Quantum Computing 1.3 -- Polarizers

A two-state quantum system is called a qubit . A qubit can be realized by a photon which has two different polarization states 0 \ket{0^{\circ}} and 9 0 \ket{90^{\circ}} which correspond to polarization along the vertical and horizontal axes, respectively. The ability to precisely and reliably manipulate qubits is key to the advent of large-scale quantum computing

A linearly polarized photon aligned with angle θ \theta can be split into its 0 \ket{0^{\circ}} and 9 0 \ket{90^{\circ}} components with amplitudes corresponding to the projection of the wave on the measurement axis. These states are orthogonal , so a pure 0 \ket{0^{\circ}} photon has no 9 0 \ket{90^{\circ}} component, and vice versa: θ = cos ( θ ) 0 + cos ( 9 0 θ ) 9 0 , θ [ 0 , 36 0 ] . \ket{\theta ^ \circ} = \cos(\theta) \ket{0^{\circ}} + \cos(90 ^ \circ -\theta) \ket{90^{\circ}}, \quad \theta \in [0^\circ,360^\circ].

One way to manipulate a photon qubit is with polarization filters. Each filter measures and absorbs all photons in the α + 9 0 \ket{\alpha+90 ^ \circ} state, while all others pass through. Consider a series of three linear polarization filters, each rotated by an angle of α = 3 0 , 6 0 , \alpha = 30 ^ \circ, 60 ^ \circ, and 9 0 , 90 ^ \circ, respectively, with respect to the 0 \ket{0^{\circ}} polarization direction.

What is the probability P P that a 0 \ket{0^{\circ}} photon passes all three polarizing filters and emerges as a 9 0 \ket{90^{\circ}} photon?

P = 0 % P = 0 \% P 33 % P \approx 33 \% P 42 % P \approx 42 \% P 87 % P \approx 87 \%

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

3 solutions

Jc 506881
Jul 2, 2018

The probability that a 0 |0\rangle photon passes through a 3 0 30^\circ filter is cos 2 ( 3 0 ) = 3 4 \cos^2 (30^\circ) = \frac{3}{4} . So, the probability the photon passes through three successive filters, each 3 0 30^\circ more than the last, is ( 3 4 ) 3 = 27 64 42 % \left( \frac{3}{4} \right)^3 = \frac{27}{64} \approx 42\% .

Minutephysics and 3Blue1Brown collaborated on a video about quantum polarization and Bell's theorem that not only details what you need to know for this problem but also talks about some interesting implications of the quantum polarization effect. I watched this video when it came out last year and it stuck with me well enough that I was able to do this problem so it's a good, clear video, well worth a look if you're interested.

I support that recommendation for the 3Blue1Brown video!

Tom Verhoeff - 2 years, 11 months ago
Markus Michelmann
Jun 23, 2018

One can represent the passage of a photon through the three polattering filters as a decision tree.

However, we start from the general case that we have n n polarization filters, each with the angle α k = k n 9 0 \alpha_k = \frac {k} {n} \cdot 90 ^ \circ , k = 1 , 2 , , n k = 1,2, \dots, n . The photon. that has already passed k k filters, has the polarization α k | \alpha_k \rangle . This state can be represented by α k = ( α k + 1 α k + 1 + α k + 1 + 9 0 α k + 1 + 9 0 ) α k = α k + 1 α k α k + 1 + α k + 1 + 9 0 α k α k + 1 + 9 0 \begin{aligned} |\alpha_k \rangle &= (|\alpha_{k+1} \rangle \langle \alpha_{k+1}| + |\alpha_{k+1} + 90^\circ \rangle \langle \alpha_{k+1} + 90^\circ|) |\alpha_k \rangle \\ &= \langle \alpha_{k+1}| \alpha_k \rangle |\alpha_{k+1} \rangle + \langle \alpha_{k+1} + 90^\circ| \alpha_k \rangle |\alpha_{k+1} + 90^\circ \rangle \end{aligned} The probability that the photon also passes the k + 1 k + 1 filter is α k + 1 α k 2 = ( cos α k + 1 0 + sin α k + 1 9 0 ) ( cos α k 0 + sin α k 9 0 ) 2 = cos α k + 1 cos α k + sin α k + 1 sin α k 2 = cos ( α k + 1 α k ) 2 = cos 9 0 n 2 \begin{aligned} |\langle \alpha_{k+1}| \alpha_k \rangle|^2 &= | (\cos \alpha_{k+1} \langle 0^\circ | + \sin \alpha_{k+1} \langle 90^\circ |) (\cos \alpha_{k} | 0^\circ \rangle + \sin \alpha_{k} | 90^\circ \rangle ) |^2 \\ &= |\cos \alpha_{k+1} \cos \alpha_{k} + \sin \alpha_{k+1} \sin \alpha_{k}|^2\\ &= |\cos(\alpha_{k+1} - \alpha_{k})|^2 \\ &= \left|\cos \frac{90^\circ}{n}\right|^2 \end{aligned} In particular, this probability is independent of k k . The total probability P n P_n that the photon passes through all n n polarization filters therefore results from multiplying all individual probabilities: P n = cos 9 0 n 2 n P_n = \left|\cos \frac{90^\circ}{n}\right|^{2n} Thus, in the case n = 3 n= 3 we get P 3 = cos 3 0 6 = 3 2 6 = 3 3 2 6 = 27 64 42 % P_3 = \left|\cos 30^\circ \right|^{6} = \left|\frac{\sqrt{3}}{2} \right|^{6} = \frac{3^3}{2^6} = \frac{27}{64} \approx 42 \,\% and the limiting case results to lim n P n = lim n cos π 2 n 2 n = lim n exp ( 2 n log ( cos π 2 n ) ) = exp ( lim n log ( cos π 2 n ) 1 2 n ) apply L’Hospital rule = exp ( lim n π 2 n 2 sin π 2 n cos π 2 n 1 2 n 2 ) = exp ( lim n π sin π 2 n cos π 2 n ) = exp ( 0 ) = 1 \begin{aligned} \lim_{n\to\infty} P_n &= \lim_{n\to\infty} \left|\cos \frac{\pi}{2 n}\right|^{2n} \\ &= \lim_{n\to\infty} \exp \left(2n \cdot\log(\cos \frac{\pi}{2 n})\right) \\ &= \exp \left( \lim_{n\to\infty} \frac{\log(\cos \frac{\pi}{2 n})}{\frac{1}{2 n}}\right) & &|\text{apply L'Hospital rule}\\ &= \exp \left( \lim_{n\to\infty} \frac{-\frac{\pi}{2n^2}\frac{\sin \frac{\pi}{2 n}}{\cos \frac{\pi}{2 n}}}{-\frac{1}{2 n^2}}\right) \\ &= \exp \left( \lim_{n\to\infty} \pi \frac{\sin \frac{\pi}{2 n}}{\cos \frac{\pi}{2 n}}\right)\\ &= \exp(0) \\ &= 1 \end{aligned} This means that the photon is not absorbed, but passes through the polarizing filters in any case. Repeated lossless measurement changes the state of the photon from 0 | 0 ^ \circ \rangle to 9 0 | 90 ^ \circ \rangle . This is an example how measurements can alter a system to the complete opposite.

An interesting demonstration about the probabilistic nature of light transmission through polarizing filters is to start with filters A and C, at 90 degrees apart. When looking through them, everything looks black. But insert filter B between A and C, 45 degrees from the others. Then you can see through all three.

Michael Mendrin - 2 years, 11 months ago

Several questions:

Why use cos ( 90 θ ) \cos(90-\theta) instead of just sin ( θ ) \sin(\theta) ?

Why doesn't the problem refer to the components being squared? The solution does, which makes sense, since sin^2+cos^2=1. This seems to indicate that every photon is accounted for. Does the problem states the rule incorrectly?

Jeremy Galvagni - 2 years, 11 months ago

Log in to reply

I, too, am not completely certain of why the components are squared; however, I believe this is because the square of the maximum amplitude of the electric field of a light wave is proportional its intensity; this fact seems to indicate that the number of photons is proportional to the light's intensity (although I do not know for certain, as I have not studied quantum mechanics). This is not mentioned in the problem, but it looks like the problems states everything correctly- however, it is slightly misleading to not state that the maximum electric field squared is proportional to light's intensity.

Gary Zhang - 2 years, 11 months ago

Log in to reply

The problem actually tells nothing about the underlying physics and is somewhat out of context. But you are quite right that the probability of measuring a photon is proportional to the intensity of the electric field, so that the field (or the amplitude of the quantum state) has to be squared.

Markus Michelmann - 2 years, 11 months ago

Originally it was really sin θ \sin \theta in the formula, but it was changed afterwards without me.

I only created this problem as a subtask of a whole quantum computing set. In this set there is a note at the beginning, which explains the qubit. But this problem has been rewritten by Brilliant Staff and provided with additional explanations, so it can also pose as a separate problem.

Nevertheless, if someone has never worked with quantum states and has no prior knowledge of the polarization of light, then he or she probably can not solve this task.

Markus Michelmann - 2 years, 11 months ago

Eh? I understood the question more than I did your solution. And I understood the question up to the word two-state.

Yogi Bear - 2 years, 11 months ago

Log in to reply

I have chosen an bit complicated approach here, since I use the Dirac notation without any explaination. But this problem was originally intended only as a subtask of a whole set on quantum computing . That's why my solution seems out of context.

Markus Michelmann - 2 years, 11 months ago
Vinod Kumar
Jul 4, 2018

At each filter the 30° component's intensity reduces by square of it's amplitude which is (cos(30°))^2=(3/4). After three successive 30° filters it will reduce by (3/4)^3~0.42, and that is the answer to this problem.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...