(Juengst, Strakovsky, Briscoe)

The goal of the first part of our program of pion photoproduction from the nucleon is the extraction of precision total and differential cross sections for both single- and double-pion production. The reactions g N ® p N and g N ® p p N are of fundamental importance for our understanding of the strong interaction [Arn02a]. In particular, the JLab single-pion data will be included in the SAID [Arn02b], MAID [Dre99], and other partial-wave analyses and will constitute a major contribution to those analyses. However, our data will only have an impact on the current models if the uncertainty of the cross section is 5% or less for the gp reactions and 7% or less for the gn reactions. The current results from other groups of the CLAS collaboration indicate a deviation of about 9% from the SAID database. This deviation is compatible with the experimental uncertainty. We intend to reduce the systematic uncertainty to ~3%. These results will have a major impact on the normalization of other CLAS analyses. The CLAS measurements will provide unique and self-consistent results from tagged photons over a broad range of angles and energies, and with few exceptions, will represent the only pion-photoproduction data above E_g = 1.8 GeV.

In the g1 and g2 series of experiments, we have measured single-pion photoproduction using CLAS and the Tagged-Photon Facility in Hall B. Electron beams of energies from 1.8 to 3.2 GeV were used to produce tagged photons with energies between 0.36 and 3.1 GeV. Cryogenic LH₂ (g1) and LD₂ (g2) targets were used for protons and neutrons, respectively. Once all contributions to our systematic uncertainties are well understood, cross sections will be determined to an estimated absolute accuracy of better than 4% for the gp reactions and 6% for the gn reactions. The relative precision of measurements using the same target will reach the 1-2% level. Angular increments of 5° for angles between 20° and 140° in the center of mass are obtainable for energy increments of 10 MeV. The larger uncertainty in the g2 measurements (that used deuterium to provide a neutron target) is a reflection of the uncertainties involved in correcting for the fact that the neutron is bound in the deuteron. Comparison of the existing data for g n ® p - p with our inverse photoproduction measurements, p^-p ® gn, (via detailed balance) performed at BNL with the Crystal Ball Spectrometer (see section II.B.1) has now provided a test of the validity of these corrections at that level [Sh03].

The extraction of cross sections and polarizations are separated into data reconstruction (calibration and event selection) and physics analysis (simulation and normalization). Event selection and physics analysis are straightforward with CLAS for reactions with one or no missing neutral particle. The 3-momentum of each charged particle is measured with the CLAS drift chambers (DC), and time-of-flight measurement is used for particle identification (PID). For the reactions g p ® p 0 p and g p ® p ⁺n, we need only to measure p or p⁺ and identify the neutral particle in the missing 4-momentum. In the later stages of our analysis, we will deal with the more complicated measurement of the p^o through its decay gammas.

The electromagnetic shower counters (EC and LAC) are useful for exclusive measurements of pion production with a neutron or decay gs of a p^o in the final state. This makes it possible to estimate the background contributions from reactions with additional gs and p^os. Right now we are only using detection of neutrals to test positive-particle trigger efficiencies. However, we could eventually produce cross sections for the reactions gn ® np^o(p^o...) from the deuteron target with spectator protons by detecting neutrals in the shower counters, albeit with rather small efficiency and over a limited range in angular acceptance.

In our extensive check of systematics, we have come across several areas in the analysis procedures that need improvement. One example of the detail required is found in the calibration and simulation of CLAS drift chamber data. The frequency distribution of the reconstructed experimental drift radius and the drift-time vs drift-radius plots indicate that the models used for tracking and simulation are too simplistic. To solve this we introduced a Tchebychev polynomial to describe the drift radius as a function of drift time and other parameters. This not only improves the drift distribution, but also allows us to use the inverse of the polynomial to perform simulations with the same calibration constants that we use for the experimental data. We now can obtain a more accurate modeling of the drift-chamber data, which reduces the uncertainties in both the tracking and acceptance calculations. Full procedures and improvements will appear in the analysis report that will be presented to the collaboration this Fall, together with our experimental results.

This site is best viewed in Internet Explorer 6.0 or higher. Send mail to Web Master with questions or comments about this web site.
Copyright © 2004 GW Experimental Nuclear Physics Research Group
Last modified: 03/31/05