Modified Mach-Zehnder Interferometer

[birulami]

The Mach-Zehnder interferometer is shown in wikipedia and on David Harrison's page.

If the second (upper right) half silvered mirror is taken away and no sample (as in the wikipedia setup) is in any of the two paths, I read that when sending individual photons through the device, roughly 50% of them end up in each detector.

What I would like to know is the following:

1. Does anyone know a paper where real counts are reported, e.g. like: "we send 100 individual photons, 49 ended up in 1 and 47 ended up in 2, 4 where lost"?
2. If I further remove one of the detectors, what percentage of photons is registered in the remaining detector? (published counts available?)

Thanks,
Harald.

 

[country boy]

1. Removing the second beamsplitter eliminates the interference and just let's the two beams from the first beamsplitter go directly to the detectors. The photon counting you are asking for is done when the efficiency of the first beamsplitter is measured (ignoring the reflectivity of the two mirrors). That is often reported in papers describing interferomters.

2. If 50% goes to each detector, then 50% will still go to the remaining detector. Removing one detector has no effect on what's measured at the other detector.

 

[DaTario]

Young's work may be a good starting point.
Mach E., The principles of physics optics
New York - Dover, 1953