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: General QM question

  1. Oct 29, 2007 #1
    Hmm well this question isn't general except in the sense that I have no idea about how to solve it and hence have no idea what physics it involves!

    Two particles

    particle one obeys H_1 = (p_1)^2/(2m_1) + V1(r_1)
    particle two obeys H_2 = (p_2)^2/(2m_2) + V2(r_2)

    The system obeys H = H_1 + H_2 + V(|r1 - r2|)

    (the latter is the relative distance between the two particles' positions)
    r_1 and r_2 are scalars, r1 and r2 are the respective vectors

    1.) Find an operator that commutes with H_1 and state its equation of motion
    2.) Find an operator that commutes with H_2 and state its equation of motion
    3.) Find an operator that commutes with H (and prove that it does)

    How do I go about doing this? I can only think of trivial examples, like H_2 commutes with H_1 and vice-versa, and nothing for H.

    There is no other information given, so I can't assume this is the helium atom or something.
  2. jcsd
  3. Oct 29, 2007 #2
    One non-trivial example of operator satisfying condition 1.) is angular momentum
    [itex]\mathbf{L}_1 = [\mathbf{r}_1 \times \mathbf{p}_1] [/itex]. This may give you a clue on how to approach questions 2.) and 3.).

  4. Oct 29, 2007 #3
    Thanks! Ugh! In fact I had considered angular momentum but I thought that its form was particular to the hydrogen atom. But I guess since there is no phi-dependence on the potential it should work fine. So L1 and L2 work for 1.) and 2.), but how about #3? Will the total angular momentum be conserved? If so, how is it defined? Is it just L = L1 + L2? I tried writing out the Hamiltonians in terms of reduced mass but that didn't seem to do anything.

    Is there any classical mechanics reasoning I can use? e.g. with Hamilton's equations?
    Last edited: Oct 29, 2007
  5. Oct 29, 2007 #4
    Yes, I think L = L1+L2 should work. You can try to calculate [L,H]=0 explicitly. There is, however, a nicer approach. Try to prove the following equivalent condition

    [tex] H = \exp(\frac{i}{\hbar} (\mathbf{L} \cdot \vec{\phi})) H \exp(-\frac{i}{\hbar} (\mathbf{L} \cdot \vec{\phi})) [/tex]

    by noticing that for any vector operator [itex] \mathbf{a} [/itex]

    [tex] \exp(\frac{i}{\hbar} (\mathbf{L} \cdot \vec{\phi})) \mathbf{a} \exp(-\frac{i}{\hbar} (\mathbf{L} \cdot \vec{\phi})) [/tex]

    is the result of rotation of this vector around axis [itex] \vec{\phi} [/itex].

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