Polarized Photons for GlueX

1. GlueX Photon Beamline-Tagger Review Nov. 14-15, 2005, Newport News Polarized Photons for GlueX presented by Richard Jones, University of Connecticut GlueX Tagged Beam Working Group University of Glasgow University of Connecticut Catholic University of America

2. Photon Beam: coherent bremsstrahlung • Design options • laser backscatter • synchrotron backscatter • bremsstrahlung • coherent bremsstrahlung • Criteria • high energy8-9 GeV • linear polarization0.5 • high flux108 s-1 • energy resolution10 MeV r.m.s. • low background “figure of merit” = (rate in 8-9 GeV window) * (polarization)2 @ constant b.g. A B C D energy polarization  flux resolution() () ()() background  () with tagging

3. incoherent (black) and coherent (red) kinematics Photon Beam: energy spectrum effects of collimation:reduce low-energy background and increase polarization effects of collimation at 80 m distance from radiator

4. Photon Beam: collimation • With increased collimator distance: • polarization grows • low-energy backgrounds shrink • tagging efficiency drops off a < 20 mr s0 < 1/3 c d > 70 m

5. Photon Beam: rates and background • Total hadronic rate is dominated by the resonance region • For a given electron beam and collimator, background is almost • independent of coherent peak energy, comes mostly from incoherent part. peak energy8 GeV 9 GeV 10 GeV 11 GeV N in peak185 M/s 100 M/s 45 M/s 15 M/s peak polarization0.54 0.41 0.27 0.11 (f.w.h.m.)(1140 MeV) (900 MeV) (600 MeV) (240 MeV) peak tagging eff.0.55 0.50 0.45 0.29 (f.w.h.m.)(720 MeV) (600 MeV) (420 MeV) (300 MeV) power on collimator5.3 W 4.7 W 4.2 W 3.8 W power on H2 target 810 mW 690 mW 600 mW 540 mW total hadronic rate385 K/s 365 K/s 350 K/s 345 K/s (in tagged peak) (26 K/s) (14 K/s) (6.3 K/s) (2.1 K/s)

6. Photon Beam: collimation effects of collimation on polarization spectrum collimator distance = 80 m effects of collimation on figure of merit: rate (8-9 GeV) * p2 @ fixed hadronic rate linear polarization 5

7. Photon Beam: coherent gain factor effects of endpoint energy on figure of merit: rate (8-9 GeV) * p2 @ fixed hadronic rate gain increases with electron beam energy gain vanishes at the end-point

8. Photon Beam: polarization • circular polarization • transfer from electron beam • reaches 100% at end-point • linear polarization • determined by crystal orientation • vanishes at end-point

9. Electron Beam: expected properties energy12 GeV r.m.s. energy spread7 MeV transverse x emittance10 mm µr transverse y emittance2.5 mm µr electron polarizationavailable minimum current100 pA maximum current5 µA x spot size at radiator 1.6 mm r.m.s. y spot size at radiator0.6 mm r.m.s. x spot size at collimator0.5 mm r.m.s. y spot size at collimator0.5 mm r.m.s. position stability±200 µm

10. Diamond crystal: properties limits on thickness from multiple-scattering rocking curve from X-ray scattering natural fwhm

11. Diamond crystal: goniometer mount temperature profile of crystal at full operating intensity oC

12. Tagging spectrometer focal plane • Requirements • spectrum coverage 25% -- 95% in 0.5% steps 70% -- 75% in 0.1% steps • energy resolution 10 MeV r.m.s. • rate capability up to 500 MHz per GeV

13. Tagging spectrometer: two dipole design radiator quadrupole dipole 1 dipole 2 full-energy electrons photon beam focal plane • final bend angle: 13.4o • dipole magnetic field: 1.5 T • focal plane length: 9.0 m • energy resolution: 5 MeV r.m.s. • gap width: 3 cm • pole length: 3.1 m • dipole weight: 38 tons • coil power: 30 kW

14. Tagger focal plane: fixed array • 141 counters, spaced every 60 MeV (0.5% E0) • plastic scintillator paddles oriented perpendicular to rays • not designed for complete recoil electron coverage • conventional phototube readout • essential for diamond crystal alignment • useful as a monitor of photon beam • can run in counting mode at full source intensity

15. Tagger focal plane: microscope • Design parameters • square scintillating fibers • size 2 mm x 2 mm x 20 mm • clear light guide readout • aligned along electron direction • exploits the maximum energy resolution of the spectrometer • readout with silicon photomultipliers (SiPM devices) • dispersion is 1.4 mm / 0.1% at 9 GeV • 2D readout to improve tagging efficiency SiPM sensors clear light fibers scintillating fibers focal plane electron trajectory

16. Beam line simulation • Detailed photon beam line description is present in HDGeant • beam photons tracked from exit of radiator • assumes beam line vacuum down to a few cm from entry to primary collimator, followed by air • beam enters vacuum again following secondary collimator and continues down to a few cm from the liquid hydrogen target • includes all shielding and sweep magnets in collimator cave • monitors background levels at several positions in cave and hall • simulation has built-in coherent bremsstrahlung generator to simulate beam line with a realistic intensity spectrum • The same simulation also includes the complete GlueX target and spectrometer, detector systems, dump etc.

17. Beam line simulation cut view of simulation geometry through horizontal plane at beam height Hall D collimator cave Fcal tagger building Cerenkov vacuum pipe spectrometer

18. Beam line simulation overhead view of collimator cave cut through horizontal plane at beam height 12 m collimators concrete air vac vac sweep magnets iron blocks lead

19. Active collimator overview • active stabilization required for collimation • distance from radiator to collimator 75 m • radius of collimator aperture 1.7 mm • size of real image on collimator face 4 mm r.m.s. • size of virtual image on collimator face 0.5 mm r.m.s. • optimum alignment of beam center on collimator aperture ±0.2 mm in x and y • steering the electron beam • BPMs on electron beam measure x and y to ±0.1 mm • BPM pairs 2 m apart gives ±4 mm at collimator • BPM technology might be pushed to reach alignment goal, under the assumption that the collimator is stationary in this ref. sys.

20. Active collimator project overview • best solution: monitor alignment of both beams • monitor on electron beam position is needed anyway to control the spot on the radiator • BPM precision in x is affected by the large beam size along this axis at the radiator • independent monitor of photon spot on the face of the collimator guarantees good alignment • photon monitor also provides a check of the focal properties of the electron beam that are not measured with BPMs. 1.9 mm 1s contour of electron beam at radiator 0.5 mm

21. Beam line simulation 3d view of primary collimator with segmented photon rate monitor in front

22. Design criteria for photon monitor • radiation hard (up to 5 W of gamma flux) • require infrequent access (several months) • dynamic range factor 1000 • good linearity over full dynamic range • gains and offsets stable for run period of days • sampling frequency at least 60 Hz at operating beam current, 600 Hz desireable • fast analog readout for use in feedback loop

23. Design choice • Segmented scintillator • used for the Hall B collimator (lower currents) • not very rad-hard • Ion chamber • requires gas system and HV • good choice for covering large area • Tungsten pin-cushion detector • used on SLAC coherent bremsstrahlung beam line since 1970’s • SLAC team developed the technology through several iterations, refined construction method • reference Miller and Walz, NIM 117 (1974) 33-37 • SLAC experiment E-160 (ca. 2002, Bosted et.al.) still uses them, required building new ones • performance is known

24. Active Collimator Simulation 12 cm 5 cm

25. Active Collimator Simulation current asymmetry vs. beam offset y (mm) 20% 40% 60% x (mm) 12 cm

26. Detector response • the photons are incident on the back side of the pin-cushions (opposite the pins) • showers start in the base plates (~2 radiation lengths) • showers develop along the pins, leaking charges into the gaps • charge flow is asymmetric (more e- than e+) due to high-energy delta rays called “knock-ons” • asymmetry leads to net current flow on the plates proportional to the photon flux that hits it • SLAC experience shows that roughly 1-2 knock-ons are produced per incident electron 1 mA * 10-4rad.len.* ln(E0/E1)Þ 1 – 2 nA detector current

27. Detector response from simulation beam centered at 0,0 10-4 radiator Ie = 1mA inner ring of pin-cushion plates outer ring of pin-cushion plates

28. Beam position sensitivity using inner ring only for fine-centering ±200 mm of motion of beam centroil on photon detector corresponds to ±5% change in the left/right current balance in the inner ring

29. Electronics and readout • tungsten plate is cathode for current loop • anode is whatever stops the knock-ons • walls of collimator housing • primary collimator • for good response, these must be in contact • tungsten plate support must be very good insulator – boron nitride (SLAC design) • uses differential current preamplifier with pA sensitivity • experience at Jlab (A. Freyberger) suggests that noise levels as low as a few pA can be achieved in the halls • requires keeping the input capacitance low (preamp must be placed near the detector) • differential readout, no ground loops

30. Beam position sensitivity • Sensitivity is greatest near the center. • Outside the central 1 cm2 region the currents are non-monotonic functions of the coordinates. • A fitting procedure has been demonstrated that could invert the eight currents to find the beam center to an accuracy of ±350 mm anywhere within 3 cm of the collimator aperture. • Using BPMs and survey data, the electron beam must be able to hit a strike zone 6 cm in diameter from a distance of 75 m.

31. Beam line: shielding overhead view of collimator cave cut through horizontal plane at beam height 12 m collimators concrete air vac vac sweep magnets iron blocks lead

32. Beam line: physics simulation • GEANT-based Monte Carlo • based on a coherent bremsstrahlung generator • good description of electromagnetic processes • extended to include hadronic photoproduction processes • complete description of beam line including collimator and shielding • integrated with detector sim. • g,N scattering • g,A single nucleon knockout • g,p pion photoproduction

33. Photon beam polarimetry • Several methods have been considered by GlueX • hadronic reactions • coherent p0 photoproduction on spin-0 target • rho or omega photoproduction (helicity conservation) • pair production from a crystal • incoherent pair production • triplet production • coherent bremsstrahlung spectrum analysis • Detailed comparisons have been carried out • A PHOTON BEAM POLARIMETER BASED ON NUCLEAR E+ E- PAIR PRODUCTION IN AN AMORPHOUS TARGET, F. Adamian, H. Hakobian, Zh. Manukian, A. Sirunian, H. Vartapetian (Yerevan Phys. Inst.), R. Jones (Connecticut U.),. 2005. 9pp. Published in Nucl.Instrum.Meth.A546:376-384,2005 • Polarimetry of coherent bremsstrahlung by analysis of the photon energy spectrum,   S. Darbinyan, H. Hakobyan, R. Jones, A. Sirunyan and H. Vartapetian,Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, In Press, Corrected Proof, Available online 26 August 2005.

34. Polarimetry: method 0 • measure azimuthal distribution of p0 photoproduction from a spin-0 target • sometimes called “coherent p0 photoproduction” • elegant – has 100% analyzing power • analysis is trivial – just N(f) ~ 1 + P cos(2f) • coherent scattering is not essential • only restriction: target must recoil in a spin-0 state • practical example: 4He scattering • must detect the recoil alpha in the ground state • requires gas target, high-resolution spectrometer for Eg ~ GeV • cross section suppressed at high energy • not competitive at GlueX energies

35. Polarimetry: method 1 • pair production from a crystal • makes use of a similar coherent process in pair production as produced the photon in CB • requires counting of pairs, but not precision tracking • analyzing power increases with energy (!) • second crystal, goniometer needed • asymmetry is in rate difference between goniometer settings • can also be done in attenuation mode, using a thick crystal • sources of systematic error • sensitive to choice of atomic form factor for pair-target crystal • shares systematic errors with calculation of CB process • in addition to theoretical uncertainty, it involves a model of the beam and crystal properties

36. Polarimetry: method 2 • angular distribution from nuclear pair production • sometimes called “incoherent pair production” • polarization revealed in azimuthal distribution of plane of pair • requires precise tracking of low-angle pairs, which becomes increasingly demanding at high energy • analyzing power is roughly independent of photon energy • analyzing power depends on energy sharing within the pair • best analyzing power for symmetric pairs • requires a spectrometer for momentum analysis • systematic errors • atomic form factor • multiple scattering in target, tracking elements or slits • simulation of geometric acceptance of detector

37. Polarimetry: method 3 • angular distribution from pair production on atomic electrons • sometimes called “triplet production” • polarization reflected in a number of observables • azimuthal distribution of large-angle recoil electrons is a preferred observable • analyzing power is roughly constant with photon energy • no need for precision tracking of forward pair • pair spectrometer still needed to select symmetric pairs • systematic errors • reduced dependence on atomic physics – “exact” calculations • analyzing power sensitive to kinematic cuts • relies on simulation to know acceptance

38. Polarimetry: method 4 • analysis of the shape of the CB spectrum • not really polarimetry – this you do anyway • probably the only way to have a continuous monitor of the beam polarization during a run • polarimetry goal would be to refine and calibrate this method for ultimate precision • takes advantage of tagging spectrometer • requires periodic measurements of tagging efficiency using a total absorption counter and reduced beam current • systematic errors • atomic form factor used to calculate CB process • model of electron beam and crystal properties • photon beam alignment on the collimator

39. Measurements at YerPhI in 2005 • beam time obtained for CB measurements • electron beam energy 4.5 GeV • 4-week run during Fall 2005 • goniometer, crystal, pair spectrometer commissioned installed to test IPP method • results anticipated soon • YerPhI measurements supported by CRDF grant