On Sun, 09 Jun 2013 13:19:09 -0700, Salmon Egg wrote:
> It is certainly possible to reduce the intensity of light in a two-slit > apparatus to where the probability of having more than one photon in the > apparatus is tiny. For example, in a three-meter device with an average > of one photon per second, the probability of having two simultaneous > photons is of the order of 1E-08 per second. While relatively slow, an > hour should give pretty good evidence of interference. > > Photomultipliers in the green can have quantum efficiency of about 20%. > So you miss 80% of the photons going through. big deal. You do have to > cool the photosurface and dynodes to minimize dark current, but liquid > nitrogen is good enough.
Salmon egg forgot to mention that what with cheap lasers and photomulitplier tubes today, this isn't millions of dollars but funding at levels high school students can manage. And if you do a step up in cost you can use night vision gear that let's you collect the diffraction pattern as an image instead of scanning for it. I'm sure you've seen night vision pictures that have that "sparkly" look. Those are individual photons arriving to make an image.
If you are interested in this you should have already read: