In short

Ampère's circuital law says \oint \vec{B}\cdot d\vec{\ell} = \mu_0 I_\text{enc}. For a capacitor being charged by a wire, you can draw two different surfaces bounded by the same Amperian loop: one pierced by the wire (so I_\text{enc} = I), one passing between the capacitor plates where no current flows (so I_\text{enc} = 0). Ampère's law gives two different answers for the same \oint \vec{B}\cdot d\vec{\ell} — a logical contradiction.

Maxwell's fix: a changing electric flux acts like a current. Define the displacement current

\boxed{\;I_d \;=\; \varepsilon_0\,\frac{d\Phi_E}{dt}\;}

and amend Ampère's law to

\boxed{\;\oint \vec{B}\cdot d\vec{\ell} \;=\; \mu_0\left(I_\text{enc} + \varepsilon_0\,\frac{d\Phi_E}{dt}\right)\;}

Between the capacitor plates, no conduction current flows, but the electric field is growing, so d\Phi_E/dt \neq 0 — and exactly that accounts for the missing current. Both surfaces now give the same answer.

Maxwell's four equations (in integral form), the compact description of all electromagnetism:

\oint \vec{E}\cdot d\vec{A} \;=\; \frac{Q_\text{enc}}{\varepsilon_0}\qquad (\text{Gauss, E-field})
\oint \vec{B}\cdot d\vec{A} \;=\; 0\qquad (\text{Gauss, B-field — no magnetic monopoles})
\oint \vec{E}\cdot d\vec{\ell} \;=\; -\frac{d\Phi_B}{dt}\qquad (\text{Faraday})
\oint \vec{B}\cdot d\vec{\ell} \;=\; \mu_0 I_\text{enc} + \mu_0\varepsilon_0\,\frac{d\Phi_E}{dt}\qquad (\text{Ampère–Maxwell})

Consequence. A changing \vec{E} creates a \vec{B} (Ampère–Maxwell). A changing \vec{B} creates an \vec{E} (Faraday). The two fields regenerate each other, propagating as a self-sustaining wave. The wave speed falls out of the two constants alone:

\boxed{\;c \;=\; \frac{1}{\sqrt{\mu_0\varepsilon_0}} \;=\; 2.998\times 10^8\text{ m/s}\;}

— the measured speed of light. Light is an electromagnetic wave, and Maxwell's equations say so.

On a monsoon evening in Mumbai, lightning flashes over the Arabian Sea. The flash is not yet at your window. A few milliseconds later, light reaches your eyes; a few seconds later, thunder reaches your ears. The difference is about 300{,}000 times — sound travels at 343 m/s, light at nearly 300 million. But how did the light travel? Light is a wave, and every wave we know about before this chapter — waves on a rope, ripples on water, sound — needed a medium. What is the medium for light? Nineteenth-century physicists invented an "ether" to hold the answer, but never measured it. The correct answer is stranger and more beautiful: light is an electromagnetic wave, and the medium is not matter but the electric and magnetic fields themselves, which regenerate each other as they propagate through empty space.

This chapter builds the physics that makes light possible. It starts with a paradox — Ampère's law breaks for a charging capacitor — and ends with four equations on a single page that describe every electromagnetic phenomenon in the universe: static charges, permanent magnets, currents, radio waves, X-rays, and light itself. The missing piece was a single term. Maxwell added it in 1865, and the number that falls out of the equations, 1/\sqrt{\mu_0\varepsilon_0}, turned out to match the measured speed of light. That coincidence is the most important result in classical physics.

The inconsistency in Ampère's law

Take a parallel-plate capacitor being charged by a steady current I flowing in a wire. (Think of a 12 V battery charging up a 100 \muF capacitor through a 1 k\Omega resistor — the current is non-zero while the voltage is rising.) Draw an Amperian loop — a small closed circle — around the wire some distance away from the plates. Apply Ampère's circuital law:

\oint \vec{B}\cdot d\vec{\ell} \;=\; \mu_0 I_\text{enc}

Why: Ampère's law says the magnetic circulation around any closed loop equals \mu_0 times the total current piercing any surface bounded by that loop. The phrase "any surface" is load-bearing here — it is what creates the paradox.

Two surfaces, two answers

Pick the Amperian loop such that it encircles the wire between the capacitor and the battery. Now choose a surface that has this loop as its boundary. Two obvious choices:

Surface 1. A flat disc through which the wire passes. Current through this surface is I_1 = I, the conduction current in the wire.

Surface 2. A bag-shaped surface that bulges outward and passes between the two capacitor plates without cutting the wire. Current through this surface is I_2 = 0 — there is no moving charge in the gap between the plates.

Two Amperian surfaces bounded by the same loop give different enclosed currents A wire carries current I left to right into the left plate of a parallel-plate capacitor. A second wire emerges from the right plate. An Amperian loop is drawn around the wire to the left of the capacitor. Surface 1 is a flat disc cut by the wire. Surface 2 is a curved surface that bulges to the right and passes between the two capacitor plates without cutting any wire. Both surfaces are bounded by the same loop. battery I +Q −Q Amperian loop Surface 1: flat disc (cuts the wire) Surface 2: bag between plates (passes between plates — no wire pierces it)
One Amperian loop, two surfaces. Surface 1 is a flat disc cut by the wire ($I_\text{enc} = I$). Surface 2 is a bag that bulges around the right plate and passes *between* the plates, cutting no wire at all ($I_\text{enc} = 0$). Ampère's law as stated cannot give both answers simultaneously.

Plug into Ampère's law:

Using Surface 1: \oint \vec{B}\cdot d\vec{\ell} = \mu_0 I.

Using Surface 2: \oint \vec{B}\cdot d\vec{\ell} = 0.

The left-hand side is the same number — it depends only on the loop, not on the surface we pick to compute I_\text{enc}. The right-hand side gives two different values. Ampère's law is inconsistent for time-varying situations. The inconsistency is not a small error; it is a logical contradiction that appears every time you charge a capacitor.

Where the "missing current" could come from

What is happening between the plates while the capacitor charges? No charge physically crosses the gap, but the electric field between the plates is growing. If the field is building up, something is changing in the gap that an honest equation should be able to detect. Let us compute.

Step 1. Surface charge density on the positive plate as the capacitor charges.

\sigma(t) = \frac{Q(t)}{A}

where A is the plate area.

Step 2. Electric field between the plates (neglecting fringing).

E(t) = \frac{\sigma(t)}{\varepsilon_0} = \frac{Q(t)}{\varepsilon_0 A}

Step 3. Electric flux through any surface that lies between the plates and has area A (the field is approximately uniform across that area).

\Phi_E(t) = E(t)\,A = \frac{Q(t)}{\varepsilon_0}

Step 4. Rate of change of flux.

\frac{d\Phi_E}{dt} = \frac{1}{\varepsilon_0}\,\frac{dQ}{dt} = \frac{I}{\varepsilon_0}

Why: the current charging the plate is by definition the rate of accumulation of charge, I = dQ/dt. So the rate of change of electric flux in the gap is proportional to the current in the wire outside it — and the proportionality constant is 1/\varepsilon_0.

Step 5. Define the displacement current to match.

I_d \;\equiv\; \varepsilon_0\,\frac{d\Phi_E}{dt} \;=\; \varepsilon_0\cdot\frac{I}{\varepsilon_0} \;=\; I

Why: defining I_d this way makes it exactly equal to the conduction current I in the wire — so whether Surface 1 (conduction current I) or Surface 2 (displacement current I_d = I) is chosen, the total "effective current" piercing the surface is the same.

The displacement current is not a real current of moving charges. Between the plates there is no charge flow, no heating, no moving electrons. But the changing electric flux plays the same role in Ampère's law that a conduction current does — and in particular, it creates a magnetic field. Maxwell's insight was to take this as a fundamental feature of nature, not a mathematical trick.

The Ampère–Maxwell law

With I_d added, Ampère's law becomes

\boxed{\;\oint \vec{B}\cdot d\vec{\ell} \;=\; \mu_0(I_\text{enc} + I_d) \;=\; \mu_0 I_\text{enc} + \mu_0\varepsilon_0\,\frac{d\Phi_E}{dt}\;}

This is the Ampère–Maxwell law. It says: a magnetic field is produced both by a conduction current and by a changing electric field. The second term is a genuine new prediction of physics, not a book-keeping device.

Verifying the consistency

Recompute our two-surface example. The Amperian loop around the wire outside the capacitor encloses:

Both surfaces now give \mu_0 I on the right-hand side. The paradox is resolved. The \vec{B}-field around the wire is the same whether you compute it from the wire's own current or from the changing electric flux between the plates — they are two faces of the same phenomenon.

The magnetic field between the plates

Maxwell's correction is not just book-keeping; it predicts a measurable magnetic field between the capacitor plates, even though no current flows there. Apply the Ampère–Maxwell law to a circular loop of radius r < R (where R is the plate radius), centred on the axis of the capacitor, lying between the plates:

B(r)\,(2\pi r) \;=\; \mu_0\varepsilon_0\,\frac{d\Phi_E}{dt}

The flux through the loop is E(t)\,\pi r^2 (uniform field), so d\Phi_E/dt = \pi r^2\,dE/dt. Then

B(r) \;=\; \frac{\mu_0\varepsilon_0 r}{2}\,\frac{dE}{dt}

This is a real, measurable, circular magnetic field. For a capacitor charged at dE/dt = 10^9 V/(m·s) — typical of a pulsed-power system — at r = 1 cm, B \approx 5.6\times 10^{-11} T. Small, but detectable with a sensitive pickup coil. And this was the experimental confirmation that the displacement current is physical, not just a bookkeeping device.

Maxwell's four equations — the complete set

With Ampère corrected, the four laws of electromagnetism settle into their final form. In integral form, for a volume V with surface \partial V, and a surface S with boundary \partial S:

\oint_{\partial V} \vec{E}\cdot d\vec{A} \;=\; \frac{Q_\text{enc}}{\varepsilon_0} \qquad \text{(Gauss's law for }\vec{E}\text{)}

Why: the total electric flux out of a closed surface equals the enclosed charge divided by \varepsilon_0. Equivalently: electric field lines start on positive charges and end on negative charges.

\oint_{\partial V} \vec{B}\cdot d\vec{A} \;=\; 0 \qquad \text{(Gauss's law for }\vec{B}\text{)}

Why: the total magnetic flux out of any closed surface is zero. Magnetic field lines are closed loops — they never start or end anywhere, which says there are no isolated magnetic monopoles in nature (so far as anyone has ever found).

\oint_{\partial S} \vec{E}\cdot d\vec{\ell} \;=\; -\frac{d\Phi_B}{dt} \qquad \text{(Faraday's law)}

Why: the circulation of \vec{E} around a closed loop equals the rate of change of magnetic flux through the loop (with a sign that encodes Lenz's law). A changing magnetic field creates a circulating electric field.

\oint_{\partial S} \vec{B}\cdot d\vec{\ell} \;=\; \mu_0 I_\text{enc} + \mu_0\varepsilon_0\,\frac{d\Phi_E}{dt} \qquad \text{(Ampère–Maxwell law)}

Why: the circulation of \vec{B} around a closed loop equals \mu_0 times the total current through the loop — where "current" now includes both real moving charges and the displacement current from a changing electric field.

Every result in classical electromagnetism — Coulomb's law, the magnetic field of a wire, the force on a moving charge, the EMF in a rotating coil, radio transmission, the pressure of sunlight on a photovoltaic panel — follows from these four equations plus the Lorentz force law \vec{F} = q\vec{E} + q\vec{v}\times\vec{B}.

Electromagnetic waves — the existence of light

Here is the punchline. Take Maxwell's equations in a region of empty space — no charges, no currents. Then Q_\text{enc} = 0 and I_\text{enc} = 0, and the four equations become:

\oint \vec{E}\cdot d\vec{A} = 0,\qquad \oint \vec{B}\cdot d\vec{A} = 0
\oint \vec{E}\cdot d\vec{\ell} = -\frac{d\Phi_B}{dt},\qquad \oint \vec{B}\cdot d\vec{\ell} = \mu_0\varepsilon_0\,\frac{d\Phi_E}{dt}

The last two are perfectly coupled. A changing magnetic field creates an electric field (Faraday). A changing electric field creates a magnetic field (Ampère–Maxwell). So a changing \vec{B} produces an \vec{E}, which is itself changing, which produces a new \vec{B}, which is changing, and so on — a self-sustaining oscillation that propagates through space without any charges or currents to drive it. An electromagnetic wave.

The wave speed from dimensional analysis

What speed does the wave propagate at? The only two constants in the equations are \mu_0 and \varepsilon_0. Their product has units:

[\mu_0\varepsilon_0] = \frac{\text{T}\cdot\text{m/A}}{1}\cdot\frac{\text{C}^2}{\text{N}\cdot\text{m}^2} = \frac{\text{s}^2}{\text{m}^2}

Why: using [\mu_0] = N/A² and [\varepsilon_0] = C²/(N·m²), the product comes out to s²/m². So \sqrt{\mu_0\varepsilon_0} has units of s/m — the reciprocal of a velocity.

Therefore 1/\sqrt{\mu_0\varepsilon_0} has units of velocity. Plug in the measured constants:

\mu_0 = 4\pi\times 10^{-7}\text{ T·m/A} \approx 1.2566\times 10^{-6}
\varepsilon_0 \approx 8.854\times 10^{-12}\text{ C}^2/(\text{N·m}^2)
\mu_0\varepsilon_0 \approx 1.2566\times 10^{-6}\times 8.854\times 10^{-12} = 1.1127\times 10^{-17}
c \;=\; \frac{1}{\sqrt{\mu_0\varepsilon_0}} \;=\; \frac{1}{\sqrt{1.1127\times 10^{-17}}} \;=\; \frac{1}{3.336\times 10^{-9}} \;\approx\; 2.998\times 10^8\text{ m/s}

That is the speed of light, measured by Fizeau in 1849 with a rotating toothed wheel and known to astronomers before that from Rømer's observations of Jupiter's moons in 1676. Maxwell's equations contain that number without any reference to light or optics. The conclusion was unavoidable: light is an electromagnetic wave.

A qualitative picture: the transverse E and B fields

An electromagnetic plane wave travelling in the +x direction has

\vec{E}(x, t) = E_0\sin(kx - \omega t)\,\hat{y},\qquad \vec{B}(x, t) = B_0\sin(kx - \omega t)\,\hat{z}

with E_0/B_0 = c and \omega/k = c. The electric and magnetic fields oscillate perpendicular to each other and perpendicular to the direction of propagation. They are in phase — both peak together, both cross zero together.

Animated: travelling electromagnetic wave A red marker traces an electric-field sine wave in space as it moves to the right, and a dark marker below traces an in-phase magnetic-field sine wave. Both move at a constant speed, representing the wave propagating in the plus x direction. x (propagation direction, arbitrary units) E (red, upward) / B (dark, downward) 0 +1 −1
Two markers moving at the same speed along $+x$. The red dot traces the electric-field amplitude as a sine wave; the dark dot traces the magnetic-field amplitude (drawn on the opposite side to make the perpendicular directions clear — in reality $\vec{B}$ is perpendicular to $\vec{E}$, both perpendicular to $\hat{x}$). $\vec{E}$ and $\vec{B}$ peak together and cross zero together: they are in phase, and the whole pattern moves together at $c$.

Three key features of the wave:

  1. Transverse: \vec{E} \perp \vec{B} \perp \hat{x}. Unlike sound (longitudinal), light wiggles sideways to its direction of travel.
  2. In phase: \vec{E} and \vec{B} peak together. Neither leads nor lags — they are partners.
  3. Self-sustaining: no medium, no charges, no currents. The fields create each other.

The wave equation derivation (sketch)

Start with Faraday and Ampère–Maxwell in differential form (without proof here — see the going deeper section of any university electrodynamics text). Take the curl of Faraday's law, substitute Ampère–Maxwell, and you obtain

\nabla^2 \vec{E} \;=\; \mu_0\varepsilon_0\,\frac{\partial^2 \vec{E}}{\partial t^2}

This is the wave equation in three dimensions, with propagation speed v = 1/\sqrt{\mu_0\varepsilon_0}. Any disturbance in \vec{E} propagates at that speed. The same procedure starting from Ampère–Maxwell gives an identical equation for \vec{B}. Light, radio, X-rays, and gamma rays all satisfy this equation; they differ only in their frequency. For a Mumbai radio station broadcasting at 94.3 MHz (Radio Mirchi), \omega = 2\pi\times 94.3\times 10^6 rad/s and the wavelength \lambda = c/f = 3\times 10^8/94.3\times 10^6 \approx 3.18 m. You tune in because the receiver's antenna picks up the oscillating electric field of that wave.

Worked examples

Example 1: Displacement current in a charging capacitor

A parallel-plate capacitor has circular plates of radius R = 5.0 cm separated by a small gap. It is being charged so that the voltage across it rises at dV/dt = 2.0\times 10^6 V/s. Find (a) the rate of change of electric field dE/dt between the plates (given plate separation d = 1.0 mm), (b) the displacement current I_d through the gap, and (c) the magnetic field at radius r = 2.0 cm from the axis, midway between the plates.

Charging capacitor with Amperian loop inside the gap Side view of a circular parallel-plate capacitor. Plates of radius 5 cm separated by 1 mm. A dashed circle of radius 2 cm is drawn between the plates, concentric with the plate axis. An arrow indicates the growing electric field pointing from top plate to bottom plate. An arrow indicates the induced circular magnetic field tangent to the Amperian loop. +Q plate (R = 5 cm) −Q plate E (growing) Amperian loop (r = 2 cm) B d = 1 mm
The growing $\vec{E}$ field inside the capacitor creates an induced circular $\vec{B}$ field, exactly as predicted by the Ampère–Maxwell law applied to an Amperian loop of radius $r$ between the plates.

Step 1. Rate of change of E.

Between the plates, E = V/d, so

\frac{dE}{dt} = \frac{1}{d}\,\frac{dV}{dt} = \frac{2.0\times 10^6}{1.0\times 10^{-3}} = 2.0\times 10^9\text{ V/(m·s)}

Why: the field depends on voltage linearly with proportionality 1/d; differentiating, the same factor carries through.

Step 2. Displacement current through the whole gap.

The electric flux through a surface the size of the plate (area A = \pi R^2) is \Phi_E = E\,A, so

I_d = \varepsilon_0\,\frac{d\Phi_E}{dt} = \varepsilon_0\,A\,\frac{dE}{dt} = 8.854\times 10^{-12}\times \pi\times (0.05)^2\times 2.0\times 10^9
= 8.854\times 10^{-12}\times 7.854\times 10^{-3}\times 2.0\times 10^9
= 1.39\times 10^{-4}\text{ A} \;\approx\; 139\ \mu\text{A}

Why: this is exactly the conduction current in the wire charging the capacitor — I = C\,dV/dt. You can check: for a 1.0 mm gap, 5 cm radius capacitor, C = \varepsilon_0 A/d = 8.854\times 10^{-12}\times 7.854\times 10^{-3}/10^{-3} = 69.5 pF, and C\,dV/dt = 69.5\times 10^{-12}\times 2\times 10^6 = 139\ \muA. The two match exactly.

Step 3. Magnetic field at r = 2 cm from the axis.

Apply Ampère–Maxwell to a circular Amperian loop of radius r centred on the axis, with a flat disc as its surface:

B(r)\,(2\pi r) = \mu_0\varepsilon_0\,\frac{d\Phi_E^{(r)}}{dt}

where \Phi_E^{(r)} = E\,\pi r^2 is the flux through the loop's disc. So

B(r) = \frac{\mu_0\varepsilon_0 r}{2}\,\frac{dE}{dt}
= \frac{(4\pi\times 10^{-7})(8.854\times 10^{-12})(0.02)}{2}\times 2.0\times 10^9
= \frac{4\pi\times 10^{-7}\times 1.7708\times 10^{-13}\times 0.02}{2}\times 2\times 10^9
\approx \frac{4.449\times 10^{-21}}{2}\times 2\times 10^9 \approx 4.45\times 10^{-12}\text{ T}

Result. dE/dt = 2.0\times 10^9 V/(m·s); I_d \approx 139\ \muA; B(r = 2\text{ cm}) \approx 4.5\times 10^{-12} T.

What this shows. The displacement current is exactly equal to the conduction current — they have to be, because the same Amperian loop sees them through different surfaces. And the induced magnetic field inside the gap is tiny but non-zero, exactly as Maxwell predicted. A sensitive SQUID magnetometer at BARC could measure fields of this order directly.

Example 2: Wavelength of a Mumbai lightning pulse

A lightning bolt near the Mumbai coast produces an electromagnetic pulse with a sharp rise time of \tau \approx 1\ \mus. Estimate (a) the characteristic frequency content of the pulse, (b) the corresponding wavelength, and (c) how long the light from the bolt takes to travel 10 km to your window.

Step 1. Frequency from rise time.

A pulse with rise time \tau has significant Fourier content up to frequencies of order f \sim 1/(2\pi\tau).

f \;\sim\; \frac{1}{2\pi\times 10^{-6}} \;\approx\; 1.6\times 10^5\text{ Hz} \;=\; 160\text{ kHz}

Why: a sharp rising edge is a superposition of sinusoids, and the highest significant frequency is roughly the reciprocal of the rise time. This is the rule of thumb used to spec oscilloscope bandwidth: to see a 1 ns edge, you need at least 1 GHz bandwidth.

Step 2. Wavelength of the dominant component.

\lambda = \frac{c}{f} = \frac{3\times 10^8}{1.6\times 10^5} \;\approx\; 1.9\text{ km}

Why: the VLF and LF portion of the spectrum (3 kHz – 300 kHz) has wavelengths of kilometres. Lightning pulses, in particular the sferics, are a major source of natural VLF radio, which propagates worldwide by bouncing between the ground and the ionosphere.

Step 3. Travel time for a 10 km path.

t = \frac{\text{distance}}{c} = \frac{10{,}000}{3\times 10^8} \;=\; 3.3\times 10^{-5}\text{ s} \;=\; 33\ \mu\text{s}

Result. f \approx 160 kHz; \lambda \approx 1.9 km; t_\text{light} \approx 33\ \mus. Sound, at 343 m/s, takes 10000/343 \approx 29 s — nearly 10^6 times longer. If you see lightning and count to 3 before hearing thunder, the bolt was about 1 km away. Light, obeying Maxwell's equations at 1/\sqrt{\mu_0\varepsilon_0}, arrives essentially instantaneously.

What this shows. The same four equations that govern the slow charging of a 100 \muF capacitor in a laboratory also describe the atmospheric VLF pulse radiated by a lightning bolt over the Arabian Sea. The speed of that pulse is set by two constants — \mu_0 and \varepsilon_0 — that were measured in electrostatics and magnetostatics experiments long before anyone suspected they had anything to do with light. That unified the three classical fields (electricity, magnetism, optics) into one.

Common confusions

You have the fix, the four equations, and the prediction of electromagnetic waves. What follows is JEE-Advanced-grade extensions — the differential forms, the Poynting vector, electromagnetic momentum, and the connection to special relativity.

Maxwell's equations in differential form

Using the divergence theorem and Stokes's theorem, the integral forms transform into differential forms that hold at every point in space:

\nabla\cdot\vec{E} = \frac{\rho}{\varepsilon_0},\qquad \nabla\cdot\vec{B} = 0
\nabla\times\vec{E} = -\frac{\partial\vec{B}}{\partial t},\qquad \nabla\times\vec{B} = \mu_0\vec{J} + \mu_0\varepsilon_0\,\frac{\partial\vec{E}}{\partial t}

where \rho is the charge density and \vec{J} is the current density. These four lines contain all of electromagnetism. The vacuum wave equation drops out in two steps: take the curl of Faraday's law, substitute Ampère–Maxwell with \vec{J} = 0, and use the identity \nabla\times(\nabla\times\vec{E}) = \nabla(\nabla\cdot\vec{E}) - \nabla^2\vec{E}. With \nabla\cdot\vec{E} = 0 in vacuum,

\nabla^2\vec{E} = \mu_0\varepsilon_0\,\frac{\partial^2\vec{E}}{\partial t^2}

which is the wave equation with speed c = 1/\sqrt{\mu_0\varepsilon_0}. The same procedure for \vec{B} gives the same equation.

The Poynting vector — energy flow

The electromagnetic field carries energy. Density:

u \;=\; \frac{\varepsilon_0}{2}|\vec{E}|^2 + \frac{1}{2\mu_0}|\vec{B}|^2

Energy flux (energy per unit area per unit time) is given by the Poynting vector

\vec{S} \;=\; \frac{1}{\mu_0}\vec{E}\times\vec{B}

For a plane wave, the time-averaged magnitude is \langle|\vec{S}|\rangle = E_0^2/(2\mu_0 c). Solar radiation reaching the top of the atmosphere at the equator has \langle|\vec{S}|\rangle \approx 1361 W/m² — the solar constant. On Indian rooftops at Bengaluru latitude, the typical incident irradiance on a south-facing panel at noon is about 950 W/m² (after atmospheric absorption). Multiply by collection efficiency (~20% for Si PV) and you get the rated power of a 1 m² solar cell.

Electromagnetic momentum and radiation pressure

The electromagnetic field also carries momentum. Density:

\vec{g} = \frac{\vec{S}}{c^2} = \varepsilon_0\vec{E}\times\vec{B}

A wave hitting a perfect absorber deposits its momentum as force per unit area — radiation pressure — equal to \langle|\vec{S}|\rangle/c. For sunlight at the Earth's surface, this is 950/(3\times 10^8) \approx 3.2\times 10^{-6} Pa, or about one-billionth of atmospheric pressure. Too small to feel, but large enough to deflect comet tails (pushing dust grains outward from the Sun) and to drive solar sails (Japan's IKAROS mission demonstrated this in 2010, and ISRO's proposed solar-sail technology demonstrator builds on the same principle).

Lorentz invariance and the non-independence of E and B

Maxwell's equations are automatically invariant under Lorentz transformations — and this is forced on them by the speed of light being frame-independent. A pure electrostatic field \vec{E} in one frame appears as a combination of \vec{E}' and \vec{B}' in another frame moving relative to the first. Specifically, for a boost in the x-direction at speed v:

E_x' = E_x,\qquad E_y' = \gamma(E_y - vB_z),\qquad E_z' = \gamma(E_z + vB_y)
B_x' = B_x,\qquad B_y' = \gamma\left(B_y + \frac{v}{c^2}E_z\right),\qquad B_z' = \gamma\left(B_z - \frac{v}{c^2}E_y\right)

These mix \vec{E} and \vec{B} in a way that preserves the combined field equations. \vec{E} and \vec{B} are not two independent vector fields; they are components of a single rank-2 antisymmetric tensor F^{\mu\nu} in four-dimensional spacetime. This unification is one of the most beautiful results in physics.

The GMRT connection

At the Giant Metrewave Radio Telescope (GMRT) near Pune, thirty 45-metre-diameter parabolic antennas pick up radio waves from deep space at frequencies between 50 MHz and 1.5 GHz — wavelengths from 6 m down to 20 cm. Each of those waves is a real-time manifestation of the Ampère–Maxwell law: billions of years ago, a changing current in a distant galaxy's ionised plasma generated an electromagnetic wave, which propagated across the universe at c, and today the wavefront arrives at Khodad as an oscillating \vec{E} field that drives an AC current in the antenna feed. The entire chain — source, propagation, reception — is Maxwell's equations from start to finish.

Dispersion relations in media

In vacuum, all frequencies travel at the same c. In a medium, electrons and ions respond to the oscillating \vec{E} with a frequency-dependent lag, and the effective propagation speed becomes v(\omega) = c/n(\omega) where n(\omega) is the refractive index. The frequency-dependence is what produces a rainbow: different colours refract through slightly different angles in water droplets because n(\text{red}) \neq n(\text{violet}). This is another face of the Ampère–Maxwell law, with bound-electron currents (polarisation) adding to the displacement current.

The four equations in one line

The entire machinery can be written in four-vector notation as just two equations:

\partial_\mu F^{\mu\nu} = \mu_0 J^\nu,\qquad \partial_{[\mu} F_{\nu\lambda]} = 0

The first contains Gauss's law for \vec{E} and the Ampère–Maxwell law; the second contains Gauss's law for \vec{B} and Faraday's law. Two equations describing all of classical electromagnetism. Maxwell wrote his original treatise in twenty equations; the modern form compresses them further, and the final tensor form is shorter still. The physics did not change — only the notation. But the compression makes the underlying unity visible.

Where this leads next