What Is a Capacitor? How It Stores Charge and What It Does
A capacitor is a part made of two metal plates facing each other. Between them sits a thin layer that does not conduct electricity (an insulator). Apply a voltage, and positive charge collects on one plate and negative charge on the other. The capacitor can then release that charge quickly when needed. It is not a battery, which makes electricity through chemical reactions. Capacitors are in every electronic device: they steady power supplies, remove noise and work in memory chips and touchscreens.
What you'll learn
- How a capacitor stores charge, and how it differs from a battery
- The three things that set how much it can store (its capacitance)
- How the main types differ: ceramic, electrolytic and film
- Why one smartphone can hold up to about 1,000 of them, and where you meet them every day
- Why large capacitors call for care
Prerequisites: a rough idea of voltage and current, at the level of school science. There is just one formula, and it is explained in words first.
In everyday life
Look at the board inside a phone or computer, the one all the parts sit on. It is packed with tiny square parts, smaller than a sesame seed. Many of them are capacitors. In 2024, the parts maker Murata said that the latest smartphones use up to about 1,000 capacitors. That count is of one type, the multilayer ceramic capacitor, explained later.
The power section of a home appliance often holds a larger capacitor shaped like a small can. An electric car also has a large capacitor, between the battery and the circuit that drives the motor. Capacitors come in many shapes and sizes, but the basic job they do is the same.
The same word, capacitor, is used for a separate part on a board and for the same structure built inside a chip. Both mean the same thing. This article mostly talks about capacitors as parts.
How it works 1: two metal plates and an insulator between them
The basic shape of a capacitor is very simple. Take two metal plates, which conduct electricity, and set them face to face with a gap between them. Fill the gap with something that does not conduct electricity, an insulator. Even air will do.
Connect a battery to the two plates. The battery pulls electrons out of one plate and pushes them into the other. Electrons carry negative charge, so the plate that gains electrons becomes negative. The plate that loses them becomes positive. The amounts of positive and negative charge are always equal.
Positive and negative charges attract each other. But the insulator in between blocks them, so they cannot cancel out. That is why the two plates stay charged even after you remove the battery. This is what "storing electricity" really means.
The oldest example of this shape is the Leyden jar. It was found separately in Germany and the Netherlands in 1745–1746 (IEEE technology history). It was a glass jar with metal on the inside and outside, using the glass as the insulator. Leiden is a university town in the Netherlands.
How it works 2: the three things that set how much it stores
Apply the same voltage to different capacitors. How much charge does each one store? This "ease of storing" is called capacitance. Its unit is the farad (F). A capacitance of 1 F stores 1 coulomb (a unit of electric charge) at 1 volt.
As a formula: stored charge = capacitance × voltage. Double the capacitance, and the same voltage stores twice the charge. One farad is a very large amount. Electronic parts often use µF (a millionth of a farad), nF (a billionth) and pF (a trillionth).
Three things set the capacitance. It grows when the plates are larger and when the gap is narrower (a physics explainer from Georgia State University in the US). Larger plates have more room for charge. Plates that are closer together make positive and negative charges attract more strongly.
The third thing is the material in the gap. Even with the same shape, the material changes how much can be stored. This property is called permittivity. It is given as how many times more charge you can store than with a vacuum in the gap (strictly, this is the "relative permittivity"). Air is almost 1. Silicon dioxide, a kind of glass widely used in chips, is about 3.9.
Changing the material to gain capacitance is an idea used inside chips too. Since 2007, advanced chips have used a higher-permittivity material for the film under the transistor gate. The gate is the electrode that turns the current on and off ("Quantum Tunneling in Transistors and Flash Memory", high-k / metal gate).
How is it different from a battery?
When you hear "storing electricity," you may think of a battery. Both can take in and give out electricity. But what happens inside is completely different. So their strengths are opposite, too.
Double the voltage, and the stored charge doubles. But the stored energy grows four times. That is because twice the charge is pushed out at twice the voltage. So the higher the voltage a capacitor works at, the more energy it holds.
How it works 3: charging and discharging
Connect a capacitor to a battery through a resistor (a part that resists the flow of current). The voltage does not jump up at once. At first, charge rushes in. As charge builds up, the flow slows down. Storing charge is called charging, and releasing it is called discharging.
A yardstick for how gradual this is: the resistance multiplied by the capacitance, called the time constant. It is also one reason signals are delayed inside chips. The detailed numbers are in "RC Time Constant Explained," so they are left out here.
The key point is that a capacitor's voltage cannot change suddenly. To change the voltage, charge has to flow in or out. This ability to soften sudden changes supports many of the uses we look at next.
Main types: ceramic, electrolytic and film
Capacitors are named after the insulating material between the plates. This insulating material is called the dielectric. The material makes a big difference to the size, the capacitance and the voltage a capacitor can withstand.
| Type | Material between the plates | Features and common uses |
|---|---|---|
| Multilayer ceramic (MLCC) | Ceramic | Small and good with fast changes. Steadying the power next to chips, removing noise |
| Aluminum electrolytic | Oxide film on the aluminum surface | Large capacitance in a fairly small body. Has a plus and a minus side. Smoothing power supply voltage |
| Film | Thin plastic film | Suited to high voltages. Circuits that drive the motors of electric cars, and others |
Multilayer ceramic capacitors (MLCCs)
The "up to about 1,000" capacitors counted in a smartphone are of this type. It is called MLCC, short for multilayer ceramic capacitor. Thin layers of ceramic and thin metal films are stacked one on top of the other and made into one small block.
Films joined to the left terminal alternate with films joined to the right terminal. That turns each ceramic layer into a small capacitor of its own. The more layers you stack, the larger the facing area, so a small body gives a large capacitance.
The ceramics come in two broad families. One is based on barium titanate, which has a very high permittivity and gives a large capacitance. The other has a lower permittivity, but its capacitance hardly changes with temperature (Murata). The race to shrink them also goes on: in 2024, Murata announced it had developed a product measuring 0.16 mm × 0.08 mm.
Aluminum electrolytic capacitors
A very thin film of aluminum oxide (a kind of rust on aluminum) is formed on the surface of aluminum foil. This film is used as the dielectric. The foil's surface is made finely rough to widen its area. A thin film combined with a wide area gives a large capacitance.
An ordinary aluminum electrolytic capacitor has a plus side and a minus side (polarity). Makers' documents clearly warn against applying voltage the wrong way round (Nippon Chemi-Con). When fitting one to a board, it is important to get the direction right.
Film capacitors
These are made by coating a thin plastic film with a metal film, then rolling or stacking it. They suit circuits that work at high voltages. In electric cars, they sit between the battery and the circuit that drives the motor, and keep the voltage steady. For the same volume, they store far less than aluminum electrolytics. So they do not suit places that need a large capacitance (a Panasonic explainer).
Use 1: smoothing power and holding up chips
The electricity from a wall outlet is alternating current (AC), which keeps switching direction. A diode can make it flow one way only, but its size still rises and falls in waves ("What Is a Diode?"). A large capacitor smooths out these waves: it stores charge when the voltage is high and gives it out when the voltage is low. A capacitor in this role is called a smoothing capacitor.
Chips also need capacitors right next to them. When a chip such as a CPU starts work, the current it needs jumps in an instant. The power circuit sits a little way off and cannot keep up with such a sudden change. A capacitor right next to the chip fills that gap with the charge it has stored.
This use is called decoupling, or bypassing. The capacitor is placed as close as possible to the chip's power pins. That is because the wiring itself holds back fast changes. An explainer from Analog Devices in the US gives an example. Even a 1 cm wire acts like a resistance of about 5 Ω (ohms, the unit of resistance) to a current that changes 100 million times a second. "Acts like a resistance" is a simplification made here.
On a real board, the roles are split: larger capacitors near the power supply, small MLCCs near the chip. The power circuits that change one voltage into another (DC-DC converters) contain capacitors too. One reason a smartphone has so many capacitors is that this support is needed for each chip and each power supply.
Use 2: passing only changes, and keeping time
If the voltage stays constant, a capacitor stops passing current once it is charged. If the voltage keeps changing, current keeps flowing in and out. So a capacitor blocks electricity that does not change direction (direct current, DC) and lets AC through.
This property lets you remove just the unwanted steady voltage from an audio signal. Faster changes pass more easily. So, combined with a resistor, a capacitor can make a circuit that removes only small, fast noise (a filter). How easily changes of different speeds pass is covered in detail in the article on AC vs DC.
The time it takes to charge is set by the resistance and the capacitance. This can be used in place of a clock. Examples are a circuit that blinks a lamp at a steady pace, and a circuit that waits a moment after power-on before it starts.
Use 3: remembering, sensing and releasing all at once
The working memory of a computer or phone (DRAM) stores each 0 or 1 (one bit) with one transistor and one capacitor. For example, a charged capacitor means 1, and an empty one means 0. Robert Dennard of IBM came up with this idea in 1966.
Since it pairs one transistor (T) with one capacitor (C), this design is called the 1T1C cell. The charge in this capacitor slowly leaks away, so it must be rewritten at regular intervals ("Memory vs Storage"). To gain capacitance in the tiny space on a chip, the capacitor may be shaped like a tall, thin cylinder, which adds area vertically. This is the "area" point from How it works 2 in action.
A phone's touchscreen senses a finger by a tiny change in the capacitance between electrodes under the screen. The change happens because a human finger conducts electricity too ("How Touchscreens Work"). Some small sensors that measure shaking or tilt work the same way. They read a change in the gap between plates as a change in capacitance.
A camera's strobe flash (a gas-filled tube that lights up) is the opposite: it releases charge all at once. It works differently from the LED light on a phone. The flash raises the battery voltage and slowly charges a capacitor, then dumps the charge at the moment of the shot to make the light. Some capacitors for photo flashes are rated for 330 V (a datasheet from Cornell Dubilier in the US).
Caution: large capacitors can stay charged
A capacitor can keep its charge even after the power is removed. It is a part built to store charge, so this is only natural. With small MLCCs, it is not a problem. Large capacitors that work at high voltages are another matter.
The power circuits of devices that plug into the wall, and camera flashes, contain capacitors that work at over a hundred volts. Touching one can give you a shock, even right after you unplug the device. Some designs include a part that slowly drains the charge, but do not rely on it to open a device.
Electrolytic capacitors have another effect: even after being discharged, voltage can come back across the terminals after a while. Nippon Chemi-Con's technical note covers this as "recovery voltage." It is safest not to take apart appliances or power supplies. Leave repairs to professionals.
Common misconceptions
Myth: Current passes straight through a capacitor
In fact, there is an insulator between the two plates, and electrons do not pass through it. As electrons enter one plate, the same number leave the other plate at the same time. From the outside, it looks as if current passed through. We say a capacitor "passes" AC because this in-and-out flow keeps repeating.
Myth: A capacitor always gives the capacitance printed on it
In fact, capacitance changes with operating conditions. Ceramic parts with a high permittivity, in particular, lose capacitance with temperature and with the DC voltage applied (a Murata explainer). Designers check the capacitance under their real operating conditions in the maker's data.
Myth: There are no capacitors inside a chip
In fact, a chip is full of capacitance. A transistor's gate is an electrode facing the silicon across a thin insulating film, so its shape is exactly that of a capacitor. Wires also face each other across insulating films. The PN junction in a diode (the boundary between two different kinds of semiconductor) has capacitance too (junction capacitance). In DRAM, capacitors are even built in on purpose.
By the numbers
| Number | What it is |
|---|---|
| Up to about 1,000 | Number of multilayer ceramic capacitors in the latest smartphones (Murata, 2024) |
| 0.16 mm × 0.08 mm | Size of the smallest multilayer ceramic capacitor at the time of the 2024 announcement (Murata) |
| About 3.9 | Permittivity of silicon dioxide, used for insulating films in chips (taking a vacuum as 1) |
| About 5 Ω | How much a 1 cm wire holds back changes at 100 MHz (100 million times a second) (Analog Devices) |
| 330 V | An example of the rated voltage of a capacitor for photo flashes (Cornell Dubilier) |
| 1745–1746 | When the first capacitor, the Leyden jar, was discovered |
0.16 mm × 0.08 mm is about the size of one grain of fine sand. Yet many films are stacked inside it. Phones keep getting thinner and smaller while holding more parts. They still fit on the board because of shrinking steps like this.
Learn more
- Related explainers: "RC Time Constant Explained" (how fast charging and discharging happen), "Memory vs Storage" (one bit of DRAM), "How Touchscreens Work"
- Compare in a table: series vs. parallel circuits (the capacitance of connected capacitors) / capacitive vs. piezoresistive MEMS sensors
- Try the numbers: RC time constant calculator / combined capacitance calculator / PDN target impedance calculator (choosing decoupling capacitors, intermediate level)
- Next terms: smoothing capacitor, 1T1C cell, capacitive sensing