Vector potential

The vector potential A is the magnetic counterpart of the electric scalar potential V. Where V satisfies E = −∇V (the electric field is the negative gradient of a scalar function), A satisfies B = ∇×A (the magnetic field is the curl of a vector function). The whole reason A exists is the divergence-free property of B: since ∇·B = 0 and the divergence of any curl is automatically zero, every magnetic field can be written as the curl of some vector field, and A is that vector field.

Operationally, A is often easier to compute than B. For a localised current distribution, A at a point r is given by the integral A(r) = (μ₀/4π) ∫ J(r') / |r−r'| dV', which has the same 1/|r−r'| structure as the electric potential integral and adds vectorially without the cross-products that make Biot–Savart messy. Once you have A, take its curl to get B. For a long straight wire, A points along the wire and falls off as ln(r); take the curl, and you get the familiar 1/r tangential B field. For a magnetic dipole, A has a clean 1/r² structure and reproduces the dipole B field directly.

The vector potential becomes essential — not optional — once you leave classical electromagnetism. In quantum mechanics, the canonical momentum of a charged particle is p − qA rather than just p, which means A appears directly in the Schrödinger equation through the gauge-covariant derivative. The Aharonov–Bohm effect (1959) shows that A produces measurable interference effects in regions where B is exactly zero — proving that A is more than just a calculational convenience. In quantum field theory, A is the photon field, and gauge symmetry built around it is the architectural principle of the Standard Model.