- #1

- 174

- 0

A common experiment used to illustrate the effect uses polarization directions of photons (http://arxiv.org/abs/1211.3498). The photons are either "vertical" or "horizontal", but one can "rotate" their polarisation direction (e.g. "by 15 degrees"). Here, one sends vertical photons down a path of six rotators (so you eventually get a 90 degree turn and hence state transition), before the photon is measured by a polarizer, which absorbs horizontal photons, transmits vertical. Now the goal is to inhibit transition from vertical to horizontal. One does this simply by measuring the polarisation after each rotator interaction (with another polarizer). The polarizer after the first rotator will re-transmit the photon down the path (with some probability?), and apparently it can be shown that if one increases the number of stages, decreasing "rotation angle" at each stage, probability of transmission to final detector increases. If there were an infinite number of stages rotation is completely inhibited yielding QZE.

If the photons can only be "vertical" or "horizontal", then how can one rotate them just by 15 degrees? What exactly do the rotators do to the photons? Do they put them into a superposition of being vertical (with high amplitude) and being horizontal (with low amp)? Then I could see how continuous measurement (collapse) would inhibit state transition. But then why call such superpositions "rotations"?

Please help! :)