1

I'm trying to prove that the Transformation of the electric and magnetic field under a boost of arbitrary direction is $$ E_i^{\prime\parallel}=E_i^\parallel\qquad B_i^{\prime\parallel}=B_i^\parallel $$ $$ E_i^{\prime\perp}=\Gamma(E_i^\perp+\epsilon_{ijk}V_jB_k^\perp)\qquad B_i^\perp=\Gamma(B_i^\perp-\epsilon_{ijk}V_jE_k^\perp) $$ To prove this what I did was to start from $$ F_{\mu\nu}^\prime=\Lambda_\mu^\alpha\Lambda_\nu^\beta F_{\alpha\beta} $$ And got $$ E_i^\prime=(\Lambda^0_0\Lambda^i_j-\Lambda^i_0\Lambda^0_j)E_j+\epsilon_{jkl}\Lambda^0_j\Lambda^i_kB_l $$ $$ B_i^\prime=\epsilon_{ijk}\Lambda^j_0\Lambda^k_lE_l+\frac{1}{2}\epsilon_{ijk}\epsilon_{lmn}\Lambda^j_l\Lambda^k_mB_n $$ The Lorentz transform for a boost in an arbitrary direction is $$ \Lambda_\mu^\nu= \begin{pmatrix} \Lambda^0_0 & \Lambda^0_j\\ \Lambda^i_0 & \Lambda^i_j \end{pmatrix}= \begin{pmatrix} \Gamma & -\Gamma V_j\\ -\Gamma V_i & P_{ij}^\perp+\Gamma P_{ij}^\parallel \end{pmatrix} $$ Substituting in the expression of the electric field I get this $$ E_i^\prime=(\Gamma P_{ij}^\perp+\Gamma^2 P_{ij}^\parallel-\Gamma^2V^2P_{ij}^\parallel)E_j+\epsilon_{kjl}\Gamma V_j(P_{ik}^\perp+\Gamma P_{ik}^\parallel)B_l $$ And my question comes here. The last term of that expression should be $$ \epsilon_{kjl}\Gamma V_j(P_{ik}^\perp+\Gamma P_{ik}^\parallel)B_l=\Gamma \epsilon_{ijk}V_jB_k^\perp $$ But I don't see why, also, how is it that only the magnetic field was multiplied by the projector and not the velocity? If I write this in another way we have $$ \Gamma(P_{ik}^\perp+\Gamma P_{ik}^\parallel)(V\times B)_k $$ That cross product will yield a vector perpendicular to the velocity, so the parallel projector times that vector will be 0 and we get $$ \Gamma P_{ik}^\perp(V\times B)_k $$ But here, why is it that the projector only affects the magnetic field?

Note: I'm using the adimensional units, where C=1.

Frobenius
  • 15,613
  • In my answer here Is it a typo in David Tong's derivation of spin-orbit interaction? for the Lorentz boost shown in the Figure, equations (03), the Lorentz transformation of the electromagnetic field is given the following equations therein.... – Frobenius Sep 04 '21 at 16:25
  • ...\begin{align} \mathbf{E}' &=\gamma \mathbf E-\dfrac{\gamma^2}{c^2 \left(\gamma+1\right)}\left(\mathbf{E}\boldsymbol{\cdot} \boldsymbol{\upsilon}\right)\boldsymbol{\upsilon},\boldsymbol{+},\gamma\left(\boldsymbol{\upsilon}\boldsymbol{\times}\mathbf{B}\right) \tag{04a}\label{04a}\ \mathbf{B}' &= \gamma \mathbf{B}-\dfrac{\gamma^2}{c^2 \left(\gamma+1\right)}\left(\mathbf{B}\boldsymbol{\cdot} \boldsymbol{\upsilon}\right)\boldsymbol{\upsilon}\boldsymbol{-}!\dfrac{\gamma}{c^2}\left(\boldsymbol{\upsilon}\boldsymbol{\times}\mathbf{E}\right) \tag{04b}\label{04b} \end{align} – Frobenius Sep 04 '21 at 16:31
  • In Lorentz transformations the sequence of indices matters. Therefore writing $\Lambda_\mu^\alpha$ is ambiguous. You probably mean $\Lambda_\mu{}^\alpha$. (In MathJax you can write it as \Lambda_\mu{}^\alpha.) – Thomas Fritsch Sep 05 '21 at 19:33
  • thanks, @ThomasFritsch I dint know that trick. But you are absolutely right. But the transform equation is right. The problem is the projection of the cross product, do you know why it only affects the magnetic field? – Carlos Andrés del Valle Sep 05 '21 at 22:37

0 Answers0