physics2u
Depth
PU-204 / L19 · lecture

Jefimenko's Equations

~6 min read · Tier toggle changes depth (T1 plain · T2 standard · T3 rigour)

BEFORE YOU START

You must be comfortable with the Lorenz gauge condition and the definition of retarded time \( t_r = t - R/c \). Familiarity with the vector potential \( \mathbf{A} \) and scalar potential \( \varphi \) as tools to generate \( \mathbf{E} \) and \( \mathbf{B} \) is assumed.

Why this matters

Jefimenko's equations provide the exact, causal solution to Maxwell's equations for arbitrary source distributions.

While Coulomb's law and the Biot-Savart law are static approximations, Jefimenko's equations describe how changes in charge and current propagate through space at the speed of light. They are the dynamic generalisation of the static laws, explicitly showing the dependence on the time derivatives of the sources. This formulation is crucial for understanding radiation, as it isolates the "velocity" and "acceleration" terms that give rise to electromagnetic waves, bridging the gap between static field theory and the relativistic field theory explored later in this unit.

The Causal Structure of Fields

Imagine a charge distribution that suddenly changes. In a static world, the field would update instantly everywhere. In reality, information travels at a finite speed, \( c \). Jefimenko's equations encode this causality by evaluating the sources not at the current time \( t \), but at the retarded time \( t_r \). This means the field at a point \( \mathbf{r} \) depends on what the sources were doing at an earlier moment, specifically when the light signal from the source reached the observation point. The equations separate the field into two conceptual parts: a "near field" term that looks like the static Coulomb or Biot-Savart law but evaluated at the retarded time, and a "radiation field" term that depends on the time derivatives of the sources. This distinction is fundamental to understanding how energy is carried away from accelerating charges.

Deriving the Electric Field

The electric field is derived from the potentials via \( \mathbf{E} = -\nabla\varphi - \partial_t \mathbf{A} \). When we substitute the retarded potentials and perform the differentiation, we must carefully account for the time dependence of the retarded time \( t_r \). The gradient \( \nabla \) acts on both the spatial separation \( R \) and the retarded time \( t_r \). This leads to a term involving \( \dot{\rho} \) (the time derivative of charge density) and a term involving \( \dot{\mathbf{J}} \) (the time derivative of current density).

\[ \mathbf{E}(\mathbf{r},t) = \frac{1}{4\pi\varepsilon_0} \int \left[ \frac{\rho(\mathbf{r}',t_r)}{R^2}\,\hat{\mathbf{R}} + \frac{\dot{\rho}(\mathbf{r}',t_r)}{cR}\,\hat{\mathbf{R}} - \frac{\dot{\mathbf{J}}(\mathbf{r}',t_r)}{c^2 R} \right] d^3r' \]

The first term is the familiar Coulomb-like contribution. The second term, proportional to \( \dot{\rho}/cR \), arises from the time variation of the charge density. The third term, proportional to \( \dot{\mathbf{J}}/c^2 R \), is the crucial radiation term. Note that the radiation terms fall off as \( 1/R \), whereas the near-field terms fall off as \( 1/R^2 \). This difference in scaling is why radiation dominates at large distances.

Deriving the Magnetic Field

The magnetic field is derived from the vector potential via \( \mathbf{B} = \nabla\times\mathbf{A} \). Similar to the electric field, the curl operation introduces dependencies on the retarded time. The resulting expression mirrors the structure of the electric field but involves the current density \( \mathbf{J} \) and its time derivative \( \dot{\mathbf{J}} \).

\[ \mathbf{B}(\mathbf{r},t) = \frac{\mu_0}{4\pi} \int \left[ \frac{\mathbf{J}(\mathbf{r}',t_r)}{R^2} + \frac{\dot{\mathbf{J}}(\mathbf{r}',t_r)}{cR} \right] \times \hat{\mathbf{R}} \, d^3r' \]

Here, the first term is the Biot-Savart-like near field. The second term is the magnetic radiation field. Notice the cross product with \( \hat{\mathbf{R}} \), which ensures that \( \mathbf{B} \) is perpendicular to the direction of propagation. The symmetry between the \( \dot{\rho} \) term in \( \mathbf{E} \) and the \( \dot{\mathbf{J}} \) term in \( \mathbf{B} \) reflects the underlying consistency of Maxwell's equations and the conservation of charge.

Connection to Conservation Laws

Jefimenko's equations are not just mathematical solutions; they embody the physical principle of local conservation of charge. The appearance of \( \dot{\rho} \) and \( \dot{\mathbf{J}} \) is a direct consequence of the continuity equation \( \nabla \cdot \mathbf{J} + \partial_t \rho = 0 \). If charge were not conserved, the terms involving time derivatives would not combine in this specific way, and the fields would not satisfy the homogeneous Maxwell equations (\( \nabla \cdot \mathbf{B} = 0 \) and \( \nabla \times \mathbf{E} = -\partial_t \mathbf{B} \)). Thus, the structure of Jefimenko's equations is a direct manifestation of the interplay between source dynamics and field propagation.

The derivation

◆ DERIVATION EMBED · D-168 · not re-derived here

Retarded Potentials and Jefimenko's Equations

Starting from the inhomogeneous wave equations that Maxwell's equations impose on the potentials in Lorenz gauge, \( \Box \varphi = -\rho/\varepsilon_0 \) and \( \Box \mathbf{A} = -\mu_0 \mathbf{J} \) with \( \Box \equiv \nabla^2 - \frac{1}{c^2}\frac{\partial^2}{\partial t^2} \), we construct the retarded Green's function of the d'Alembertian and obtain the retarded potentials — each source element contributing at the retarded time \( t_r = t - |\mathbf{r}-\mathbf{r}'|/c \). Differentiating these potentials, \( \mathbf{E} = -\nabla\varphi - \partial_t \mathbf{A} \) and \( \mathbf{B} = \nabla\times\mathbf{A} \), yields Jefimenko's equations: the exact, manifestly causal fields of an arbitrary prescribed charge and current distribution in vacuum.

\[ \varphi(\mathbf{r},t) = \frac{1}{4\pi\varepsilon_0} \int \frac{\rho(\mathbf{r}',t_r)}{R}\, d^3r', \qquad \mathbf{A}(\mathbf{r},t) = \frac{\mu_0}{4\pi} \int \frac{\mathbf{J}(\mathbf{r}',t_r)}{R}\, d^3r' \] \[ \mathbf{E}(\mathbf{r},t) = \frac{1}{4\pi\varepsilon_0} \int \left[ \frac{\rho(\mathbf{r}',t_r)}{R^2}\,\hat{\mathbf{R}} + \frac{\dot{\rho}(\mathbf{r}',t_r)}{cR}\,\hat{\mathbf{R}} - \frac{\dot{\mathbf{J}}(\mathbf{r}',t_r)}{c^2 R} \right] d^3r' \] \[ \mathbf{B}(\mathbf{r},t) = \frac{\mu_0}{4\pi} \int \left[ \frac{\mathbf{J}(\mathbf{r}',t_r)}{R^2} + \frac{\dot{\mathbf{J}}(\mathbf{r}',t_r)}{cR} \right] \times \hat{\mathbf{R}}\; d^3r', \qquad t_r = t - \frac{R}{c} \]
Open full derivation →
Tier 3 · watch out

A common misconception is that the \( 1/R \) radiation terms are always dominant. In reality, for non-relativistic, quasi-static sources, the \( 1/R^2 \) near-field terms often dominate at distances not vastly larger than the source size. The radiation terms only become significant when the observation distance \( R \) is much larger than the wavelength of the radiation, or when the source acceleration is high. Confusing the near-field and radiation-field regimes leads to incorrect predictions of energy flow and field strength. C

Worked example

Problem. Why does the Biot–Savart law work so well for slowly varying currents? Expand Jefimenko's magnetic field for a current density varying as \(\mathbf{J} \propto \cos\omega t\), to second order in the delay \(\tau = R/c\), and estimate the relative error of using Biot–Savart (with the present-time current) for 50 Hz mains wiring seen from \(R = 10\ \mathrm{m}\).

  1. Taylor-expand about the present time \(t\): \(\mathbf{J}(t_r) \approx \mathbf{J} - \tau\dot{\mathbf{J}} + \tfrac12\tau^2\ddot{\mathbf{J}}\) and \(\dot{\mathbf{J}}(t_r) \approx \dot{\mathbf{J}} - \tau\ddot{\mathbf{J}}\), all evaluated at \(t\).
  2. Combine the two terms in the bracket: \(\dfrac{\mathbf{J}(t_r)}{R^2} + \dfrac{\dot{\mathbf{J}}(t_r)}{cR} \approx \dfrac{\mathbf{J}}{R^2} - \dfrac{\dot{\mathbf{J}}}{cR} + \dfrac{\ddot{\mathbf{J}}}{2c^2} + \dfrac{\dot{\mathbf{J}}}{cR} - \dfrac{\ddot{\mathbf{J}}}{c^2} = \dfrac{\mathbf{J}}{R^2} - \dfrac{\ddot{\mathbf{J}}}{2c^2}\).
  3. The first-order retardation corrections cancel exactly: the delay in the Biot–Savart term is compensated by the \(\dot{\mathbf{J}}\) term. Only a second-order term survives.
  4. For harmonic time dependence \(|\ddot{\mathbf{J}}| = \omega^2 |\mathbf{J}|\), so the relative error is \(\dfrac{\omega^2 R^2}{2c^2} = \tfrac12\left(\dfrac{\omega R}{c}\right)^2\).
  5. At 50 Hz and 10 m: \(\omega R/c = 2\pi \times 50 \times 10/(3.00 \times 10^{8}) = 1.05 \times 10^{-5}\), so the relative error is \(5.5 \times 10^{-11}\).
\[ \frac{\Delta B}{B} \approx \frac12\left(\frac{\omega R}{c}\right)^2 \approx 5.5 \times 10^{-11} \]

Units check. \(\ddot{\mathbf{J}}/c^2\) has units \(\mathrm{A\,m^{-2}\,s^{-2}}/(\mathrm{m^2\,s^{-2}}) = \mathrm{A\,m^{-4}}\), the same as \(\mathbf{J}/R^2\), so the two terms can be compared directly; \(\omega R/c\) is dimensionless.

Biot–Savart remains excellent for any circuit much smaller than the wavelength (\(6000\ \mathrm{km}\) at 50 Hz); it fails only when \(\omega R/c\) approaches 1.

CHECK YOURSELF

Three quick tests

  1. What is the physical significance of the retarded time \( t_r \) in Jefimenko's equations?
    AnswerIt ensures causality, meaning the field at \( \mathbf{r} \) depends on the source state at the time the light signal left the source, not the current time.
  2. How do the radiation terms in \( \mathbf{E} \) and \( \mathbf{B} \) scale with distance \( R \)?
    AnswerThey scale as \( 1/R \), unlike the near-field terms which scale as \( 1/R^2 \).
  3. Why do the time derivatives \( \dot{\rho} \) and \( \dot{\mathbf{J}} \) appear in the equations?
    AnswerThey arise from differentiating the retarded potentials with respect to time and space, reflecting the dynamic nature of the sources and the finite propagation speed of changes in the field.
PROBLEMS

Problem set for PU-204 →

Each derivation in this unit ends with five graded problems with full solutions.

NOTES

Lecture notes checked line by line by the editor; every number in the worked example was recomputed.

Derivations cited: D-168 · Threads: force, energy, light