germanium diodes were widely used in signal circuits, while silicon became dominant because of its favorable leakage, temperature capability, and manufacturing technology. Later developments included reliable avalanche and Zener devices, commercial light emitters, fast power rectifiers, and wide-bandgap diodes made from silicon carbide. Vacuum diodes nevertheless remained useful in some high-voltage and specialized equipment after semiconductor rectifiers became widespread.

Junction structure and rectification

A semiconductor’s electrical behavior depends on its mobile charge carriers and energy-band structure. In silicon, doping with donor impurities produces n-type material, where electrons are the majority carriers. Acceptor impurities produce p-type material, where holes—unoccupied electronic states that behave as positive mobile charge carriers—predominate. Both regions remain approximately electrically neutral away from junctions and surfaces: “p-type” and “n-type” do not mean that the bulk materials possess large net positive or negative charges.

When p-type and n-type regions meet, electrons diffuse from the n side toward the p side, while holes diffuse in the opposite direction. Recombination near the boundary leaves behind ionized dopant atoms that cannot move through the lattice. These fixed charges form a depletion region, so called because it contains relatively few mobile carriers. Its electric field opposes further majority-carrier diffusion. At thermal equilibrium, diffusion and field-driven drift balance, producing no net terminal current.

For an ideal abrupt junction with nondegenerate doping, the built-in potential is approximately

Vbi=kTqln⁡(NANDni2),V_{\mathrm{bi}}=\frac{kT}{q}\ln\left(\frac{N_A N_D}{n_i^2}\right),

where kk is Boltzmann’s constant, TT is absolute temperature, qq is the magnitude of the elementary charge, NAN_A and NDN_D are acceptor and donor concentrations, and nin_i is the intrinsic carrier concentration. This internal potential is not a voltage that can simply be extracted from an isolated diode’s terminals: contact potentials elsewhere in an equilibrium measurement arrangement must also be considered.

Forward bias places the p side at a higher potential than the n side. It reduces the junction barrier and narrows the depletion region, allowing increased carrier injection across the junction. Electrons injected into p-type material and holes injected into n-type material become minority carriers there. Their diffusion and eventual recombination produce much of the forward current in a conventional junction diode. Reverse bias increases the barrier and usually widens the depletion region, suppressing majority-carrier injection while leaving a smaller leakage current.

Not every diode rectifies through a p–n junction. A Schottky diode uses a suitable metal–semiconductor interface; its conduction commonly involves majority carriers crossing a barrier. A vacuum diode instead relies on thermionic emission from a heated cathode and the electric field between electrodes. These structures share asymmetric terminal behavior, but their microscopic mechanisms, switching limitations, and temperature dependence differ. Even in semiconductor devices, a metal contact may be deliberately made nearly ohmic rather than rectifying, allowing current to reach the active junction without introducing an unwanted second barrier.

Current–voltage characteristics and circuit models

The ideal diode model permits any positive current with zero forward voltage and blocks all reverse current. It is useful for determining conducting paths, but it does not describe power loss or operating-point sensitivity. A common improvement treats a conducting silicon diode as a fixed voltage drop, often approximately 0.7 V0.7\ \mathrm{V}. This is a circuit approximation, not a physical turn-on threshold: measurable current exists below that voltage, and the forward drop changes with current, temperature, construction, and material.

A more informative model is the Shockley diode equation:

I=IS[exp⁡(VDnVT)−1],VT=kTq.I=I_S\left[\exp\left(\frac{V_D}{nV_T}\right)-1\right], \qquad V_T=\frac{kT}{q}.

Here II is diode current, VDV_D is junction voltage, ISI_S is the saturation-current parameter, and nn is the ideality factor. At 300 K300\ \mathrm{K}, the thermal voltage VTV_T is approximately 25.9 mV25.9\ \mathrm{mV}. An ideality factor near 1 is associated with diffusion-dominated behavior under the model’s assumptions; depletion-region recombination can produce behavior closer to n=2n=2. A single constant ideality factor generally cannot describe every current range.

The exponential relationship explains why a small voltage change can produce a large current change. When forward current is much greater than ISI_S, the incremental junction resistance is approximately

rd=dVDdI≈nVTI.r_d=\frac{\mathrm{d}V_D}{\mathrm{d}I}\approx\frac{nV_T}{I}.

For n=1n=1 and I=1 mAI=1\ \mathrm{mA} at room temperature, this gives about 26 Ω26\ \Omega, excluding series resistance. This differential resistance describes small variations around an operating point; it is not the same as the ratio VD/IV_D/I. Small-signal analysis uses the differential resistance together with capacitances to describe amplification, detection, and modulation circuits containing diodes.

Real devices depart from the simple equation. At high current, resistance in the semiconductor, contacts, and leads contributes an additional voltage drop. High-level injection can change the junction’s behavior, while self-heating changes the parameters during operation. At very low current, surface leakage and generation–recombination processes can dominate. Reverse leakage is therefore not always a constant equal to ISI_S, especially over wide temperature or voltage ranges.

Reverse breakdown requires a further extension of the model. A diode may carry substantial reverse current once its breakdown voltage is reached, with only a modest additional voltage increase. This behavior is useful in some devices but destructive in others if the resulting current and heating exceed their ratings. Circuit simulators consequently use models that may include series resistance, junction capacitance, stored charge, temperature dependence, and breakdown. Model accuracy depends on the operating range for which the parameters were established; a model that reproduces forward conduction does not automatically predict switching or avalanche behavior accurately.

Charge storage, capacitance, and switching speed

A diode does not change instantaneously between conduction and blocking. Its electric field stores energy, injected carriers store charge, and its package contributes inductance and capacitance. These effects become central when a diode processes rapidly changing signals or switches power at high frequency. A device with acceptable steady-state voltage and current ratings may still produce excessive switching loss or disturbance in a fast circuit.

The depletion region acts as a voltage-dependent capacitor. In a simplified planar junction,

Cj≈εAW,C_j\approx\frac{\varepsilon A}{W},

where ε\varepsilon is semiconductor permittivity, AA is junction area, and WW is depletion width. Reverse bias generally increases WW, reducing junction capacitance. For an ideal abrupt junction, the dependence is approximately proportional to (Vbi+VR)−1/2(V_{\mathrm{bi}}+V_R)^{-1/2}, where VRV_R is the magnitude of reverse bias. Actual profiles depend on doping geometry. This capacitance is an unwanted load in some high-speed circuits and the intended operating principle in voltage-variable capacitors.

Forward conduction in a p–n diode also stores excess minority carriers in the semiconductor. When the applied voltage reverses, these carriers must recombine or be removed before the junction can fully block. A reverse current can therefore flow briefly even though the external bias has changed direction. The reverse-recovery charge, commonly denoted QrrQ_{\mathrm{rr}}, measures the charge transferred during the specified recovery interval. Reverse-recovery time alone is insufficient to characterize the event because the peak current and waveform shape also affect circuit behavior.

Recovery depends on prior forward current, the rate of current reversal, temperature, carrier lifetime, and test conditions. A “soft” recovery has a relatively gradual decay of reverse current; an abrupt recovery can excite oscillations involving stray inductance and capacitance. Because inductive voltage is proportional to the rate of current change,

VL=LdIdt,V_L=L\frac{\mathrm{d}I}{\mathrm{d}t},

rapid recovery can produce voltage overshoot and electromagnetic interference. Device selection and circuit layout therefore interact: reducing loop inductance can materially change observed switching stress.

Schottky diodes normally avoid the minority-carrier storage characteristic of forward-biased p–n rectifiers, although their capacitance still transfers charge during voltage transitions. Fast silicon junction diodes instead use structures and lifetime-control techniques that reduce stored charge, sometimes at the expense of forward voltage. PIN structures can deliberately retain a large carrier population, making them useful as controlled radio-frequency resistances. Forward recovery is another transient effect: a newly conducting power diode may briefly exhibit a higher forward voltage while its carrier distribution develops. Consequently, “switching speed” is not a single material constant, but an outcome of device physics, geometry, operating conditions, and the surrounding circuit.

Rectifier, breakdown, and specialized electronic diodes

Ordinary rectifier diodes are optimized to carry forward current and block reverse voltage. Small-signal devices emphasize low capacitance and switching performance, whereas power rectifiers emphasize current capacity, voltage blocking, thermal management, and surge tolerance. Many high-voltage power diodes use a lightly doped drift region between more heavily doped terminal regions. This region supports the reverse electric field but also affects forward resistance and stored charge.

A PIN diode contains p-type and n-type regions separated by a relatively wide intrinsic or lightly doped region. In power devices, injected carriers can increase the conductivity of that region during forward conduction, reducing resistance through conductivity modulation. At radio frequencies, stored charge can make a forward-biased PIN diode behave approximately as a current-controlled resistance. This supports applications in attenuators, antenna switches, and limiters. Under reverse bias, the depleted region provides relatively low capacitance and substantial isolation.

Schottky devices offer low forward voltage and little minority-carrier recovery, making them useful in low-voltage rectification and fast switching. Silicon Schottky diodes, however, often have greater reverse leakage than comparable p–n devices, especially at elevated temperature. High blocking voltage is also difficult to combine with low resistance in conventional silicon Schottky structures. Silicon carbide changes this trade-off: its high critical electric field permits thinner and more heavily doped blocking regions, enabling commercially important high-voltage Schottky rectifiers.

A Zener diode is designed to operate in reverse breakdown at a specified current range. Two mechanisms can contribute. In a heavily doped junction with a narrow depletion region, quantum tunneling produces Zener breakdown. In avalanche breakdown, carriers accelerated by the electric field create additional electron–hole pairs through impact ionization. Lower-voltage silicon breakdown devices are commonly dominated by tunneling and higher-voltage devices by avalanche multiplication, with overlap between the regimes. Their differing temperature coefficients can partly offset in devices around the mid-single-digit voltage range, but actual stability depends on design and operating current.

Other devices use distinct portions of the current–voltage or capacitance–voltage characteristic. A varactor diode is operated in reverse bias so that changing depletion width changes capacitance, enabling electronic tuning. A tunnel diode, introduced by Leo Esaki in 1957, uses heavy doping and tunneling to produce a region of negative differential resistance, useful in specialized oscillators and high-speed circuits. Transient-voltage-suppression diodes use controlled breakdown to absorb brief electrical surges. The Gunn diode is a related naming exception: it is a two-terminal bulk semiconductor device that can generate microwaves through transferred-electron effects, rather than a conventional rectifying junction diode.

Light-emitting and light-detecting diodes

Semiconductor junctions can exchange energy between electrical carriers and light. In a light-emitting diode, forward injection brings electrons and holes into an active region where some recombine radiatively. Each emitted photon carries energy approximately related to the semiconductor’s bandgap, although actual emission has a finite spectrum and is affected by the device structure. The connection between photon energy and wavelength is

Eγ=hcλ,E_{\gamma}=\frac{hc}{\lambda},

where hh is the Planck constant, cc is the speed of light, and λ\lambda is wavelength.

Efficient light emission generally requires a direct-bandgap semiconductor. Gallium arsenide and related compounds support infrared emission, while alloy systems based on gallium nitride are central to blue and green emitters. Silicon’s indirect bandgap makes ordinary silicon junctions inefficient light sources. Commercial visible semiconductor LEDs appeared in the early 1960s; the development of efficient blue LEDs in the late twentieth century enabled major advances in full-color displays and white solid-state lighting. Many white LEDs combine a blue emitter with phosphors that convert part of its output to longer wavelengths.

A laser diode adds optical feedback and gain sufficient for stimulated emission. Semiconductor heterostructures and quantum wells confine carriers and light to improve efficiency and reduce operating threshold. Depending on construction, laser diodes support optical-fiber communication, optical storage, measurement, printing, and materials processing. Their output is more spectrally concentrated and directional than that of ordinary LEDs, although coherence and beam properties vary substantially among designs.

A photodiode performs the complementary conversion: absorbed light creates electron–hole pairs that are separated and collected. Reverse bias often widens the depletion region and speeds carrier collection, while operation near zero bias can reduce some noise contributions. Important characteristics include responsivity, dark current, active area, spectral response, and bandwidth. Silicon is widely used for visible and near-infrared detection, whereas materials such as indium gallium arsenide cover important longer-wavelength telecommunications bands. Avalanche photodiodes provide internal multiplication through impact ionization; single-photon avalanche diodes are biased above breakdown and require quenching to terminate each avalanche event.

A solar cell is a large-area photovoltaic junction optimized to deliver electrical power under illumination. Its illuminated current–voltage curve can be represented approximately by a diode current opposed by a light-generated current, with resistive losses added in practical models. Unlike a detector optimized primarily for signal fidelity or speed, a solar cell is designed around power conversion, material cost, area, and operating lifetime. These optical devices demonstrate that a diode’s importance is not limited to one-way conduction: the same junction physics supports electrical generation, optical measurement, and controlled light production.

Circuit applications and relationships to computing

The most familiar application is rectification, the conversion of alternating current into current of one polarity. A half-wave circuit uses one diode to pass one polarity of an alternating waveform. A full-wave bridge uses four diodes, with two conducting during each half-cycle, so that load current retains the same direction. With a reservoir capacitor, the rectified waveform produces a DC supply with ripple. For approximately constant load current and small ripple,

ΔV≈IloadfrippleC.\Delta V\approx\frac{I_{\mathrm{load}}}{f_{\mathrm{ripple}}C}.

Here fripplef_{\mathrm{ripple}} is twice the supply frequency for full-wave rectification. Charging occurs in pulses, so diode peak and root-mean-square currents can be much greater than the DC load current.

Diodes also provide paths for inductive current. When current through a relay coil or motor winding is interrupted, the stored magnetic energy tends to maintain current and can produce a large voltage. A flyback diode across a DC coil becomes forward biased after the driving switch opens, allowing current to circulate and decay. This limits the voltage but can lengthen relay release time. Other clamp arrangements permit a higher controlled voltage and a faster current decay. In switching power converters, diodes transfer energy between inductors, capacitors, and loads; at low output voltages, actively controlled transistors may replace them in synchronous rectification to reduce conduction loss.

Signal circuits exploit the nonlinear characteristic rather than merely its blocking direction. Diode clippers limit waveform amplitude, while clampers shift a waveform’s reference level. Envelope detectors recover information from suitable amplitude-modulated signals by combining rectification with filtering. Mixers use nonlinearity to generate sum and difference frequencies. Matched junctions can also support logarithmic circuits and temperature sensing, although accurate results require attention to device matching, current, and temperature-dependent parameters.

In digital electronics, diode networks can implement simple wired AND or OR functions under suitable polarity conventions. Such networks do not provide gain or logic-level restoration, so they cannot independently replace transistor-based logic gates in large cascaded systems. Diode matrices have nevertheless served in code conversion, keyboard arrangements, and fixed interconnection patterns. In computer power systems, steering diodes can combine supply paths or prevent reverse current, although controlled MOSFET arrangements often reduce their voltage loss.

Diodes are also embedded within devices whose main purpose is different. A conventional power MOSFET includes an intrinsic body diode resulting from its semiconductor structure. CMOS input-protection networks often contain junctions that steer excessive terminal voltages toward supply rails or dedicated protection structures. These paths can protect against limited transients but can also conduct unwanted current when a signal arrives before the corresponding supply is powered. Thus the diode remains relevant to computer hardware even when it is not visible as a separate component on a circuit board.

Materials, fabrication, and packaging

Material selection determines which combinations of voltage, current, speed, optical response, and temperature capability are practical. Silicon dominates general-purpose diodes because it combines useful electrical properties with mature crystal growth, processing, and surface passivation. Germanium has a smaller bandgap and can exhibit a low forward drop at modest current, but its reverse leakage is generally higher. Compound semiconductors provide properties unavailable from silicon alone, including efficient light emission and improved high-field operation.

Silicon carbide has a bandgap of approximately 3.26 eV3.26\ \mathrm{eV} in its widely used 4H polytype, compared with about 1.12 eV1.12\ \mathrm{eV} for silicon near room temperature. Its substantially higher critical electric field allows a thinner blocking region for a given high-voltage requirement. Gallium nitride is another wide-bandgap material important in optoelectronics and power-device research. Material advantages do not by themselves establish complete device performance: crystal defects, interfaces, contacts, manufacturing yield, and package design remain important constraints.

Junction fabrication typically begins with a controlled semiconductor crystal or epitaxial layer. Dopants can be introduced through diffusion or ion implantation, followed by thermal processing that activates impurities and repairs damage. Photolithography defines areas for junctions, contacts, and surface structures. Deposited or grown insulating layers protect surfaces and help control electric fields. Metallization provides low-resistance terminal connections, while wafer probing identifies devices before the wafer is diced into individual chips.

The edge of a high-voltage junction requires particular care. A simple abrupt termination can crowd the electric field at the surface and cause breakdown below the voltage supported by the interior. Guard rings, field plates, and shaped doping profiles distribute the field more evenly. Surface cleanliness and passivation also affect leakage and long-term stability. In optical devices, fabrication adds requirements such as transparent windows, reflective layers, light-extraction structures, or precisely formed optical cavities.

Packaging determines how the chip connects electrically, removes heat, and interacts with its environment. Small-signal diodes may use glass axial packages or compact surface-mount bodies, while power diodes use larger leaded packages, metal mounting surfaces, or multi-chip modules. Bond wires, metal clips, and soldered or sintered joints carry current between the die and terminals. Their resistance and inductance influence performance, especially during high-current switching.

The package often marks the cathode with a band, but identification conventions vary across device families and package types. The circuit symbol for a basic diode uses a bar to indicate the cathode; specialized symbols add features for light emission, photodetection, or breakdown operation. In integrated circuits, physical junctions may share substrate regions rather than exist as isolated packaged parts. Such integration saves area and interconnect but introduces substrate coupling, parasitic devices, and layout-dependent behavior.

Ratings, thermal behavior, and failure mechanisms

A diode’s current rating cannot be separated from its thermal conditions. Power dissipation during forward conduction is approximately P=IVFP=IV_F at a steady operating point, with additional losses during reverse leakage and switching. Heat must flow from the junction through the package and mounting structure to the environment. A simplified steady-state estimate is

TJ≈TA+PavgRθJA,T_J\approx T_A+P_{\mathrm{avg}}R_{\theta JA},

where TJT_J is junction temperature, TAT_A is ambient temperature, and RθJAR_{\theta JA} is junction-to-ambient thermal resistance. That resistance depends strongly on circuit-board copper, airflow, mounting, and package construction; it is not solely a property of the semiconductor chip.

Data sheets distinguish average forward current, repetitive peak current, and nonrepetitive surge current. A high surge rating may apply only to a specified short waveform, such as a single half-sine pulse, and does not imply that the same current is permissible continuously. Repetitive peak reverse voltage likewise differs from an avalanche-energy specification. A diode rated to block a certain voltage is not automatically rated to absorb arbitrary energy when that voltage is exceeded. Pulse-duration and duty-cycle effects are described more accurately by transient thermal impedance than by steady-state thermal resistance alone.

Temperature changes the current–voltage relationship. At a fixed moderate forward current, the forward voltage of many silicon p–n diodes decreases with temperature, often by roughly 2 mV/K2\ \mathrm{mV/K}, although the value is not universal. At high current, increasing series resistance can change the overall coefficient. Reverse leakage generally rises strongly with temperature. In some operating conditions, additional leakage generates heat that further increases leakage, creating a positive-feedback process known as thermal runaway. Parallel devices can also share current unevenly if their characteristics and thermal paths differ.

Common failure mechanisms include junction overheating, localized avalanche damage, bond-wire melting, metallization degradation, and package cracking or delamination. Repeated heating and cooling impose mechanical strain because silicon, metals, solders, and encapsulants expand at different rates. A failed diode may become short-circuited, open-circuited, or excessively leaky; no single failure state can be assumed. Optical devices may instead show gradual loss of output, increased dark current, or changes in wavelength and efficiency before complete electrical failure.

Measurement must distinguish these operating regimes. A handheld meter’s diode-test mode applies a small test current and reports an approximate forward voltage; it does not establish rated reverse blocking, surge capability, or high-frequency recovery. Curve tracers and semiconductor parameter analyzers characterize current–voltage behavior, while switching measurements require controlled current, voltage, temperature, and circuit inductance. For fast power devices, probe capacitance and connection inductance can materially alter the measured waveform, making the test arrangement part of the reported result.