Aligning the collimator

When the angles $\gamma$ and $\epsilon$ are adjusted to the correct values, the collimator is placed in the vacuum chamber behind the He* source. The collimator creates a beam of He* atoms parallel to the central axis of the collimator. So it is important that this central axis and the beam line from the He* source to the channeltron coincide. One end of the collimator can be moved in horizontal en vertical direction with the use of adjustment screws. To align the collimator we used the setup shown in fig 2.6.

Figure 2.6: Decreasing the angle $\phi$ between the collimator axis and the beam axis.

The laser light is shone in from one side under a large angle $\alpha$ with the collimator axis, so that the number of reflections inside the collimator is small. If the collimator axis is misaligned by an angle $\phi$, the light coming from one side of the collimator makes an angle $\beta_1 = \alpha - \phi$ with the beam axis and the light coming from the other side of the collimator makes an angle $\beta_2 = \alpha + \phi$ with the beam axis. Since the angles with the beam axis are not the same, the detuning for which the laser light would optimally deflect the He* atoms from the beam axis is not the same for the two sides. We periodically changed the frequency of the laser light around resonance, by putting a current with a sawtooth shape on the laser diode. If the detuning is optimal for one of the two angles to interact with the He* atoms, the number of atoms detected at the channeltron will drop significantly. If we scan over the frequency and the angle $\phi$ is not zero, we will see two dips in the signal.

Figure 2.7: Channeltron signal with sawtooth current on diode laser, with large angle $\phi$ (above) and small angle $\phi$ (below).

The distance between these dips is an indication of the angle $\phi$. If the angle $\phi$ is zero, the two angles $\beta_1$ and $\beta_2$ are equal, and instead of two dips, we will see only one, much smaller dip. The dip will be smaller since the atoms are now in resonance with the laser light coming from both mirrors at the same time. Since the laser light leaves the collimator at the same side as it enters the collimator, we know that the number of times the laser light overlaps the He* beam is even, so that the number of times the laser light interacts with the beam from one side is equal to the number of times the laser light interacts with the beam from the other side. This means that when the angle $\phi$ is zero, the two forces on the He* atoms are almost equal and opposite, and the net force on the He* atoms is almost zero.

The width of the dip is not only caused by the width of the transition given by equation 1.4, but is mainly caused by the velocity distribution of the He* source. Since the channeltron is placed at a distance of 4.5 meters behind the collimator, the spread in the velocity of 350 m/s will cause a spread in flight time of 1.3 ms. The signal received at the channeltron is a convolution of the resonant frequency range of the laser and the time distribution. This makes it more difficult to find the correct position of the collimator for which the angle $\phi$ is zero. When the angle $\phi$ is minimized as well as can be done with this method, we shine the lasers in from both sides under the angle $\alpha_0$ and optimize the angles $\phi_{horizontal}$ and $\phi_{vertical}$ for the two pairs of mirrors independently until the channeltron signal is optimal.