What Is a Diode? How It Lets Current Flow Only One Way
A diode is a semiconductor part that lets current flow easily in one direction and hardly at all in the other. Inside, it is the boundary where two different kinds of semiconductor meet (a PN junction). Diodes turn the AC from a wall outlet into DC in chargers, and they protect circuits. LEDs and solar cells belong to the same family.
What you'll learn
- How a diode makes current flow "one way only"
- Why the number "about 0.6–0.7 volts" comes up with silicon diodes
- What "breakdown" is, which happens when the reverse voltage gets too high
- "Rectification," which turns AC into DC, and how diodes protect circuits
- How Zener diodes, Schottky diodes, LEDs and solar cells are related
Prerequisites: it helps to know n-type and p-type (semiconductors with extra electrons, and with extra holes, the empty seats for electrons) ("What Is a Semiconductor?"). No equations are used.
Where you see it
Phone chargers and laptop power adapters contain diodes. They help change the electricity from the wall outlet into a form the device can use. The glowing part of an LED bulb is a kind of diode. So is the part at the tip of a TV remote that sends out infrared light.
The circuit boards in cars and home appliances carry many diodes. Chips such as CPUs also have diodes built in to protect their circuits from static electricity.
In circuit diagrams, a diode is drawn as a triangle with a bar across its tip. The triangle points in the direction that current flows easily. The two terminals have names. Current enters at the anode and leaves at the cathode.
How it works 1: a one-way street for current
Let's connect a diode to a battery and try to send a current. If the current flows the way the triangle points, it flows well. This direction is called forward. Connect it the other way, and almost no current flows. This is the reverse direction.
The ease of current flow changes greatly with the direction of the voltage. This property is called "rectification." A diode has no terminal that tells it to turn on or off. Only the direction and size of the voltage across it decide whether it passes current or blocks it. That is the difference from a transistor, which is switched on and off by a voltage.
How it works 2: inside is a PN junction
An ordinary diode is made by joining p-type and n-type semiconductors. N-type has many electrons to carry electricity. P-type has many "holes," the empty seats for electrons. The boundary where the two meet is called a PN junction.
Right near the boundary, some electrons from the n-type side and some holes from the p-type side cross over. They meet and cancel out. This leaves a thin layer at the boundary with almost no free electrons or holes. This layer is called the depletion layer.
Inside the depletion layer, the n-type side, which lost electrons, becomes positively charged. The p-type side, which lost holes, becomes negatively charged. This imbalance of charge pushes electrons and holes back, so that no more cross the boundary. It is as if a "hill" for electrons and holes has formed at the boundary.
The PN junction was found in 1940 by Russell Ohl at Bell Labs in the US, inside a sample of silicon (Computer History Museum). This discovery led to later solar cells and transistors.
Connecting the p-type side to the battery's plus and the n-type side to its minus is the forward direction. The outside voltage lowers the hill at the boundary. Electrons and holes cross the boundary one after another, then meet and cancel out on the other side. The battery replaces the ones that cancel out, so the current keeps flowing.
Connect it the other way, and electrons and holes are pulled toward the electrodes, away from the boundary. The depletion layer widens, and the hill gets higher. Almost no electrons or holes can get over it, so almost no current flows.
How it works 3: what "0.6–0.7 volts" means
Even in the forward direction, a very small voltage gives only a tiny current. Few electrons and holes can get over the hill. In a silicon diode, current starts to flow clearly at about 0.6–0.7 volts. The voltage across a diode while it carries current is called the "forward voltage."
This number is linked to the height of the hill at the boundary. The hill's height with no outside voltage, expressed as a voltage, is called the built-in potential. In silicon, it is about 0.6–0.9 volts. When an outside voltage lowers this hill enough, electrons and holes can cross in large numbers.
The key point is that 0.6–0.7 volts is not a line where current "suddenly starts." The current rises smoothly with voltage, but very steeply. Take Toshiba's datasheet for the 1SS181, a diode for small currents (a small-signal diode). Its typical value at 1 milliamp (one thousandth of an amp) is 0.61 volts. At 10 milliamps it is 0.74 volts, and at 100 milliamps, 0.92 volts. Each time the current rises tenfold, the voltage grows by only about 0.1–0.2 volts.
So for rough circuit estimates, "about 0.6–0.7 volts" is a handy number to remember. At large currents, though, the resistance inside the part adds to it, giving around 1 volt.
Forward voltage also changes with temperature. Keep the current through a silicon diode the same and raise the temperature by 1 °C. The voltage drops by about 2 millivolts (0.002 volts). Some chips contain diodes that use this property to measure temperature (Toshiba).
A different material gives a different forward voltage. LED materials have a larger energy "step" for electrons (the band gap) than silicon, so the hill at the boundary is higher. Their forward voltage is higher too. For ROHM products at 20 milliamps, the typical value is 1.9 volts for a red LED and 3.2 volts for a blue LED.
How it works 4: the reverse limit, breakdown
Even in the reverse direction, the current is not exactly zero. Heat creates a few electrons and holes, and they pass through the boundary. For example, the small-signal diode 1N4148 is rated for a reverse current of at most 25 nanoamps at 20 volts and 25 °C. One nanoamp is one billionth of an amp. At 150 °C, however, the limit is 2000 times higher.
Raise the reverse voltage further, and at a certain voltage a large current suddenly flows. This is called "breakdown." The voltage at which it happens is set by each product's design. For example, the 1N4007 rectifier diode is rated to take up to 1000 volts in reverse, again and again.
Breakdown happens in two main ways. In the first, the strong electric force in the depletion layer speeds up electrons. They crash into atoms and knock loose new electrons and holes. The new electrons also speed up and crash, so the number grows like an avalanche. This is called avalanche breakdown.
The second happens when the depletion layer is made very thin by adding a lot of impurities. Electrons slip through a thin wall they should not be able to cross, thanks to quantum mechanics ("Quantum Tunneling in Transistors and Flash Memory"). Breakdown caused by this slipping through is called Zener breakdown.
Breakdown by itself does not destroy a diode. What destroys it is the heat when a large voltage and a large current come together. If the heat is kept within safe limits, a diode can keep working in breakdown. The Zener diode, covered later, is built to do just this.
Use 1: turning AC into DC (rectification)
The electricity from a wall outlet switches its direction of flow back and forth. It does this 50 or 60 times a second, depending on the region. Japan, for example, uses 50 Hz in the east and 60 Hz in the west. Electricity that keeps reversing like this is called alternating current (AC). Phone batteries and electronic circuits run on electricity that flows one way. This is called direct current (DC).
Put one diode in the path of AC, and only the half flowing one way gets through. The other half is blocked. This is called half-wave rectification. The current now flows one way, but no electricity arrives for half of the time.
Four diodes arranged in a diamond can also pass the reverse half, flipped around. They are connected so that current in either direction leaves the output flowing the same way. This is called full-wave rectification.
Rectified electricity flows one way, but its size still goes up and down in waves. So a part that stores electricity for a moment and then releases it (a capacitor) is added to smooth the waves. This brings it close to DC, steady in both direction and size.
Most chargers today first rectify the outlet's AC with a set of four diodes (a diode bridge), then smooth it with a capacitor. Next, they switch it on and off very fast, tens of thousands of times a second or more. It then passes through a small transformer that lowers the voltage. Finally, it is rectified and smoothed again into the DC sent to the phone (ROHM). The faster the switching, the smaller the transformer can be, so the charger gets smaller and lighter.
Use 2: protecting circuits
Diodes also protect circuits. For example, a diode can be placed in the battery's path so that no current flows if the battery is put in backward. A little electricity is lost, equal to the forward voltage. But a wrongly inserted battery will not harm the circuit inside.
Diodes are also essential for parts that use a coil (a wound wire), such as motors and relays (switches moved by an electromagnet). A coil tends to keep its current flowing. So the moment a switch turns off, a high voltage appears, and it can destroy other parts. A diode connected alongside the coil gives the current a way out and keeps this high voltage down.
Static electricity built up on your body can rush from a fingertip into a terminal in an instant. This is called ESD (electrostatic discharge), and it can break thin films inside a chip. So chip terminals have built-in diodes that steer excess current away from the circuit. Boards may also carry separate diodes just for protection (Toshiba).
The diode family
Diodes come in many types, each built differently for its purpose.
| Type | Operating state | Main use |
|---|---|---|
| Rectifier (ordinary PN junction) | Passes forward, blocks reverse | Chargers, power supplies |
| Zener diode | Reverse breakdown | Making a fixed voltage, protection |
| Schottky diode | Forward (conducts at a low voltage) | Making power supplies more efficient |
| LED (light-emitting diode) | Forward | Giving off light |
| Photodiode | Reverse, or no voltage | Measuring the amount of light |
| Solar cell | No outside voltage applied | Making power from light |
Zener diodes: using breakdown to hold a voltage
A Zener diode is made so that breakdown happens at a chosen voltage. When current flows through it in reverse, the voltage across it stays close to that value. This is used to make a fixed voltage, or to keep a voltage from rising too high.
It is named after the physicist Clarence Zener. In 1934, he explained in theory a breakdown in insulators caused by electrons slipping through a wall. But Zener breakdown mainly works in parts whose breakdown voltage is below about 5 volts. In parts above that, avalanche breakdown is mostly what happens (Toshiba, Nexperia). Even so, by custom, both kinds are called Zener diodes.
Schottky diodes: low voltage, fast switching
A Schottky diode uses a junction between a metal and a semiconductor instead of a PN junction. Toshiba gives its forward voltage as about 0.4–0.7 volts at a set current, depending on the metal. That is lower than a PN junction under the same conditions. Less electricity is lost to this voltage, so power supplies become more efficient.
A PN junction diode keeps conducting backward for a short time right after it is switched to reverse. This lasts until the electrons and holes left inside cancel out. A Schottky diode has almost no such delay, so it suits power supplies that switch fast. Its weak points are a large reverse leakage current and, in silicon, a low tolerance for high voltage. For high voltages, Schottky diodes made of silicon carbide (SiC) are used ("What Is a Power Semiconductor?").
LEDs, photodiodes and solar cells: the light family
An LED is a diode that gives off light. Electrons and holes sent in the forward direction meet at the boundary, and their energy comes out as light. "How Do LEDs Work?" explains how the material sets the color.
A photodiode and a solar cell work the other way around. Light creates pairs of electrons and holes near the boundary. The electric force in the depletion layer pulls them apart, making a current. Camera pixels use this to measure light ("How Do Image Sensors Work?"). Solar cells use it to make power ("How Do Solar Cells Work?").
Common misconceptions
Myth: No current flows below 0.7 volts, and then it all flows at once
In fact, the current rises smoothly and very steeply with voltage. 0.6–0.7 volts is a rough guide to the voltage at which the currents used in ordinary circuits flow. Circuit lessons often simplify this to "on at 0.7 volts." That makes people think there is a real threshold.
Myth: A diode in breakdown is broken
In fact, breakdown itself is reversible. A diode is destroyed only when a large current flows and too much heat builds up. Zener diodes and protection diodes are made to work in breakdown. The name "breakdown" suggests damage, which leads to this myth.
Myth: Passing AC through a diode gives clean DC
In fact, a diode only makes the current flow one way. Right after rectification, its size still goes up and down in waves. A capacitor smooths the waves, and a voltage-control circuit evens it out. Only then is it DC that electronics can use. Diodes are often described simply as "parts that turn AC into DC," which leads to this myth.
By the numbers
| Number | What it is |
|---|---|
| About 0.6–0.7 V | Rough forward voltage at which a silicon diode starts to conduct clearly |
| About −2 mV/°C | Change in a silicon diode's forward voltage for each 1 °C rise (at the same current) |
| At most 25 nA | Current through a 1N4148 with 20 V applied in reverse (25 °C). One 40-millionth of an amp |
| 0.4–0.7 V | Forward voltage of a Schottky diode (Toshiba; at a set current) |
| 1.9 V / 3.2 V | Forward voltage of red and blue LEDs (typical values for ROHM products, 20 mA) |
With 10 milliamps flowing forward through a 1N4148, the voltage is at most 1 volt. In reverse, even at 20 volts, the current is at most 25 nanoamps. The ratio of the currents is about 400,000. That is what "current flows only one way" really means.
Learn more
- Compare in a table: Forward vs reverse bias, PN junction vs Schottky diode, Zener vs avalanche breakdown
- Try it with diagrams and numbers: the PN junction diagram (how the depletion layer width changes with voltage), the PN junction depletion calculator (change the doping and voltage and calculate)
- Terms for the next step: Rectifier circuit (the detailed forms of rectifier circuits), Reverse recovery time (the time current flows backward right after switching), Synchronous rectification (rectifying with transistors to cut losses)