EC-6.3 The P-N Junction
The standard p-n junction treatment as taught in semiconductor device courses, September 2026
What this is and why it exists
One structure, and almost every device in electronics is built from it.
Put a p-type region against an n-type one. Carriers diffuse across, leaving behind fixed charge that sets up a field opposing further diffusion. What you end up with is a region emptied of mobile carriers, with a voltage across it that nobody applied.
Everything a diode does follows from what an applied bias does to that region. So does the transistor in two topics' time, and the solar cell in one.
Draw the band diagram until you can draw it from memory in all three states. Nothing else in this module repays the effort as well.
The vocabulary
- Junction — the boundary where p-type meets n-type.
- Depletion region — the zone either side of it, emptied of mobile carriers.
- Built-in potential — the voltage across that region with nothing connected.
- Forward bias — an applied voltage that opposes the built-in field.
- Reverse bias — an applied voltage that adds to it.
- Saturation current — the small current that flows under reverse bias.
- Thermal voltage — the temperature-dependent quantity in the diode equation's exponent.
- Junction capacitance — the charge storage of the depletion region, which varies with voltage.
The mental model
Imagine the two regions brought together, and follow what happens.
The n side has far more electrons, so electrons diffuse into the p side. The p side has far more holes, so holes diffuse the other way. Each departing carrier leaves behind an impurity ion it can no longer neutralise — fixed, immobile, charged.
Those fixed charges build an electric field pointing from the n side to the p side. That field opposes exactly the diffusion that created it. Diffusion continues until the field is strong enough to stop it, and then everything settles.
What is left is the depletion region: a zone with almost no mobile carriers, containing fixed charge, with a built-in potential across it. Nothing is connected and there is a voltage there. It cannot be measured with a voltmeter across the terminals, because the contacts add their own offsets that cancel it exactly. That is a good thing. A diode supplying free voltage would be a perpetual motion machine.
Now bias it.
Forward bias applies a voltage opposing the built-in field. The barrier drops, the depletion region narrows, and diffusion resumes. Because the barrier appears in an exponent, current rises very steeply once the barrier is nearly cancelled. That steepness is why a silicon diode appears to switch on at around two thirds of a volt. It is not a threshold. It is an exponential that was negligible and suddenly is not.
Reverse bias applies a voltage adding to the built-in field. The barrier grows, the depletion region widens, diffusion stops almost completely. A tiny current still flows, carried by minority carriers generated inside the depletion region, and it barely changes with voltage.
The diode equation puts this in one line: current depends exponentially on applied voltage, with the thermal voltage in the exponent. Two consequences follow. First, a diode has no single resistance — the ratio of voltage to current is different at every point. Second, temperature enters through that thermal voltage, which is why diode circuits drift.
One last property, and it is the one most often overlooked. The depletion region separates charge across a zone with no mobile carriers in it, which is exactly what a capacitor is. So a junction stores charge, and because the region's width changes with applied voltage, the capacitance changes too. That is deliberately exploited to make voltage-controlled tuning elements, and it is an unwanted parasitic everywhere else, limiting how fast a diode can switch.
What you should now be able to explain or do
- Describe what forms when p-type and n-type meet, and why it stops.
- Say what the depletion region and built-in potential are.
- Explain forward and reverse bias in terms of the barrier and the region width.
- Say why a diode appears to turn on at a particular voltage without having a threshold.
- Explain why a junction is also a voltage-dependent capacitor.
Check yourself
What stops carriers diffusing across a junction indefinitely?
The fixed charge left behind builds a field opposing the diffusion. When the two balance, the net flow stops.
Why can you not measure the built-in potential with a voltmeter across the diode?
The metal contacts introduce their own offsets that cancel it exactly. If they did not, the diode would supply energy from nothing.
A silicon diode is said to turn on at about two thirds of a volt. Is that a threshold?
No. The current is exponential in voltage, so it is negligible below that and large above it. The apparent threshold is the exponential becoming visible.
Why does a junction have capacitance, and why does it change with voltage?
The depletion region separates charge across a carrier-free zone, which is a capacitor. Reverse bias widens the region, and a wider separation means less capacitance.
Go deeper
We haven't checked most of these for screen reader use yet.