Skip to content

The Inner Layer Sets the Limit

Irodov 3.107DifficultyDifficulty 2 of 5UniquenessUniqueness 2 of 5

Problem Irodov 3.107

A cylindrical capacitor has an inner electrode of radius R1R_1 and an outer electrode of inner radius R3R_3. The gap holds two coaxial dielectric layers: permittivity ε1\varepsilon_1 from R1R_1 to R2R_2, and permittivity ε2\varepsilon_2 from R2R_2 to R3R_3. The breakdown fields of the two dielectrics are E1E_1 and E2E_2 respectively. Find the breakdown voltage of the capacitor, given that ε1R1E1<ε2R2E2\varepsilon_1 R_1 E_1 < \varepsilon_2 R_2 E_2.

The double-layer cylindrical capacitor in cross-section about its axis O: the inner electrode of radius R_1, the layer ε_1 out to R_2, the layer ε_2 out to the outer electrode at R_3.

The double-layer cylindrical capacitor in cross-section about its axis OO: the inner electrode of radius R1R_1, the layer ε1\varepsilon_1 out to R2R_2, the layer ε2\varepsilon_2 out to the outer electrode at R3R_3.

Given
  • radii R1<R2<R3R_1 < R_2 < R_3
  • layers ε1\varepsilon_1 (inner), ε2\varepsilon_2 (outer)
  • breakdown fields E1E_1, E2E_2
  • ε1R1E1<ε2R2E2\varepsilon_1 R_1 E_1 < \varepsilon_2 R_2 E_2
Find
the breakdown voltage VV

Solution

Step 1The field in the two layers

Let the inner electrode carry the charge λ\lambda per unit length. Far from the ends, by symmetry D\vect D is radial and the same all round the axis, and Gauss’s theorem on a coaxial cylinder of radius rr and length ℓ\ell gives

D⋅2πrℓ=λℓD\cdot 2\pi r\ell = \lambda\ell

Dividing by ε0ε\varepsilon_0\varepsilon of each layer:

E(r)=λ2πε0ε1r(R1<r<R2)E(r)=λ2πε0ε2r(R2<r<R3)(1) \begin{gathered} E(r) = \frac{\lambda}{2\pi\varepsilon_0\varepsilon_1 r}\quad (R_1 < r < R_2) \\ E(r) = \frac{\lambda}{2\pi\varepsilon_0\varepsilon_2 r}\quad (R_2 < r < R_3) \end{gathered} \tag{1}
The field with the inner electrode positive, drawn for ε_1 = 2, ε_2 = 3: the outer layer carries fewer lines, and the rest end on the negative bound charge at the boundary R_2.

The field with the inner electrode positive, drawn for ε1=2\varepsilon_1 = 2, ε2=3\varepsilon_2 = 3: the outer layer carries fewer lines, and the rest end on the negative bound charge at the boundary R2R_2.

Step 2Which layer breaks down first

In each layer the field falls as 1/r1/r, so its largest value is at the layer’s inner surface. By (1) the two layers reach breakdown at the charges

λ1=2πε0 ε1R1E1(at r=R1)λ2=2πε0 ε2R2E2(at r=R2)\begin{gathered} \lambda_1 = 2\pi\varepsilon_0\,\varepsilon_1 R_1 E_1 \quad (\text{at } r = R_1) \\ \lambda_2 = 2\pi\varepsilon_0\,\varepsilon_2 R_2 E_2 \quad (\text{at } r = R_2) \end{gathered}

The given inequality ε1R1E1<ε2R2E2\varepsilon_1 R_1 E_1 < \varepsilon_2 R_2 E_2 says that λ1<λ2\lambda_1 < \lambda_2. As the voltage rises, λ\lambda reaches λ1\lambda_1 first, so the inner layer breaks down first, at its surface next to the inner electrode, and the capacitor fails there:

λb=2πε0 ε1R1E1(2) \lambda_{\mathrm{b}} = 2\pi\varepsilon_0\,\varepsilon_1 R_1 E_1 \tag{2}
Step 3The voltage at that charge

The voltage is the line integral of (1) across both layers:

V=∫R1R2λ dr2πε0ε1r+∫R2R3λ dr2πε0ε2r=λ2πε0(1ε1ln⁡R2R1+1ε2ln⁡R3R2)\begin{aligned} V &= \int_{R_1}^{R_2}\frac{\lambda\,dr}{2\pi\varepsilon_0\varepsilon_1 r} + \int_{R_2}^{R_3}\frac{\lambda\,dr}{2\pi\varepsilon_0\varepsilon_2 r}\\ &= \frac{\lambda}{2\pi\varepsilon_0}\left(\frac{1}{\varepsilon_1}\ln\frac{R_2}{R_1} + \frac{1}{\varepsilon_2}\ln\frac{R_3}{R_2}\right) \end{aligned}

Setting λ\lambda to (2), the factor 2πε02\pi\varepsilon_0 cancels and ε1\varepsilon_1 multiplies through the bracket:

V=R1E1[ln⁡R2R1+ε1ε2ln⁡R3R2]V = R_1 E_1\left[\ln\dfrac{R_2}{R_1} + \dfrac{\varepsilon_1}{\varepsilon_2}\ln\dfrac{R_3}{R_2}\right]
(3)
Check

Dimensions: R1E1R_1 E_1 is a length times a field, a voltage, and the bracket is a pure number. Single layer: with ε2=ε1\varepsilon_2 = \varepsilon_1 the bracket becomes ln⁡(R3/R1)\ln(R_3/R_1) and V=R1E1ln⁡(R3/R1)V = R_1 E_1\ln(R_3/R_1), the breakdown voltage of a uniformly filled cylindrical capacitor whose peak field V/(R1ln⁡(R3/R1))V/(R_1\ln(R_3/R_1)) sits at the inner electrode. And with R3=R2R_3 = R_2 (no outer layer) the result reduces to R1E1ln⁡(R2/R1)R_1E_1\ln(R_2/R_1) in the same way.

Remark

Only the inner layer’s strength E1E_1 enters the answer; E2E_2 matters only through the inequality that picks which layer fails. A higher permittivity ε2\varepsilon_2 outside lowers VV, since the outer layer then takes a smaller share of the voltage and more of it falls on the inner layer, which is already the weak one. When the inequality becomes an equality, both layers fail at the same voltage and neither is stronger than it needs to be.