1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Proof of a Lorentz Transformation

  1. Oct 17, 2012 #1
    1. The problem statement, all variables and given/known data
    Show that the following is a Lorentz Transform:
    [itex] \Lambda _{j}^{i}=\delta _{j}^{i}+v^iv_j\frac{\gamma -1}{v^2}[/itex]
    [itex] \Lambda _{j}^{0}=\gamma v_j , \Lambda _{0}^{0}=\gamma , \Lambda _{0}^{i}=\gamma v^i[/itex]

    where [itex]v^2 =\vec{v}\cdot \vec{v}[/itex], and [itex]\delta _{j}^{i}[/itex] is the Kronecker Delta.
    2. Relevant equations
    [itex]\eta_{\mu\nu}=\eta_{\mu'\nu'}\Lambda_{\mu}^{\mu'} \Lambda_{\nu}^{\nu'}[/itex]
    [itex]\eta = \Lambda^T \eta \Lambda[/itex]

    3. The attempt at a solution
    I know how to go about proving a transform is a Lorentz transform, based on my "relevant equations", but I'm having a hard time setting the [itex]\Lambda[/itex] matrix up correctly. When I set up the matrix, I have terms in every cell, such as
    [itex]\Lambda_{1}^{1}=1+v^1 v_1 \frac{\gamma -1}{v^2}[/itex]
    [itex]\Lambda_{1}^{2}=v^2 v_1 \frac{\gamma -1}{v^2}[/itex]

    and so on and so forth, but this feels wrong. I end up having to multiply two exceedingly complicated matrices along the way, which I know to be wrong (the professor hinted that excessive matrix multiplication was a sign you were doing the problem wrong.) How do I set things us? What I really want to know is, what is [itex]\Lambda_{j}^{i}[/itex]? How do I handle the vector indices (vi, vj)?
  2. jcsd
  3. Oct 17, 2012 #2
    I think you should do this, let [itex]g_{\mu\nu}=\eta_{\mu'\nu'}\Lambda_{\mu}^{\mu'} \Lambda_{\nu}^{\nu'}[/itex] then you sholud show that [itex]g_{\mu\nu}=\eta_{\mu\nu}[/itex]. This you can do calculating for each case [itex]g_{00},g_{0k}\, and\, g_{kl}[/itex] using
    [tex]g_{\mu\nu}=\eta_{\mu'\nu'}\Lambda_{\mu}^{\mu'} \Lambda_{\nu}^{\nu'}=\eta_{00}\Lambda_{\nu}^{0}\Lambda_{\mu}^{0}+\eta_{ij}\Lambda_{\nu}^{i}\Lambda_{\mu}^{j}[/tex]
    [tex]=-\Lambda_{\nu}^{0}\Lambda_{\mu}^{0}+\delta_{ij}\Lambda_{\nu}^{i} \Lambda_{\mu}^{j}=-\Lambda_{\nu}^{0}\Lambda_{\mu}^{0}+\Lambda_{\nu}^i\Lambda_{\mu}^i[/tex]
    Last edited: Oct 17, 2012
  4. Oct 17, 2012 #3
    Ok, that makes some sense to me. I'll give it a try, thank you!

    The only part I couldn't follow is where you came up with the [itex]\delta_{ij} \Lambda_{\nu}^{i} \Lambda_{\mu}^{j}[/itex]. Where did the delta come from? Sorry, I'm quite new at this sort of math.
  5. Oct 18, 2012 #4
    assuming that latin indices take on values 1,2,3 while greek indices 0,1,2,3 then [itex]\eta_{ik}=\delta_{ik}[/itex] while [itex]\eta_{00}=-1[/itex]
  6. Oct 19, 2012 #5
    Ah, right, thank you!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook