Opening the book…
The Biot–Savart law gives the field of any current, but the integrals can be quite challenging. Just as Gauss’s law made electric fields easy to find when the charges have enough symmetry (Chapter 17), a second law makes magnetic fields easy to find when the currents are symmetric. It concerns the circulation of \(\vec{B}\) around a closed path.
Circulation around a wire. Take a circle of radius \(r\) around a long straight wire. The field of the wire is tangent to the circle everywhere, with constant magnitude \(\mu_0 I/2\pi r\). Divide the circle into short steps \(d\vec{s}\) and add up \(\vec{B}\cdot d\vec{s}\) all the way around:
\[\oint\vec{B}\cdot d\vec{s} = B\,(2\pi r) = \frac{\mu_0 I}{2\pi r}(2\pi r) = \mu_0 I.\]
The radius has dropped out. It turns out that the result depends only on the current passing through the path, not on the shape of the path at all, and that currents outside the path contribute nothing. This is Ampère’s law:
Key Equation: Ampère’s law
\[\oint\vec{B}\cdot d\vec{s} = \mu_0 I_{\text{enc}}. \tag{23.8}\]
The integral is taken around any closed path, called an Ampèrian loop, and \(I_{\text{enc}}\) is the net current passing through any surface bounded by that path. Signs follow the right-hand rule: curl the fingers of your right hand around the loop in the direction you traverse it, and currents in the direction of your thumb count as positive (Figure 23.7).
Ampère’s law is true for any steady current and any loop, but, like Gauss’s law, it is useful for finding \(\vec{B}\) only when the symmetry lets us choose a loop along which \(\vec{B}\) is either tangent with constant magnitude, or perpendicular, so that \(\vec{B}\) can be taken outside the integral. The main cases are the long straight wire, the solenoid, and the toroid. In its current form, Ampère’s law is begging for a modification. It will make its return in Chapter 27, where its complete form tells us something profound about the underlying nature of our universe.
Inside a wire. Let a long straight wire of radius \(R\) carry current \(I\) spread uniformly over its cross-section (Figure 23.8). By symmetry, the field lines are circles centered on the axis. Outside the wire, a circular Ampèrian loop encloses the whole current, and we recover \(B = \mu_0 I/2\pi r\). Inside, a circle of radius \(r < R\) encloses only the fraction \(\pi r^2/\pi R^2\) of the current, so
\[B(2\pi r) = \mu_0 I\,\frac{r^2}{R^2} \qquad\Longrightarrow\qquad B = \frac{\mu_0 I r}{2\pi R^2} \quad (r \le R).\]
The field grows linearly from zero at the axis to its maximum at the surface, then falls off as \(1/r\) outside.
Example 23.5: Inside and outside a wire
A copper wire of radius \(1.0\,\mathrm{mm}\) carries a uniform current of \(10\,\mathrm{A}\). Find the field (a) at its surface, (b) halfway to its axis, and (c) \(5.0\,\mathrm{mm}\) from its axis.
Solution
(a) At the surface, an Ampèrian loop of radius \(R\) just encloses the whole current, so the inside and outside formulas agree: \(B = \mu_0 I/2\pi R\). It helps to note that \(\mu_0/2\pi = 2\times 10^{-7}\,\mathrm{T\cdot m/A}\). Then
\[B = \frac{\mu_0 I}{2\pi R} = \frac{(2\times10^{-7}\,\mathrm{T\cdot m/A})(10\,\mathrm{A})}{1.0\times10^{-3}\,\mathrm{m}} = \boxed{2.0\,\mathrm{mT}}.\]
(b) Halfway to the axis, \(r = R/2 = 0.50\,\mathrm{mm}\). A loop of this radius encloses only the fraction \((r/R)^2 = 1/4\) of the current, \(2.5\,\mathrm{A}\), but it is also only half as long as the loop at the surface. Using the inside result \(B = \mu_0 I r/2\pi R^2\),
\[B = \frac{(2\times10^{-7}\,\mathrm{T\cdot m/A})(10\,\mathrm{A})(0.50\times10^{-3}\,\mathrm{m})}{(1.0\times10^{-3}\,\mathrm{m})^2} = \boxed{1.0\,\mathrm{mT}}.\]
Because \(B\) is proportional to \(r\) inside the wire, halving \(r\) halves the field.
(c) At \(r = 5.0\,\mathrm{mm}\) the loop is outside the wire and encloses all \(10\,\mathrm{A}\), so
\[B = \frac{\mu_0 I}{2\pi r} = \frac{(2\times10^{-7}\,\mathrm{T\cdot m/A})(10\,\mathrm{A})}{5.0\times10^{-3}\,\mathrm{m}} = \boxed{0.40\,\mathrm{mT}},\]
one-fifth of the surface value, since this point is five times as far from the axis.
Check. The field is largest right at the surface and smaller on both sides of it, just as the graph in Figure 23.8b shows.
Example 23.6: A coaxial cable
A coaxial cable has an inner conductor carrying current \(I\) one way and an outer cylindrical conductor, surrounding it, carrying the same current \(I\) back. Use Ampère’s law to find the field (a) between the conductors and (b) outside the cable.
Solution
Take circular Ampèrian loops centered on the axis.
(a) Between the conductors, the loop encloses only the inner current, so \(B = \mu_0 I/2\pi r\), the field of the inner wire alone.
(b) Outside the cable, the loop encloses \(I - I = 0\). By symmetry \(B\) has the same magnitude all around the loop, so \(B(2\pi r) = 0\) and \(\boxed{B = 0}\). A coaxial cable confines its magnetic field entirely between its conductors. This is why coax is used to carry television and radio-frequency signals: its fields neither leak out to interfere with other equipment nor pick up interference from outside.
Checkpoint 23.3
Three long parallel wires pass through a closed Ampèrian loop, carrying \(5\,\mathrm{A}\) upward, \(3\,\mathrm{A}\) downward, and \(2\,\mathrm{A}\) upward; a fourth wire, carrying \(10\,\mathrm{A}\) upward, lies outside the loop. (a) What is \(\oint\vec{B}\cdot d\vec{s}\) around the loop, taking upward current as positive? (b) Does the fourth wire affect the field at points on the loop? Does it affect the integral?
Answers to Checkpoints
Checkpoint 23.3. (a) Ampère’s law counts only the current passing through the loop: \(\oint\vec{B}\cdot d\vec{s} = \mu_0 I_{\rm enc} = \mu_0(5 - 3 + 2)\,\mathrm{A} = \mu_0(4\,\mathrm{A})\). (b) The fourth wire does contribute to the field at every point on the loop: its field lines pass right through that region. But it does not change the integral: its field adds as much to \(\oint\vec{B}\cdot d\vec{s}\) along one part of the loop as it subtracts along another, so its net contribution is zero. The field depends on all currents; the integral depends only on the enclosed ones.
This page titled 23.4: Ampère’s Law was last modified on Thu, 12 Nov 2026 04:26:50 GMT and is shared under a CC BY 4.0 license and was authored, remixed, and/or curated by Michael Ragsdale.
The book could not be loaded just now.