Science
In reply to the discussion: Quantum Entanglement, Dark Counts, Coincidence Detection [View all]caraher
(6,278 posts)Today I was getting roughly 2000 coincidence counts per second per milliwatt of pump power. That's pumping a 0.5 mm thick BBO crystal cut for Type II, with about 80 mW of blue light centered on 400 nm (I'm actually using frequency-doubled broadband light from an ultrafast laser running at around 800 nm), which was giving me up to 180,000 coincidence counts per second. I'm collecting pretty much all the downconverted light, not just the photons from the overlap region, but I also have a 20 nm bandpass filter and the 750 nm longpass filter I mentioned before, as well as a dichroic mirror at 45 degrees to reject the pump light; this is what's left over after those losses, losses due to other optics, and ~55-60% single detector efficiency at 800 nm.
I've pumped the same crystal with a 50 mW 407 nm diode laser and get very comparable count rates. The diode laser is a rather nice one made by Power Technology, but costs several thousand dollars. I think if I were doing your experiment I'd at least consider the ~$1k 405 nm laser system Roithner sells; it should give comparable performance. But I'd try with a cheap laser first, since if that fails you can always buy a pricier one, whereas if it works you've saved a bunch of money! 50 mW is plenty of power if you have a nice mode. It's also *relatively* safe, certainly much more so than a 250 mW unit!