Scalar and Vector Potentials, Gauge Freedom

The Idea in Plain Language

Imagine trying to describe the flow of water in a complex pipe system. Instead of tracking the velocity of every water molecule (the field), you might describe the pressure difference between points (the potential). The pressure is a simpler quantity; the flow is derived from how the pressure changes in space and time. Similarly, the electric and magnetic fields are the "flow" of the electromagnetic interaction. The scalar potential \(\phi\) and vector potential \(\mathbf{A}\) are the underlying "pressure" variables. Crucially, adding a constant to the pressure everywhere does not change the flow. In electromagnetism, we can add specific mathematical functions to the potentials without changing the physical fields at all. This redundancy is called gauge freedom. It tells us that the potentials are not directly observable; only the fields \(\mathbf{E}\) and \(\mathbf{B}\) are physical. This distinction is vital because it means the potentials are mathematical tools, not physical entities in the classical sense, though they play a crucial role in quantum mechanics.

Defining the Potentials

The homogeneous Maxwell equations are \(\nabla \cdot \mathbf{B} = 0\) and \(\nabla \times \mathbf{E} = -\frac{\partial \mathbf{B}}{\partial t}\). The first equation states that the magnetic field has no sources or sinks; it is solenoidal. A fundamental theorem of vector calculus states that if the divergence of a vector field is zero, that field can be expressed as the curl of another vector field. Therefore, we define the vector potential \(\mathbf{A}\) such that:

Substituting this into the second homogeneous equation, we get \(\nabla \times \mathbf{E} = -\frac{\partial}{\partial t} (\nabla \times \mathbf{A})\). Rearranging, we find \(\nabla \times (\mathbf{E} + \frac{\partial \mathbf{A}}{\partial t}) = 0\). A vector field with zero curl is conservative and can be written as the negative gradient of a scalar function. Thus, we define the scalar potential \(\phi\):

These definitions automatically satisfy the two homogeneous Maxwell equations. The inhomogeneous equations, \(\nabla \cdot \mathbf{E} = \rho/\epsilon_0\) and \(\nabla \times \mathbf{B} = \mu_0 \mathbf{J} + \mu_0 \epsilon_0 \frac{\partial \mathbf{E}}{\partial t}\), now serve as the equations of motion for the potentials \(\phi\) and \(\mathbf{A}\).

Gauge Freedom

The potentials \(\phi\) and \(\mathbf{A}\) are not unique. Suppose we have a valid pair \((\phi, \mathbf{A})\). Consider a new pair defined by an arbitrary smooth scalar function \(\chi(\mathbf{r}, t)\):

Let us verify that the physical fields remain unchanged. For the magnetic field:

Since the curl of a gradient is always zero (\(\nabla \times \nabla \chi = 0\)), we have \(\mathbf{B}' = \mathbf{B}\). For the electric field:

Expanding the terms:

Assuming the spatial and temporal derivatives commute (which is true for smooth functions), \(\nabla (\frac{\partial \chi}{\partial t}) = \frac{\partial}{\partial t} (\nabla \chi)\). These two terms cancel exactly, leaving \(\mathbf{E}' = -\nabla \phi - \frac{\partial \mathbf{A}}{\partial t} = \mathbf{E}\). Thus, the transformation leaves the physics invariant. This is the gauge freedom.

Choosing a Gauge

While the potentials are not unique, we can choose a specific "gauge" to simplify the equations of motion. The most common choice is the Coulomb gauge, defined by \(\nabla \cdot \mathbf{A} = 0\). In this gauge, the scalar potential satisfies Poisson's equation, \(\nabla^2 \phi = -\rho/\epsilon_0\), which is instantaneous and resembles electrostatics. The vector potential satisfies a wave equation with a source term. Another important choice is the Lorenz gauge (note the 'z'), defined by:

In the Lorenz gauge, both \(\phi\) and \(\mathbf{A}\) satisfy the same inhomogeneous wave equation:

This symmetry is crucial for special relativity, as it treats space and time on equal footing. The choice of gauge does not change the physical predictions, but it changes the mathematical form of the solutions.