I want to calculate the potential of a point charge via Gauss's law
My Idea:
we can find an expression for $4\pi$ that has the form of an integral,
by using spherical coordinates we can derive:
$$\int_{V}\vec{\nabla}^2 \left( \frac{1}{r} \right)dV =\int_V \vec{\nabla}\cdot\left( \vec{\nabla} \frac{1}{r} \right)dV$$
by using the Divergence theorem and applying spherical coordinates we obtain: $$\oint_S \vec{\nabla} \left(\frac{1}{r}\right) \cdot \vec{n}dS = \oint_S \partial_r \left( \frac{1}{r} \right)r^2\sin(\theta) d\theta dr = -4\pi$$
$$\Rightarrow \vec{\nabla}\cdot \vec{E}=-\int_V \vec{\nabla}^2 \frac{\rho(\vec{r})}{|\vec{r}-\vec{r}'|}dV = -\vec{\nabla}^2\int_V \frac{\rho(\vec{r})}{|\vec{r}-\vec{r}'|}dV$$
since following expression is known: $$\vec{\nabla}\cdot \vec{E}=-\vec{\nabla}^2\phi$$
we can deduce that:
$$\Rightarrow \phi(\vec{r})=\int_V \frac{\rho(\vec{r}')}{|\vec{r}-\vec{r}'|}dV$$
My problem is that I'm not sure how to solve this integral. I'm aware that the charge density for discrete charges is:
$$\rho(\vec{r})=\sum_i^nq_i \delta(\vec{r}-\vec{r}')$$
and since we are talking about a single charge at the origin ($\vec{r}'=0$) this sum will equal $q$.
$$\Rightarrow \phi(\vec{r})=\int_V \frac{q}{r}dV=q\int_0^{2\pi}\int_0^\pi \int_0^R \frac{1}{r}r^2\sin(\theta) \space d\theta d\varphi dr$$
However, this delivers:
$$\phi(\vec{r})=2q\pi r^2$$
Which is of course wrong. Any help is appreciated


Your reasoning is correct until your equation $$\rho(\vec{r})=\sum_i^nq_i \delta(\vec{r}-\vec{r}'_i)$$ (I've added an index $i$ here for clarity, because $\vec{r}'_i$ denotes the place of charge $q_i$)

But then, for the special case of a single charge $q$ at the origin ($\vec{r}' = \vec{0}$) this sum will be equal $$\rho(\vec{r})=q \delta(\vec{r})$$

From this it follows by integration $$\begin{align} \phi(\vec{r})&=\int_V \frac{\rho(\vec{r}')}{|\vec{r}-\vec{r}'|}dV \\ &=\int_V \frac{q\delta(\vec{r}')}{|\vec{r}-\vec{r}'|}dV \\ &=\frac{q}{|\vec{r}|} \end{align}$$ (The last step here used the definition of the $\delta$ function)

| cite | improve this answer | |
  • $\begingroup$ If I am not mistaken we should have $\int_V \frac{q}{r} \delta(\vec{r})dV$ at $\vec{r}'=0$. what happened to the integral? It seems like $\int_V \frac{q}{r} \delta (\vec{r})dV = \frac{q}{r}$ but that doesn't seem right $\endgroup$ – Alessio Popovic Nov 4 '19 at 21:52
  • $\begingroup$ @AlessioPopovic See my edit, I've refined the last part into some smaller steps $\endgroup$ – Thomas Fritsch Nov 4 '19 at 22:11

Note that $\mathbf{\nabla}\cdot\mathbf{E}=4\pi\rho(\mathbf{r})$ is equivalent to $$\begin{align} \int_V \mathbf{\nabla}\cdot\mathbf{E}\, d\tau & = 4\pi\int_V \rho\,d\tau \\ \oint_S \mathbf{E}\cdot d\mathbf{a}&=4\pi \,Q_{enc} \end{align}$$ by the divergence theorem. Define a sphere around the point. Then, $\mathbf{E}$ is spherically symmetric, constant, and comes out of the integral. If we say the point has charge $q$, then we have $$ \begin{align} \oint_S |\mathbf{E}|\,da &= 4\pi q \\ |\mathbf{E}|\oint da &= 4\pi q \\ |\mathbf{E}|\cdot 4\pi r^2 &= 4\pi q \\ |\mathbf{E}| & = \frac{q}{r^2} \\ \mathbf{E} & = \frac{q}{r^2}\hat{\mathbf{r}} \end{align} $$ Now, we have $$V(\mathbf{r})=-\int^\mathbf{r}_\infty \mathbf{E}\cdot d\mathbf{l}$$ Therefore, $$\begin{align} V(\mathbf{r})&=-\int^\mathbf{r}_\infty \frac{q}{r'^2}\, dr' \\ &= \frac{q}{r'}\bigg|_\infty^r \\ &= \frac{q}{r} \end{align}$$

| cite | improve this answer | |
  • 1
    $\begingroup$ you are absolutely right but I was actually hinting with my approach hoping to find a more general solution which is compatible with multiple charges. In other words a system consisting of a dipole, quadrupole etc. $\endgroup$ – Alessio Popovic Nov 5 '19 at 8:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.