1. Not finding help here? Sign up for a free 30min 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!

Prove Vector Quadruple Product with Levi-Civita/Index Notation

  1. Sep 6, 2014 #1
    I'm asked to prove the following using Levi-Civita/index notation:
    (\mathbf{a \times b} )\mathbf{\times} (\mathbf{c}\times \mathbf{d}) = [\mathbf{a,\ b, \ d}] \mathbf c - [\mathbf{a,\ b, \ c}] \mathbf d \

    I'm able to prove it using triple product identities, but I'm completely stuck with the index notation. I was previously able to prove Lagrange's Identity with index notation, but applying similar concepts I just get stuck on the first step with the quadruple product.

    Using the same first step of proving Lagrange's identity, I transformed [itex]
    (\mathbf{a \times b} )[/itex] into [itex]\varepsilon_{ijk} a^j b^k[/itex] and [itex]
    (\mathbf{c \times d} )[/itex] into [itex]\varepsilon_{imn} c^m d^n[/itex] but then I'm just left with [itex](\varepsilon_{ijk} a^j b^k) \mathbf{\times} (\varepsilon_{imn} c^m d^n)[/itex] which is seemingly unhelpful.

    I also tried letting AxB = W and CxD = Z and transforming WxZ to index notation. Then I tried to 'un-nest' the original cross products in index notation, but it quickly ended up in a place where I couldn't understand what the different indexes represented.

    Any help would be appreciated. Thanks.
  2. jcsd
  3. Sep 6, 2014 #2
    I am not entirely sure how you defined ##[\mathbf{a},\mathbf{b},\mathbf{c}]##. Probably as a determinant? Either way, you must have ##[\mathbf{a},\mathbf{b}, \mathbf{c}] = \mathbf{a} \cdot (\mathbf{b} \times \mathbf{c})##. That might be helpful.

    There is also an identity that relates ##\epsilon_{ijk}\epsilon_{lmk}## to Kronecker deltas. It might also prove useful.
  4. Sep 7, 2014 #3
    Yes, [a,b,c]=a⋅(b×c). I'm sure it will be helpful, but only after I can get past the initial steps.

    I'm aware of the Kronecker delta identity you refer to, I used it to prove the scalar quadruple product/Lagrange's Identity as part of the same assignment. I don't doubt that that will be involved as well, but again, only after I can get past this initial step.

    I don't know how to combine [itex](\varepsilon_{ijk} a^j b^k) \mathbf{\times} (\varepsilon_{imn} c^m d^n)[/itex] without the index notation blowing up on me to a point where I don't understand it. It could just be that I'm approaching it from the wrong direction in the one step I did make in re-writing the two inner cross products in index notation.
  5. Sep 7, 2014 #4
    Thought I had a breakthrough after referencing another forum post. Was able to work through this:

    [(A\times B)\times(C\times D)]_{i} &= \varepsilon_{ijk}(A\times B)_{j}(C\times D)_{k} \\
    &= \varepsilon_{ijk} \varepsilon_{jmn} A_{m}B_{n} \varepsilon_{kpq} C_{p}D_{q} \\
    &= -\varepsilon_{jik} \varepsilon_{jmn} A_{m}B_{n} \varepsilon_{kpq} C_{p}D_{q} \\
    &= -(\delta_{im}\delta_{kn} - \delta_{in}\delta_{km})A_{m}B_{n} \varepsilon_{kpq} C_{p}D_{q}\\
    &= (-A_{i}B_{k} + A_{k}B_{i}) \varepsilon_{kpq} C_{p}D_{q}\\
    &= -A_{i} \varepsilon_{kpq} B_{k}C_{p}D_{q} + B_{i} \varepsilon_{kpq} A_{k}C_{p}D_{q} \\
    &= (\mathbf{A} \cdot \mathbf{C} \times \mathbf{D})\mathbf{B} - (\mathbf{B} \cdot \mathbf{C} \times \mathbf{D})\mathbf{A}\end{align*}[/itex]

    This has to be the right approach...too many things went right while I was working through it, but then I came to then end and compared it to the identity:

    [itex](\mathbf{A} \cdot \mathbf{B} \times \mathbf{D})\mathbf{C} - (\mathbf{A} \cdot \mathbf{B} \times \mathbf{C})\mathbf{D}[/itex]

    So close, yet so far. Anybody see any mistakes? My brain is fried.
  6. Sep 7, 2014 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    You didn't make a mistake, but to get the identity you want, try contracting ##\varepsilon_{ijk}## with ##\varepsilon_{kpq}## instead.
  7. Sep 7, 2014 #6

    Thanks, that does it.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted