5
$\begingroup$

Prove that $L_i$ ; i = 1, 2, 3 of the ang. momentum operator are related to the Cartesian components of the position and mometum operators by $L_i = \epsilon_{ijk} x_ j p_ k $ (summation convention: sum over repeated indices).

I got to the part following link how $$L_x=-\iota\hbar(y\frac{\partial}{\partial z}-z\frac{\partial}{\partial y})$$ $$L_y=-\iota\hbar(z\frac{\partial}{\partial x}-x\frac{\partial}{\partial z})$$ $$L_z=-\iota\hbar(x\frac{\partial}{\partial y}-y\frac{\partial}{\partial x})$$ but, the next step is a little too quick

$$L_i=-\iota\hbar\epsilon_{ijk}x_j\frac{\partial}{\partial x_k} $$

How did we reach this last step ? And what trasnformation do I need to do to make the RHS look like $$ L_i = \epsilon_{ijk} x_ j p_ k$$

$\endgroup$
2
  • $\begingroup$ I think you're missing understanding the Levi-Civita symbol - the epsilons $\epsilon_{ijk}$. $\endgroup$ Commented Aug 28, 2017 at 21:48
  • 4
    $\begingroup$ I think there is nothing to prove since $\vec L=\vec r\times\vec p$ is a definition. $\endgroup$
    – Diracology
    Commented Aug 28, 2017 at 23:17

2 Answers 2

5
$\begingroup$

It is simplest to do this by brute force, with $(x,y,z)=(1,2,3)$ and using the explicit values $\epsilon_{123}=\epsilon_{231}=\epsilon_{312}=1$, $\epsilon_{213}=\epsilon_{132}=\epsilon_{321}=-1$, with the other combinations $0$.

Then, for instance: \begin{align} L_x&=L_1=-i\hbar\epsilon_{1jk}x_j\frac{\partial }{\partial x_k}\, ,\\ &=-i\hbar\left(\epsilon_{123}x_2\frac{\partial }{\partial x_3}+\epsilon_{132} x_3\frac{\partial}{\partial x_2}\right)\, ,\\ &=-i\hbar\left( x_2\frac{\partial }{\partial x_3}-x_3\frac{\partial }{\partial x_2}\right)\, ,\\ &= -i\hbar\left(y\frac{\partial }{\partial z} - z\frac{\partial }{\partial y}\right). \end{align} $L_y$ and $L_z$ are done in the same way as $L_x.$

Finally, since $-i\hbar \partial/\partial x_k=p_k$, the last step follows.

$\endgroup$
2
  • $\begingroup$ corrected my expressions, if you could remove the part for the mistake to prevent others from being confused $\endgroup$
    – fireball.1
    Commented Aug 29, 2017 at 6:58
  • $\begingroup$ @fireball.1 done. $\endgroup$ Commented Aug 29, 2017 at 8:25
3
$\begingroup$

This the definition of the cross product: the $i^{th}$ component of the cross product between $x$ and $p$ is $\epsilon_{i\,j\,k}\,x_j\,p_k$. As in ZeroTheHero's answer, you're not really proving anything here, but rather simply expanding the definition out.

The directed area of the parallelogram defined by $x$ and $p$ is the two form $x\wedge p$. Then the Hodge dual, implemented with $\epsilon$ converts this directed plane element into a vector, which is equivalent in three dimensions because the vector fully defines the plane it is normal to. That's how we define the cross product: as a "vectorized" version of the directed area of the parallelogram defined by two vectors.

$\endgroup$
2
  • $\begingroup$ This is all true, but I think discussing things such as Hodge dual would just confuse the average physics student. $\endgroup$
    – Omry
    Commented Aug 29, 2017 at 4:51
  • $\begingroup$ @Omry Possibly: I'm just trying to motivate the bald statement that "this is simply the definition" and give a reader a lead for further reading. "That's just how it is" is always a bit unsatisfying without some motivation at least. If you can think of a better way, I'd love to see an answer: there's surely some intuition we can add here and to my mind the directed parallelogram is the best geometrical motivation one can think of. $\endgroup$ Commented Aug 29, 2017 at 6:07

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.