Alignment
The toughest part of these experiments is the alignment. The down converted beams are too weak to be seen so we have to use a HeNe laser beam to align all the components. In our first layout, shown schematically in Fig. 1, we set up the equipment so that we work with the down converted photons leaving the crystal at +3 and -3 degrees.
Fig. 1 Layout of the experiments that use a Mach Zehnder
Interferometer. The components shown are: BBO crystal (C),
mirrors (M), beam splitters (B), mirror moved by the piezo
(Mp), lenses (L), band pass filters (F), avalanche photodiodes
(APD).
Given the position of the crystal we calculate the position of the detectors and put irises in those locations. We then send a HeNe laser beam so that it traces the desired path of the two beams. The detectors are subsequently located such that the laser beams reach them. In Fig. 2 we show the path of the pilot HeNe laser beam that we use to align the components. Initially both beams are sent to the detectors. Then with the HeNe laser off we look for down conversions. The lenses in front of the detectors are then displaced transversely with XY mounts (Thorlabs model LM1XY) to maximize the down conversion counts. Once we have down conversions we may tune the phase matching angle slightly also to maximize the counts.
Once the detectors are in the correct positions we then place two mirrors to divert the beam to the Mach Zehnder interferometer. We do this so that if something goes wrong with the alignment we can always make sure that we are having down conversions by sending these photons directly to the detectors.
Fig. 2 Picture of our layout with an overlay of the pilot HeNe laser beam that
we use for
aligning all the optical components.
Aligning the Mach Zehnder interferometer is not easy. As a general rule we force the beam to go square with the holes in the optical breadboard/table. We have our own systematic way of doing this:
Fig. 3 Steps to align a laser beam to be parallel to the rows of holes. We need
an iris mounted to a plate. We put screws
along a row of holes in two positions. Through an iterative process of changing
the position of the iris on its mount and
the direction of the beam via a mirror mount we can get the beam aligned.
The interferometer is set up by putting the components one by one, making sure that all the beams are aligned with rows of holes. In Fig. 2 we show our Mach-Zehnder interferometer. If we redo this setup again we would not use the tall mirror mounts. To reduce vibrations we connected the posts with rods. This was kind of clumsy. Better is to set up the path of the beams to be less than two inches from the table and using the pedestal mounts, as shown in Fig. 3. This makes the interferometer much more stable.
E.J. Galvez/Colgate U.