The Shockley diode equation describes the current–voltage relationship of an ideal semiconductor diode. It predicts that forward current increases approximately exponentially with applied voltage, while reverse current approaches a small saturation value before electrical breakdown occurs. Named after William Shockley, the equation is a fundamental model in electrical engineering and semiconductor device physics. Its strict theoretical form describes an ideal p–n junction; a commonly used extension includes an ideality factor to represent departures from ideal behavior.
Mathematical form
The equation is usually written as
where:
- is the conventional current through the diode, positive in the forward direction.
- is the reverse saturation current, a device- and temperature-dependent parameter.
- is the voltage across the junction, positive for forward bias.
- is the positive magnitude of the elementary charge.
- is the Boltzmann constant.
- is the absolute junction temperature.
- is the dimensionless ideality factor, also called the emission coefficient.
- is the thermal voltage.
At ,
The strict ideal diffusion-current equation has . In practical modeling, is often between 1 and 2 over a useful operating range, although measured values can lie outside that range when other transport mechanisms or parasitic effects are important.
The thermal voltage is not a diode’s forward-voltage drop or turn-on voltage. It is the temperature-dependent voltage scale governing the exponential relation.
Physical basis
Junction equilibrium and carrier injection
A p–n junction joins p-type and n-type semiconductor regions. Carrier diffusion near the interface establishes a depletion region containing fixed ionized dopants. Its internal electric field opposes further diffusion. At thermal equilibrium, opposing carrier flows balance, so the net terminal current is zero.
Forward bias reduces the junction’s potential barrier. This increases the concentration of minority carriers at the depletion-region boundaries: electrons enter the p-type material, and holes enter the n-type material. Under ideal conditions, these boundary concentrations increase exponentially with applied voltage.
Injected carriers then diffuse through the approximately charge-neutral regions and recombine. Solving the steady-state carrier transport equations gives the exponential current–voltage relationship.
Reverse bias increases the barrier and suppresses majority-carrier injection. In the ideal model, the remaining current is limited by the supply of thermally generated minority carriers and approaches .
Ideal assumptions
The elementary derivation assumes:
- Steady-state operation.
- A uniform junction temperature.
- Nondegenerate carrier statistics, so Boltzmann approximations apply.
- Low-level injection, meaning injected minority carriers do not substantially change majority-carrier concentrations.
- Negligible generation and recombination within the depletion region.
- Negligible voltage loss in the semiconductor bulk and electrical contacts.
- Operation below reverse breakdown.
For a simple, long-base junction with uniform properties, the saturation current is
where is junction area, is intrinsic carrier concentration, and are acceptor and donor concentrations, and are minority-carrier diffusion coefficients, and and are minority-carrier diffusion lengths.
This expression shows why depends on semiconductor material, doping, geometry, carrier lifetime, and temperature. Devices with short neutral regions or different boundary conditions require modified expressions.
Operating regimes
Forward bias
When is several times , the exponential term is much larger than unity:
Solving for voltage gives
Consequently, equal multiplicative increases in current require approximately equal voltage increases. At , a tenfold current increase requires
provided temperature, ideality factor, and the dominant transport mechanism remain approximately constant.
The equation contains no sharp forward-voltage threshold. The familiar approximation of a silicon diode as having a fixed drop near is a circuit simplification for particular operating conditions, not a fundamental switching boundary.
Near zero voltage
For , expanding the exponential gives
Thus, the ideal characteristic is approximately linear very close to the origin. Retaining the minus-one term is essential here: it ensures that the net current is zero at zero applied voltage.
Reverse bias
For a sufficiently negative voltage, but below breakdown,
Actual reverse leakage may vary substantially with voltage because of depletion-region generation, surface leakage, defects, and field-assisted transport. The model’s therefore need not equal the total reverse leakage measured in a real device.
The Shockley equation does not describe avalanche breakdown or tunneling breakdown. These mechanisms produce large reverse currents that require additional models.
Ideality factor and practical departures
The ideality factor summarizes the slope of an exponential current–voltage region. In forward bias, when the minus-one term is negligible,
A value near 1 is associated with ideal minority-carrier diffusion. Recombination through defect states within the depletion region can produce a component with near 2 under appropriate conditions. High-level injection can also produce slopes corresponding to an apparent ideality factor near 2, so the factor alone does not uniquely identify a mechanism.
Real diodes may exhibit several operating regions with different apparent ideality factors. A single fitted and generally describe only a limited current and temperature range.
At high forward current, bulk and contact resistance become important. A common extension is
where represents series resistance. The junction voltage is then smaller than the measured terminal voltage. At still higher currents, self-heating and changes in carrier transport further limit the simple model.
Temperature dependence
Temperature affects both the thermal voltage and the saturation current. Although increases linearly with absolute temperature, generally increases much more strongly, largely because intrinsic carrier concentration depends strongly on temperature and semiconductor band gap.
For many silicon p–n diodes operated at constant forward current, the forward voltage decreases as temperature rises. A coefficient near is a common engineering approximation, but the actual value depends on current, device construction, and temperature range.
Treating as temperature-independent can therefore give a misleading prediction. Practical models must represent both its temperature dependence and the changing thermal voltage.
Small-signal behavior and applications
Differentiating the equation gives the incremental conductance:
The corresponding small-signal resistance is
In appreciable forward conduction,
For example, at , , and , the ideal incremental resistance is approximately . This is the local slope resistance, not the static ratio , and it excludes external series resistance.
The equation supports diode bias calculations, rectifier analysis, logarithmic circuits, temperature-sensing circuits, and small-signal models used with an electronic amplifier. Its exponential relationship also underlies closely related transistor junction models.
In a solar cell, a common single-diode model combines the diode current with a light-generated current source and parasitic resistances. Using the convention that delivered output current is positive,
where is photocurrent and is shunt resistance.
Historical context and scope
Shockley developed the theoretical description of p–n junction behavior during the emergence of junction transistors in the late 1940s. His 1949 work on p–n junctions and junction transistors established the diffusion-based treatment associated with this equation.
Despite its name, the Shockley diode equation should not be confused with the Shockley diode, a four-layer switching device.
The equation is a steady-state current–voltage law, not a complete dynamic device model. Junction capacitance, stored minority-carrier charge, reverse recovery, noise, and thermal transients require additional descriptions. More detailed models are particularly important in high-frequency circuits and power electronics, where switching behavior and energy loss cannot be inferred from the static exponential relationship alone.