Relativistic Larmor Formula (Liénard's Result)

Larmor's formula was derived under a flag of convenience: $v\ll c$ . That flag excludes precisely the machines and objects where radiation matters most — storage rings pushing electrons to $\gamma\sim2\times10^5$ , cosmic-ray electrons spiraling in galactic fields, jets launched by black holes. This article removes the restriction. The result, obtained by Alfred-Marie Liénard in 1898 — before relativity existed to explain why it is beautiful — gives the power radiated by a charge at any speed, and it carries a shock: a factor $\gamma^6$ in front, moderated to $\gamma^4$ for circular motion. Radiation does not just grow with energy; it grows with the fourth to sixth power of it. The derivation route matters as much as the result, because it teaches the two deepest bookkeeping lessons in radiation theory. First: the power that leaves the source per unit of the source's own time differs from the flux crossing a distant sphere per unit observer time — the retarded-time Jacobian $\kappa=1-\hat{\mathbf{n}}\cdot\boldsymbol\beta$ , inherited from the Liénard–Wiechert construction, converts one to the other, and its fifth power in the angular distribution is what squeezes the radiation into the famous forward cone of half-angle $1/\gamma$ . Second: total radiated power is a Lorentz invariant — radiated energy and elapsed time pick up the same $\gamma$ , which cancels — so the whole formula can be written in one covariant stroke using the four-acceleration, and evaluated in whatever frame is cheapest. The payoff table is immediate and expensive: LEP's electrons paid three GeV per turn in synchrotron light — the bill that capped circular electron colliders and rerouted the field toward linacs and proton rings — while the same formula, run as a light source rather than a tax, powers every synchrotron beamline in the world.

Key takeaways

Liénard's generalization of Larmor (1898): $P=\dfrac{\mu_0q^2\gamma^6 c}{6\pi}\Big[\dot{\boldsymbol\beta}^2-(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2\Big]$ — exact at any speed, reducing to Larmor as $\beta\to0$ .
Total radiated power is a Lorentz invariant: in the instantaneous rest frame the radiation carries no net momentum, so energy and time transform with the same $\gamma$ and $P=P'$ .
Covariant form (metric $+,-,-,-$ ): $P=-\dfrac{\mu_0q^2}{6\pi m^2c}\,\dfrac{dp^\mu}{d\tau}\dfrac{dp_\mu}{d\tau}$ , with the minus sign because four-acceleration is spacelike.
Geometry rules the exponents: acceleration parallel to velocity radiates with $\gamma^6\dot\beta^2$ but is tame when expressed via force; perpendicular (circular) acceleration gives the synchrotron law $P\propto\gamma^4/\rho^2$ .
The angular distribution carries $\kappa^{-5}=(1-\hat{\mathbf{n}}\cdot\boldsymbol\beta)^{-5}$ : emission beams into a forward cone of half-angle $\sim1/\gamma$ — $5\ \mu$ rad at LEP.
Anchor numbers: LEP (100 GeV, $\rho=3026$ m) lost 2.9 GeV per electron per turn; a 100 MV/m linac loses $4\times10^{-5}$ eV per meter to radiation — nothing.
Key idea: radiated power is an invariant built from the four-acceleration; expressed in lab variables it explodes as $\gamma^6$ (linear) or $\gamma^4$ (circular), and the mismatch between those two geometries designs the world's accelerators.

The core insight

Total radiated power is a Lorentz invariant equal to $-\frac{\mu_0q^2}{6\pi m^2c}(dp^\mu/d\tau)(dp_\mu/d\tau)$ ; in 3-vector form the geometry of $\boldsymbol\beta$ relative to $\dot{\boldsymbol\beta}$ controls everything through the $\gamma^6$ and cross-product structure.

Where this lives today

The formula's two faces run two industries. As a tax, it writes the boundary conditions of high-energy physics: the ILC and CLIC linear-collider designs exist because (2) makes linear acceleration radiation-free while (3) forbids circular $e^+e^-$ machines beyond LEP's scale; the proposed FCC-ee lives inside the loophole of a 100 km circumference (bigger $\rho$ , smaller tax) and actually uses the radiation as a built-in beam cooler. Muon-collider studies split the difference: at $m_\mu=207m_e$ the $\gamma^4$ tax drops by $(207)^4\approx1.8\times10^9$ , buying circular machines back — if the muons can be used before they decay. As a source, equation (3) powers more than fifty synchrotron-light facilities. The $1/\gamma$ headlight sweeping past a beamline delivers femtosecond-scale X-ray pulses whose spectrum extends to a critical frequency $\sim\gamma^3c/\rho$ (the spectral story — how the searchlight's sweep compresses the received pulse by another $\gamma^2$ beyond naive expectation — belongs to the advanced synchrotron article); protein crystallography, battery tomography, and semiconductor metrology queue for that light. The same $\gamma^4$ physics, running wild at astrophysical scale, makes synchrotron emission the universal tracer of relativistic electrons: the radio glow of supernova remnants, the jets of active galactic nuclei, the Crab Nebula's eerie blue — all Liénard's formula applied to cosmic accelerators, with the observed power-law spectra decoding the electrons' energy distribution.

Origins · 1 minAlfred-Marie Liénard and how this was discovered

Alfred-Marie Liénard

1869–1958 · École des Mines, Paris — who generalized Larmor seven years before relativity could explain the result

Liénard published the arbitrary-velocity radiation formula in 1898 in L'Éclairage Électrique — an electrical-engineering journal — as part of the same investigation that produced the Liénard–Wiechert potentials. Working entirely within pre-relativistic electrodynamics, he correctly assembled what we now read as $\gamma^6[\dot{\boldsymbol\beta}^2-(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2]$ : every Lorentz factor in its proper place, seven years before Einstein supplied the kinematics that makes the structure inevitable and thirty before "Lorentz invariant" became the standard compliment for it. (This article's Derivation B, run in reverse, is the modern reader's way of discovering how much relativity was already hiding inside Maxwell's theory — a theme the next three articles make explicit.) A mining engineer by training who rose to direct the École des Mines, Liénard published sparingly but lastingly: the potentials and power formula that bear his name, and the Liénard equation of nonlinear oscillator theory, which still organizes limit-cycle analysis. He declined most honors, served as an artillery-ballistics consultant through the First World War, and lived to see his 1898 formula operating both the storage rings of particle physics and the radio astronomy of the relativistic universe.

History

1897

— Larmor's nonrelativistic power formula: the rest-frame anchor

1898

— Liénard extends it to arbitrary velocity; the γ⁶ and cross-product structure appear pre-relativity

1900

— Wiechert independently completes the point-charge potential framework

1905

— Einstein's relativity recasts Liénard's factors as kinematic inevitabilities

1912

— Schott's Electromagnetic Radiation works out the angular distributions and beaming in detail

1944

— Ivanenko and Pomeranchuk predict the γ⁴ energy ceiling for betatrons; Schwinger's synchrotron theory follows

1947

— synchrotron light seen at General Electric: the headlight made visible

2000

— LEP shuts down at 104.5 GeV per beam, the γ⁴ bill unpayable — the formula retires its own machine

The physical picture

Keep the kink picture from the Larmor article, and turn up the speed. A relativistic charge nearly keeps pace with its own outgoing wavefronts in the forward direction: the news of each kink, emitted at $c$ , barely outruns the charge that emitted it. Wavefronts pile up ahead and stretch out behind — and the transverse kink field, crowded into that thin forward-piled region, is intensified enormously in the direction of motion. This geometric crowding is the entire origin of the huge $\gamma$ factors: not new physics, but the old acceleration field viewed from a frame where its source almost catches its own light. The crowding has a precise bookkeeping name, carried over from the Liénard–Wiechert construction: $\kappa=1-\hat{\mathbf{n}}\cdot\boldsymbol\beta$ , the factor by which source time compresses into observer time along the line of sight. Ahead of an ultrarelativistic charge $\kappa\approx1/2\gamma^2$ — a signal emitted over one nanosecond of the charge's journey arrives packed into a fraction $1/2\gamma^2$ of a nanosecond. Powers of $\kappa$ in the denominator are powers of $\gamma^2$ of forward brightening, and the angular distribution below carries five of them. The other pillar is an invariance argument of arresting simplicity. Sit in the charge's instantaneous rest frame: Larmor applies exactly, and the $\sin^2\theta$ donut is fore-aft symmetric, so the radiated burst carries energy but zero net momentum. A packet with $d\mathbf{p}'=0$ transforms like the time component of a four-vector: $dW=\gamma\,dW'$ . But the time elapsed also dilates: $dt=\gamma\,dt'$ . Their ratio — the power — is untouched. Every inertial observer, whatever angular pattern and Doppler shifts they record, agrees on the single number $P$ . That number, evaluated once in the rest frame, is Larmor with the proper acceleration; all the $\gamma$ 's of the lab formula are then just the dictionary between proper and lab acceleration.

Figure

The headlight effect — rest-frame donut, lab-frame searchlight: kappa powers convert one into the other

The math: emitted power, received power, and the Jacobian between them

The raw material is inherited by name from Fields of a Moving Point Charge: the acceleration field of an arbitrarily moving charge,

\mathbf{E}_{\text{rad}}=\frac{q}{4\pi\varepsilon_0c}\,\frac{\hat{\mathbf{n}}\times\big[(\hat{\mathbf{n}}-\boldsymbol\beta)\times\dot{\boldsymbol\beta}\big]}{\kappa^3\,r}\Bigg|_{\text{ret}},\qquad \kappa=1-\hat{\mathbf{n}}\cdot\boldsymbol\beta, \qquad (1)

where: everything on the right is evaluated at the retarded time, and the three powers of $\kappa$ already in the field are the Liénard–Wiechert compression factors that article derived. Radiated flux goes as $E^2$ , so six powers of $\kappa$ appear in the energy arriving per unit observer time — but one power is spent converting to the source's clock, and the distinction is the derivation's one genuinely subtle move.

Derivation AFrom the acceleration field to Lienard's formula

The goal is the exact radiated power of a charge with arbitrary $\boldsymbol\beta$ , $\dot{\boldsymbol\beta}$ — Liénard's 1898 result. The strategy: form the Poynting flux of (1); convert flux-per-observer-time to energy-lost-per-source-time with the Jacobian $dt_{\text{obs}}=\kappa\,dt_{\text{ret}}$ (this is what "power radiated by the charge" must mean if it is to appear in the charge's energy budget); then integrate the resulting $\kappa^{-5}$ distribution over the sphere. The angular integral is standard but heavy; we organize it so the two geometric invariants $\dot\beta^2$ and $(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2$ emerge and quote the tabulated integrals for the final assembly, with the rest-frame limit and the invariance theorem (Derivation B) as independent audits.

Step 1. Flux at the sphere. With the radiation triad ( $B=E/c$ , $\mathbf{S}$ along $\hat{\mathbf{n}}$ ):

\mathbf{S}\cdot\hat{\mathbf{n}}=\frac{|\mathbf{E}_{\text{rad}}|^2}{\mu_0c}=\frac{\mu_0q^2}{16\pi^2c}\,\frac{\big|\hat{\mathbf{n}}\times[(\hat{\mathbf{n}}-\boldsymbol\beta)\times\dot{\boldsymbol\beta}]\big|^2}{\kappa^6\,r^2}.

Step 2. Convert to the source's clock. Energy crossing the sphere during observer interval $dt$ was emitted during source interval $dt_{\text{ret}}=dt/\kappa$ ; the power drained from the charge per unit of its own coordinate time therefore carries one $\kappa$ upstairs:

\frac{dP}{d\Omega}=r^2(\mathbf{S}\cdot\hat{\mathbf{n}})\,\kappa=\frac{\mu_0q^2}{16\pi^2c}\,\frac{\big|\hat{\mathbf{n}}\times[(\hat{\mathbf{n}}-\boldsymbol\beta)\times\dot{\boldsymbol\beta}]\big|^2}{(1-\hat{\mathbf{n}}\cdot\boldsymbol\beta)^5}.

Why this bookkeeping and not the other: both quantities are legitimate — flux per observer time is what a detector reads; power per source time is what enters $dE_{\text{particle}}/dt$ . They differ by exactly the factor that, for periodic motion, redistributes the energy within a cycle without changing the total. For the energy-loss physics this chapter is heading toward (radiation reaction, accelerator budgets) the source-clock version is the meaningful one — and forgetting the distinction is the classic way to lose a power of $\kappa$ and get the beaming exponent wrong.

Step 3. The forward cone, read before integrating. For $\beta\to1$ and small angle $\theta$ from the velocity, $\kappa\approx(1+\gamma^2\theta^2)/2\gamma^2$ : the denominator's fifth power concentrates emission within $\theta\sim1/\gamma$ , brightening the forward direction by $\sim\gamma^{10}$ relative to naive Larmor while the solid angle shrinks as $1/\gamma^2$ — net growth $\sim\gamma^{4\text{–}6}$ , previewing the total.

Step 4. Integrate over the sphere. Expanding the numerator and using the tabulated angular integrals of $\kappa^{-n}$ (equivalently: boost the rest-frame donut), the cross terms assemble into two invariant combinations:

P=\frac{\mu_0q^2\gamma^6 c}{6\pi}\Big[\dot{\boldsymbol\beta}^2-\big(\boldsymbol\beta\times\dot{\boldsymbol\beta}\big)^2\Big].

Why these two structures and no others: $P$ is a scalar under rotations built quadratically from $\dot{\boldsymbol\beta}$ with coefficients depending on $\boldsymbol\beta$ — the only candidates are $\dot\beta^2$ , $(\boldsymbol\beta\cdot\dot{\boldsymbol\beta})^2$ and $\beta^2\dot\beta^2$ , of which the cross-product combination packages the allowed pair. The overall $\gamma^6$ then follows from the parallel-acceleration special case alone.

Step 5. Audit the limits. $\beta\to0$ : bracket $\to\dot\beta^2=a^2/c^2$ , prefactor $\to\mu_0q^2c/6\pi\cdot(a^2/c^2)\cdot(1/c^0)$ ... assembling, $P\to\mu_0q^2a^2/6\pi c$ — Larmor, exactly. $\dot{\boldsymbol\beta}\parallel\boldsymbol\beta$ : cross product dies, $P=\mu_0q^2\gamma^6\dot\beta^2c/6\pi\cdot c^{-1}\cdot c^2$ — the $\gamma^6$ law. $\dot{\boldsymbol\beta}\perp\boldsymbol\beta$ : $(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2=\beta^2\dot\beta^2$ , bracket $=\dot\beta^2(1-\beta^2)=\dot\beta^2/\gamma^2$ : the $\gamma^6$ demotes to $\gamma^4$ — the synchrotron law. Verified: matches Jackson Eq. 14.26 (route and result) and Griffiths §11.2.2; the $\beta\to0$ , parallel, and perpendicular reductions land on the standard forms; Derivation B reproduces the same total from pure invariance.

Go deeper · Derivation B · 2 minThe invariance theorem and the covariant form

Derivation BThe invariance theorem and the covariant form

The goal is the formula's deepest property — $P$ is a Lorentz scalar — and its payoff: a one-line covariant expression that expands into Liénard's 3-vector result, tying this article to the four-vector machinery whose full development belongs to the relativity chapter this track is about to open (and, as a companion treatment, to the mechanics article Four-Vectors and Minkowski Spacetime). Strategy: prove invariance physically from the rest-frame symmetry, then write the unique scalar with the right rest-frame limit and check it reproduces Derivation A.

Step 1. Rest-frame symmetry kills the radiated momentum. In the instantaneous rest frame $S'$ the charge is momentarily at rest; Larmor applies exactly, and the $\sin^2\theta'$ donut is symmetric under $\hat{\mathbf{n}}'\to-\hat{\mathbf{n}}'$ . Integrating the momentum flux over the sphere: $d\mathbf{p}'_{\text{rad}}=0$ . The emitted burst has energy $dW'$ and no net momentum.

Step 2. Energy and time dilate together. Transforming the burst's four-momentum $(dW'/c,\mathbf{0})$ to the lab: $dW=\gamma\,dW'$ . Transforming the emission interval (a proper-time interval on the charge's worldline): $dt=\gamma\,dt'$ . Hence

P=\frac{dW}{dt}=\frac{\gamma\,dW'}{\gamma\,dt'}=P'.

Why the momentum-zero step was essential: had the burst carried net momentum $d\mathbf{p}'\ne0$ , the energy transformation would read $dW=\gamma(dW'+v\,dp'_x)$ and the $\gamma$ 's would not cancel. The fore-aft symmetry of the rest-frame donut is not decoration — it is the entire theorem.

Step 3. Write the invariant. $P$ is a scalar that reduces, in the rest frame, to Larmor with the proper acceleration. The four-acceleration $a^\mu=du^\mu/d\tau$ (with $u^\mu$ the four-velocity, metric $\eta_{\mu\nu}=\mathrm{diag}(+,-,-,-)$ throughout this track, matching the mechanics companions) has rest-frame components $(0,\mathbf{a}')$ , so $a^\mu a_\mu=-a'^2$ — spacelike, negative square. The unique scalar with the right limit:

P=-\frac{\mu_0q^2}{6\pi c}\,a^\mu a_\mu=-\frac{\mu_0q^2}{6\pi m^2c}\,\frac{dp^\mu}{d\tau}\frac{dp_\mu}{d\tau},

the second form using $p^\mu=mu^\mu$ — handy because $dp^\mu/d\tau$ is the four-force accelerators quote.

Step 4. Expand and recover Liénard. Writing $u^\mu=\gamma(c,\mathbf{v})$ and differentiating with $d/d\tau=\gamma\,d/dt$ , the invariant square evaluates (a half-page of chain rule, verified numerically for random $\boldsymbol\beta,\dot{\boldsymbol\beta}$ in this edition's audit) to

a^\mu a_\mu=-\gamma^6c^2\Big[\dot{\boldsymbol\beta}^2-(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2\Big],

and substitution reproduces Derivation A's boxed result identically — two independent routes, one law. Verified: rest-frame reduction gives $P=\mu_0q^2a'^2/6\pi c$ (Larmor with proper acceleration); the $a^\mu a_\mu$ identity checked symbolically/numerically; matches Jackson §14.1–14.2 and the covariant form in Griffiths Problem 12.71 territory.

The two clean geometries deserve their engineering spellings. Linear ( $\dot{\boldsymbol\beta}\parallel\boldsymbol\beta$ ): since $dp/dt=m\gamma^3\dot v$ for parallel force, the $\gamma$ 's absorb perfectly into the force,

P_\parallel=\frac{\mu_0q^2}{6\pi m^2c}\left(\frac{dp}{dt}\right)^2 \qquad (2)

where: the radiated power depends only on the force, not on the energy already attained — which is why linacs scale to arbitrary energy with negligible radiative tax (worked example 2 prices it at $10^{-13}$ of the gain). Circular (radius $\rho$ , $\dot\beta=\beta^2c/\rho...$ i.e. $|\dot{\boldsymbol\beta}|=\beta^2c/\rho$ ):

P_\perp=\frac{\mu_0\,q^2c^3\,\gamma^4\beta^4}{6\pi\rho^2},\qquad \Delta E_{\text{per turn}}=\frac{q^2\gamma^4\beta^3}{3\varepsilon_0\rho}, \qquad (3)

where: the per-turn loss follows from $P_\perp\times(2\pi\rho/\beta c)$ , and the $\gamma^4=(E/mc^2)^4$ is the formula that ended an era of machines: double a ring's beam energy and the radiation bill rises sixteenfold, with the electron's small mass making $\gamma^4$ astronomically larger than a proton's at the same energy — the entire strategic logic of LEP versus LHC in one exponent. (A frequent typo in the literature drops the $c^3$ in the power form of (3); dimensional audit — done for this edition — restores it.)

Figure

Two geometries, two destinies — the linac loophole and the synchrotron tax

Limiting cases

case	result	meaning
$\beta\to0$	$P=\mu_0q^2a^2/6\pi c$	Larmor recovered; the rest-frame anchor of the invariance proof
$\dot{\boldsymbol\beta}\parallel\boldsymbol\beta$	$P=\mu_0q^2\gamma^6\dot\beta^2c/6\pi$ $=\mu_0q^2(dp/dt)^2/6\pi m^2c$	fierce in $\gamma$ , tame in force: the linac loophole
$\dot{\boldsymbol\beta}\perp\boldsymbol\beta$	$P=\mu_0q^2c^3\gamma^4\beta^4/6\pi\rho^2$	synchrotron law; per-turn loss $q^2\gamma^4\beta^3/3\varepsilon_0\rho$
equal proper acceleration	$P_\perp/P_\parallel=\gamma^2$	geometry beats magnitude: bending radiates more than pushing
$\beta\to1$ angular pattern	cone $\theta\sim1/\gamma$ , brightening $\propto\kappa^{-5}$	the headlight; $5\ \mu$ rad at LEP
same $E$ , protons vs electrons	$P_e/P_p=(m_p/m_e)^4\approx1.1\times10^{13}$	why proton rings reach 7 TeV where electron rings stalled at 0.1
observer-time bookkeeping	drop the $\kappa$ of Step 2	received-flux distribution ( $\kappa^{-6}$ ): a detector statement, not an energy-budget one

Common trapGamma-sixth is the law, gamma-fourth the special case — and the Jacobian kappa is not optional

Two errors, both fatal at relativistic speeds. First, the exponent hierarchy: the general scalar formula carries $\gamma^6$ ; the famous synchrotron $\gamma^4$ arises only because perpendicular geometry cancels two powers via $(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2=\beta^2\dot\beta^2$ . Quoting " $\gamma^4$ " as the universal radiation law, or applying $\gamma^6$ with the lab acceleration to circular motion, each misprices the physics by $\gamma^2$ — a factor of $10^{10}$ at LEP. Second, the retarded-time Jacobian: the power that debits the particle's energy is $r^2S\kappa$ , not $r^2S$ . Forgetting the $\kappa$ (or double-counting it) shifts the angular distribution between $\kappa^{-5}$ and $\kappa^{-6}$ and corrupts the total. The two bookkeepings answer different questions — energy lost per source time versus flux received per observer time — and every relativistic radiation calculation must declare, before integrating, whose clock it is using.

Worked examples

1 · LEP's three-GeV toll — the formula that capped an accelerator dynasty (confidence builder)

Setup. LEP circulated $E=100$ GeV electrons ( $mc^2=0.511$ MeV) on bending radius $\rho=3026$ m. Compute the per-turn radiative loss and the instantaneous synchrotron power per electron.

Step 1. The Lorentz factor: $\gamma=E/mc^2=10^{11}/5.11\times10^5=1.96\times10^5$ , so $\gamma^4=1.47\times10^{21}$ and $\beta\approx1$ to thirteen digits.

Step 2. Per-turn loss from (3):

\Delta E=\frac{e^2\gamma^4}{3\varepsilon_0\rho}=\frac{(1.60\times10^{-19})^2(1.47\times10^{21})}{3(8.85\times10^{-12})(3026)}=4.68\times10^{-10}\ \text{J}=\mathbf{2.9\ GeV\ per\ turn}

— three percent of the beam energy, every $89\ \mu$ s, per electron; the RF system existed almost entirely to pay this bill, and pushing past $\sim104$ GeV per beam was economically impossible. LEP was dismantled and the LHC installed in its tunnel.

Step 3. Instantaneous power: $P=\Delta E/T$ with $T=2\pi\rho/c=6.34\times10^{-5}$ s gives $P=7.4\times10^{-6}$ W per electron — microwatts from one particle, delivered as X-rays into a $5\ \mu$ rad headlight ( $1/\gamma$ ).

Step 4. The proton control. At the LHC's 7 TeV in nearly the same tunnel ( $\rho=2804$ m): $\gamma_p=7460$ , $\gamma_p^4=3.1\times10^{15}$ , and $\Delta E\approx\mathbf{6.7\ keV}$ per turn — five orders of magnitude below LEP's despite seventy times the energy. The mass ratio to the fourth power, $(m_p/m_e)^4\approx1.1\times10^{13}$ , is the whole story of why hadron colliders own the energy frontier.

The lesson: $\gamma^4/\rho$ with the right mass is a one-line machine-design verdict.

2 · The linac loophole — radiation under linear acceleration (exam level)

Setup. An electron rides a linac gradient of $100$ MV/m ( $F=eE_{\text{acc}}=1.6\times10^{-11}$ N, constant). Compute the radiated power, the energy radiated per meter, and compare with the energy gained — at any $\gamma$ .

Step 1. Apply (2), which is exact for parallel acceleration and already contains all the $\gamma^6$ physics inside $dp/dt=F$ :

P_\parallel=\frac{\mu_0e^2F^2}{6\pi m^2c}=\frac{(1.26\times10^{-6})(1.60\times10^{-19})^2(1.60\times10^{-11})^2}{6\pi(9.11\times10^{-31})^2(3.0\times10^8)}=\mathbf{1.8\times10^{-15}\ W},

independent of the electron's current energy — the force alone sets the bill.

Step 2. Radiated energy per meter (at $v\approx c$ , one meter takes $1/c$ seconds):

\frac{dE_{\text{rad}}}{dx}=\frac{P_\parallel}{c}=5.9\times10^{-24}\ \text{J/m}=\mathbf{3.7\times10^{-5}\ eV\ per\ meter}.

Step 3. The comparison. The gradient delivers $10^8$ eV per meter: the radiative tax is $3.7\times10^{-13}$ of the gain — thirteen orders of magnitude of loophole. Even at exawatt-laser gradients the parallel-acceleration channel stays negligible; equivalently, the general criterion (radiation matters when the energy gained over a distance $mc^2/F$ ... i.e. when the gradient approaches $mc^2/r_e\sim2\times10^{13}$ MV/m) sits eleven orders beyond current technology.

Step 4. Reconcile with the fierce $\gamma^6$ . No contradiction: $P_\parallel=\mu_0q^2\gamma^6\dot\beta^2c/6\pi$ is huge at fixed $\dot\beta$ — but a fixed force produces an ever-smaller $\dot\beta\propto1/\gamma^3$ (relativistic inertia), and the two effects cancel exactly. The physical invariant is the force; expressed in it, linear radiation is a spectator. Circular machines enjoy no such cancellation because bending at fixed $\rho$ demands $\dot\beta=\beta^2c/\rho$ regardless of $\gamma$ .

The lesson: always ask what is held fixed. Fixed force: linear radiation is free. Fixed radius: circular radiation is ruinous. That asymmetry — a factor $\gamma^2$ at equal proper acceleration — is the blueprint of every collider strategy meeting since 1960.

Going further

References. Jackson, Classical Electrodynamics, 3rd ed., §14.1–14.3 · Griffiths, Introduction to Electrodynamics, 4th ed., §11.2.2 · Zangwill, Modern Electrodynamics, §23.4 · Schott, Electromagnetic Radiation (1912), Ch. IV

Provenance. Liénard formula and angular distribution verified against Jackson §14.2–14.3 (Eqs. 14.26, 14.38) and Griffiths §11.2.2 (Eqs. 11.73–11.74); Lorentz-invariance argument follows the Griffiths/Jackson rest-frame derivation. A verifier should: (1) check the $\beta\to0$ limit yields $P=\mu_0q^2a^2/6\pi c$ ; (2) confirm $a^\mu a_\mu=-\gamma^6c^2[\dot{\boldsymbol\beta}^2-(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2]$ ; (3) verify the synchrotron per-turn loss $\Delta E=q^2\gamma^4\beta^3/3\varepsilon_0\rho$ against Jackson Eq. 14.32. Addendum for this edition: the acceleration field (1) is quoted by name from Fields of a Moving Point Charge (which owns its derivation and the $\kappa^3$ anatomy); the nonrelativistic anchor is owned by Larmor Formula; the synchrotron spectrum (critical frequency, pulse compression) is deferred by name to the advanced synchrotron article, and the covariant radiation-field route to the E11 twin. Metric signature $(+,-,-,-)$ locked to this track's relativity chapter and the mechanics companions. Base-error note: the base's synchrotron eqCard $P_\perp=\mu_0q^2c\gamma^4\beta^4/6\pi\rho^2$ is dimensionally short a factor $c^2$ ; corrected here to $\mu_0q^2c^3\gamma^4\beta^4/6\pi\rho^2$ and cross-checked numerically ( $P\times T=\Delta E$ to 8 digits). Worked numbers for this edition (LEP: $\gamma=1.96\times10^5$ , $\Delta E=2.9$ GeV/turn, $P=7.4\ \mu$ W, cone $5.1\ \mu$ rad; LHC protons 6.7 keV/turn; linac at 100 MV/m: $P=1.8\times10^{-15}$ W, $3.7\times10^{-5}$ eV/m $=3.7\times10^{-13}$ of gain; $a^\mu a_\mu$ identity verified numerically) recomputed for this edition.

Builds on: Fields of a Moving Point Charge · Liénard–Wiechert Potentials · Larmor Formula · Retarded Potentials Leads to: Radiation Reaction Force · Abraham–Lorentz Equation and Its Pathologies · Postulates of Special Relativity

How could Liénard get every relativistic factor right in 1898, before relativity?

Because the factors were never relativity's property — they belong to Maxwell's equations, which were relativistic all along. The Liénard–Wiechert potentials and the fields derived from them encode the full finite- $c$ , retarded structure of the theory; computing the Poynting flux honestly, as Liénard did, mechanically produces the $\kappa$ powers and the combination $\gamma^6[\dot{\boldsymbol\beta}^2-(\boldsymbol\beta\times\dot{\boldsymbol\beta})^2]$ . What 1898 could not supply was the reading: that these factors express a symmetry of space and time itself, that $P$ is a scalar of that symmetry, and that the same $\gamma$ 's must govern mechanics too. That reading arrived in 1905, and the next articles of this track retrace exactly this discovery arc — electrodynamics first exhibiting the structure, relativity then explaining it.

Is the radiated power really the same for all observers even though the radiation is beamed forward?

Yes — invariance of $P$ and wild frame-dependence of its distribution coexist without tension. A lab observer sees a $\kappa^{-5}$ searchlight with blue-shifted, pulse-compressed radiation ahead and a feeble red tail behind; the co-moving observer sees a placid symmetric donut. But total energy per total time is a ratio of two quantities that transform identically (given the rest-frame momentum-zero property), so the single number $P$ survives. The practical consequence is enormous: you may compute $P$ in the instantaneous rest frame with schoolbook Larmor and quote it in any frame. Only when you need where the energy goes — detector placement, spectral shaping, shielding — must you pay for the full angular machinery.

The linac loophole says radiation depends only on force for parallel acceleration. Where did the gamma-sixth actually go?

Into relativistic inertia, exactly cancelling. For parallel force, $F=dp/dt=m\gamma^3\dot v$ : as the particle stiffens, a fixed force produces $\dot\beta\propto1/\gamma^3$ . Insert into $P_\parallel=\mu_0q^2\gamma^6\dot\beta^2c/6\pi$ and $\gamma^6\times\gamma^{-6}=1$ — the power collapses to the force-only form (2). Circular motion enjoys no such mercy: maintaining a fixed orbit radius requires the same $\dot\beta=\beta^2c/\rho$ at every $\gamma$ (the transverse "inertia" is only $m\gamma$ , and geometry, not force budget, dictates the acceleration), so the $\gamma^4$ survives at full strength. The deep version: $a^\mu a_\mu$ for parallel force is $F^2/m^2$ exactly, while for transverse force it is $\gamma^2F^2/m^2$ — the $\gamma^2$ that separates the two accelerator destinies.

What is the difference between the power computed here and what a detector actually measures during a flyby?

Three distinct quantities, one event. (i) $P_{\text{source}}$ — energy drained from the charge per unit of its coordinate time: Liénard's formula, the $\kappa^5$ distribution, the accelerator designer's number. (ii) Received flux — energy per unit observer time at the detector: carries $\kappa^{-6}$ , and during the headlight sweep of a circular-motion flyby the received pulse is compressed by a further factor $\sim\gamma^2$ beyond the emission-cone estimate (the sweep and the light chasing itself compound), which is why synchrotron pulses are so short and their spectra reach the critical frequency $\sim\gamma^3c/\rho$ — the advanced synchrotron article's opening calculation. (iii) The spectrum — the Fourier transform of (ii), not of (i). Conflating (i) and (ii) is the standard route to wrong pulse widths and wrong spectral cutoffs; the $\kappa$ Jacobian of Derivation A Step 2 is precisely the dictionary between them.

Does the invariance of P mean radiated four-momentum is a four-vector with zero spatial part?

No — and the distinction is instructive. Over an extended stretch of accelerated motion the charge radiates both energy and net momentum in the lab (the beamed radiation obviously carries forward momentum; that recoil is a piece of the radiation-reaction story ahead). The invariance theorem is an instantaneous, rest-frame-anchored statement: during an infinitesimal proper-time interval, the burst's four-momentum is $(dW'/c)\,u^\mu/c$ — proportional to the charge's own four-velocity, because the rest-frame emission is momentum-free. Contracting with $u_\mu$ gives the invariant $dW'$ , and dividing by $d\tau$ gives the scalar rate; but the burst's lab momentum $\gamma\beta\,dW'/c$ is perfectly real. So: $P$ is a scalar; radiated four-momentum is a genuine four-vector along $u^\mu$ ; and the recoil force it implies is $-P u^\mu/c^2$ — the covariant seed of the Abraham–Lorentz–Dirac analysis two articles ahead.

Why do astrophysicists instantly attribute power-law radio spectra to synchrotron radiation?

Because the formula plus a power-law electron population makes an unmistakable fingerprint. A single relativistic electron in a magnetic field radiates $P\propto\gamma^2B^2$ (the $\gamma^4$ law with $\rho=\gamma mc/eB$ folding two powers into the field), peaked near its critical frequency $\nu_c\propto\gamma^2B$ . Cosmic accelerators generically produce electron energy distributions $N(\gamma)\propto\gamma^{-p}$ (shock acceleration's signature), and integrating single-electron emission over that population yields a flux $S_\nu\propto\nu^{-(p-1)/2}$ — a featureless power law over decades of frequency, strongly linearly polarized (up to $\sim70\%$ for ordered fields), steepening with age as the fastest electrons radiate away first. Thermal mechanisms cannot mimic that combination of shape, polarization, and evolution. Reading $p$ , the break frequency, and the polarization off a radio spectrum hands you the accelerator physics of a source megaparsecs away — Liénard's 1898 engineering journal result, repurposed as the universe's electron spectrometer.