Uniqueness of charge distribution

85 Views Asked by At

Some exercises in general physics ask for students to find specific charge distributions on the boundaries of given conductors in various of situations. Usual answers goes like follows; firstly using symmetries or other techniques, e.g. image charges, to 'guess' how the distribution looks like, and then completes solution testing the boundary conditions, vanishing electric fields, etc. Almost surely, a mathematically oriented one may suspect that other possibilities maybe possible. So this question is almost surely duplicated. So let me state precise questions in the following form.

Question 0 Is question 1 duplicates or nonsense? If not, ignore question 0. If so please indicate the places where this question appeared or the reason it is trivially wrong or true.

Question 1 Is the following statement is true: Let $V_e: \mathbb{R}^3 \longrightarrow \mathbb{R}$ be a smooth function which vanishes at infinity uniformly. Let $C$ be a finite union of a collection of pairwise disjoint bounded domains say $C_1$, $C_2$, $\ldots$, $C_n$ in $\mathbb{R}^3$ whose boundary $S$ is a union of smooth surfaces $S_1$, $S_2$, $\ldots$, $S_n$ which are mutually non-overlapping(that is for any $i \neq j$, $S_i \cap S_j$ has no interior in $S_i$ with respect to the induced topology on $S_i$). For each $j=1,2, \ldots , n$, let $Q_j$ be a real number which represents the total charge on $S_j$. Then there is unique surface distribution $\sigma$ on $S$ so that the potential $$V(y) = \int_S \frac{\sigma(x)}{\hbox{dist}(x,y)} dx + V_e(y) $$ is constant on each of connected component of $C$ where $\int_{S_j} \sigma (x) dx =Q_j$ for each $j=1,2, \ldots, n$