1. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 1/30 Muon (g-2) to 0.2 ppm B. Lee Roberts Department of Physics Boston University roberts @ bu.edu http://physics.bu.edu/roberts/html

2. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 2/30 (in modern language) (and in English)

3. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 3/30 Dirac + Pauli moment

4. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 4/30 Standard Model Value for (g-2) e vrs.  : relative contribution of heavier things ?

5. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 5/30 One reason that I’m here is the relationship between e + e - annihilation and a 

6. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 6/30 When we started in 1983, theory and experiment were known to about 10 ppm. Theory uncertainty was ~ 9 ppm Experimental uncertainty was 7.3 ppm

7. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 7/30 E821 achieved 0.5 ppm and the e + e - based theory is also at the 0.6 ppm level. Both can be improved. All E821 results were obtained with a “blind” analysis. world average

8. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 8/30 With an apparent discrepancy at the level of 2.6 ... it’s interesting and you have to work harder to improve the measurement and the theory value ….

9. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 9/30 A (g-2)  Experiment to ± 0.2 ppm Precision –BNL E969 Collaboration R.M. Carey, I. Logashenko, K.R. Lynch J.P. Miller B.L. Roberts- Boston University; G. Bunce W. Meng, W. Morse, P. Pile, Y.K. Semertzidis -Brookhaven; D. Grigoriev B.I. Khazin S.I. Redin Yuri M. Shatunov, E. Solodov – Budker Institute; F.E. Gray B. Lauss E.P. Sichtermann – UC Berkely and LBL; Y. Orlov – Cornell University; J. Crnkovic,P. Debevec D.W. Hertzog, P. Kammel S. Knaack, R. McNabb – University of Illinois UC; K.L. Giovanetti – James Madison University; K.P. Jungmann C.J.G. Onderwater – KVI Groningen; T.P. Gorringe, W. Korsch – U. Kentucky, P. Cushman – Minnesota; Y. Arimoto, Y. Kuno, A. Sato, K. Yamada – Osaka University; S. Dhawan, F.J.M. Farley – Yale University

10. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 10/30 We measure the difference frequency between the spin and momentum precession 0 With an electric quadrupole field for vertical focusing

11. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 11/30 Inflector Kicker Modules Storage ring Central orbit Injection orbit Pions Target Protons (from AGS)p=3.1GeV/c Experimental Technique Spin Momentum Muon polarization Muon storage ring injection & kicking focus by Electric Quadrupoles 24 electron calorimeters R=711.2cm d=9cm (1.45T) Electric Quadrupoles polarized 

12. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 12/30 Pedestal vs. Time Near sideFar side E821: used a “forward” decay beam Pions @ 3.115 GeV/c Decay muons @ 3.094 GeV/c This baseline limits how early we can fit data

13. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 13/30 The Production Target proton beam top view of target

14. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 14/30 Decay Channel Plenty of room to add more quadrupoles to increase the acceptance of the beamline.

15. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 15/30 E969: will use a “backward” decay beam Pions @ 5.32 GeV/c No hadron-induced prompt flash Approximately the same muon flux is realized x 1 more muons Expect for both sides Then we quadruple the number of quadrupoles in the decay channel > x 2 new front-end increase of proton beam Decay muons @ 3.094 GeV/c

16. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 16/30 The incident beam must enter through the magnet yoke and through an inflector magnet

17. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 17/30 The mismatch between the inflector exit and the storage aperture + imperfect kick causes coherent beam oscillations Upper Pole Piece

18. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 18/30 The E821 inflector magnet had closed ends which lost half the beam. Length = 1.7 m Central field = 1.45 T Open end prototype, built and tested → X2 Increase in Beam

19. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 19/30 The 700 ton (g-2)  precision storage ring Muon lifetime t  = 64.4  s (g-2) period t a = 4.37  s Cyclotron period t C = 149 ns Scraping time (E821) 7 to 15  s Total counting time ~700  s Total number of turns ~4000

20. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 20/30 The fast kicker is the major new feature not in the CERN experiment. Kicker Modulator is an LCR circuit, with V ~ 95 kV, I 0 ~ 4200 A oil Fluorinert (FC40)

21. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 21/30 E821 Electron Detectors were Pb-scintillating fiber calorimeters read-out by 4 PMTs. New experiment needs segmented detectors for pileup reduction.

22. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 22/30 We count high-energy e - as a function of time.

23. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 23/30 New segmented detectors of tungsten/scintillating- fiber ribbons to deal with pile-up System fits in available space Prototype under construction Again the bases will be gated.

24. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 24/30 The magnetic field is measured and controlled using pulsed NMR and the free-induction decay. Calibration to a spherical water sample that ties the field to the Larmor frequency of the free proton  p. So we measure  a and  p

25. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 25/30 The ± 1 ppm uniformity in the average field is obtained with special shimming tools. We can shim the dipole, quadrupole sextupole independently 0.5 ppm contours

26. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 26/30 E969 needs 10 times more muons than E821 stored. Open Inflector X2 Backward Beam X1 Quadruple the Quadrupoles X 2-3 Beam increase design factor X 5 Absence of injection flash will permit us to begin analyzing data much earlier

27. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 27/30 The error budget for E969 represents a continuation of improvements already made during E821 Field improvements: better trolley calibrations, better tracking of the field with time, temperature stability of room, improvements in the hardware Precession improvements will involve new scraping scheme, lower thresholds, more complete digitization periods, better energy calibration Systematic uncertainty (ppm)1998199920002001E969 Goal Magnetic field –  p 0.50.40.240.170.1 Anomalous precession –  a 0.80.30.310.210.1 Statistical uncertainty (ppm)4.91.30.620.660.14 Total Uncertainty (ppm)5.01.30.730.720.20

28. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 28/30 Summary E821 Achieved a precision of ± 0.5 ppm There appears to be a discrepancy between experiment and e + e - based theory E969 proposes to push the precision down to ± 0.2 ppm There is lots of work worldwide on the hadronic theory piece, both experimental and theoretical. Thanks to all of you who are working on these problems! !

29. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 29/30 Outlook: E969 will be considered by the national U.S. Particle Physics Project Prioritization Panel (P5) at the end of March We hope that our friends in the theory, e + e - and  communities will continue to work on the hadronic contribution to a  If both theory can improve by a factor of 2, and experiment can improve by a factor of 2.5, the stage is set for another showdown.

30. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 30/30 Thanks to the organizers for this excellent workshop! Thank you СПАСИБО

31. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 31/30

32. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 32/30 σ systematic 199 9 2000 200 1 E969 Pile-up0.13 0.080.07 AGS Background 0.10 * Lost Muons0.10 0.090.04 Timing Shifts0.100.02 E-Field, Pitch0.080.03*0.05 Fitting/Binning0.070.06* CBO0.050.210.070.04 Beam Debunching 0.04 * Gain Change0.020.13 0.03 total0.30.310.210.11 Systematic errors on ω a (ppm) Σ* = 0.11 Backward beam Beam manipulation Detector segmentation and lower energy- threshold required for pile-up rejection with higher rates

33. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 33/30 Systematic errors on ω p (ppm) *higher multipoles, trolley voltage and temperature response, kicker eddy currents, and time-varying stray fields. E969 (i ) (I) (II) (III) (iv)

34. (g – 2)  B. Lee Roberts e + e - collisions  to  : Novosibirsk 1 March 2006 - p. 34/30 E969 Builds on the apparatus and Experience of E821 1.AGS Proton Beam 12 – bunches from the AGS 60 Tp total intensity 2.0 o  Beam  decay channel  Beam injected into the ring through a superconducting inflector 5.Fast Muon Kicker 6.Precision Magnetic Storage Ring 7.Electron calorimeters, custom high-rate electronics and wave-form digitizers