Divergence of B and No Magnetic Monopoles

The Idea in Plain Language

Imagine a bar magnet. If you draw the magnetic field lines outside the magnet, they emerge from the north pole and curve around to enter the south pole. Inside the magnet, the lines continue from the south pole back to the north pole. There are no "ends" to these lines. You cannot find a point in space where a magnetic field line starts or stops. This is different from an electric field, where lines start on positive charges and stop on negative charges. The statement that magnetic field lines never begin or end is the physical content of the equation \(\nabla \cdot \mathbf{B} = 0\). It means that if you were to place a tiny closed surface anywhere in space, the total amount of magnetic field entering the surface must exactly equal the total amount leaving it. There is no "source" or "sink" of magnetic field inside the volume.

Mathematical Statement and Flux

The divergence of a vector field measures the net flux per unit volume. For the magnetic field \(\mathbf{B}\), the condition that there are no magnetic monopoles is expressed as:

By the divergence theorem, this local statement is equivalent to the global statement that the net magnetic flux through any closed surface \(S\) is zero:

Here, \(d\mathbf{A}\) is the vector area element pointing outward from the surface. The dot product \(\mathbf{B} \cdot d\mathbf{A}\) represents the flux of \(\mathbf{B}\) through that element. The integral sums this flux over the entire closed surface. The result being zero means that any field line entering the surface must exit it somewhere else. This is a topological constraint on the field lines: they must be closed loops or extend to infinity, but they cannot terminate in empty space.

Derivation from Biot–Savart

We do not assume \(\nabla \cdot \mathbf{B} = 0\) as a separate postulate. Instead, it follows from the Biot–Savart law for a steady current distribution. The magnetic field at a point \(\mathbf{r}\) due to a current element \(I d\mathbf{l}'\) at \(\mathbf{r}'\) is:

To find the divergence, note that \(\nabla\) acts on the field point \(\mathbf{r}\), so \(d\mathbf{l}'\) is a constant vector as far as \(\nabla\) is concerned. Using \(\nabla \cdot (\mathbf{A} \times \mathbf{C}) = \mathbf{C} \cdot (\nabla \times \mathbf{A}) - \mathbf{A} \cdot (\nabla \times \mathbf{C})\) with \(\mathbf{A} = d\mathbf{l}'\) (so \(\nabla\times\mathbf{A} = 0\)) and \(\mathbf{C} = (\mathbf{r} - \mathbf{r}')/|\mathbf{r} - \mathbf{r}'|^3\): this kernel is the gradient \(-\nabla(1/|\mathbf{r}-\mathbf{r}'|)\), and the curl of a gradient is zero. Both terms vanish, so \(\nabla\cdot d\mathbf{B} = 0\); summing over the whole current distribution gives \(\nabla \cdot \mathbf{B} = 0\) everywhere.

Comparison with Electrostatics

In electrostatics, Gauss's law states:

Here, \(\rho\) is the electric charge density. If \(\rho\) is non-zero, electric field lines can begin or end. This allows for the existence of electric monopoles (isolated charges). In contrast, the magnetic analogue has no source term. There is no magnetic charge density \(\rho_m\) such that \(\nabla \cdot \mathbf{B} = \mu_0 \rho_m\). The absence of such a term is what we mean by "no magnetic monopoles." This asymmetry is one of the key differences between electric and magnetic fields in classical electromagnetism. It implies that magnetic fields are always solenoidal (divergence-free), whereas electric fields are generally not.