1
$\begingroup$

I am trying to solve an exercise, but I can't seem to get it to work. I have to show $$\langle \sigma_x \rangle^2 + \langle \sigma_y \rangle^2+\langle \sigma_z \rangle^2 =1 $$

for $|\psi\rangle = \cos\theta|0\rangle + \sin\theta|1\rangle$. I first expand upon the rule, by actually computing the probabilities for each base.

enter image description here

This then leads me to this: enter image description here

This leads me to this: enter image description here

I assume my mistake lies here.

When inserting $a_0 = \cos\theta$ and $a_1 = \sin\theta$ I get this: enter image description here

This is incorrect though, because the relationship at the beginning should, according to the exercise, hold for all $\theta$ and not just when $\theta = n\pi$ $\forall n \in \mathbb N$, because it asks me to verify it for that specific state ($|\psi\rangle= \cos\theta|0\rangle + \sin\theta|1\rangle$). Can anyone help me out here? Thanks, Tom

$\endgroup$

1 Answer 1

2
$\begingroup$

I think that you made a mistake at

$$\frac{\vert a_0 - a_1 \vert^2}{2} = \frac{\vert \cos^2(\theta)-\sin^2(\theta)\vert^2}{2} = \frac{1}{2}(\cos^2(\theta) -\color{red}{2}\cos(\theta)\sin(\theta)+\sin^2(\theta))$$ and similarly for the other term where there seems to be a missing factor of $2$. Therefore I think that the calculations should proceed as follows, after the third image.

$$\frac{1}{4}((a_0-a_1)^2-(a_0+a_1)^2)^2 + (0)^2 + (a_0^2-a_1^2)^2=$$

$$\frac{1}{4}(a_0^2-2a_0a_1+a_1^2-(a_0^2+2a_0a_1+a_1^2))^2 + (a_0^2-a_1^2)^2=$$

$$\frac{1}{4}(-4a_0a_1)^2 + (a_0^2-a_1^2)^2 = a_0^4-2a_0^2a_1^2+a_1^4+4a_0^2a_1^2=a_0^4+2a_0^2a_1^2+a_1^4$$

$$=(a_0^2+a_1^2)^2=(\sin^2(\theta)+\cos^2(\theta))^2=1^2=1$$

So in conclusion you were correct on where you did the mistake. No worries, these calculations are big and mistakes like that are common.

$\endgroup$
4
  • $\begingroup$ Thanks for the answer! I just have one problem: I think the factor of $\frac14$ at the beginning is incorrect. $\endgroup$ Commented Mar 26, 2022 at 9:34
  • $\begingroup$ @tommasopeduzzi I don't think it is because you have ((1/sqrt(2))^2)^2 = 1/4. The first 1/sqrt(2) is the state, the first ^2 is the overlap and the second ^2 is the exp. squared. $\endgroup$
    – R.W
    Commented Mar 26, 2022 at 13:35
  • 1
    $\begingroup$ Thanks a lot! I often overlook these kinds of details, even if I stare at my monitor for hours. $\endgroup$ Commented Mar 26, 2022 at 17:44
  • $\begingroup$ Could you please accept the answer if possible and if you think it was the correct answer for the question. Att. R.W. $\endgroup$
    – R.W
    Commented Mar 31, 2022 at 11:09

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.