Spin operator in an arbitrary direction

let $$\ \hat{n} = (\sin{\phi}\cos{\theta},\sin{\phi} \sin{\theta},\cos{\phi})$$

How to prove $$\vec{S}\cdot\hat{n}$$ is the spin operator in an arbitrary axis rigorously?

You just have to find an eigenvalues of this operator. Suppose you have the case of spin $\frac{1}{2}$ particle. Then $$\vec{S} = \frac{1}{2}\left( \sigma_{x}, \sigma_{y}, \sigma_{z}\right),$$ where $\sigma_{i}$ is Pauli matrices.

Now you have that $$\hat{h}\equiv \vec{S}\cdot \vec{n} =\sum_{\sigma = \pm \frac{1}{2}}\sigma| v_{\sigma}\rangle\langle v_{\sigma} | ,$$ where $v_{\sigma}$ is the eigenvector of $\hat{h}$ operator (which, by the correspondence principle, defines projection of spin on $\vec{n}$ axis) and $\sigma$ is helicity. Physical sense of eigenvector is that expansion coefficient near it for an arbitrary state $|\psi \rangle$, $$|\psi\rangle = \sum_{\sigma}c_{\sigma}|v_{\sigma}\rangle, \quad \sum_{\sigma}|c_{\sigma}|^{2} = 1$$ defines the probability to obtain value $\sigma$ when we measure average value of $\hat{h}$, $$\langle \psi|\hat{h} | \psi\rangle =\sum_{\sigma}\sigma |c_{\sigma}|^{2}$$ It is clear that $|v_{\sigma}\rangle$ is the the state in which the value of the spin in the direction given by $\vec{n}$. Really, by setting $\theta = \varphi = 0$ you obtain that $\hat{h} = \sigma_{z}$, while $v_{\sigma}= \begin{pmatrix} 1 \\ 0\end{pmatrix}, \begin{pmatrix} 0\\ 1\end{pmatrix}$ are eigenvectors of $\sigma_{z}$.

• Sorry, I think you haven't proven that it is a spin operator in an arbitrary axis rigorously, because you just have proven it is a spin operator along the z axis. Of course, it is easy to see what happens along the x or y axis. Generally, in an arbitrary direction, how to prove it? Feb 8, 2016 at 14:20
• @ZhuWei : as I've written, just find an explicit form of eigenvectors $v_{\sigma}$ for arbitrary dimensions. The other is completely equal to case of $z$ direction. If you strongly need it, I can write down the result. Feb 8, 2016 at 14:22
• I know the explicit form of eigenvectors for arbitrary directions. But I can't understand the relation of the eigenvectors to my question. Feb 8, 2016 at 14:40
• @ZhuWei : by the correspondence principle, $\hat{h} = \vec{S}\cdot\vec{n}$ is just operator of projection of spin on the direction given by $\vec{n}$. Thus eigenvectors of $\hat{h}$ play the role of generalized eigenvectors of $\frac{1}{2}\sigma_{z}$ operator. Feb 8, 2016 at 14:45
• What's the correspondence principle? The behavior of systems described by the theory of quantum mechanics reproduces classical physics in the limit of large quantum numbers? Feb 8, 2016 at 15:17