where U is a scalar potential and C a vector potential. The first term (the gradient of U) is irrotational since taking its curl
yields zero. The second term is solenoidal since taking its divergence yields zero. Thus, the general process of defining scalar
and vector potentials for vector fields is as follows:
(1) If the vector field is curl-free (irrotational), a scalar potential may be defined which completely describes the vector field.
(2) If the vector field is divergence-free (solenoidal), a vector potential may be defined which completely describes
the vector field.
(3) For a general vector field, both a scalar and a vector potential are required to describe the vector field. The gradient of the
scalar potential is used to describe the irrotational part of the field, whereas the vector potential is used to describe
the solenoidal part of the field.
The potentials we define need not have any physical meaning, although they often do. Their definition is based on the
vector properties of the fields and may be viewed as transformations. As such, as long as the transformation is unique and is
properly defined, the potentials are valid. We will discuss here the electric scalar potential and the magnetic vector potential;
these are needed for our discussion of electromagnetic fields. There are, however, many other potential functions that may be
defined. We will only touch on some of these as examples.