1
$\begingroup$

$\hat{L_z}$ has the eigenvalue m$\hbar$ for a given eigenfunction with non-zero $m_l$ values. This is the component/projection of the angular momentum vector on the $z$ axis. Can i say that the $\hat{L_z}$ operator is a projection operator?

$\endgroup$

2 Answers 2

2
$\begingroup$

No, a projection operator in quantum mechanics has the specific condition that it is idempotent. That is, if $\hat P$ is a projection operator, then

$$\def\bra#1{\left\langle#1\right|}\def\ket#1{\left|#1\right\rangle} \hat P^2=\hat P $$

Intuitively, the idea is that if you project a vector along another vector, then doing the projection again shouldn't change the result. $\hat L_z$ does not have this property, except for the trivial case of system with no angular momentum, where it is the zero operator.

As an example of a projection operator associated with angular momentum, consider a spin-1/2 system, with the normalized eigenstates of $\hat S_z$ being $\ket{+}$ and $\ket{-}$ for the $+1/2$ and $-1/2$ eigenvalues, respectively. Then $\ket{+}\bra{+}$ and $\ket{-}\bra{-}$ are both projection operators, for the spin-up and spin-down states. $\hat S_z$ can be written as a sum of these projection operators like (in units where $\hbar=1$)

$$\hat S_z=\frac{1}{2}\ket{+}\bra{+}-\frac{1}{2}\ket{-}\bra{-}$$

$\hat S_z^2$ is instead given by

$$\hat S_z^2=\frac{1}{4}\ket{+}\bra{+}+\frac{1}{4}\ket{-}\bra{-}=\frac{1}{4}\hat{I}$$

and is just proportional to the identity operator.

$\endgroup$
2
  • $\begingroup$ Good answer. Just a small thing regarding notation: measuring spin with $\hat{L}_z$ might frustrate readers, as it is used to denote the angular momentum operator in the $z$-direction. In your answer, it looks like you are using $\hat{L}_z$ to measure spin in the $z$-direction. Usually, spin eigenvalues are extracted by the use of the \textbf{spin operator} $\hat{S}_z$. Also, the eigenvalues of $\hat{L}_z$ should have units of angular momentum. $\endgroup$
    – Andrew
    Commented Oct 1, 2020 at 0:06
  • 1
    $\begingroup$ @AndreasMastronikolis Good point, I've changed it to $\hat S_z$, and made clear I'm using natural units. $\endgroup$
    – Chris
    Commented Oct 1, 2020 at 0:13
2
$\begingroup$

For an operator to be a projection, it must be the identity operator for states in its image. That is $P$ is a projection if and only if $P(P(x)) = P(x)$ for all $x$ in the domain of $P$. Consider now the action of $L_z$ on a state $\psi$ with some angular momentum in the $z$ direction. Then $L_z \psi = \ell \psi$ where $\ell$ is the eigenvalue. Therefore $L_z(L_z(\psi)) = \ell^2\psi\neq \ell\psi = L_z(\psi)$ so $L_z$ is not a projection.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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