Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Is partial trace still cyclic?

  1. Sep 11, 2008 #1

    I know trace is usually cyclic, but is partial trace cyclic too? Why?


  2. jcsd
  3. Sep 11, 2008 #2
    Ok... so I know that it isn't cyclic now ... just by picking a random example, but if anyone knows the reason why it's not cyclic, and has a general proof as to why it's not, I'd be very grateful to hear it!

  4. Sep 12, 2008 #3


    User Avatar
    Science Advisor
    Homework Helper

    What's the definition of a partial trace?
  5. Sep 12, 2008 #4
    Normal trace is equivalent to the sum of the eigenvalues (or diagonal elements) of a matrix. Partial trace acts only on part of the system, so for a density matrix.. say it's a pure state but entangled,

    [tex]\rho_{AB}=\frac{1}{2}(|01\rangle +|10\rangle )(\langle 01|+\langle 10 |)[/tex]

    The partial trace over subsystem B gives the reduced density matrix [tex]\rho_A[/tex], so [tex]Tr_B(\rho_{AB})=\rho_A[/tex]


    [tex]\rho_A=_B\langle 0 |\rho_{AB}|0\rangle_B +_B\langle 1 |\rho_{AB}|1\rangle_B[/tex]
    [tex]\rho_A=|1\rangle \langle 1 | + |0\rangle \langle 0 | [/tex]

    My question is how do I prove that

    [tex]Tr_B (\rho \sigma) = Tr_B (\sigma \rho)[/tex]

    where [tex]\rho[/tex] and [tex]\sigma[/tex] are both density matrices of a system AB.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook