Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. It's 100% free.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have three Spin 1/2 Particles and a Hamiltonian given by $$H=A(S_1\cdot S_2)+B(S_2\cdot S_3+S_1\cdot S_3)$$ In order to find the energy spectrum, I want to diagonalize H in terms of $(S_1+S_2+S_3)^2$ and a coupling of 2 of them (I already have spectrum of $H=A(S_1\cdot S_2)$). However, every combination I have tried if terms of coefficients and couplings have not been successful, so I was hoping someone could give me a hint as to how to couple these guys. Thanks.

share|cite|improve this question
up vote 3 down vote accepted


  1. Notice that if we define $\mathbf S_{123} = \mathbf S_1 + \mathbf S_2 + \mathbf S_3$ and $\mathbf S_{12} = \mathbf S_1 + \mathbf S_2$, then we have \begin{align} \mathbf S_{123}^2 = \mathbf S_{12}^2 + \mathbf S_3^2 + 2\mathbf S_{12}\cdot\mathbf S_3 \end{align}

  2. Notice that your hamiltonian can be written as follows: \begin{align} H = \frac{A}{2}(\mathbf S_{12}^2- \mathbf S_1^2 - \mathbf S_2^2) + B(\mathbf S_{12}\cdot\mathbf S_3) \end{align}

  3. Combine the last two hints.

  4. Recall that the representation theory of the angular momentum algebra (aka addition of angular momentum) tells us that a tensor product of three spin-$1/2$ reprsentations splits into a direction sum as follows: \begin{align} \tfrac{1}{2}\otimes\tfrac{1}{2}\otimes\tfrac{1}{2} &= (1 \oplus 0)\otimes\tfrac{1}{2} \\ &= (1\otimes \tfrac{1}{2}) \oplus (0\otimes \tfrac{1}{2}) \\ &= \underbrace{(\tfrac{3}{2}\oplus\tfrac{1}{2})}_{s_{12}=1, s_3 = \frac{1}{2}}\oplus \underbrace{\tfrac{1}{2}}_{s_{12}=0, s_3=\frac{1}{2}} \end{align}

  5. (Addendum) To clarify some discussion in the comments, the last line in number 4 means that the Hilbert space admits an orthonormal basis of states which I'll label $|s_{123}, s_{12}\rangle$ for which \begin{align} \mathbf S_{123}^2|s_{123}, m_{123},s_{12}\rangle &= \hbar^2 s_{123}(s_{123}+1)|s_{123}, m_{123},s_{12}\rangle \\ S_{123}^z |s_{123}, m_{123},s_{12}\rangle &= \hbar m_{123}|s_{123}, m_{123},s_{12}\rangle\\ \mathbf S_{12}^2|s_{123}, m_{123},s_{12}\rangle &= \hbar^2 s_{12}(s_{12}+1)|s_{123}, m_{123},s_{12}\rangle \\ \mathbf S_{1}^2|s_{123}, m_{123},s_{12}\rangle &= \hbar^2 \tfrac{1}{2}(\tfrac{1}{2}+1)|s_{123}, m_{123},s_{12}\rangle \\ \mathbf S_{2}^2|s_{123}, m_{123},s_{12}\rangle &= \hbar^2 \tfrac{1}{2}(\tfrac{1}{2}+1)|s_{123}, m_{123},s_{12}\rangle \\ \mathbf S_{3}^2|s_{123}, m_{123},s_{12}\rangle &= \hbar^2 \tfrac{1}{2}(\tfrac{1}{2}+1)|s_{123}, m_{123},s_{12}\rangle \end{align} and explicitly, the basis is as follows: \begin{align} \left.\begin{array}{l} |\tfrac{3}{2}, \tfrac{3}{2},1\rangle \\ |\tfrac{3}{2}, \tfrac{1}{2},1\rangle \\ |\tfrac{3}{2}, -\tfrac{1}{2},1\rangle \\ |\tfrac{3}{2}, -\tfrac{3}{2},1\rangle \end{array}\right\} s_{123} = \tfrac{3}{2}, s_{12} = 1, \dim = 4\\ \left.\begin{array}{l} |\tfrac{1}{2}, \tfrac{1}{2},1\rangle \\ |\tfrac{1}{2}, -\tfrac{1}{2},1\rangle \end{array}\right\} s_{123} = \tfrac{1}{2}, s_{12} = 1, \dim = 2\\ \left.\begin{array}{l} |\tfrac{1}{2}, \tfrac{1}{2},0\rangle \\ |\tfrac{1}{2}, -\tfrac{1}{2},0\rangle \end{array}\right\} s_{123} = \tfrac{1}{2}, s_{12} = 0, \dim = 2 \end{align} This is a basis for the entire three spin-$1/2$ particle Hilbert space; it is eight-dimensional. There are therefore a total of 8 orthonormal eigenvectors for the hamiltonian.

share|cite|improve this answer
You are actually the best. As a followup, from yesterday, I know the spectrum of $S_{12}$ and then $S_{123}$ can be 3/2 or 1/2? – yankeefan11 Mar 28 '14 at 2:34
Haha ;). Yup, I think you got it. Basically the last line of 4 is saying that the Hilbert space decomposes into orthogonal subspaces, two of which have total spin quantum number $s_{123} = \tfrac{1}{2}$ and one of which has $s_{123} = \tfrac{3}{2}$, and within each of these subspaces, $s_{12}$ has a certain value based on where that subspace came from when we first combined the first and second spins. – joshphysics Mar 28 '14 at 2:39
For degeneracies, I know that I can have 4 $S_{123}=3/2$ states, and that would be coupled with 3 $S_{12}=1$ states and just one $S_{12}=0$ state. Same goes for $S_{123}=1/2$, except I only have two possible states for that to coupled with $S_{12}$? – yankeefan11 Mar 28 '14 at 3:26
If I do that, I only get 24 states. Shouldn't I have 27? – yankeefan11 Mar 28 '14 at 3:51
@yankeefan11 I don't quite follow. The spin-$\frac{1}{2}$ Hilbert space is $2$-dimensional, so its threefold tensor product has dimension $2^3 = 8$. This matches the dimension of the direct sum since $\dim(\frac{3}{2}\oplus\frac{1}{2}\oplus\frac{1}{2}) = 4 + 2 + 2 = 8$. – joshphysics Mar 28 '14 at 5:11

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.