Sign of the Asymmetry

R. Michaels, Nov 2009

Experiments take great care to get the asymmetry right to 1% accuracy but what about the sign ? Getting the sign wrong is a 100% error. It would be dishonest to normalize the sign to the Standard Model.

Ingredients in the Sign

Conventions for Moller sign.
Physics of electron-electron scattering: the cross-section is larger when the spin of the electrons are anti-parallel, and smaller when they are parallel.
g-2 precession of the spin in the accelerator and effect of the Wien filter. It requires the BMT equations (see Jackson) plus knowledge of the laser polarization. There is long experience at the lab.
Sign conventions of the DAQ and software. Now, and for 10 years, this has been opposite for Moller and HAPPEX DAQs. This means that using a standard convention that A = (sigma_R - sigma_L) / (sigma_R + sigma_L) for Right(R) and Left((L) handed longtidunally polarized electrons, we need to flip our sign.

Example Applied to Nov 2009 PVDIS Data

Do not take my word for it ! Decide yourself !

Nov 6, 2009. Moller results
- Look at bottom of Moller results . click on detailed info ...
- detailed info from Nov 6.
- Halfwave plate state: OUT
- Expected asymmetry for R-L with present Wien setting (says Sasha, but we should check): POSITIVE
- Coil = -15.9 and A_raw = +0.0509 consistent with POSITIVE expectation.
PVDIS Asymmetries
- Our Slug 1 Plots.
- A = +240 ppm POSITIVE
- Halfwave plate state OUT
- HAPPEX Sign check: Our "R" is Moller's "L". (Now, and for 10 years previous.)
- Therefore, PVDIS Asymmetry: A = -240 ppm (NEGATIVE).
- Theory: A (R-L) = -113 ppm. (NEGATIVE).
- Therefore Signs are correct !
- The odds are 50%. :)