Paper 4, Section II, D

Electrodynamics | Part II, 2017

A dielectric material has a real, frequency-independent relative permittivity ϵr\epsilon_{r} with ∣ϵr−1∣≪1\left|\epsilon_{r}-1\right| \ll 1. In this case, the macroscopic polarization that develops when the dielectric is placed in an external electric field Eext (t,x)\mathbf{E}_{\text {ext }}(t, \mathbf{x}) is P(t,x)≈ϵ0(ϵr−1)Eext (t,x)\mathbf{P}(t, \mathbf{x}) \approx \epsilon_{0}\left(\epsilon_{r}-1\right) \mathbf{E}_{\text {ext }}(t, \mathbf{x}). Explain briefly why the associated bound current density is

Jbound (t,x)≈ϵ0(ϵr−1)∂Eext (t,x)∂t\mathbf{J}_{\text {bound }}(t, \mathbf{x}) \approx \epsilon_{0}\left(\epsilon_{r}-1\right) \frac{\partial \mathbf{E}_{\text {ext }}(t, \mathbf{x})}{\partial t}

[You should ignore any magnetic response of the dielectric.]

A sphere of such a dielectric, with radius aa, is centred on x=0\mathbf{x}=0. The sphere scatters an incident plane electromagnetic wave with electric field

E(t,x)=E0ei(k⋅x−ωt)\mathbf{E}(t, \mathbf{x})=\mathbf{E}_{0} e^{i(\mathbf{k} \cdot \mathbf{x}-\omega t)}

where ω=c∣k∣\omega=c|\mathbf{k}| and E0\mathbf{E}_{0} is a constant vector. Working in the Lorenz gauge, show that at large distances r=∣x∣r=|\mathbf{x}|, for which both r≫ar \gg a and ka2/r≪2πk a^{2} / r \ll 2 \pi, the magnetic vector potential Ascatt (t,x)\mathbf{A}_{\text {scatt }}(t, \mathbf{x}) of the scattered radiation is

Ascatt(t,x)≈−iωE0ei(kr−ωt)r(ϵr−1)4πc2∫∣x′∣⩽aeiq⋅x′d3x′\mathbf{A}_{\mathrm{scatt}}(t, \mathbf{x}) \approx-i \omega \mathbf{E}_{0} \frac{e^{i(k r-\omega t)}}{r} \frac{\left(\epsilon_{r}-1\right)}{4 \pi c^{2}} \int_{\left|\mathbf{x}^{\prime}\right| \leqslant a} e^{i \mathbf{q} \cdot \mathbf{x}^{\prime}} d^{3} \mathbf{x}^{\prime}

where q=k−kx^\mathbf{q}=\mathbf{k}-k \hat{\mathbf{x}} with x^=x/r\hat{\mathbf{x}}=\mathbf{x} / r.

In the far-field, where kr≫1k r \gg 1, the electric and magnetic fields of the scattered radiation are given by

Escatt (t,x)≈−iωx^×[x^×Ascatt (t,x)]Bscatt (t,x)≈ikx^×Ascatt (t,x)\begin{aligned} &\mathbf{E}_{\text {scatt }}(t, \mathbf{x}) \approx-i \omega \hat{\mathbf{x}} \times\left[\hat{\mathbf{x}} \times \mathbf{A}_{\text {scatt }}(t, \mathbf{x})\right] \\ &\mathbf{B}_{\text {scatt }}(t, \mathbf{x}) \approx i k \hat{\mathbf{x}} \times \mathbf{A}_{\text {scatt }}(t, \mathbf{x}) \end{aligned}

By calculating the Poynting vector of the scattered and incident radiation, show that the ratio of the time-averaged power scattered per unit solid angle to the time-averaged incident power per unit area (i.e. the differential cross-section) is

dσdΩ=(ϵr−1)2k4(sin⁡(qa)−qacos⁡(qa)q3)2∣x^×E^0∣2\frac{d \sigma}{d \Omega}=\left(\epsilon_{r}-1\right)^{2} k^{4}\left(\frac{\sin (q a)-q a \cos (q a)}{q^{3}}\right)^{2}\left|\hat{\mathbf{x}} \times \hat{\mathbf{E}}_{0}\right|^{2}

where E^0=E0/∣E0∣\hat{\mathbf{E}}_{0}=\mathbf{E}_{0} /\left|\mathbf{E}_{0}\right| and q=∣q∣q=|\mathbf{q}|.

[You may assume that, in the Lorenz gauge, the retarded potential due to a localised current distribution is

A(t,x)=μ04π∫J(tret,x′)∣x−x′∣d3x′,\mathbf{A}(t, \mathbf{x})=\frac{\mu_{0}}{4 \pi} \int \frac{\mathbf{J}\left(t_{\mathrm{ret}}, \mathbf{x}^{\prime}\right)}{\left|\mathbf{x}-\mathbf{x}^{\prime}\right|} d^{3} \mathbf{x}^{\prime},

where the retarded time tret =t−∣x−x′∣/c.]\left.t_{\text {ret }}=t-\left|\mathbf{x}-\mathbf{x}^{\prime}\right| / c .\right]

Typos? Please submit corrections to this page on GitHub.