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

Which of these statements about tensor products is incorrect?

  1. Jun 23, 2013 #1


    User Avatar
    Gold Member

    I have read the following three simplifications in various places, but together they give a contradiction, so at least one of them must be an oversimplification. Which one?
    (a) Interaction between two systems A and B is described by A[itex]\otimes[/itex]B
    (b) An entangled state C is a pure state, and hence there does not exist A and B such that A[itex]\otimes[/itex]B = C
    (c) By appropriate interactions one can produce entangled states.

  2. jcsd
  3. Jun 23, 2013 #2
    (A) is not correct. The correct statement should be
    If [itex]H_A[/itex] is the hilbertspace of system A and [itex]H_B[/itex] is space of system B, then the hilbertspace of composite system is [itex]H_A[/itex][itex]\otimes[/itex][itex]H_B[/itex]

    (B) is not correct. The correct statement should be
    If [itex]|C\rangle[/itex][itex]\in[/itex][itex]H_A[/itex][itex]\otimes[/itex][itex]H_B[/itex] represents entangled state then it can not be written in the form [itex]|A\rangle[/itex][itex]\otimes[/itex][itex]|B\rangle[/itex] where [itex]|A\rangle[/itex][itex]\in[/itex][itex]H_A[/itex] and [itex]|B\rangle[/itex][itex]\in[/itex][itex]H_B[/itex]

    (C) is correct.
  4. Jun 23, 2013 #3
    (a) interaction between two systems whose individual Hilbert spaces are [itex]H_A[/itex] and [itex]H_B[/itex] can be describe by states [itex]|\psi\rangle[/itex] belonging to the direct product of these Hilbert spaces [itex]H=H_A\otimes H_B[/itex].

    (b) an entangled state [itex]|\psi\rangle_C[/itex] belongs to this direct product Hilbert space, however, it cannot be expressed simply as a tensor product of states belonging to [itex]H_A[/itex] and [itex]H_B[/itex]. That is, for [itex]|\psi\rangle_A\in H_A[/itex] and [itex]|\psi\rangle_B\in H_B[/itex], one cannot write [itex]|\psi\rangle_C=|\psi\rangle_A\otimes|\psi\rangle_B[/itex], rather [itex]|\psi\rangle_C=\sum_{i,j}\alpha_{ij}|\psi_i\rangle_A\otimes|\psi_j \rangle_B[/itex] where [itex]\{|\psi_i\rangle_A\}\subset H_A[/itex], [itex]\{|\psi_i\rangle_B\}\subset H_B[/itex] and there MUST be more than one non-vanishing terms in the sum for an entangled state [itex]|\psi\rangle_C[/itex].

    (c) this is correct. (Edit: Actually, we can get an entangled state simply by appropriately partitioning our total system into subsystems A and B.)
    Last edited: Jun 23, 2013
  5. Jun 23, 2013 #4


    User Avatar
    Gold Member

    Thank you very much, Ravi Mohan and conana. That clears up a lot of confusion on my part. Very helpful.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook