Backgrounds

The biggest problem with detecting the protons is the presence of background processes that generate protons close to the elastic peak. In particular, photoproduction of neutral pions will cause a background of high energy protons. For the low Q² data, the threshold for pion production is far enough below the elastic peak that it can be easily cut away. For the higher Q² value, these protons can have momenta less than 1% below the elastic peak. For these kinematics, we will need to subtract away these contributions. Figure 4 shows a measured spectrum of proton elastic singles from a Hall A measurement at Q²=3.0 GeV², along with a monte carlo simulation of the contributions from elastic scattering from protons (smeared to match the HRS resolution), quasielastic scattering from the aluminum target windows, and protons coming from pion photoproduction (using d$\sigma$/d$t \propto s^{-7}$, and normalizing to the measured distribution). For a cut of $\vert\delta p/p\vert < 1$%, the pion photoproduction background is a 3% contribution to the yield, and is well reproduced by the calculated photoproduction spectrum.

While the calculated elastic spectrum plus pion photoproduction background does a good job of reproducing the proton spectrum, we will make additional tests of our photoproduction background calculation. For roughly half of the kinematics, the pion photoproduction threshold is well separated from the elastic peak, and we can test our photoproduction spectrum with only the tail of the elastic peak as background. We will have coincidence runs at three kinematics, which will allow us to separate the elastic and the photoproduction in order to test our calculations of the lineshapes. As an additional test we will put a hodoscope in the hall to tag electrons corresponding to the detected elastic protons for all forward angle kinematics (where the electrons are scattered at larger angles). By rejecting events where an electron is detected, we can examine inclusive protons with the elastic peak suppressed in order to compare the photoproduction background to our calculated lineshape. By rejecting events with no detected electron, we can generate a sample of events with a suppressed photoproduction background. We will not use the electron hodoscope to remove background events in the analysis, because it would reject events due to radiation of the outgoing electron and because any difference in efficiency or solid angle matching between the high and low $\epsilon$ points could introduce a large uncertainty in the result.

  

Figure 4: HRS Proton elastic singles spectrum for E₀=3.4 GeV, Q²=3.0 GeV² (crosses). The curves show simulations of the elastic scattering (dotted), endcap contributions (dash-dot), and pion photoproduction (dashed), and the histogram shows the sum of the simulated contributions.

There will also be a background of charged pion photoproduction. For several kinematics (including most of the high Q² kinematics), the pion production threshold is far enough below the elastic peak to cleanly separate the pions. For the other kinematics, time of flight will efficiently remove pions for the low Q² data, and an Aerogel detector will be used to reject pions where the time of flight is not fully efficient. In addition, having both Aerogel and time of flight separation will allow us to determine the efficiency of the particle identification cuts.