Hard Two-Photon Contribution to Elastic Lepton-Proton Scattering Determined by the OLYMPUS Experiment

The OLYMPUS Collaboration reports on a precision measurement of the positron-proton to electron-proton elastic cross section ratio, $R_{2\gamma }$, a direct measure of the contribution of hard two-photon exchange to the elastic cross section. In the OLYMPUS measurement, 2.01 GeV electron and positron beams were directed through a hydrogen gas target internal to the DORIS storage ring at DESY. A toroidal magnetic spectrometer instrumented with drift chambers and time-of-flight scintillators detected elastically scattered leptons in coincidence with recoiling protons over a scattering angle range of $\approx 20°$ to 80°. The relative luminosity between the two beam species was monitored using tracking telescopes of interleaved gas electron multiplier and multiwire proportional chamber detectors at 12°, as well as symmetric Møller or Bhabha calorimeters at 1.29°. A total integrated luminosity of $4.5{\mathrm{fb}}^{-1}$ was collected. In the extraction of $R_{2\gamma }$, radiative effects were taken into account using a Monte Carlo generator to simulate the convolutions of internal bremsstrahlung with experiment-specific conditions such as detector acceptance and reconstruction efficiency. The resulting values of $R_{2\gamma }$, presented here for a wide range of virtual photon polarization $0.456<\epsilon <0.978$, are smaller than some hadronic two-photon exchange calculations predict, but are in reasonable agreement with a subtracted dispersion model and a phenomenological fit to the form factor data.

Figure 1

Approximate effects of radiative corrections versus $\epsilon$ are on the order of several percent.

Figure 2

OLYMPUS result for $R_{2\gamma }$ using the Mo-Tsai [21] prescription for radiative corrections to all orders. Uncertainties shown are statistical (inner bars), uncorrelated systematic (added in quadrature, outer bars), and correlated systematic (gray band). Note the 12° data point at $\epsilon =0.978$ is completely dominated by systematic uncertainties.

Figure 3

Comparison of the recent results to the calculation by Blunden. The data are in good agreement, but generally fall below the prediction. Please note that data at similar $\epsilon$ values have been measured at different $Q^2$. Also note that the VEPP-3 data have been normalized to the calculation at high $\epsilon$.