4
$\begingroup$

I tried to proof that the Maxwell equations are invariant under parity transformations. Therefore I used the covariant formulation of the Maxwell equations

\begin{align} \partial_{\nu}F^{\nu\mu} &= \frac{4\pi}{c}j^{\mu}\\ \partial_{\nu}\tilde{F}^{\nu\mu} &= 0 \end{align} and the parity transformation given by

\begin{align} P = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & -1 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & -1 \end{pmatrix} \end{align}

Regarding only the first equation $\partial_{\nu}F^{\nu\mu} = \frac{4\pi}{c}j^{\mu}$ we have

\begin{align} \begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix} \cdot \begin{pmatrix} \frac{1}{c}\frac{\partial}{\partial \text{t}} \\ \vec{\nabla} \end{pmatrix} = \begin{pmatrix} \frac{1}{c}\frac{\partial}{\partial \text{t}} \\ -\vec{\nabla} \end{pmatrix} \end{align} as well as \begin{align} \begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix} \cdot \begin{pmatrix} c\rho \\ \vec{j} \end{pmatrix} = \begin{pmatrix} c\rho \\ -\vec{j} \end{pmatrix} \end{align}

and

\begin{align} P \cdot F^{\nu\mu} &= \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & -1 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & -1 \end{pmatrix}\begin{pmatrix} 0 & -E^1 & -E^2 & -E^3 \\ E^1 & 0 & -B^3 & B^2 \\ E^2 & B^3 & 0 & -B^1 \\ E^3 & -B^2 & B^1 & 0 \end{pmatrix} = \begin{pmatrix} 0 & -E^1 & -E^2 & -E^3 \\ -E^1 & 0 & B^3 & -B^2 \\ -E^2 & -B^3 & 0 & B^1 \\ -E^3 & B^2 & -B^1 & 0 \end{pmatrix} \end{align}

Based on these calculations, is there a way to see that Maxwell equations are invariant under parity transformations and if so how do I see it?

$\endgroup$

2 Answers 2

8
$\begingroup$

You are not applying the transformations correctly. Your transformation, $P$, is linear map that changes a vector into another vector. Well, $F^{\mu\nu}$ is a rank (2,0) tensor, not a vector (rank (1,0) tensor). This all becomes much clearer if you use index notation, rather than writing matricies. I will work in Cartesian basis.

So, let us denote your $P$ with a rank (1,1) tensor, $P^\mu_\nu$, such that for a vector $V^\mu=\left(V^0, V^1, V^2, V^3\right)^\mu$, the effect of $P$ would be:

$V^\mu\to\bar{V}^\mu = P^\mu_\nu V^\nu = (V^0, -V^1, -V^2, -V^3)$

Now the transformation for the electromagnetic tensor would be then:

$F^{\mu\nu}\to\bar{F}^{\mu\nu}=P^\mu_\eta P^\nu_\sigma F^{\eta\sigma}$

In this case, because your transformation is simple, you can rewrite this as a matrix equation (in general, it may not work so flawlesly):

$\bar{F}=P\cdot F \cdot P = \left(\begin{matrix} 1 & 0 & 0 & 0 \\ 0 & -1 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & -1 \\ \end{matrix}\right) \left(\begin{matrix} 0 & -E_x & -E_y & -E_z \\ E_x & 0 & -B_z & B_y \\ E_y & B_z & 0 & -B_x \\ E_z & -B_y & B_x & 0 \\ \end{matrix}\right) \left(\begin{matrix} 1 & 0 & 0 & 0 \\ 0 & -1 & 0 & 0 \\ 0 & 0 & -1 & 0 \\ 0 & 0 & 0 & -1 \\ \end{matrix}\right)=\left(\begin{matrix} 0 & E_x & E_y & E_z \\ -E_x & 0 & -B_z & B_y \\ -E_y & B_z & 0 & -B_x \\ -E_z & -B_y & B_x & 0 \\ \end{matrix}\right)$

So electric field transforms as a polar vector, whilst magnetic as axial. As is well known. You will also need to apply transform $J^\mu\to\bar{J}^\mu=P^\mu_\nu J^{\nu}$, and $\partial_\mu \to \bar{\partial}_\mu = \left(P^{-1}\right)^\nu_\mu \partial_\nu$ (i.e. you need to apply inverse transforms to co-variant tensors). When you do that, you will see that the equation for the new quantities will be the same as for the old ones, i.e. $\bar{\partial}_\mu \bar{F}^{\mu\nu}={4\pi}{c}\bar{J}^{\nu}$, so the equation does not change (vectors&tensors do)


Extra following the comment.

I really advise against matrix notation in these calculations, so I will stop using it. Here is how you work out $\bar{F}^{\mu\nu}$:

$F^{0\{1,2,3\}}=\{-E_x, -E_y, -E_z\}$

$F^{12}=-B_z,\, F^{13}=B_y,\, F^{23}=-B_x$

So, using $P^0_0=1, P^1_1=P^2_2=P^3_3=-1$, and zero otherwise:

$\bar{F}^{0i}=P^0_0 P^i_j F^{0j} = (1)(-1)F^{0i}$, so electric field gets a minus

$\bar{F}^{ij}=P^i_s P^j_k F^{sk}=(-1)^2 F^{ij}$, so magnetic field is unchanged

Regarding the equation, the term on LHS goes as $\partial_{\mu}F^{\mu\nu}\to\left(P^{-1}\right)^{\kappa}_\mu\partial_{\kappa}P^\mu_\sigma P^\nu_\rho F^{\sigma\rho}=\delta^\kappa_\sigma\partial_{\kappa} P^\nu_\rho F^{\sigma\rho}=P^\nu_\rho \partial_{\mu} F^{\mu\rho}$. You will find that the RHS transforms the same way, so the effect of transformation is to multiply the whole equation by an invertible operator $P$. Thus equation does not change.

$\endgroup$
1
  • $\begingroup$ Why do you get $\bar{F} = P \cdot F \cdot P$ out of $F^{\mu\nu}\to\bar{F}^{\mu\nu}=P^\mu_\eta P^\nu_\sigma F^{\eta\sigma}$? Afterwards you say that I have to apply $J^{\mu} \to \bar{J}^{\mu} = P^\mu_\nu J^\nu$ and $\partial_\mu \to \bar{\partial}_\mu = (P^{-1})^\nu_\mu\partial_\nu$. As $P^{-1} = P$ that's what I did, when I asked the question, didn't I? So taking your result for $\bar{F}$ and my results for the other two transformations I still don't quite see why the Maxwell equations are invariant under parity transformation. Could you explain it a bit more detailed? $\endgroup$
    – offline
    Feb 9, 2019 at 6:48
-3
$\begingroup$

As current is a vector, it is not invariant under parity. Therefore neither is Ampère's law.

$\endgroup$
0

Your Answer

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

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