Capacitance & the Parallel-Plate Capacitor

Every tap on your phone screen is a capacitance measurement. Under the glass sits a grid of tiny electrodes, and when your fingertip — a salty, conducting bag of water — hovers over one, it changes the amount of charge that electrode can hold at a given voltage by a few trillionths of a farad. The screen's controller notices the change and reports: touch, here. No button moves. The physics doing the work is the subject of this module: capacitance, the measure of how much charge a conductor arrangement can store per volt. You have spent two chapters building exactly the tools this idea needs. You know from Electrostatics of Conductors & Shielding that a conductor in equilibrium sits at one potential. You know from Gauss's Law at Work what field a charged sheet makes, and from Equipotentials & the Field–Potential Link that a uniform field and a voltage are linked by $E = V/d$ . Put those three facts in a row and the most important circuit component after the resistor — the capacitor — falls out in four lines of algebra. For your exams this module is a gift that keeps giving. The definition $C = Q/V$ and the parallel-plate formula $C = \varepsilon_0 A/d$ appear every single year across CBSE, ISC, JEE, NEET, AP Physics 2 and A-Level papers — as one-mark definitions, as three-mark derivations, and as the launch pad for the series-parallel networks and energy questions of the next two modules. Learn what the farad really measures, and the whole capacitor cluster becomes bookkeeping.

Key takeaways

A capacitor is two conductors separated by an insulator; capacitance $C = Q/V$ measures the charge stored per volt of potential difference. Unit: the farad, $1\ \mathrm{F} = 1\ \mathrm{C/V}$ .
$C$ depends only on geometry and the material in the gap — never on $Q$ or $V$ themselves. Double the charge and the voltage doubles with it; the ratio cannot move.
The workhorse: parallel plates of area $A$ and gap $d$ give $C = \varepsilon_0 A/d$ — big plates and a thin gap store more.
An isolated sphere of radius $R$ has $C = 4\pi\varepsilon_0 R = R/k$ — even the entire Earth manages only about $709\ \mu\mathrm{F}$ .
One farad is enormous: real electronics lives in microfarads, nanofarads and picofarads.
Key idea: capacitance is the size of a charge container — fixed by shape and spacing alone, so a capacitor stores charge in strict proportion to the voltage you apply.

The core insight

Capacitance $C = Q/V$ is the geometry-fixed ratio between the charge $\pm Q$ held on a pair of conductors and the potential difference $V$ between them — for parallel plates, $C = \varepsilon_0 A/d$ .

Where this lives today

The capacitance you learned to compute is now mostly a sensor. Your phone's touchscreen resolves the few-picofarad shift a fingertip causes; digital calipers read position from sliding capacitor plates; the accelerometer that rotates your screen measures a proof mass moving between electrodes as a change in $C$ . Meanwhile each of the billions of memory cells in a DRAM chip stores one bit as roughly $25\times10^{-15}$ F of charged capacitor — leaking constantly, refreshed dozens of times per second, which is why your laptop's memory forgets everything when power dies. At the other extreme sit supercapacitors: thousands of farads in a soda-can package, built by shrinking $d$ to a molecular layer and inflating $A$ into porous carbon with the internal area of several football fields. Buses in Shanghai and Belgrade brake into supercapacitor banks and reuse the energy at the next acceleration — a straight application of $C = \varepsilon_0 A/d$ pushed to its engineering limits, with the energy arithmetic you will meet in Energy Stored in a Capacitor.

The physical picture

Think of a capacitor as a bucket for charge. A battery is the pump: it pushes charge from one plate to the other until the voltage between the plates matches the battery's own. The voltage is the level the bucket has filled to; the charge is how much is inside; and the capacitance is the width of the bucket. A wide bucket — large plates, tiny gap — swallows a lot of charge before the level reaches the battery's voltage. A narrow bucket fills to the same level on a whisper of charge.

Figure

A capacitor is a charge bucket — capacitance is its width

The bucket picture carries one promise that deserves to be made precise, because examiners test it every year: the ratio $Q/V$ never changes for a given capacitor. Push in twice the charge and the field between the plates doubles everywhere (each bit of charge contributes its share, and shares add — the superposition idea you met in Superposition: Forces & Fields of Many Charges). Double field across the same gap means double voltage. Charge doubled, voltage doubled, ratio untouched. That is why $C$ deserves to be called a property of the device: it is fixed the moment the shapes and spacing are fixed, before any battery arrives. What does change $C$ ? Geometry and the gap material, nothing else. Move the plates closer, overlap them more, slide an insulating slab between them (the story of Dielectrics & Polarisation, two modules ahead) — or bring a fingertip near a screen electrode. Every capacitive touchscreen, hovering drone altimeter and digital caliper is a machine for noticing that $C$ changed because geometry changed.

Defining capacitance — and the enormous farad

Take any two conductors carrying charges $+Q$ and $-Q$ with a potential difference $V$ between them. Whatever their shape, the discussion above guarantees $Q \propto V$ , so the ratio is a constant worth naming:

C = \frac{Q}{V} \qquad (1)

where: $Q$ is the magnitude of the charge on either plate (not the total, which is zero), $V$ is the potential difference between the plates in volts, and $C$ is the capacitance in farads: $1\ \mathrm{F} = 1\ \mathrm{C/V}$ . Read it as an appetite: a 1 F capacitor swallows one full coulomb for every volt applied. Since a coulomb is itself a monstrous amount of charge — remember from Electric Charge & Its Properties that rubbing produces nanocoulombs — the farad is a monstrous unit. How monstrous? Suppose you demand $C = 1\ \mathrm{F}$ from two plates separated by 1 mm of air. The formula we derive below requires a plate area $A = Cd/\varepsilon_0 = (1)(10^{-3})/(8.85\times10^{-12}) \approx 1.13\times10^8\ \mathrm{m^2}$ — square plates about 10.6 km on a side, roughly the footprint of Mumbai's western suburbs. Real components therefore live far below the farad:

unit	size	where you meet it
microfarad, $1\ \mu\mathrm{F} = 10^{-6}\ \mathrm{F}$	camera-flash and power-supply capacitors	the fat cylinders on circuit boards
nanofarad, $1\ \mathrm{nF} = 10^{-9}\ \mathrm{F}$	filter and timing circuits	small ceramic discs
picofarad, $1\ \mathrm{pF} = 10^{-12}\ \mathrm{F}$	radio tuning, touchscreens, cables	your fingertip's effect on a screen

The exceptions that prove the rule are "supercapacitors" rated at thousands of farads — they cheat magnificently, using molecule-thin charge layers and sponge-like electrodes with football fields of internal area, exactly the two levers ( $d$ down, $A$ up) that equation (2) is about to expose.

The parallel-plate capacitor

The parallel-plate capacitor — two flat plates of area $A$ facing each other across a small gap $d$ — is the workhorse of the subject and the geometry every board asks you to derive. The beauty is that you already own every ingredient. When the plates carry $+Q$ and $-Q$ , the charge spreads uniformly with surface density $\sigma = Q/A$ , and you saw in Gauss's Law at Work: Wire, Sheet & Shell that two such facing sheets produce a uniform field confined to the gap, of strength $E = \sigma/\varepsilon_0$ . And you saw in Equipotentials & the Field–Potential Link that a uniform field across a gap $d$ corresponds to a voltage $V = Ed$ . Chain them together:

Derivation ACapacitance of a Parallel Plate Capacitor

We want the capacitance of two parallel plates, each of area $A$ , separated by a small air gap $d$ , carrying charges $+Q$ and $-Q$ . The strategy is the standard three-step recipe that works for every capacitor: write the field in the gap using results we already own, convert that field into a potential difference, then divide charge by voltage. The step to watch is the first one — we quote the two-sheet field $\sigma/\varepsilon_0$ from Gauss's Law at Work rather than re-deriving it, and the "small gap" assumption is what lets us treat the field as perfectly uniform, ignoring the slight bulge at the edges.

Step 1. Write the field in the gap. Each plate carries surface charge density $\sigma = Q/A$ . Between two facing sheets of opposite charge, the fields add in the gap and cancel outside, giving a uniform field pointing from the positive plate to the negative plate:

E = \frac{\sigma}{\varepsilon_0} = \frac{Q}{\varepsilon_0 A}

Why not σ/2ε₀? That is the field of one isolated sheet. Here both sheets contribute equally in the gap — the factor-of-two bookkeeping was settled in Gauss's Law at Work, and quoting its result is exactly what it is for.

Step 2. Convert field to voltage. The field is uniform, so the potential drops steadily across the gap, and the total drop is field times distance:

V = Ed = \frac{Qd}{\varepsilon_0 A}

Why field times distance? E is the slope of the potential landscape — a constant slope over a run d falls by Ed, the E = V/d link from Equipotentials & the Field–Potential Link, used in reverse.

Step 3. Divide. By the definition $C = Q/V$ :

C = \frac{Q}{Qd/\varepsilon_0 A} = \frac{\varepsilon_0 A}{d}

Notice that $Q$ cancelled — it had to, because capacitance cannot depend on how much charge you happened to load. Verified: A = 0.04 m² and d = 2.0 mm give C = (8.85×10⁻¹²)(0.04)/(0.002) = 1.77×10⁻¹⁰ F = 177 pF; loading it to 9.0 V stores Q = CV = 1.59 nC, and Q/V returns 177 pF exactly — the numbers reused in the worked example below.

Figure

Anatomy of the parallel-plate capacitor — the two levers of the formula

The result deserves a slow read:

C = \frac{\varepsilon_0 A}{d} \qquad (2)

where: $A$ is the overlapping plate area in $\mathrm{m^2}$ , $d$ the gap in metres, and $\varepsilon_0 = 8.85\times10^{-12}\ \mathrm{C^2/N\,m^2}$ the permittivity of free space. Big plates give more room for charge to spread out, so $C \propto A$ . A thin gap means the negative plate sits close enough to reward every extra charge with strong attraction, so charge piles on easily: $C \propto 1/d$ . This double lever — grow $A$ , shrink $d$ — is the entire engineering history of the capacitor, from Leyden jars to supercapacitors.

The isolated sphere — a capacitor with one plate?

A single charged sphere also has a capacitance, which sounds like it violates the "two conductors" rule until you ask where its field lines end: at infinity, which acts as the far-away second plate. A sphere of radius $R$ carrying charge $Q$ sits at potential $V = kQ/R$ (its surface potential, from Potential of Point Charges & Charge Systems — outside a sphere the charge acts as if concentrated at the centre). So

C = \frac{Q}{V} = \frac{Q}{kQ/R} = \frac{R}{k} = 4\pi\varepsilon_0 R \qquad (3)

where: $R$ is the sphere's radius and $k = 9.0\times10^9\ \mathrm{N\,m^2/C^2}$ . The mental-math form $C = R/k$ is a JEE favourite: a sphere of radius 9.0 cm has $C = 0.09/(9.0\times10^9) = 10\ \mathrm{pF}$ , instantly. A 1 cm lab sphere: about 1.1 pF. And the whole Earth, $R = 6371$ km? $C = 6.371\times10^6/(9.0\times10^9) \approx 7.1\times10^{-4}\ \mathrm{F}$ — about 709 μF. A capacitor the size of your thumb from an electronics shop out-stores the planet, because the planet's "gap" to infinity is the worst geometry imaginable while the component's gap is micrometres. That comparison is the farad's enormity made vivid, and it turns up as an assertion-reason question surprisingly often.

Limiting cases

case	result	meaning
gap $d \to$ smaller	$C$ grows as $1/d$	thinner gaps store more — the touchscreen and supercapacitor lever
area $A \to$ larger	$C$ grows as $A$	more room for charge; series-parallel networks will exploit this
$Q$ doubled	$V$ doubles, $C$ unchanged	capacitance is a property of the device, not the charging
sphere $R \to \infty$	$C \to \infty$	an infinite conductor absorbs charge without its potential rising
plates pulled far apart	$C \to 0$	distant conductors barely notice each other — almost no storage
insulating slab inserted	$C$ multiplied by $\kappa$	preview of Dielectrics & Polarisation: the gap material is the third lever

Pitfall

Worked examples

Board level — the standard plate capacitor (CBSE/ISC pattern)

Two square plates of side 20 cm are separated by a 2.0 mm air gap. (a) Find the capacitance. (b) The capacitor is connected to a 9.0 V battery: find the charge stored and the electric field between the plates. Take $\varepsilon_0 = 8.85\times10^{-12}\ \mathrm{C^2/N\,m^2}$ .

Step 1. Collect the geometry in SI units: $A = (0.20)^2 = 0.040\ \mathrm{m^2}$ , $d = 2.0\times10^{-3}\ \mathrm{m}$ . (Writing the unit conversions explicitly is a mark on most schemes — and skipping them is the classic way to be off by $10^4$ .)

Step 2. Capacitance: $C = \varepsilon_0 A/d = (8.85\times10^{-12})(0.040)/(2.0\times10^{-3}) = 1.77\times10^{-10}\ \mathrm{F}$ . $\boxed{C = 177\ \mathrm{pF}}$ .

Step 3. Charge at 9.0 V: $Q = CV = (1.77\times10^{-10})(9.0) = 1.6\times10^{-9}\ \mathrm{C} \approx 1.6\ \mathrm{nC}$ .

Step 4. Field between the plates — two equivalent routes, quote either: $E = V/d = 9.0/(2.0\times10^{-3}) = 4500\ \mathrm{V/m}$ , or $E = \sigma/\varepsilon_0$ with $\sigma = Q/A$ , which returns the same 4500 V/m. $\boxed{E = 4.5\times10^3\ \mathrm{V/m}}$ , uniform, directed from the positive plate to the negative plate.

Step 5. Sanity check the sizes: hundreds of picofarads, nanocoulombs, kilovolts-per-metre — exactly the scales the unit table promised for a hand-sized air capacitor.

Competitive level — pulling the plates apart, battery attached (JEE-Main/NEET pattern)

The 177 pF capacitor above stays connected to its 9.0 V battery while the plates are slowly pulled apart to double the gap, $d \to 2d$ . What happens to the capacitance, the charge, and the field between the plates?

Step 1. Identify what is held fixed. Battery attached means $V = 9.0$ V is pinned — this single observation drives the entire problem. (Its twin, "battery disconnected so $Q$ is frozen", is the other half of the question bank; deciding which quantity is pinned is always Step 1.)

Step 2. Capacitance is pure geometry: $C = \varepsilon_0 A/d$ , so doubling $d$ halves it. $C' = 88.5\ \mathrm{pF}$ .

Step 3. Charge must obey $Q' = C'V = (88.5\times10^{-12})(9.0) = 0.80\ \mathrm{nC}$ — half the original 1.6 nC. Physically, the battery pulled charge back off the plates as you separated them: capacitors connected to batteries exchange charge with them whenever geometry changes.

Step 4. Field: $E' = V/(2d) = 9.0/(4.0\times10^{-3}) = 2250\ \mathrm{V/m}$ — also halved. Consistency check via the other route: $\sigma' = Q'/A$ halved, so $\sigma'/\varepsilon_0$ halves too. ✓

Step 5. Boxed summary: $\boxed{C, Q, E\ \text{all halve};\ V\ \text{fixed}}$ . Contrast with the disconnected version (which the next paper will ask): there $Q$ freezes, $E = \sigma/\varepsilon_0$ stays constant, and $V = Ed$ doubles. Timing note: 90 seconds as an MCQ. The entire problem is the first decision — battery attached pins V. Train the reflex: read "remains connected"/"is disconnected" before touching any formula, underline it, then let $C = \varepsilon_0 A/d$ plus either $Q = CV$ or $V = Q/C$ do the rest.

Exam traps

Unit conversion in $A$ and $d$ . Plate areas arrive in cm², gaps in mm. $1\ \mathrm{cm^2} = 10^{-4}\ \mathrm{m^2}$ — the square is the killer. A 20 cm side means $A = 0.04\ \mathrm{m^2}$ , not $0.4$ . This single slip produces answers off by factors of 100 to 10,000 and is the most common lost mark in the whole chapter.

Answering in farads without prefixes. Writing $1.77\times10^{-10}$ F is correct but invites transcription errors; boards accept and prefer 177 pF. Know the ladder cold: $\mu = 10^{-6}$ , n $= 10^{-9}$ , p $= 10^{-12}$ .

"Capacitance increases with voltage." See the pitfall box. In assertion-reason format: assertion "a capacitor at 20 V has the same capacitance as at 12 V" is TRUE; reason "capacitance depends only on geometry and the medium" is TRUE and explains it.

Sphere capacitance: mixing $R/k$ and $4\pi\varepsilon_0 R$ . They are the same formula ( $k = 1/4\pi\varepsilon_0$ ). Use $R/k$ with $k = 9.0\times10^9$ for speed; use $4\pi\varepsilon_0 R$ when the paper supplies $\varepsilon_0$ . Do not multiply by both.

Fixed- $V$ vs fixed- $Q$ confusion. "Remains connected to the battery" pins $V$ ; "charged and then disconnected" pins $Q$ . Every geometry-change question hinges on this reading. CBSE and JEE both set paired versions specifically to catch students who memorised one outcome.

Constants per board. CBSE/ISC/NEET supply $\varepsilon_0 = 8.85\times10^{-12}$ and accept $k = 9\times10^9$ ; JEE-Main answer grids assume $k = 9\times10^9$ unless stated; AP Physics 2 and A-Level provide constant sheets — use the sheet's values, not your memorised ones.

Origins · 1 minPieter van Musschenbroek and how this was discovered

Pieter van Musschenbroek

1692–1761 · Leiden

The man who invented the capacitor by accident and reported it with magnificent honesty. In January 1746, Musschenbroek — a physics professor from a famous Leiden family of instrument makers — was trying to "bottle" electricity by charging water in a glass jar held in his hand. When he touched the wire dipping into the water, the world's first capacitor discharged through his body. He wrote to a colleague in Paris that his arm and body were affected "in a terrible manner that I cannot express; I thought it was all over with me," adding that he would not take another shock "for the whole kingdom of France." The Leyden jar — glass as the gap, water and hand as the plates — swept through Europe within months: abbés discharged it through chains of two hundred monks, showmen sold shocks at fairs, and Benjamin Franklin used linked jars to dissect what "charged" really meant. Every capacitor since — ceramic chip, electrolytic can, touchscreen electrode — is Musschenbroek's jar with better manners.

History

1746

— Musschenbroek in Leiden (and, independently, von Kleist in Pomerania) discovers the Leyden jar, the first capacitor, via a memorable shock

1749

— Franklin links Leyden jars into the first "battery" of capacitors, the word borrowed from artillery

1837

— Faraday shows the insulating material between the conductors sets the capacity — the dielectric constant, honoured later in the unit's name

1881

— The farad is adopted internationally as the unit of capacitance, named for Faraday

1957

— The first patent for the electrochemical double-layer capacitor opens the road to farad-scale supercapacitors

Going further

Texts. NCERT Physics Part 1 (Class 12), Ch. 2 §2.11–2.12 — the board-canonical definition and parallel-plate derivation · H.C. Verma, Concepts of Physics 2, Ch. 31 · Halliday, Resnick & Walker, Ch. 25 · For the full UG treatment with spheres, cylinders and the uniqueness theorem: Capacitance (UG deep dive).

Provenance. All numbers recomputed: 1 F at 1 mm gap needs $A = 1.13\times10^8\ \mathrm{m^2}$ (side 10.6 km); $C = 177$ pF for $A = 0.04\ \mathrm{m^2}$ , $d = 2.0$ mm; $Q = 1.59$ nC and $E = 4500$ V/m at 9.0 V; sphere $R = 9$ cm → 10 pF; Earth → $7.1\times10^{-4}$ F; battery-attached gap doubling → 88.5 pF, 0.80 nC, 2250 V/m.

Prerequisites. Electric Potential & Potential Difference · Electrostatics of Conductors & Shielding · Gauss's Law at Work: Wire, Sheet & Shell. Leads to. Capacitors in Series & Parallel · Energy Stored in a Capacitor · Dielectrics & Polarisation.

Is the charge on a capacitor Q or is it zero, since the plates carry +Q and −Q?

Both statements are about different questions. The net charge of the whole device is zero: $+Q$ on one plate, $-Q$ on the other. But "the charge on a capacitor" always means $Q$ , the magnitude on either plate — that is the charge the battery actually shifted from one plate to the other, and the $Q$ in $C = Q/V$ . Never add the plates to get $2Q$ and never cancel them to get zero; both are classic one-mark losses. Think of the battery as a pump moving water from one tank to another: the "amount pumped" is $Q$ .

Why does capacitance increase when the plates come CLOSER? I keep feeling it should be the opposite.

Bring the negative plate closer and every positive charge on the other plate feels the attraction of its opposite partner more strongly — the pile becomes more comfortable, so the same voltage can hold more charge. Alternatively, run the formula chain: for a fixed voltage $V$ , the field must satisfy $E = V/d$ , so a smaller $d$ demands a bigger field, which requires more surface charge ( $E = \sigma/\varepsilon_0$ ). Either way, small gap means big storage. The feeling of "closer = less room" comes from picturing charge stored in the gap volume — but charge sits on the plates, and the gap only sets how strongly the plates talk to each other.

Where exactly does the charge sit on each plate?

On the two facing surfaces. Field lines must leave positive charge and land on negative charge across the gap; charge on the outer faces would produce field with nowhere to go. This has a consequence exams like: making the plates thicker changes nothing, because only the facing surfaces and the gap between them do any work. It also explains why $\sigma = Q/A$ uses the overlap area — charge on non-overlapping plate regions barely participates.

If the farad is so huge, why do shops sell 3000 F supercapacitors for a few hundred rupees?

Because $C = \varepsilon_0 A/d$ has two levers and supercapacitors pull both to absurd extremes. Charge in a supercapacitor sits in an electrochemical double layer about a nanometre thick — a million times thinner than our 1 mm example — over porous carbon electrodes whose internal surface area reaches thousands of square metres per gram. Tiny $d$ , colossal $A$ : the same formula, driven to its limits. The catch is the voltage rating, only about 2.7 V per cell, because the nanometre "gap" breaks down easily — which is why supercapacitor banks appear in series stacks, the subject of the next module.

Does the electric field between the plates depend on the gap distance?

Depends on what is held fixed — the distinction that runs the whole chapter. At fixed charge (battery disconnected), $E = \sigma/\varepsilon_0$ knows only the charge density: pull the plates apart and $E$ stays constant while $V = Ed$ grows. At fixed voltage (battery attached), $E = V/d$ : pull the plates apart and the field thins as charge drains back into the battery. Same device, opposite answers, decided entirely by the circuit connection — reread the competitive worked example with this in mind.

Is the capacitance of an isolated sphere a real thing or just a definition trick?

Completely real — it is a two-conductor capacitance in disguise, with the second "plate" pushed to infinity where the sphere's field lines terminate. Charge the sphere and its potential rises as $V = kQ/R$ ; the ratio $Q/V = R/k$ behaves exactly like any other capacitance. It even matters practically: the dome of a Van de Graaff generator stores charge on precisely this capacitance, and the body of an aircraft accumulating static in flight charges up its own few hundred picofarads. You will cash this formula in the E2 gym's earthing problems.

My textbook writes C = ε₀A/d but also C = Aε₀κ/d. Which is it?

Both — the second is the first with the gap filled by an insulating material of dielectric constant $\kappa$ instead of air. For air or vacuum, $\kappa = 1.0006 \approx 1$ and the formulas coincide. Fill the gap with paper ( $\kappa \approx 3.7$ ) or ceramic ( $\kappa$ in the hundreds) and the capacitance multiplies by $\kappa$ — that is why real capacitors are never air-gapped. Why an insulator boosts storage is a genuinely beautiful piece of physics involving molecular alignment, and it gets its own module: Dielectrics & Polarisation, at the end of this chapter.