In this experiment we take a high intensity electron beam and break it up into 33 ms long “windows” with opposite polarizations — spin pointing forward in one window, backwards in the next. We scatter the electrons off the protons in a hydrogen target (or the nuclei in a helium target, for the HAPPEX-Helium experiment) and measure the difference in the scattering rates for the two polarizations. At the energy and angle we work at, the rates in hydrogen differ by just about one part per million (ppm) — as we measured last year. For helium, it’s about 6.7 ppm.
To measure such small rate differences, or asymmetries as we call them, you have to collect a lot of data — we ran helium this year from July through September, and hydrogen from October to now — and you have to be careful with systematics; among other things, you have to measure and keep track of differences in the position, angle, energy, and intensity of the beam between one polarization and the other. We not only measured the beam intensity difference but suppressed it, by feeding back our measurements to control the beam source. Averaged over our 5-week hydrogen run, the intensity difference was about 170 parts per billion. Feeding back on beam position is possible — we did it in SLAC experiment E158 — but tricky and to be avoided if possible. So we just set up the source carefully and let it run, and measured average beam position differences that were zero within a precision of 1 to 2 nanometers. That’s about 20 to 40 times the radius of a single hydrogen atom.
And why do all this? Not for any known practical application, of course. But, through what looks at first like black magic but is really rather simple reasoning, measuring these asymmetries gives you information about the contributions made by strange quarks to the radius and magnetic moment of the proton. If you know what a strange quark is, you probably were taught that there aren’t any in a proton, which leaves you wondering how they could contribute to anything. The answer is that there are no valence strange quarks in a proton — none that persist independently, that is — but there’s a “sea” of virtual quark-antiquark pairs continually appearing and annihilating. Some of these sea quarks (and antiquarks) should be strange, and they could influence the properties of the proton.
That’s what we’re trying to measure. The first HAPPEX experiment, six years ago, measured a strange quark contribution consistent with zero, as did last year’s run of HAPPEX-II and a couple of other similar experiments, but taken together they suggest a nonzero effect — and recently another JLab experiment, G Zero, claimed to have established a strange quark contribution to the proton radius and magnetic moment. I think that claim is a little too strong, though certainly their results are tantalizing.
G Zero measured asymmetries at several different angles. HAPPEX-II took data at only one angle, but with higher precision. If there’s an effect there as big as the other experiments suggest, we should be able to measure it cleanly. G Zero will no doubt continue to claim they got there first, but we may be able to produce a numerical value for the strange quark contribution with a high confidence level, something G Zero cannot yet do.
In the pipeline, HAPPEX-III will return to an angle and energy near those of HAPPEX-I and measure that point with higher precision. The HAPPEX collaboration will also use similar techniques in PREX, the Pb Radius Experiment, to get at a completely different piece of physics: the thickness of the “neutron skin” of the lead nucleus. The asymmetry in that experiment will be only 0.5 ppm; that and other issues (for example, how do you put a 0.5 mm lead foil into a 400 watt electron beam without turning your accelerator into the world’s largest and most expensive soldering iron? Answer: cool the edges of the foil cryogenically, and sandwich it between two thin layers of diamond.) will make it a serious challenge. We’re up to it.
But that’s to come… this week, we’re done. I spent the last ten days as Run Coordinator, meaning I was in charge of day to day conduct of the experiment, meaning I carried the official cell phone and pager, spent a lot of time in the “counting house” where we control data taking, and went to a lot of meetings. Miraculously no one ever had to call me in the middle of the night — all our catastrophes happened on day and swing shifts. Things actually went very well, and our data look excellent. We spent the last couple of shifts doing tests for PREX, putting a lead target in the beam (and not melting it). Today we had an end of run party, which was still going strong when I decided I was about to topple over so I went back to my room and slept. A couple more meetings tomorrow, then Thursday I fly home for family reunion and turkey dinner. It’s been a long haul but I feel good about it.