Extension to E93-021: the Charged Pion Form Factor

At JLab, we have a unique opportunity to dramatically improve the F database. Much can be learned about the usefulness of QCD sum rules and relativistic potential models for understanding the structure of the pion in the (presumably) diicult and non-perturbative Q 2 regime of 1-5 (GeV=c) 2. This proposal deals with the continuation of our successful F program. Using the Hall C spectrometers and beam with energy up to 4.045 GeV, we h a ve recently obtained a result (E93-021) for the charged pion form factor (F) up to Q 2 =1.6 (GeV=c) 2 at W=1.95 GeV. Based on this result, a n umber of issues arise which w e address in this proposal: 1. The E93-021 result extends the region of high quality F data from Q 2 = 0.7 (GeV=c) 2 to 1.6 (GeV=c) 2. Even at Q 2 = 1.6 (GeV=c) 2 , the old Cornell F values are widely scattered, and are not based on a true L/T separation. (A certain recipe was taken for the transverse cross section, which according to the E93-021 results gives too small T values.) The higher energy beam that is now a vailable allows us to perform high quality F measurements with the existing Hall C instrumentation up to Q 2 = 2.5 (GeV=c) 2. This is the region where the theoretical calculations for F begin to diverge, and data are used as input to several of the QCD-related models of F to constrain the treatment of the soft contributions. 2. The higher-energy beam also allows us to perform high quality measurements at higher W than in E93-021. As extraction of F from the data inherently depends upon a model of the p(ee e 0 +)n reaction, the higher W is advantageous because it allows measurements to be taken closer to the + pole than otherwise, where t-channel contributions dominate. In addition, for the E93-021 result, the Regge model used in the extraction of the form factor had a shallower dd L =dt dependence than the data, resulting in a F model dependence comparable to the experimental uncertainty. A l i k ely reason for this is that the value W = 1 :95 GeV of the measurement was a bit low, and resulted in resonance contributions to the cross section. This model dependence is expected to be reduced if the measurements are performed at higher W, as proposed …