Put a diode in a circuit the "wrong" way round and almost nothing flows — a few nanoamps at most. Turn it around and the same part conducts amps with well under a volt across it. Nothing mechanical moves inside, so what is doing the blocking? The answer is a thin layer of silicon that has been emptied of free charge: the depletion region.
The short answer
A diode is a PN junction: a piece of silicon that is doped one way on one side and the opposite way on the other. Where the two halves meet, free charges cancel each other out and leave behind a narrow zone with no carriers and a built-in electric field. That zone is an insulator.
- Forward bias pushes against the built-in field, shrinks the zone and lets current flood across once you pass roughly 0.6–0.7 V (silicon).
- Reverse bias pulls in the same direction as the built-in field, makes the zone wider and leaves almost nothing that can carry current across it.
Everything else — leakage, capacitance, breakdown — follows from that one picture.
Inside the PN junction
Pure silicon is a poor conductor. Doping changes that:
- N-type silicon gets a tiny amount of phosphorus. Each phosphorus atom brings one spare electron that is free to move. Electrons are the majority carriers.
- P-type silicon gets boron. Each boron atom is one electron short, leaving a "hole" that behaves like a mobile positive charge. Holes are the majority carriers here.
Both pieces are electrically neutral on their own — every free electron is balanced by a fixed positive phosphorus ion, every hole by a fixed negative boron ion. The interesting part happens when the two are made in one crystal.
Near the boundary, electrons from the N side diffuse into the P side and fill holes; holes diffuse the other way. Each crossing removes a pair of mobile carriers but leaves the fixed ions behind: positive ions on the N side, negative ions on the P side. Those uncovered ions create an electric field pointing from N to P that pushes back on any further diffusion. Very quickly the two effects balance.
The field corresponds to a built-in potential of roughly 0.6–0.8 V for silicon. You cannot measure it with a voltmeter across a bare diode (the contacts cancel it out), but every carrier that wants to cross the junction has to climb it.
Forward bias: lowering the barrier
Connect the P side (anode) to positive and the N side (cathode) to negative. The external voltage opposes the built-in field, so the depletion region narrows and the barrier drops. Majority carriers can now cross in huge numbers: holes into the N side, electrons into the P side.
The current rises exponentially with voltage, which is described by the Shockley diode equation:
Here IS is the tiny saturation current, n is an ideality factor between 1 and 2, and VT ≈ 26 mV at room temperature. Because of the exponential, every extra ~60 mV multiplies the current by about ten. That is why a silicon diode seems to "turn on" at 0.6–0.7 V: below that the current is negligible, above it the current is limited mainly by the rest of the circuit.
Reverse bias: widening the barrier
Now swap the polarity: anode negative, cathode positive. The external field adds to the built-in one. Electrons on the N side are pulled towards the positive terminal, holes on the P side towards the negative terminal — both away from the junction. More fixed ions are uncovered, the depletion region gets wider, and the barrier gets taller.
The widening stops when the uncovered ion charge matches the applied voltage. After that brief rearrangement, majority carriers simply cannot climb the barrier, so they contribute no steady current. The diode blocks.
Why a little current still leaks
"Almost nothing" is not "nothing". Heat constantly breaks a few electrons free from silicon atoms, creating electron–hole pairs everywhere in the crystal. Inside the P side these thermally generated electrons are rare, so they are called minority carriers (likewise holes on the N side).
If a minority carrier wanders into the depletion region, the field is pointing the right way to sweep it across. That trickle is the reverse saturation current, IS — the same constant that appears in the Shockley equation. Two properties are worth remembering:
- It barely depends on voltage. The supply of minority carriers is set by temperature, not by how hard you pull. Doubling the reverse voltage hardly changes it.
- It depends strongly on temperature. A common rule of thumb is that silicon leakage roughly doubles every 10 °C.
For a small-signal silicon diode such as the 1N4148 the datasheet leakage is in the nanoamp range at room temperature. Schottky diodes trade a lower forward drop for much higher leakage — microamps to milliamps on larger parts, rising fast with heat — which matters in battery-powered and high-temperature designs.
The diode as a capacitor
A reverse-biased diode is two conductive regions separated by an insulating layer — the exact recipe for a capacitor. As reverse voltage increases, the depletion region widens, so the "plates" move apart and the junction capacitance falls.
Varactor (varicap) diodes are built to exploit this: a DC reverse voltage tunes their capacitance and therefore the frequency of an oscillator or filter. In switching circuits the same capacitance is a nuisance — it has to be charged and discharged every cycle, which is one reason fast diodes are physically small.
Breakdown: when blocking stops
Keep raising the reverse voltage and eventually the current shoots up abruptly. This is reverse breakdown, and it happens by two mechanisms:
- Avalanche breakdown. The field becomes strong enough that a minority carrier gains enough energy between collisions to knock new electrons free. Those do the same, and the current multiplies like an avalanche. Dominant in lightly doped junctions and at higher voltages.
- Zener breakdown. In heavily doped junctions the depletion region is extremely thin, and the field can pull electrons straight out of their bonds by quantum tunnelling. Dominant below roughly 5 V.
Breakdown is not automatically destructive. What destroys a diode is heat: high reverse voltage times breakdown current is a lot of power in a tiny junction. A rectifier like the 1N4007 is rated for 1000 V of peak reverse voltage and must never reach breakdown in normal use. A Zener diode, on the other hand, is designed to sit in breakdown permanently, with a resistor limiting its current — that is how it holds a stable reference voltage.
What this means in real circuits
| Use | Which property it relies on |
|---|---|
| Rectifier (AC → DC) | Blocks the reverse half-cycle; peak reverse voltage rating must exceed the AC peak. |
| Reverse-polarity protection | Blocks if the battery is connected backwards. |
| Flyback diode across a relay or motor | Normally reverse-biased (idle); conducts only when the coil's voltage spike flips polarity. |
| Zener reference / clamp | Operates deliberately in controlled breakdown. |
| Varactor tuning | Junction capacitance that changes with reverse voltage. |
| Photodiode | Reverse-biased; light creates carriers in the depletion region, so leakage becomes signal. |
Try it with a multimeter
- Set the meter to diode mode (the diode symbol).
- Red probe on the anode, black on the cathode (the band). A silicon diode reads about 0.5–0.7 V; a Schottky about 0.15–0.4 V; an LED may read higher or light up faintly.
- Swap the probes. The meter should show
OL— open — because the diode is reverse-biased and blocking. - A reading near 0 V both ways means a shorted diode;
OLboth ways means it is open.
Summary
- A PN junction forms a depletion region with no free carriers and a built-in field.
- Forward bias shrinks it and current rises exponentially past ~0.6–0.7 V.
- Reverse bias widens it; only thermally generated minority carriers leak across — nanoamps for small silicon diodes, more for Schottky parts and at high temperature.
- At high enough reverse voltage, avalanche or Zener breakdown lets large current flow; heat is what causes damage.