M distributions:

The following graph shows the CO2 laser signal as a function of laser angle as the stage was scanned over the 10-29 transition in He. The laser was on line 10R(6). The angles shown are in degrees from copropogating. For this scan, the signal was recorded synchronous with the CO2 laser chopper, and the background was recorded synchronous with the target chopper, both on the off-axis channeltron with a bias of –800 V. Plotted is the ratio of signal to background. (The background is an indicator of the neutralization from the target.)

There is a scale factor thrown in because of the way the computer records the different signals. Thus the peak does not really mean that the signal is 50 times greater than the background. The background was taken on one of the small lock ins and the signal on the main li. I think the correct raw numbers are that the max signal is about 10 mV and the background is 200 mV.

Since the noise in these measurements depends mostly the number of background Rydberg atoms in the detector, I looked at how the background intensity depends on the voltage used in the stripper. In the experiment, I recorded the background signal on the off-axis channeltron and the neutral beam current on axis which is synchronous with the target chopping. After each change of the stripping field, the detector lens and y deflections were tuned for maximum background signal. The ratio of background to neutrals is shown below. The stripper gap used was 0.8 mm.

This data was taken before finding any signal, so I do not know how the signal depends on the stripper field. Actually, it was taken after failing to find the 10-30 transition. Most likely, that transition occures at too great a laser angle to work in the present LIR and z stack arrangement. The 10-29 data shown above was taken at a stripper voltage of 900 V. The 10-31 transition should also be accessible to me. Using that will allow me to use a lower stripper field than for the 29 state.

I also tried to use the z-field stack as a prestripper to empty out the high-n states before the laser excitation. The way this was done was to raise the stack to a potential Vz while leaving the snout at ground potential. Thus, a large field was created between the snout exit flange and the first plate of the stack. There was nominally no field in the LIR region in this setup. But because of fringing fields around the LIR mirror, I could not observe the transition with a potential applied to the stack. What is shown, again, is the ratio of background to target neutrals.

It does not look like this technique is immediately useful. For now I will have to operate the system with the LIR near to cround potential. In a future apparatus, however, it might be possible to build it so that the LIR region is shielded better from ground, or better yet, so the target can be at high potential and the LIR at ground. Then this technique might be able to do some good. I have not yet looked into what fields are permitted to not mix up the collision distribution.

The next step is to see how the z field system works at low fields. I will try to trim the potential on the LIR mirror to narrow the line widths as much as possible. Then I will start putting a voltage across the z stack to look at how the line shapes change as a function of electric field. I don't know if any individual m states will be visible from the F or G manifolds before everything gets smeared out, but it looks like a good idea to keep working with helium for this phase of the testing.

If it looks like the apparatus is working and signal-to-noise is large enough to hope to see the m states, then I will see if I can get S out of the source once again.