January 30, 2014 - Special Theory Meeting

Meeting Logistics

OUR EXPERIMENTAL PROGRAM: To precisely measure scattering asymmetry of spin polarized electrons from thin pure Z foils; Accessible energy range is about 3 to 8 MeV; Targets on hand are gold, silver, copper foils of varying thickness, but can install others readily; Measure polarimeter systematics with goal <0.5%

OUR SIMULATION PROGRAM: Build GEANT4 model of polarimeter and benchmark detector response against measurement; Implement physics (cross-section, Sherman function, spin transfer functions) provided by theory; Develop ab initio simulation dependence of asymmetry on target thickness (multiple scattering, radiation effects, etc) to measurement; By calibrating model to experimental data predict zero-thickness asymmetry with high precision <1%

DESIRED THEORY PROGRAM: Provide physics tables for simulation (cross-section, Sherman function, spin transfer functions); Describe theoretical basis, corrections, uncertainties; Advise which measurements may be best tests on leading corrections or limit absolute knowledge of physics; Consider calculation to improve uncertainty in physics below 1%

Question #1 (Joe 1/13/2014): Coulomb screening is a leading effect for electron energies <1 MeV, and finite nuclear size is a leading effect for energies >10MeV (when DeBrogile wavelength is comparable to nuclear radius). Although our polarimeter is optimized for 5MeV we can operate with beam energies typically from about 3-8 MeV. What is the size of the uncertainty on the corrections in this energy range and for Z we use such as gold, silver, copper? Are we sensitive enough in this region to perform a test on the uncertainty of the physics calculations, e.g. using suitable Z or extending the energy reach?

Question #2 (Joe 1/15/2014): Is the biggest uncertainty in the theoretical calculations radiative corrections? Can a sound theoretical calculation be made that calculates this contribution with relative accuracy of ~30%?

Question #3 (Joe 1/15/2014): Although we will pester the theorists about the precision on the Sherman function calculation, also the cross-section and spin transfer functions are derived from the same F & G functions. In particular, all four are relevant for our GEANT4 simulations. Are the uncertainties for the latter functions similar or are there any arguments why they would be larger or smaller? We need to understand the uncertainties that are folded into the physics for the target thickness simulation.

Question #4 (Joe 1/15/2014): What would be the most interesting or useful measurements we could attempt to help appreciate or validate theory calculations?

Question #5 (Tim 1/15/2014): For Mott scattering at 120 keV, where radiative corrections (through, e.g., bremsstrahlung emission) are thought to affect the Sherman function S well below the 1% level [G.Holzwarth and H.J.Meister, Nucl.Phys. 59, 56 (1964)], the only way an electron can lose energy is in an “elastic” scattering event through nuclear recoil, which is kinematicaly well-defined for a fixed scattering angle. The Mott scattering process at 120 keV is thus pretty well defined. At 5 MeV, energy loss is about 50% due to bremsstrahlung [L.Pages et al., At.Data 4, 1 (1972)], so my assumption is that radiative corrections may significantly affect the Sherman function. Will these corrections be taken into account and by how much do you estimate they will affect the value of S?

Question #6 (Tim 1/15/2014): How accurate do you estimate your calculations will be?

Question #7 (Tim 1/15/2014): The only way we will have to assess the accuracy of your calculations is by measuring relative values of S for various values of Z and E (and possibly θ). Do you anticipate doing calculations for a variety of targets and energies between, e.g., 3 and 6 MeV?

Question #8 (Charlie 1/16/2014): Are there any effects from inelastic scattering that affect the measured polarization?

Question #9 (Charlie 1/16/2014): Is it a correct statement that the (complex) F and G functions for scattering an electron from a point nucleus are calculable with uncertainties small enough to be unimportant for our project?

Question #10 (Charlie 1/16/2014): Would there be any benefit from improving the energy resolution of our detection scheme (say, to perhaps 1%)?