Skip to content

Radiation from Moving Charges and Dipoles, Retarded Potentials

CSIR NET radiation: retarded and Liénard–Wiechert potentials, Larmor formula, synchrotron loss, dipole antennas, scattering and Cherenkov radiation.

  • 19 min read
  • Updated
On this page · 18 sections
  1. Prerequisites
  2. The Hook: The Atom That Should Collapse
  3. Retarded Potentials
  4. Liénard–Wiechert Potentials of a Point Charge
  5. Larmor Formula (Non-Relativistic)
  6. Liénard Formula (Relativistic)
  7. Electric Dipole Radiation
  8. Magnetic Dipole Radiation
  9. Antennas and Radiation Resistance
  10. Scattering: Thomson and Rayleigh
  11. Cherenkov Radiation
  12. Memory Tricks
  13. Don’t Confuse
  14. Common Mistakes
  15. Practice
  16. Quick Revision Table
  17. Key Takeaways
  18. What’s Next

Prerequisites

The Hook: The Atom That Should Collapse

Retarded Potentials

In the Lorenz gauge each potential obeys an inhomogeneous wave equation, \( \Box V = \rho/\varepsilon_0 \) and \( \Box\mathbf A = \mu_0\mathbf J \). The Green’s function that respects causality is

$$G(\mathbf r, t; \mathbf r', t') = \frac{\delta\left(t - t' - |\mathbf r - \mathbf r'|/c\right)}{4\pi|\mathbf r - \mathbf r'|}$$

an expanding spherical shell. Integrating over the sources gives the retarded potentials:

$$\boxed{V(\mathbf r, t) = \frac{1}{4\pi\varepsilon_0}\int\frac{\rho(\mathbf r', t_r)}{\mathscr R}\,d\tau', \qquad \mathbf A(\mathbf r, t) = \frac{\mu_0}{4\pi}\int\frac{\mathbf J(\mathbf r', t_r)}{\mathscr R}\,d\tau', \qquad t_r = t - \frac{\mathscr R}{c}}$$

with \( \mathscr R = |\mathbf r - \mathbf r'| \). Points to hold on to:

  • The signal from each source element left it at the retarded time \( t_r \) and travels at \( c \).
  • The advanced solution (\( t_a = t + \mathscr R/c \)) also solves the equations; it is discarded on causal grounds.
  • The retarded potentials automatically satisfy the Lorenz condition, by the continuity equation.
  • The fields are not the static Coulomb and Biot–Savart fields evaluated at \( t_r \). Differentiating \( t_r \) inside the integrals produces extra terms (Jefimenko’s equations) involving \( \dot\rho \), \( \dot{\mathbf J} \) and \( 1/\mathscr R \): these are the radiation terms.

Liénard–Wiechert Potentials of a Point Charge

For a point charge on a trajectory \( \mathbf w(t) \), the retarded time is defined implicitly by \( |\mathbf r - \mathbf w(t_r)| = c(t - t_r) \); exactly one such \( t_r \) exists for a charge moving slower than light. Doing the integral carefully (a point charge moving towards the observer “appears” larger because the field point sweeps across it as the light-cone condition is applied) gives

$$\boxed{V = \frac{1}{4\pi\varepsilon_0}\frac{q}{\mathscr R(1 - \hat{\mathbf n}\cdot\boldsymbol\beta)}\bigg|_{t_r}, \qquad \mathbf A = \frac{\mathbf v}{c^2}V}$$

where \( \hat{\mathbf n} \) is the unit vector from the retarded position to the field point and \( \boldsymbol\beta = \mathbf v/c \) at \( t_r \). The factor \( \kappa = 1 - \hat{\mathbf n}\cdot\boldsymbol\beta \) is a purely geometric (Doppler-like) effect, present even at low speeds; it is not a relativistic contraction.

The fields

$$\mathbf E = \frac{q}{4\pi\varepsilon_0}\left[\underbrace{\frac{(\hat{\mathbf n} - \boldsymbol\beta)(1 - \beta^2)}{\kappa^3\mathscr R^2}}_{\text{velocity field}} + \underbrace{\frac{\hat{\mathbf n}\times\left[(\hat{\mathbf n} - \boldsymbol\beta)\times\dot{\boldsymbol\beta}\right]}{c\,\kappa^3\mathscr R}}_{\text{acceleration (radiation) field}}\right]_{t_r}, \qquad \mathbf B = \frac{\hat{\mathbf n}\times\mathbf E}{c}$$
PartFalls asDepends onRole
Velocity field\( 1/\mathscr R^2 \)\( \mathbf v \) onlyboosted Coulomb field (previous page); carries no energy to infinity
Acceleration field\( 1/\mathscr R \)\( \dot{\mathbf v} \)transverse to \( \hat{\mathbf n} \), \( E = cB \); Poynting flux \( \propto 1/\mathscr R^2 \), so finite power crosses an arbitrarily large sphere

Only acceleration radiates. A charge in uniform motion has only the velocity field.

graph TD
    A["Sources ρ, J"] --> B["Lorenz gauge wave equations
□V = ρ/ε₀, □A = μ₀J"] B --> C["Retarded Green's function
δ(t − t' − R/c)/4πR"] C --> D["Retarded potentials"] D --> E["Point charge:
Liénard–Wiechert"] D --> F["Oscillating dipole:
multipole expansion"] E --> G["Velocity field ∝ 1/R²"] E --> H["Acceleration field ∝ 1/R
Larmor, Liénard"] F --> I["Electric dipole: P ∝ p₀²ω⁴
Magnetic dipole: P ∝ m₀²ω⁴/c²"]

Larmor Formula (Non-Relativistic)

For \( \beta \ll 1 \), the radiation field reduces to \( \mathbf E_{\text{rad}} = \dfrac{q}{4\pi\varepsilon_0c^2\mathscr R}\hat{\mathbf n}\times(\hat{\mathbf n}\times\mathbf a) \), magnitude \( \propto a\sin\theta \) where \( \theta \) is measured from \( \mathbf a \). The Poynting vector gives

$$\frac{dP}{d\Omega} = \frac{q^2a^2}{16\pi^2\varepsilon_0c^3}\sin^2\theta, \qquad \boxed{P = \frac{q^2a^2}{6\pi\varepsilon_0c^3} = \frac{\mu_0q^2a^2}{6\pi c}}$$
  • Pattern: a doughnut, zero along \( \mathbf a \), maximum perpendicular to it.
  • Polarisation: \( \mathbf E_{\text{rad}} \) lies in the plane containing \( \hat{\mathbf n} \) and \( \mathbf a \).
  • Scaling: for a charge driven by a given external field, \( a = qE/m \), so \( P \propto q^4E^2/m^2 \). A proton radiates \( (m_p/m_e)^2 \approx 3.4\times10^6 \) times less than an electron in the same field.
  • Numerically, for an electron \( P = 5.71\times10^{-54}\,a^2 \) W (with \( a \) in m/s\( ^2 \)).

Radiation reaction

Energy balance requires a recoil force on the charge. Averaged over periodic motion it is the Abraham–Lorentz force

$$\mathbf F_{\text{rad}} = \frac{\mu_0q^2}{6\pi c}\dot{\mathbf a} = m\tau\dot{\mathbf a}, \qquad \tau = \frac{\mu_0q^2}{6\pi mc} \approx 6.3\times10^{-24}\ \text{s (electron)}$$

The time \( \tau \) is about the time light takes to cross the classical electron radius. The force is small unless the motion changes on that time scale; taken literally it has unphysical “runaway” solutions, a known defect of the classical point-charge theory.

Liénard Formula (Relativistic)

The Larmor result is the rest-frame limit of a Lorentz-invariant expression. For arbitrary velocity:

$$\boxed{P = \frac{q^2\gamma^6}{6\pi\varepsilon_0c^3}\left[a^2 - \frac{|\mathbf v\times\mathbf a|^2}{c^2}\right]}$$
MotionPower
\( \mathbf a\parallel\mathbf v \) (linear accelerator, bremsstrahlung)\( \dfrac{q^2\gamma^6a^2}{6\pi\varepsilon_0c^3} \)
\( \mathbf a\perp\mathbf v \) (circular motion)\( \dfrac{q^2\gamma^4a^2}{6\pi\varepsilon_0c^3} \)

For linear acceleration the angular distribution is \( \dfrac{dP}{d\Omega} \propto \dfrac{\sin^2\theta}{(1 - \beta\cos\theta)^5} \) (power per unit retarded time, \( \theta \) from \( \mathbf v \)): the doughnut is tilted forward into a cone, and still zero exactly along \( \mathbf v \). This is the angular shape of bremsstrahlung from a decelerating fast electron.

Synchrotron radiation

In circular motion at relativistic speed the radiation is beamed forward into a cone of half-angle about \( 1/\gamma \) along the instantaneous velocity. An observer sees a short flash once per turn, which is why the spectrum extends to high harmonics of the orbital frequency (a broad continuum for large \( \gamma \)).

With \( a = v^2/\rho \) and period \( 2\pi\rho/v \), the energy lost per turn is

$$\Delta E = \frac{q^2\beta^3\gamma^4}{3\varepsilon_0\rho}, \qquad \text{for electrons: } \Delta E\,[\text{keV}] \approx 88.5\frac{(E\,[\text{GeV}])^4}{\rho\,[\text{m}]}$$

At fixed energy \( \gamma \propto 1/m \), so a proton loses \( (m_e/m_p)^4 \approx 10^{-13} \) times as much as an electron of the same energy and orbit. That is why the highest-energy circular colliders accelerate protons, and why electron rings are built with large radii.

Electric Dipole Radiation

Take \( \mathbf p(t) = p_0\cos\omega t\,\hat z \), of size \( d \). Three length scales separate cleanly when \( d \ll \lambda \ll r \):

ZoneConditionFields
Near (static)\( r \ll \lambda \)quasi-static dipole field \( \propto 1/r^3 \)
Intermediate\( r \sim \lambda \)mixture
Radiation (far)\( r \gg \lambda \)transverse, \( \propto 1/r \)

In the radiation zone, from the retarded potentials expanded to first order in \( d \):

$$\mathbf E = -\frac{\mu_0p_0\omega^2}{4\pi}\frac{\sin\theta}{r}\cos\left[\omega\left(t - \frac rc\right)\right]\hat\theta, \qquad \mathbf B = \frac{E}{c}\hat\phi$$$$\langle\mathbf S\rangle = \frac{\mu_0p_0^2\omega^4}{32\pi^2c}\frac{\sin^2\theta}{r^2}\hat r, \qquad \boxed{\langle P\rangle = \frac{\mu_0p_0^2\omega^4}{12\pi c}}$$

For an arbitrary small source, \( P = \dfrac{\mu_0\ddot p^2}{6\pi c} \), which is Larmor’s formula with \( q\mathbf a \to \ddot{\mathbf p} \).

Key features: \( \omega^4 \) dependence; \( \sin^2\theta \) pattern (no radiation along the dipole axis); linear polarisation along \( \hat\theta \); intensity \( \propto 1/r^2 \).

Magnetic Dipole Radiation

A small loop of radius \( b \) with current \( I_0\cos\omega t \) has \( m(t) = \pi b^2I_0\cos\omega t \). In the radiation zone the fields are those of the electric dipole with \( \mathbf E \) and \( c\mathbf B \) roles exchanged (\( \mathbf E \) along \( \hat\phi \)), and

$$\boxed{\langle P\rangle = \frac{\mu_0m_0^2\omega^4}{12\pi c^3}}$$

The same \( \sin^2\theta \) pattern, the other polarisation. Comparing sources of similar size \( d \sim b \):

$$\frac{P_{\text{mag}}}{P_{\text{elec}}} = \left(\frac{m_0}{p_0c}\right)^2 \sim \left(\frac{\omega b}{c}\right)^2 \ll 1$$

Magnetic dipole (and electric quadrupole) radiation is weaker by \( (d/\lambda)^2 \); it matters only when electric dipole radiation is forbidden, which is the classical root of selection rules in atomic spectra.

Antennas and Radiation Resistance

An antenna fed with current amplitude \( I_0 \) radiates \( \langle P\rangle = \tfrac12R_{\text{rad}}I_0^2 \), defining the radiation resistance.

Antenna\( R_{\text{rad}} \)Directivity
Short (Hertzian) dipole, uniform current, length \( d \ll \lambda \)\( \dfrac{2\pi}{3}\eta_0\left(\dfrac d\lambda\right)^2 = 80\pi^2\left(\dfrac d\lambda\right)^2 \approx 790\left(\dfrac d\lambda\right)^2 \) Ω1.5
Short dipole, triangular current (centre-fed, open ends)one quarter of the above, \( 20\pi^2(d/\lambda)^2 \)1.5
Half-wave dipoleabout 73 Ωabout 1.64
Small loop, area \( A \)\( 320\pi^4(A/\lambda^2)^2 \approx 3.1\times10^4(A/\lambda^2)^2 \) Ω1.5

For the uniform short dipole \( p_0\omega = I_0d \), so \( \langle P\rangle = \mu_0I_0^2d^2\omega^2/12\pi c \), which gives the first row. Short antennas have tiny radiation resistance compared with their ohmic loss, so they radiate inefficiently; that is why practical antennas are a sizeable fraction of a wavelength. Directivity is the peak of \( dP/d\Omega \) divided by its average, \( 4\pi(dP/d\Omega)_{\max}/P \); for a \( \sin^2\theta \) pattern it is \( 3/2 \).

Scattering: Thomson and Rayleigh

An incident wave \( E_0\cos\omega t \) drives a charge, which re-radiates as a dipole.

  • Free electron (Thomson): \( a = eE_0/m_e \), independent of \( \omega \). The total scattering cross-section is \( \sigma_T = \dfrac{8\pi}{3}r_e^2 = 6.65\times10^{-29} \) m\( ^2 \), frequency-independent.
  • Bound electron (natural frequency \( \omega_0 \)): \( \sigma = \sigma_T\dfrac{\omega^4}{(\omega_0^2 - \omega^2)^2} \) (light damping). For \( \omega \ll \omega_0 \) this is Rayleigh scattering, \( \sigma \propto \omega^4 \propto \lambda^{-4} \): blue light (400 nm) scatters \( (700/400)^4 \approx 9.4 \) times more than red (700 nm), so the daytime sky is blue and the setting Sun red.
  • Light scattered at \( 90^\circ \) to an unpolarised beam is linearly polarised, because the dipole has no component along the line of sight.

Cherenkov Radiation

A charge in uniform motion through a medium of refractive index \( n \) radiates if its speed exceeds the phase velocity of light in the medium, \( v > c/n \). Wavelets emitted along the path add up on a cone (an electromagnetic shock wave) at angle

$$\cos\theta_C = \frac{1}{n\beta}$$

measured from the velocity. Threshold: \( \beta > 1/n \). In water (\( n = 1.33 \)) this is \( \beta > 0.752 \), \( \gamma > 1.52 \), a kinetic energy of about 0.26 MeV for an electron. This does not contradict “only acceleration radiates”: that statement is for vacuum, and in a medium the radiation comes from the polarisation response of the material.

Memory Tricks

Don’t Confuse

ThisvsThat
Retarded time \( t_r = t - \mathscr R/c \)Advanced time (discarded) \( t + \mathscr R/c \)
Velocity field \( \propto 1/R^2 \): no net radiationAcceleration field \( \propto 1/R \): carries energy away
Electric dipole: \( \mathbf E\parallel\hat\theta \)Magnetic dipole: \( \mathbf E\parallel\hat\phi \)
Larmor (\( \beta \ll 1 \))Liénard: \( \gamma^6 \) (linear), \( \gamma^4 \) (circular)
Thomson: free electron, \( \sigma \) constantRayleigh: bound, \( \omega \ll \omega_0 \), \( \sigma \propto \omega^4 \)
Bremsstrahlung: \( \mathbf a\parallel\mathbf v \)Synchrotron: \( \mathbf a\perp\mathbf v \)
Cherenkov: uniform motion, \( v > c/n \) in a mediumUniform motion in vacuum: no radiation

Common Mistakes

Practice

All questions below are original practice questions written for this page. They are not from any CSIR NET paper. Use \( c = 3.00\times10^8 \) m/s, \( e = 1.602\times10^{-19} \) C, \( m_ec^2 = 0.511 \) MeV, \( \varepsilon_0 = 8.854\times10^{-12} \) F/m.

Level 1: Foundation

  1. Define the retarded time.
  2. Write the Larmor formula.
  3. Along which direction does an oscillating electric dipole radiate no power?
Answers
  1. \( t_r = t - |\mathbf r - \mathbf r'|/c \): the time at which a signal must leave the source point to reach the field point at \( t \).
  2. \( P = q^2a^2/(6\pi\varepsilon_0c^3) = \mu_0q^2a^2/(6\pi c) \).
  3. Along its own axis (\( \theta = 0 \) and \( \pi \)).

Level 2: Part B Style

Practice Q1. An oscillating electric dipole of fixed amplitude \( p_0 \) has its frequency doubled. Its time-averaged radiated power becomes:

(a) twice as large (b) four times as large (c) eight times as large (d) sixteen times as large

Answer: (d)

\( \langle P\rangle = \mu_0p_0^2\omega^4/12\pi c \propto \omega^4 \), so \( 2^4 = 16 \). Option (b) is the fixed-current-amplitude case (\( P \propto \omega^2 \)).

Practice Q2. A straight 1.0 m antenna carries a uniform current oscillating at 30 MHz. Its radiation resistance is about:

(a) 0.8 Ω (b) 7.9 Ω (c) 73 Ω (d) 79 Ω

Answer: (b)

\( d/\lambda = 1/10 \); \( R = 80\pi^2(0.1)^2 = 7.9 \) Ω. Option (d) forgets to square \( d/\lambda \); option (c) is the half-wave dipole value.

Practice Q3. An electron and a proton are placed in the same uniform electric field. The ratio of the power radiated by the electron to that radiated by the proton (non-relativistic) is about:

(a) 1 (b) 1836 (c) \( 3.4\times10^6 \) (d) \( 1.1\times10^{13} \)

Answer: (c)

\( a = eE/m \), so \( P \propto 1/m^2 \): \( (m_p/m_e)^2 = 1836^2 \approx 3.37\times10^6 \). Option (d) is the fourth power, which applies to synchrotron loss at equal energy.

Practice Q4. The minimum kinetic energy for an electron to emit Cherenkov radiation in water (\( n = 1.33 \)) is about:

(a) 0.26 MeV (b) 0.51 MeV (c) 0.78 MeV (d) 1.5 MeV

Answer: (a)

\( \beta > 1/1.33 = 0.752 \Rightarrow \gamma = 1/\sqrt{1 - 0.565} = 1.517 \). \( K = (\gamma - 1)m_ec^2 = 0.517\times0.511 = 0.264 \) MeV. Option (c) is the total energy \( \gamma m_ec^2 \); option (d) is \( \gamma \) mistaken for MeV.

Level 3: Part C Style

Practice Q5. An electron storage ring stores 3.0 GeV electrons on a bending radius of 8.0 m. Consider:

  1. The energy lost to synchrotron radiation per turn is about 0.90 MeV.
  2. The radiation is concentrated in a forward cone of half-angle about \( 1.7\times10^{-4} \) rad.
  3. Raising the energy to 6.0 GeV at the same radius would double the loss per turn.
  4. Protons of 3.0 GeV on the same radius would lose about \( 10^{-13} \) times as much per turn.

Which are correct?

(a) 1, 2 and 4 only (b) 1 and 2 only (c) 1, 3 and 4 only (d) 2, 3 and 4 only

Answer: (a) 1, 2 and 4 only

Statement 1: \( \Delta E = 88.5\times3.0^4/8.0 = 88.5\times81/8 = 896 \) keV \( \approx 0.90 \) MeV. (From \( q^2\beta^3\gamma^4/3\varepsilon_0\rho \) with \( \gamma = 3000/0.511 = 5871 \): the same.) True.

Statement 2: \( 1/\gamma = 1.7\times10^{-4} \) rad. True.

Statement 3: \( \Delta E \propto E^4 \): doubling the energy multiplies the loss by 16. False.

Statement 4: at equal energy \( \gamma \propto 1/m \), so the loss scales as \( (m_e/m_p)^4 = (1/1836)^4 \approx 8.8\times10^{-14} \). (A 3 GeV proton is only mildly relativistic, \( \gamma \approx 3.2 \), so \( \beta^3 \) changes the number slightly, but the order of magnitude stands.) True.

Practice Q6. For a classical electron on a circular orbit around a proton, starting at \( r_0 = 5.29\times10^{-11} \) m, consider (\( r_e = 2.82\times10^{-15} \) m):

  1. The radiated power increases as the orbit shrinks, as \( r^{-4} \).
  2. The collapse time is \( r_0^3/(4r_e^2c) \approx 1.6\times10^{-11} \) s.
  3. Starting from twice the radius would double the collapse time.
  4. The radiation at the start is at the orbital frequency, about \( 6.6\times10^{15} \) Hz.

Which are correct?

(a) 1 and 2 only (b) 1, 2 and 4 only (c) 2 and 3 only (d) 1, 2, 3 and 4

Answer: (b) 1, 2 and 4 only

Statement 1: \( a \propto 1/r^2 \), \( P \propto a^2 \propto r^{-4} \). True.

Statement 2: \( t = (5.29\times10^{-11})^3/[4(2.82\times10^{-15})^2(3.00\times10^8)] = 1.56\times10^{-11} \) s. True.

Statement 3: \( t \propto r_0^3 \): doubling \( r_0 \) multiplies \( t \) by 8. False.

Statement 4: \( v = \alpha c = 2.19\times10^6 \) m/s, \( f = v/2\pi r_0 = 2.19\times10^6/(2\pi\times5.29\times10^{-11}) = 6.6\times10^{15} \) Hz. A non-relativistic charge in circular motion radiates at the orbital frequency. True.

Practice Q7. Consider the following statements about the Liénard–Wiechert potentials and fields of a point charge:

  1. The potentials are those of a static charge evaluated at the retarded position, multiplied by \( 1/(1 - \hat{\mathbf n}\cdot\boldsymbol\beta) \).
  2. The factor \( 1 - \hat{\mathbf n}\cdot\boldsymbol\beta \) arises from Lorentz contraction of the charge distribution.
  3. The acceleration part of \( \mathbf E \) is perpendicular to \( \hat{\mathbf n} \) and falls off as \( 1/\mathscr R \).
  4. The retarded potentials satisfy the Lorenz gauge condition.

Which are correct?

(a) 1, 3 and 4 only (b) 1, 2 and 3 only (c) 3 and 4 only (d) 1, 2, 3 and 4

Answer: (a) 1, 3 and 4 only

Statement 1: \( V = q/[4\pi\varepsilon_0\mathscr R(1 - \hat{\mathbf n}\cdot\boldsymbol\beta)] \) at \( t_r \), \( \mathbf A = \mathbf vV/c^2 \). True.

Statement 2: the factor is geometric. It comes from the light-cone condition: a charge moving towards the observer is “seen” for a longer stretch of its path. It is present even at low speed and contains no \( \gamma \). False.

Statement 3: \( \hat{\mathbf n}\times[(\hat{\mathbf n} - \boldsymbol\beta)\times\dot{\boldsymbol\beta}] \) is perpendicular to \( \hat{\mathbf n} \), with a \( 1/\mathscr R \) prefactor. True.

Statement 4: they are the retarded solutions of the Lorenz-gauge wave equations and satisfy \( \nabla\cdot\mathbf A + c^{-2}\partial_tV = 0 \) by charge conservation. True.

Quick Revision Table

ResultFormula
Retarded potentials\( V = \frac{1}{4\pi\varepsilon_0}\int\rho(t_r)/\mathscr R \), \( \mathbf A = \frac{\mu_0}{4\pi}\int\mathbf J(t_r)/\mathscr R \)
Liénard–Wiechert\( V = q/[4\pi\varepsilon_0\mathscr R(1 - \hat{\mathbf n}\cdot\boldsymbol\beta)] \), \( \mathbf A = \mathbf vV/c^2 \)
Larmor\( P = q^2a^2/6\pi\varepsilon_0c^3 \); \( dP/d\Omega \propto \sin^2\theta \)
Liénard\( P = \frac{q^2\gamma^6}{6\pi\varepsilon_0c^3}(a^2 - \vert \mathbf v\times\mathbf a\vert ^2/c^2) \)
Synchrotron loss\( \Delta E = q^2\beta^3\gamma^4/3\varepsilon_0\rho \); \( 88.5E^4/\rho \) keV (GeV, m)
Beaminghalf-angle \( \approx 1/\gamma \)
Electric dipole\( \langle P\rangle = \mu_0p_0^2\omega^4/12\pi c \)
Magnetic dipole\( \langle P\rangle = \mu_0m_0^2\omega^4/12\pi c^3 \)
Radiation resistance\( 80\pi^2(d/\lambda)^2 \) (short, uniform); about 73 Ω (half-wave)
Thomson\( \sigma_T = \frac{8\pi}{3}r_e^2 = 6.65\times10^{-29} \) m\( ^2 \)
Rayleigh\( \sigma \propto \omega^4 \) for \( \omega \ll \omega_0 \)
Abraham–Lorentz\( \mathbf F = m\tau\dot{\mathbf a} \), \( \tau = 6.3\times10^{-24} \) s
Cherenkov\( \cos\theta_C = 1/n\beta \), threshold \( \beta > 1/n \)

Key Takeaways

What’s Next

Advertisement
Advertisement