On the four corners of the table one finds diode lasers. On the corners nearest and furthest are grazing incidence lasers. The light from these lasers first goes across the face of a grating. Most of the light goes on to the experiment, but a little bit gets split off and sent to a mirror mounted on a piezo-electric transducer. This light bounces back into the laser, making what is known as an external cavity laser. The piezo and hence the mirror are maneuvered such that the laser is forced to a particular wavelength. In order to keep the laser on that wavelength we use saturation spectroscopy, in which some of the laser light is sent counterpropagating through a glass cell filled with Rubidium. The light of one direction goes into a photodiode, which in turn is connected to a lock-in amplifier and some home-built electronics. This allows us to put the laser on the wavelength of a particular atomic transition; the laser is then said to be "locked." One of these grazing incidence lasers serve as our "repump" laser, and we lock it to the 5S1/2 F=2 -> 5P1/2 F=3 transition at 795 nm.
Two other lasers are of what is know as the Littrow configuration. In this type there is no mirror, but rather light diffracted by the grating is sent directly back to the laser diode. One of these serves as our trapping laser, driving the 5S/1/2 F=3 -> 5P3/2 F=4 transition at 780 nm. The other is locked to the 5P3/2 F=4 -> 5D5/2 F=5 transition at 776 nm and is used as a master laser in an injection locking scheme. In such a scheme the light from a locked laser is sent into another (presumeably more powerful) laser in order to lock it at the desired wavelength. We use two master and two slave lasers in the STIRAP setup, the other master laser being a grazing incidence laser locked to the same transition as the trapping laser.
Near the center of the table is the cell trap itself, a glass structure supported by steel and connected to an ion pump. The trapping and repumping beams are split into three orthogonal beams which pass through the center of the cell and are reflected back on themselves. There are also two wire coils arranged in an anti-Helmholtz configuration. These produce an inhomogeneous magnetic field which, among other things, defines the center of the trap, lifts the hyperfine state degeneracy and provides the restoring force for the trap.