Question about double slit experiment with detectors

[o-man]

I've read that when one performs the double slit experiment (dse) without any detectors to "see" which slit the photon goes through one gets an interference pattern. When one places a detector to see which slit the photon goes through the interference pattern disappears. My question are these.

1. Has the DSE been done with a detector on the ingoing side of the slits and then the outgoing side of the slits?

2. What exactly do the detectors do? As near as i can figure for the detector to detect it has to interact with the photon and the only way it can do that is by exchange of energy. If that is true then the photon in the experiment with the detector and the photon in the experiment without the detector likely don't have the same energy to begin with or/and end with. Or am i missing something entirely?

As always thanks for reading.

 

[Simon Bridge]

1. yes - and much more convoluted approaches too.
2. depends on the detector and what it is supposed to detect.
For photon - yes: a photon detector has to destroy the photon it detects.

In the case of doing the experiment with photons, then, it is not surprising to find the interference pattern destroyed. The important observation, then, is that the photons are detected at one slit or the other - they are never half-and-half or continuous as expected from a classical wave.

We could use a detector which only detects a photon half (or some other fraction) of the time ... what do you think happens then?

It's misleading anyway - each photon passes randomly through one or the other slit but the way we calculate the interference patters includes terms related to both slits. Not for each individual photon but for the pattern as a whole.

It's a bit like how the pattern in dice tossing can be predicted only by taking account of all the possible numbers that could be rolled even though each individual toss only shows one number.

The following lecture is fast becoming a fave:
http://www.scottaaronson.com/democritus/lec9.html
... it's very abstract, but shows up how the math works.