I am currently reading through my electrodynamics lecture notes and I can't understand a calculation.

$$ E = \frac{1}{2} \int \mathrm{d}^3x \, \phi(\mathbf{x}) \rho(\mathbf{x}) = - \frac{\epsilon_0}{8\pi} \int \mathrm{d}^3x \, \phi(\mathbf{x}) \Delta \phi(\mathbf{x}) \\ = \frac{\epsilon_0}{8\pi} \int \mathrm{d}^3x \, \nabla \phi(\mathbf{x}) \nabla \phi(\mathbf{x}) = \frac{1}{8\pi} \int \mathrm{d}^3x \, \mathbf{D} \cdot \mathbf{E}$$

I know the identity $\nabla ( f \nabla g) = \nabla f \cdot \nabla g + f \Delta g$ which I could easily prove by looking at the components and summing. I think I have to use this identity from the first to second line in the above equation but I don't know why $\int \mathrm{d}^3x \, \nabla (\phi \nabla \phi) = 0$. I tried using $\nabla \phi = -\mathbf{E}$ and the divergence theorem but I couldn't show that.

The question is pretty specific and you don't get much insight from it but I would really like some help.


That integral is zero because it's over all space. As you mentioned, by the divergence theorem:

$$\int_{\mathbb R^3} d^3 \mathbf x \boldsymbol\nabla . \big(\phi(\mathbf x) \boldsymbol \nabla \phi(\mathbf x)\big) = \oint_{\partial {\mathbb R^3}} d \mathbf a \ .\phi(\mathbf x) \boldsymbol \nabla \phi(\mathbf x)$$ Now if your integral is over all space, which it is, the boundary is "at infinity". The potential goes to zero at infinity; making the right boundary integral go to zero. Thus, $$\int_{\mathbb R^3} d^3 \mathbf x \boldsymbol\nabla . \big(\phi(\mathbf x) \boldsymbol \nabla \phi(\mathbf x)\big)=0$$


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.