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!

Raising and Lowering Operators in the Lipkin Model

  1. Feb 22, 2014 #1
    1. The problem statement, all variables and given/known data

    I am trying to calculate the expectation value of an operator in the Lipkin model of nuclear physics. The background isn't important because my problem in really just a math problem.

    2. Relevant equations

    The anticommutation relation

    \begin{align*}
    a_{p\sigma} a_{p'\sigma'}^{\dagger} + a^{\dagger}_{p'\sigma'} a_{p\sigma} = \delta_{pp'}\delta_{\sigma\sigma'}
    \end{align*}

    Whenever an annihilation operator acts on the vacuum, you get 0.

    3. The attempt at a solution

    I will be very explicit in what I am doing.

    \begin{align*}
    \langle \Psi | H_{0} | \Psi \rangle
    &= \left[ \langle 0 | \prod_{p} \sum_{\sigma} \left( C_{\alpha\sigma}^{*} a_{p\sigma} \right) \right] \left[ \left( \frac{1}{2} \right) \epsilon \sum_{p'} \sum_{\sigma'} \left( \sigma' a^{\dagger}_{p'\sigma'} a_{p'\sigma'} \right) \right] \left[ \prod_{p''} \sum_{\sigma''} \left( C_{\alpha\sigma''} a_{p''\sigma''}^{\dagger} \right) | 0 \rangle \right] \\
    %
    &= \left( \frac{1}{2} \right) \epsilon \langle 0 | \prod_{p} \prod_{p''} \sum_{p'} \sum_{\sigma} \sum_{\sigma'} \sum_{\sigma''} C_{\alpha\sigma}^{*} C_{\alpha\sigma''} \sigma' a_{p\sigma} a^{\dagger}_{p'\sigma'} a_{p'\sigma'} a_{p''\sigma''}^{\dagger} | 0 \rangle \\
    %
    &= \left( \frac{1}{2} \right) \epsilon \langle 0 | \prod_{p} \prod_{p''} \sum_{p'} \sum_{\sigma} \sum_{\sigma'} \sum_{\sigma''} C_{\alpha\sigma}^{*} C_{\alpha\sigma''} \sigma' \left( \delta_{pp'}\delta_{\sigma\sigma'} - a_{p'\sigma'}^{\dagger} a_{p\sigma} \right) \left( \delta_{p'p''}\delta_{\sigma'\sigma''} - a_{p''\sigma''}^{\dagger} a_{p'\sigma'} \right) | 0 \rangle \\
    %
    &= \left( \frac{1}{2} \right) \epsilon \langle 0 | \prod_{p} \prod_{p''} \sum_{p'} \sum_{\sigma} \sum_{\sigma'} \sum_{\sigma''} C_{\alpha\sigma}^{*} C_{\alpha\sigma''} \sigma' \left( \delta_{pp'} \delta_{p'p''} \delta_{\sigma\sigma'} \delta_{\sigma'\sigma''} \right) | 0 \rangle \\
    &= \left( \frac{1}{2} \right) \epsilon \langle 0 | \prod_{p} \prod_{p''} \sum_{p'} \sum_{\sigma'} \sum_{\sigma} C_{\alpha\sigma}^{*} C_{\alpha\sigma'} \sigma' \left( \delta_{pp'} \delta_{p'p''} \delta_{\sigma\sigma'} \right) | 0 \rangle \\
    &= \left( \frac{1}{2} \right) \epsilon \langle 0 | \prod_{p} \prod_{p''} \sum_{p'} \sum_{\sigma'} C_{\alpha\sigma'}^{*} C_{\alpha\sigma'} \sigma' \left( \delta_{pp'} \delta_{p'p''} \right) | 0 \rangle \\
    &= \left( \frac{1}{2} \right) \epsilon \langle 0 | \prod_{p} \prod_{p''} \sum_{p'} \left( \delta_{pp'} \delta_{p'p''} \right) \sum_{\sigma'} \left( C_{\alpha\sigma'}^{*} C_{\alpha\sigma'} \sigma' \right) | 0 \rangle \\
    \end{align*}

    The possible values for p are 1, 2, 3, and 4. The possible values of sigma are -1 and 1.

    The final total answer should be

    \begin{align*}
    2 \epsilon \left( |C_{\alpha+}|^{2}-|C_{\alpha-}|^{2}\right)
    \end{align*}

    When I sum over sigma', I will get

    \begin{align*}
    |C_{\alpha+}|^{2}-|C_{\alpha-}|^{2}
    \end{align*}

    This means to get the answer I am supposed to get, everything that has to do with p, p', and p'' must equal 4. I have tried numerous times writing out explicitly with p,p',p'' = 1, 2, 3, 4 and end up with something that is very messy and that will not equal 4. I also tried moving the sums and products of the p's around, and that doesn't seem to help either.

    Does anyone see what I am doing wrong?
     
  2. jcsd
  3. Feb 22, 2014 #2

    TSny

    User Avatar
    Homework Helper
    Gold Member

    If p' must simultaneously equal p and p'', what can you say about p and p''?
     
  4. Feb 22, 2014 #3
    They must be equal... I know that. I'm trying to be very explicit in all my steps though. So if the kronecker deltas say p=p'=p'', how does that affect the products?
     
  5. Feb 22, 2014 #4

    TSny

    User Avatar
    Homework Helper
    Gold Member

    ## \prod_{p} \prod_{p''} \sum_{p'} \left( \delta_{pp'} \delta_{p'p''} \right)##

    If p ≠ p'', then what is the value of ##\sum_{p'} \left( \delta_{pp'} \delta_{p'p''} \right)##

    If you let ##a_{pp''} = \sum_{p'} \left( \delta_{pp'} \delta_{p'p''} \right)##, then your expression is of the form ## \prod_{p} \prod_{p''} a_{pp''}##. This is a product of terms ##a_{pp''}##. If any one of the ##a_{pp''}## is zero, what is the value of the overall expression?
     
  6. Feb 22, 2014 #5
    0

    0. I'm still not seeing how I am going to get the value of 4 I need.

    Let me show you what my problem is

    \begin{align*}
    \prod_{p} \prod_{p''} \sum_{p'} \left( \delta_{pp'} \delta_{p'p''} \right)
    &=\prod_{p} \prod_{p''} \left( \delta_{p1}\delta_{1p''} + \delta_{p2}\delta_{2p''} + \delta_{p3}\delta_{3p''} + \delta_{p4}\delta_{4p''} \right) \\
    &= \prod_{p} \left( \delta_{p1}\delta_{11} + \delta_{p2}\delta_{21} + \delta_{p3}\delta_{31} + \delta_{p4}\delta_{41} \right)
    \left( \delta_{p1}\delta_{12} + \delta_{p2}\delta_{22} + \delta_{p3}\delta_{32} + \delta_{p4}\delta_{42} \right)
    \left( \delta_{p1}\delta_{13} + \delta_{p2}\delta_{23} + \delta_{p3}\delta_{33} + \delta_{p4}\delta_{43} \right)
    \left( \delta_{p1}\delta_{14} + \delta_{p2}\delta_{24} + \delta_{p3}\delta_{34} + \delta_{p4}\delta_{44} \right) \\
    &= \prod_{p} \left( \delta_{p1} \delta_{11} \right) \left( \delta_{p2} \delta_{22} \right) \left( \delta_{p3} \delta_{33} \right) \left( \delta_{p4} \delta_{44} \right) \\
    &= \prod_{p} \left( \delta_{p1} \right) \left( \delta_{p2} \right) \left( \delta_{p3} \right) \left( \delta_{p4} \right) \\
    &=0 \neq 4
    \end{align*}

    Sorry if I am not understanding you or am missing something that is totally obvious.
     
  7. Feb 22, 2014 #6

    TSny

    User Avatar
    Homework Helper
    Gold Member

    Yes, I agree that ## \prod_{p} \prod_{p''} \sum_{p'} \left( \delta_{pp'} \delta_{p'p''} \right) = 0.##

    I don't understand how you justify reordering the products and summation when you wrote

    For example I think you can easily check that in general

    ##( \prod_{i=1}^2 \sum_{j=1}^2A_{ij})(\prod_{i\,'=1}^2 \sum_{j\,'=1}^2B_{i\,'j\,'}) \neq \prod_{i=1}^2\prod_{i\,'=1}^2\sum_{j=1}^2\sum_{j\,'=1}^2A_{ij}B_{i\,'j\'}##.
     
  8. Feb 23, 2014 #7
    Ok...if I write out the original order (product, sum, product), I will still get zero

    \begin{align*}
    \prod_{p}\sum_{p'}\prod_{p''}\delta_{pp'}\delta_{p'p''}
    &=\prod_{p}\sum_{p'}\delta_{pp'}\delta_{p'1}\delta_{p'2}\delta_{p'3} \delta_{p'4} \\
    &=\prod_{p}\left(\delta_{p1}\delta_{11}\delta_{12}\delta_{13}\delta_{14}+\delta_{p2}\delta_{21}\delta_{22}\delta_{23}\delta_{24}+\delta_{p3} \delta_{31}\delta_{32}\delta_{33}\delta_{34}+\delta_{p4}\delta_{41} \delta_{42}\delta_{43}\delta_{44} \right) \\
    &= \prod_{p} \left(0 + 0 + 0 + 0 \right) \\
    &= 0 \neq 4
    \end{align*}

    So something is still wrong...
     
  9. Feb 23, 2014 #8

    TSny

    User Avatar
    Homework Helper
    Gold Member

    I was able to get the result, but only by writing things out explicitly and assuming ##\small |C_{\alpha+}|^{2}+|C_{\alpha-}|^{2} = 1##.

    Note that the expression
    ## \left[ \langle 0 | \prod_{p} \sum_{\sigma} \left( C_{\alpha\sigma}^{*} a_{p\sigma} \right) \right] \left[ \left( \frac{1}{2} \right) \epsilon \sum_{p'} \sum_{\sigma'} \left( \sigma' a^{\dagger}_{p'\sigma'} a_{p'\sigma'} \right) \right] \left[ \prod_{p''} \sum_{\sigma''} \left( C_{\alpha\sigma''} a_{p''\sigma''}^{\dagger} \right) | 0 \rangle \right]##

    has three main factors. The factor on the right is

    ##\left[ \prod_{p''} \sum_{\sigma''} \left( C_{\alpha\sigma''} a_{p''\sigma''}^{\dagger} \right) | 0 \rangle \right] = \left(C_{\alpha+} a_{1+}^{\dagger} + C_{\alpha-} a_{1-}^{\dagger} \right) \left(C_{\alpha+} a_{2+}^{\dagger} + C_{\alpha-} a_{2-}^{\dagger} \right) \left(C_{\alpha+} a_{3+}^{\dagger} + C_{\alpha-} a_{3-}^{\dagger} \right) \left(C_{\alpha+} a_{4+}^{\dagger} + C_{\alpha-} a_{4-}^{\dagger} \right)##

    I found it convenient to let

    ##\left[1 \right] = \left(C_{\alpha+} a_{1+}^{\dagger} + C_{\alpha-} a_{1-}^{\dagger} \right) ## ##\;\;\; ## ##\left[2 \right] = \left(C_{\alpha+} a_{2+}^{\dagger} + C_{\alpha-} a_{2-}^{\dagger} \right) ## ##\;\;\;## etc.

    so that

    ## \prod_{p''} \sum_{\sigma''} \left( C_{\alpha\sigma''} a_{p''\sigma''}^{\dagger} \right) | 0 \rangle = [1][2][3][4] | 0 \rangle##

    The middle main factor of the general expression contains

    ## \sum_{p'} \sum_{\sigma'} \left( \sigma' a^{\dagger}_{p'\sigma'} a_{p'\sigma'} \right) = \left( n_{1+}-n_{1-} \right) + \left( n_{2+}-n_{2-} \right) + \left( n_{3+}-n_{3-} \right) + \left( n_{4+}-n_{4-} \right) ##

    where ##n_{1+} = a_{1+}^{\dagger}a_{1+}\;\;## etc.

    You can show that ## \left( n_{1+}-n_{1-} \right) [1][2][3][4]| 0 \rangle = [\overline{1}][2][3][4]| 0 \rangle##, where ##[\overline{1}]= \left(C_{\alpha+} a_{1+}^{\dagger} - C_{\alpha-} a_{1-}^{\dagger} \right) ##

    Then see if you can show

    ## \left[ \sum_{p'} \sum_{\sigma'} \left( \sigma' a^{\dagger}_{p'\sigma'} a_{p'\sigma'} \right) \right] \prod_{p''} \sum_{\sigma''} \left( C_{\alpha\sigma''} a_{p''\sigma''}^{\dagger} \right) | 0 \rangle = \left( [\overline{1}][2][3][4] + [1][\overline{2}][3][4] + [1][2][\overline{3}][4] + [1][2][3][\overline{4}] \right) | 0 \rangle ##

    Finally you can try to see what happens when you apply the final factor

    ## \langle 0 | \prod_{p} \sum_{\sigma} \left( C_{\alpha\sigma}^{*} a_{p\sigma} \right) = \langle 0 | \{1\}\{2\}\{3\}\{4\}## where ##\{1\} = \left(C_{\alpha+}^{*} a_{1+} + C_{\alpha-}^{*} a_{1-} \right) ## ##\;\;\;## etc.

    Note that ##\{1\}[1]| 0 \rangle = \left(|C_{\alpha+}|^2 + |C_{\alpha-}|^2 \right) | 0 \rangle = | 0 \rangle## and ##\{1\}[\overline{1}] | 0 \rangle = \left(|C_{\alpha+}|^2 - |C_{\alpha-}|^2 \right) | 0 \rangle##

    Sorry, I don't see an elegant, compact way using Kronecker deltas, etc. I get confused if I try to move around product and sum symbols.
     
  10. Feb 23, 2014 #9
    Thanks for the reply! I will try to work through everything you have written. By the way, your assumption ##\small |C_{\alpha+}|^{2}+|C_{\alpha-}|^{2} = 1## is correct because the operators were constructed from a unitary transformation.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Raising and Lowering Operators in the Lipkin Model
Loading...