Two Photon Exchange Contributions to Elastic e + p --> e + p Process in a Nonlocal Field Formalism

TL;DR

This paper calculates two photon exchange effects in elastic electron-proton scattering using a nonlocal field theory, revealing their potential to explain discrepancies between different experimental methods and predicting epsilon dependence of form factor ratios.

Contribution

The work introduces a gauge-invariant nonlocal formalism for two photon exchange calculations and explores its implications for experimental discrepancies in proton form factor measurements.

Findings

Two photon exchange corrections depend on an unknown parameter .

Corrections can reconcile differences between polarization transfer and Rosenbluth measurements.

Predicted epsilon dependence of form factor ratios can be tested experimentally.

Abstract

We compute the two photon exchange contributions to elastic scattering of polarized electrons from target protons. We use a nonlocal field theory formalism for this calculation. The formalism maintains gauge invariance and provides a systematic procedure for making this calculation. The results depend on one unknown parameter \bar b. We compute the two photon exchange correction to the ratio of electric to magnetic form factors extracted using the polarization transfer experiments. The correction is found to be small if \bar b ~ 1. However for larger values of \bar b>3, the correction can be quite significant. The correction to the polarization transfer results goes in the right direction to explain their difference with the ratio measured by Rosenbluth separation method. We find that the difference between the two experimental results can be explained for a wide range of values of the parameter \bar b. We also find that the corrections due to two photon exchange depend on the photon longitudinal polarization epsilon. Hence we predict an epsilon dependence of the form factor ratio extracted using the polarization transfer technique. Finally we obtain a limit on \bar b by requiring that the non-linearity in epsilon dependence of the unpolarized reduced cross section is within experimental errors.