Back

Capacitors

Capacitors are devices that store electrical energy. The charge is proportional to the voltage between them:

Q=CU,C=QU. \begin{align*} Q &= C|U|, \\ C &= \frac{Q}{|U|}. \end{align*}

The unit of capacitance is farad (FF).

Consider two infinite parallel planes at a distance dd with opposite charges of equal magnitude QQ:

Two parallel planes

The electric field between the planes as derived in Gauss's law chapter is equal to:

E=σϵ0z^,E=σϵ0. \begin{align*} \boldsymbol{E} &= -\frac{\sigma}{\epsilon_0}\boldsymbol{\hat{z}}, \\ E &= \frac{\sigma}{\epsilon_0}. \end{align*}

The voltage is equal to:

U=CEds=Ed,U=Ed. \begin{align*} U &= - \int_C \boldsymbol{E} \cdot d\boldsymbol{s} \\ &= -Ed, \\ |U| &= Ed. \end{align*}

The capacitance is equal to:

C=QU=QEd=Qϵ0σd=Aϵ0d, \begin{align*} C &= \frac{Q}{|U|} \\ &= \frac{Q}{Ed} \\ &= \frac{Q \epsilon_0}{\sigma d} \\ &= \frac{A \epsilon_0}{d}, \end{align*}

where AA is the area of each of the planes.

Cylindrical capacitor consists of a cylinder of radius aa and a cylindrical shell of radius bb of equal length LL. The inner cylinder has a positive charge QQ and the outer shell has a negative charge Q-Q:

Cylindrical conductor illustration

We choose cylinder with radius rr and equal length LL as the gaussian surface:

Cylindrical conductor illustration 2

The electric field is constant on the cylinder. Using Gauss's law to get the electric field:

ΦE=SEdA=EA=2EπrL=qϵ0,E=q2ϵ0πrL, \begin{align*} \Phi_E &= \oiint_S \boldsymbol{E} \cdot d\boldsymbol{A} \\ &= E A \\ &= 2 E \pi r L = \frac{q}{\epsilon_0}, \\ E &= \frac{q}{2 \epsilon_0 \pi r L}, \end{align*}

where qq is the charge enclosed by the gaussian surface (black cylinder). If r<ar < a then q=0q = 0. If r>br > b then q = Q + (Q)=0q~=~Q~+~(-Q) = 0. This implies a<r<ba < r < b and thus q=Qq = Q. The voltage is equal to:

U=CEds=Q2ϵ0πLab1rdr=Q2ϵ0πL[lnr]ab=Q2ϵ0πLlnba=U. \begin{align*} U &= \int_C \boldsymbol{E} \cdot d\boldsymbol{s} \\ &= \frac{Q}{2 \epsilon_0 \pi L} \int_a^b \frac{1}{r} dr \\ &= \frac{Q}{2 \epsilon_0 \pi L} \left[\ln r\right]_a^b \\ &= \frac{Q}{2 \epsilon_0 \pi L} \ln \frac{b}{a} = |U|. \end{align*}

The capacitance is equal to:

C=QU=2ϵ0πLlnba \begin{align*} C &= \frac{Q}{|U|} \\ &= \frac{2 \epsilon_0 \pi L}{\ln \frac{b}{a}} \end{align*}

Spherical capacitor consists of two cocentric spheres. The inner sphere has radius aa and a positive charge QQ. The outer sphere has a negative charge Q-Q and a radius bb. We choose a cocentric sphere of radius rr as our gaussian surface:

Spherical cylinder illustration

The electric field is constant on the surface of the gaussian surface. Using Gauss's law, the electric field is equal to:

ϕE=SEdA=EA=4Eπr2=qϵ0,E=q4ϵ0πr2, \begin{align*} \phi_E &= \oiint_S \boldsymbol{E} \cdot d \boldsymbol{A} \\ &= EA \\ &= 4 E \pi r^2 \\ &= \frac{q}{\epsilon_0}, \\ E &= \frac{q}{4 \epsilon_0 \pi r^2}, \end{align*}

where qq is the charged enclosed by the gaussian surface. Similarly to the cylindrical capacitor, the charge is nonzero only when a<r<ba < r < b, thus q=Qq = Q. We can now calculate the voltage:

U=CEds=Q4ϵ0πab1r2dr=Q4ϵ0π[1r]ba=Q4ϵ0π(1a1b)=Q(ba)4ϵ0πab=U. \begin{align*} U &= \int_C \boldsymbol{E} \cdot d\boldsymbol{s} \\ &= \frac{Q}{4 \epsilon_0 \pi} \int_a^b \frac{1}{r^2} dr \\ &= \frac{Q}{4 \epsilon_0 \pi} \left[\frac{1}{r}\right]_b^a \\ &= \frac{Q}{4 \epsilon_0 \pi} \left(\frac{1}{a} - \frac{1}{b}\right) \\ &= \frac{Q (b - a)}{4 \epsilon_0 \pi ab} = |U|. \end{align*}

The capacitance is equal to:

C=QU=4ϵ0πabba. \begin{align*} C &= \frac{Q}{|U|} \\ &= \frac{4 \epsilon_0 \pi ab}{b - a}. \end{align*}

An isolated capacitor is similar to a setup where the negative capacitor is placed at infinity:

C=limb4ϵ0πabba=4ϵ0πlimba1ab=4ϵ0πa. \begin{align*} C &= \lim_{b \to \infty} \frac{4 \epsilon_0 \pi ab}{b - a} \\ &= 4 \epsilon_0 \pi \lim_{b \to \infty} \frac{a}{1 - \frac{a}{b}} \\ &= 4 \epsilon_0 \pi a. \end{align*}

Capacitors can be used to store energy. Consider a capacitor with initial charge equal to zero. The potential energy is the external work done to charge the capacitor is given by:

Ep=Wext=0QUdq=0QqCdq=Q22C=12CU2, \begin{align*} E_p &= W_{ext} = \int_0^Q |U| dq \\ &= \int_0^Q \frac{q}{C} dq \\ &= \frac{Q^2}{2 C} \\ &= \frac{1}{2} C |U|^2, \end{align*}

where QQ charge after the charging process and CC is the capacitance.

Consider a parallel plate capacitor, where C=Aϵ0/dC = A \epsilon_0/d and U=Ed|U| = Ed then the potential energy is equal to:

Ep=12Aϵ0dE2d2=12ϵ0E2Ad=12ϵ0E2V, \begin{align*} E_p &= \frac{1}{2} \frac{A \epsilon_0}{d} E^2 d^2 \\ &= \frac{1}{2} \epsilon_0 E^2 Ad \\ &= \frac{1}{2} \epsilon_0 E^2 V, \end{align*}

where VV is the volume.

The energy density is the energy per unit volume:

ep=12ϵ0E2e_p = \frac{1}{2} \epsilon_0 E^2

For example, consider a spherical conductor of radius aa. As derived with Gauss's law, the electric field outside is equal to:

E=a2σr2ϵ0=4πrc2σ4πr2ϵ0=Q4πr2ϵ0. \begin{align*} E &= \frac{a^2 \sigma}{r^2 \epsilon_0} \\ &= \frac{4 \pi r_c^2 \sigma}{4 \pi r^2 \epsilon_0} \\ &= \frac{Q}{4 \pi r^2 \epsilon_0}. \end{align*}

Then the energy density is equal to:

ep=Q232π2r4ϵ0.e_p = \frac{Q^2}{32 \pi^2 r^4 \epsilon_0}.

The potential energy is equal to:

Ep=Vep dV=02π0πaQ232π2r4ϵ0r2sinθ dr dθ dϕ=Q232π2ϵ002πdϕ0πsinθ dθ a1r2dr=Q232π2ϵ04π[1r]a=Q28πϵ0a. \begin{align*} E_p &= \iiint_V e_p\ dV \\ &= \int_0^{2\pi} \int_0^{\pi} \int_a^{\infty} \frac{Q^2}{32 \pi^2 r^4 \epsilon_0} r^2 \sin \theta\ dr\ d\theta\ d\phi \\ &= \frac{Q^2}{32 \pi^2 \epsilon_0} \int_0^{2\pi} d\phi \int_0^{\pi} \sin \theta\ d\theta\ \int_{\infty}^a -\frac{1}{r^2} dr \\ &= \frac{Q^2}{32 \pi^2 \epsilon_0} 4 \pi \left[\frac{1}{r}\right]_{\infty}^a \\ &= \frac{Q^2}{8 \pi \epsilon_0 a}. \end{align*}

The electric potential at the shell is φ=Q4πϵ0a\varphi = \frac{Q}{4 \pi \epsilon_0 a}. The potential energy can be simplified to:

Ep=12Qφ.E_p = \frac{1}{2} Q \varphi.

We can arrive at the same potential energy by calculating the external work required to charge the capacitor:

Ep=Wext=0Qφ dq=0Qq4πϵ0a dq=Q28πϵ0a. \begin{align*} E_p &= W_{ext} = \int_0^Q \varphi\ dq \\ &= \int_0^Q \frac{q}{4 \pi \epsilon_0 a}\ dq \\ &= \frac{Q^2}{8 \pi \epsilon_0 a}. \end{align*}

There are two ways to connect capacitors:

Parallel capacitors illustrationSerial capacitors illustration

The first is parallel and the second is serial. For parallel, the voltage across each capacitor is the same - U|U|, each capacitor has the capacitance CiC_i and charge QiQ_i:

Parallel capacitors illustration
Ci=QiU    Qi=UCi.C_i = \frac{Q_i}{|U|} \iff Q_i = |U|C_i.

The capacitors can be replaced by a single capacitor with capacitance CC and charge QQ:

Q=iQi=iUCi=UiCi,C=QU=iCi. \begin{align*} Q &= \sum_i Q_i = \sum_i |U|C_i = |U| \sum_i C_i, \\ C &= \frac{Q}{|U|} = \sum_i C_i. \end{align*}

For serial, capacitors with capacitance CiC_i are connected to a battery with total voltage U|U| each capacitor has voltage Ui|U_i|:

Parallel capacitors illustration

The outer plates have charges ±Q\pm Q. The inner plates are charged to opposite charge. The voltage of each capacitor Ui=QCi|U_i| = \frac{Q}{C_i} and the total voltage is given by:

U=iUi=iQCi.|U| = \sum_i |U_i| = \sum_i \frac{Q}{C_i}.

Again, the capacitors can be replaced with capacitor with capacitance CC:

U=QC=iQCi,1C=i1Ci. \begin{align*} |U| = \frac{Q}{C} = \sum_i \frac{Q}{C_i}, \\ \frac{1}{C} = \sum_i \frac{1}{C_i}. \end{align*}

Parallel connection

C=0 FC = 0\ F