The formulae for the force on an electric dipole and the force on a magnetic dipole seem to be closely related:

$$\mathbf{F}_{e} = (\mathbf{p} \cdot \mathbf{\nabla}) \mathbf{E}, $$

$$ \mathbf{F}_{m} = \mathbf{\nabla}(\mathbf{m} \cdot \mathbf{B}) $$

where $\mathbf{p}$ is the electric dipole moment and $\mathbf{m}$ is the magnetic dipole moment.

However, we cannot rewrite $\mathbf{F}_e$ generally as $\mathbf{\nabla} (\mathbf{p} \cdot \mathbf{E})$, so these two formulae are not analogous. Why must the $\mathbf{m}$ be operated on by $\nabla$, but the $\mathbf{p}$ must not? What fundamental difference between $\mathbf{p}$ and $\mathbf{m}$ causes this?

  • $\begingroup$ Remark: $\nabla$ does not operate on $\vec m$ either, as $\vec m$ is not a function of space. $\endgroup$ Nov 23, 2015 at 17:30

1 Answer 1


They are the same (at least in the static limit). Using the rule $\vec a \times (\vec b \times \vec c) = \vec b (\vec a \vec c) - \vec c (\vec a \vec b)$ one can write: $$\vec F_m = \nabla (\vec m \cdot \vec B) = \vec m \times (\nabla \times \vec B) + (\vec m \cdot \nabla)\vec B.$$ The first term is zero, as $\nabla \times \vec B = \frac 1 {c^2} \partial_t \vec E + \mu_0 \vec j$ and

  1. These equations only hold in the quasi-static case anyway (that is, we ignore radiation effects, which amounts to approximating $\frac 1 {c^2} \partial_t \vec E \approx 0$),

  2. The $\vec B$ in this equation is actually the external magnetic field (not including the field of the dipole which is divergent at the dipole's position), therefore the source term $\vec j$ will be zero.


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.