1
$\begingroup$

I'm trying to understand E&M in the language of forms, and my goal is to get from the Maxwell equations $\mathrm dF=0$ and $\mathrm d\!\star\!F=\star J$ to the integral form of Ampère-Maxwell's law, namely something that looks like: $$\oint_{\partial S}\vec{B}\cdot\mathrm d\vec{\ell}=\int_S\vec J\cdot\mathrm d\vec s+\int_S\partial_t\vec E\cdot\mathrm d\vec s$$

I want to do this by integrating the second Maxwell equation, not by figuring out the differential equations. I already achieved this for Gauß's law by integrating $\mathrm d\!\star\!F$ in a volume, but i couldn't figure it out in the case of Ampère, since it looks like on the left we have the integral of a 2-form in a curve, and on the right the integral of 1-form on a surface. In 3D this wouldn't be a problem because of the duality between 1-forms and 2-forms, but in 4D, 2-forms are dual to 2-forms.

I'm clearly going about it the wrong way, any help would be appreciated.

$\endgroup$

1 Answer 1

3
$\begingroup$

It seems like you have too many misconceptions to get the answer.

You should be using the differential forms version, and the theorems in that framework, to relate everything together. You should be starting with [note sign change between Equation (1) and (2)] \begin{align} \tag1F=&E_x\,\mathrm dx\wedge\mathrm dt+E_y\,\mathrm dy\wedge\mathrm dt+E_z\,\mathrm dz\wedge\mathrm dt+B_x\,\mathrm dy\wedge\mathrm dz+B_y\,\mathrm dz\wedge\mathrm dx+B_z\,\mathrm dx\wedge\mathrm dy\\ \tag2\star F=&B_x\,\mathrm dt\wedge\mathrm dx+B_y\,\mathrm dt\wedge\mathrm dy+B_z\,\mathrm dt\wedge\mathrm dz+E_x\,\mathrm dy\wedge\mathrm dz+E_y\,\mathrm dz\wedge\mathrm dx+E_z\,\mathrm dx\wedge\mathrm dy\\ \mathrm d\!\star\!F=&+(\partial_xE_x+\partial_yE_y+\partial_zE_z)\,\mathrm dx\wedge\mathrm dy\wedge\mathrm dz-(\partial_yB_z-\partial_zB_y-\partial_tE_x)\,\mathrm dt\wedge\mathrm dy\wedge\mathrm dz\\ \tag3&-(\partial_zB_x-\partial_xB_z-\partial_tE_y)\,\mathrm dt\wedge\mathrm dz\wedge\mathrm dx-(\partial_xB_y-\partial_yB_x-\partial_tE_z)\,\mathrm dt\wedge\mathrm dx\wedge\mathrm dy\\ \tag4\star J=&-\rho\star\!\mathrm dt+j_x\star\!\mathrm dx+j_y\star\!\mathrm dy+j_z\star\!\mathrm dz\\ \tag5\therefore\ \rho=&\vec\nabla\cdot\vec E\qquad\bigwedge\qquad\vec J+\partial_t\vec E=\vec\nabla\times\vec B \end{align} where we have compared Equations (4) and (3) to get (5). Now, we can try to use $\int_\Omega\mathrm d\!\star\!F=\int_{\partial\Omega}\star F$ $$\tag{6, Gauß's Law}\int_\Omega\rho\,\mathrm dx\wedge\mathrm dy\wedge\mathrm dz=\oint_{\partial\Omega} E_x\,\mathrm dy\wedge\mathrm dz+E_y\,\mathrm dz\wedge\mathrm dx+E_z\,\mathrm dx\wedge\mathrm dy$$ $$\tag7\mathrm dt\wedge\oint_{\partial\Omega}B_x\,\mathrm dx+B_y\,\mathrm dy+B_z\,\mathrm dz=\mathrm dt\wedge\int_\Omega(j_x+\partial_tE_x)\,\mathrm dy\wedge\mathrm dz+(j_y+\partial_tE_y)\,\mathrm dz\wedge\mathrm dx\\+(j_z+\partial_tE_z)\,\mathrm dx\wedge\mathrm dy\qquad\text{(Ampère-Maxwell)}$$ That is, these are a mix of 3-forms and 2-forms, and 1-forms are not supposed to be appearing in them.

$\endgroup$
2
  • $\begingroup$ Thanks for your answer. In eq (7), are you thinking about $B_i dx^i$ as a 1-form, and the taking the wedge product with $dt$ ? Also what tipo of surface and curve are you taking? $\endgroup$ Commented Jul 21, 2023 at 22:51
  • $\begingroup$ No, I am extracting $\mathrm dt\wedge$ out of the integrals on both sides so that you can see how they connect with the usual expressions. In Gauß's Law the $\Omega$ has to be a volume and $\partial\Omega$ its enclosing surface. In Ampère-Maxwell's Law the $\Omega$ is a surface area at a particular time slice, and $\partial\Omega$ its enclosing loop at the same time slice. They are always 2-forms and 3-forms, just that it looks like 1-forms and 2-forms because we are working on one time slice. $\endgroup$ Commented Jul 22, 2023 at 5:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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