Did the detector change the spin states?

Question

I am wondering why every detectors that try to measure the spin states of quantum particle had only 2 outcomes for example either spin up or spin down, can we measure the actual spin states or are we just forcing the spin states into either of the expected spin states?

Answer 1

Using a polarisation filter - which is nothing more than a part of your measuring device -it is possible to let through about 50% of the light (for equally distributed electric fields of the involved photons). Using a second filter, 90° rotated to the first, no light (of suitable wavelength) goes through.

The amazing fact is that using a third filter between the other two - best under 45° - some light goes through. That means that there has to be an influence of of the slits. The slits rotate the photon's electric field.

See also Relationship between the material properties of an edge and the fringes behind this edge

Answer 2

By your wording, I can tell you mean specifically spin-1/2 particles.

I am unaware of a direct measurment of spin. In the case of the Stern-Gerlach experiment, a measurement of the magnetic dipole moment is the link to the spin. So, let's just say we do measure the spin.

Spin-1/2 is a special case that allows us in principle to measure something akin to the "actual spin state". For every unentangled spin-1/2 particle, there are two orientations, 180 degrees apart, of the the SG apparatus that will be guaranteed to return a deterministic result. That means the spin that goes in is the spin that comes out of the apparatus. However, if you don't know the state of your particle, good luck guessing those orientations. You would have to have a beam of identical particles to find the orientation.

If you don't have this special case, then I think most would say that when the wave function interacts with the detector and a measurement is made, the detector forces the particle into a spin state.

Answer 3

For photons we can use a laser source or an incoherent source with polarizers ... with the laser source we can get 100% of the photons to go thru in one orientation, 0% in another and any % between these 2 positions. This would imply that the photons have the spin to begin with.

Answer 4

A measurement is an interaction between a measuring device $M$ and a system $S$ that produces a record of some information from $S$ in $M$.

Suppose you have a spin 1/2 particle in a state where it hasn't interacted with $M$: $$|\psi(0)\rangle_{SM}=(a|\uparrow\rangle_{Sz}+b|\downarrow\rangle_{Sz})|0\rangle_M$$ where $|\uparrow\rangle_{Sz},|\downarrow\rangle_{Sz}$ are the up and down spin states on the $z$ axis. A perfect measurement of the spin on the $z$ axis does the following $$|\uparrow\rangle_{Sz}|0\rangle_M\to|\uparrow\rangle_{Sz}|\uparrow\rangle_M\\ |\downarrow\rangle_{Sz}|0\rangle_M\to|\downarrow\rangle_{Sz}|\downarrow\rangle_M$$

So if we measure $S$ in the state $|\psi\rangle_S$ we get the state $$|\psi(1)\rangle_{SM}=a|\uparrow\rangle_{Sz}|\uparrow\rangle_M+b|\downarrow\rangle_{Sz}|\downarrow\rangle_M$$

This measurement changes the state of $S$ and in general prevents interference between different values of the spin that would have been possible before the measurement. If you don't modify the equations of quantum mechanics with collapse then the two different values evolve approximately independently of one another and there are two versions of the measurement apparatus and any system that gets information from it such as a person. There is a large literature on this kind of suppression of interference under the name of decoherence:

https://arxiv.org/abs/1111.2189

https://arxiv.org/abs/0707.2832

The measurement changes the state of the measured system but this is not particularly extraordinary since there is no reason to expect an interaction to leave the state of a system unchanged. The measurement does give you some information about the state and multiple measurements of systems in the same state gives you information about the square amplitudes $|a|^2,|b|^2$ of each possible value.