Precision Electron Beam Polarimetry

1. Wolfgang Lorenzon (Michigan) SPIN 2008 Symposium 6-October 2008 Precision Electron Beam Polarimetry

2. How to measure polarization of e-/e+ beams? Three different targets used currently: 1. e- - nucleus: Mott scattering 30 – 300 keV (5 MeV: JLab)spin-orbit coupling of electron spin with (large Z) target nucleus 2. e - electrons: Møller (Bhabha) scat. MeV – GeVatomic electron in Fe (or Fe-alloy) polarized by external magnetic field 3. e - photons: Compton scattering > 1 GeVlaser photons scatter off lepton beam Goal: measureDP/P ≈ 1% (Hall C, EIC)DP/P ≈ 0.25% (0.1%) (ILC)realistic?

3. Electron Polarimetry Many polarimeters are, have beenin use, or a planned: Compton Polarimeters: ILC 45.6 – 500 GeV EIC 3 – 20 GeV LEP mainly used as machine tool for resonant depolarization SLAC SLD 46 GeV DESY HERA, storage ring 27.5 GeV (three polarimeters) JLab Hall A < 8 GeV / Hall C < 12 GeV Bates South Hall Ring < 1 GeV Nikhef AmPS, storage ring < 1 GeV Møller / Bhabha Polarimeters: Bates linear accelerator < 1 GeV Mainz Mainz Microtron MAMI < 1 GeV Jlab Hall A, B, C

4. Polarimeter Roundup

5. Phys. Rev. ST Accel. Beams 7, 042802 (2004) The “Spin Dance” Experiment (2000) Source Strained GaAs photocathode (l = 850 nm, Pb >75 %) Accelerator 5.7 GeV, 5 pass recirculation Wien filter in injector was varied from -110o to 110o to vary degree of longitudinal polarization in each hall → precise cross-comparison of JLab polarimeters

6. “Spin Dance” 2000 Data Pmeas cos(hWien+f)

7. Polarization Results Results shown include statistical errors only → some amplification to account for non-sinusoidal behavior Statistically significant disagreement Systematics shown: Mott Møller C 1% Compton Møller B 1.6% Møller A 3% Even including systematic errors, discrepancy still significant

8. Polarization Results- Reduced Data Set Hall A, B Møllers sensitive to transverse components of beam polarization Normally – these components eliminated via measurements with foil tilt reversed, but some systematic effects may remain closed circles = full data set open circles = reduced data set Agreement improves, but still statistically significant deviations  when systematics included, discrepancy less significant

9. Lessons Learned Providing/proving precision at 1% level challenging Including polarization diagnostics and monitoring in beam lattice design is crucial Measure polarization at (or close to) IP Measure beam polarization continuously protects against drifts or systematic current-dependence to polarization Flip electron and laser polarization fast enough to protect against drifts Multiple devices/techniques to measure polarization cross-comparisons of individual polarimeters are crucial for testing systematics of each device at least one polarimeter needs to measure absolute polarization, others might do relative measurements (fast and precise) absolute measurement does not have to be fast New ideas?

10. Compton vs Møller Polarimetry HERA EIC Jlab -7/9 ILC • Detect g at 0°, e-< Ee • Strong  need <<1 • at Ee< 20 GeV • Plaser~100% • non-invasive measurement • syst. Error: 3 → 50 GeV (~1 → 0.5%) hard at < 1 GeV: (Jlab project: ~0.8%) • rad. corr. to Born < 0.1% • Detect e-at qCM~90° •  good systematics • beam energy independent • ferromagnetic target PT~8% • beam heating (Ie < 2-4 mA), Levchuck eff. • invasive measurement • syst. error 2-3% typically0.5% (1%?) at high magn. field • rad. corr. to Born < 0.3%

11. E. Chudakov et al., IEEE Trans. Nucl. Sci. 51, 1533 (2004) Polarized atomic hydrogen in a cold magnetic trap use ultra-cold traps (at 300 mK: Pe~ 1-10-5, density ~ 3∙1015 cm-3 , stat. 1% in 10 min at 100 mA) expected depolarization for 100 mA CEBAF < 10-4 limitations: beam heating → “continuous” beam & complexity of target advantages: expected accuracy < 0.5% & non-invasive, continuous, the same beam Problem: very unlikely to work for high beam currents for EIC (due to gas and cell heating) Jet Target: avoids these problems VEPP-3 100 mA, transverse stat 20% in 8 minutes (5 ∙ 1011 e- /cm2 , 100% polarization) What is electron polarization in a jet? • Jet Target: need to address these problems • Breit-Rabi measurement analyzes only part of jet • → uniformity of jet has to be understood • large background from ions in the beam: most of them associated with jet (hard to measure) • origin of background observed in Novosibirsk still unclear? • clarification of depolarization by beam RF needed • → might be considerable

12. Jeff Martin (Winnipeg) Electron Beam LaserBeam New Fiber Laser Technology (Hall C) 30 ps pulses at 499 MHz Gain switched • external to beamline vacuum → easy access • huge lumi boost when phase locked • excellent stability, low maintenance

13. Compact, Off-The Shelf, Rack Mountable… RF locked low-power 1560 nm fiber diode ErYb-doped fiber amplifier Frequency doubler 13

14. Fiber Laser for Hall C Compton • Seed laser at 1064 nm • Fiber amplifier (50 W output at 1064 nm) • Frequency doubling cavity • Result: 25 W, 532 nm, 30 ps pulses at 499 MHz • JLab Polarized source group is building laser (J. Grames) • problems with amplification 14

15. Dominant Challenge: determine Az • Best tool to measure e- polarization • → Compton e- (integrating mode) • Challenge • accurate knowledge of ∫Bdl • must calibrate the electron detector • fit the asymmetry shape or use Compton Edge 15

16. W. Lorenzon PSTP 2007 Kent Paschke Electron Polarimetry

17. Future Efforts • EIC • e-p and e-ion collisions at c.m. energies: 20 - 100 GeV • 10 GeV (~3 - 20 GeV) electrons/positrons • Longitudinal polarization at IP: ~70% or better • Needed accuracy: DP/P = 1% • Bunch separation: 3 - 35 ns • Luminosity: L(ep) ~1033 - 1034 cm-2 s-1 per IP • ILC • e- - e+ collisions at c.m. energies: 45.6 - 500 GeV • Longitudinal polarization at (IP) • P(e- ) > 80% • P(e+ ) > 50% • Needed accuracy: DP/P = 0.25% (0.1%) → new territory 17

18. EIC Compton Polarimeter • No serious obstacles are foreseen to achieve 1% precision for electron beam polarimetry at the EIC (3-20 GeV) • JLAB at 12 GeV will be a natural testbed for future EIC e-/e+ Polarimeter tests • evaluate new ideas/technologies for the EIC • There are issues that need attention (crossing frequency 3-35 ns; beam-beam effects at high currents; crab crossing effect on polarization) 18

19. ILC Polarimeters Goal:DP/P ≈ 0.25% (0.1%) upstream polarimeter downstream polarimeter ~150 m behind IP to e+e- IP: 1.8 km • Three ways to measure polarization at the ILC • upstream Compton polarimeter • downstream Compton polarimeter • (e+e- →) W+W- production (has large cross section & very sensitive to polarization) • Complication • polarization at IP = lumi-weighted polarization ≠ polarization at polarimeter • depolarization and spin transport effects estimated at 0.1%-0.4% levels! • same level as required accuracy • keep errors in these effects small • Need upstream, downstream & e+e- physics measurements • determine best values for each polarimeter separately (hide from each other) • compare and see whether they agree • final calibration with e+e- → W+W- 19

20. Summary • Electron beam polarization can be measured with high precision • no serious obstacles are foreseen to achieve 1% • imperative to include polarimetry in beam lattice design • use multiple devices/techniques to control systematics • Big challenge to reach DP/P = 0.25% • no fundamental show stoppers (but high beam energy needed) • DP/P = 0.1% enters new territory • goes beyond traditional polarimetry • use physics processes directly (W+W- production) • Build on experience, but be open to new ideas • maybe use e-p elastic scattering at low energies (measures PePT) • … • Extensive modeling necessary • AZ, depolarization, BMT, etc. • much work done, much work still ahead 20