1. The Strange Vector Current of the Nucleon Forward Angle Experiment Jianglai Liu University of Maryland Strangeness, very briefly Parity violation as an experimental probe The G 0 experiment Emerging physics picture Fermi Lab seminar, 11-15-2005

2. Strangeness in the Nucleon? Quark models: Only u and d quarks in nucleons. No strangeness! Quark-antiquark pairs and gluons make up the QCD vacuum (“sea”). ss are “virtual pairs”, so the net strangeness is zero. s and s might not have identical distributions. So strangeness might manifest locally. Analogous to the charge distribution in neutron! “Full QCD description”

3. What’s the Big Deal of Strangeness? s quark belongs to the 2 nd generation. So “   Lamb shift in QCD”!!! Would have been hugely suppressed in a perturbative theory. Tough to calculate!!! Different from QED, however, QCD is non-perturbative! So the vacuum fluctuation could be sizable, so does strangeness! Nucleons are the “hydrogen atom” of QCD. ss gives direct access to the “loops” of QCD sea. Recall the QED loops and the famous Lamb shift, g-2 …

4. What Do We “Know” Already? Contribution of s quark to the longitudinal momentum Contribution to the nucleon mass Contribution to the nucleon spin DIS N charm production:  N scattering + hyperon mass splitting: likely 100% uncertainty  polarized inclusive DIS  elastic N scattering  polarized semi-inclusive DIS All these indicate s quarks contribute sizably in nucleon structure, however with large uncertainties.

5. Strange Vector Current EM quark current of the nucleon Define vector (EM) form factors:  Neglected heavier quarks  Charge symmetry Strange quark contributes to nucleon charge and magnetism? Distribution of nucleon’s charge and magnetization. Need one more constraint …

6. Neutral-weak Current  Additional Constraint NC contains a vector and axial piece QcVcV cAcA e -1+4sin 2  W 1 u2/3 1-8/3sin 2  W d,s-1/3 -1+4/3sin 2  W 1 So define NC vector form factor: Charges in the unified electroweak theory in SM sin 2  W = .2312 ± 0.00015 Kaplan and Manohar, 1988

7. s-quark form factors calculations at Q 2 =0 N N 

8. N e N e Elastic e-N scattering Measuring the NC Form Factor: Parity Violation  NC amplitude suppressed by ~10 -4  Difficult to see in cross- section measurement However, if one measures the parity violation in the elastic scattering, one accesses the interference between EM and NC interactions  “amplify” the relative experimental sensitivity to NC interaction. PCPV Mckeown and Beck, 1989

9. 60 Co  60 Ni* + e - + e C. S. Wu Measurement of Parity Violation 60 Co B  detector First observation of parity violation in weak interaction; Madam Wu’s famous 1957 60 Co beta decay experiment. In parity violating e-p scattering, the spin (helicity) of the electron is flipped back and forth. ep spin R L p’ e’ detector

10. Parity Violating Asymmetry kinematic factors forward ep backward ep backward ed Assuming EM and axial form factors are known (with errors), each measurement yield G E s +  G M s where

11. publishing, running x2, publishing, running published x2, running Summary of PV Electron Scattering Experiments From D.H. Beck

12. Strange FF: Results at Q 2 =0.1  2 =1 95% c.l. Combining world data (backward and forward) at Q 2 =0.1 GeV 2 allows one to separate G E s and G M S G M s = 0.55  0.28 G E s = -0.01  0.03

13. The Jefferson Laboratory A B C G0G0 polarized source linac

14. The G 0 Collaboration D.S.Armstrong 1, J.Arvieux 2, R.Asaturyan 3, T.Averett 1, S.L.Bailey 1, G.Batigne 4, D.H.Beck 5, E.J.Beise 6, J.Benesch 7, L.Bimbot 2, J.Birchall 8, A.Biselli 9, P.Bosted 7, E.Boukobza 2,7, H.Breuer 6, R.Carlini 7, R.Carr 10, N.Chant 6, Y.-C.Chao 7, S.Chattopadhyay 7, R.Clark 9, S.Covrig 10, A.Cowley 6, D.Dale 11, C.Davis 12, W.Falk 8, J.M.Finn 1, T.Forest 13, G.Franklin 9, C.Furget 4,D.Gaskell 7, J.Grames 7, K.A.Griffioen 1, K.Grimm 1,4,B.Guillon 4, H.Guler 2, L.Hannelius 10, R.Hasty 5, A. Hawthorne Allen 14, T.Horn 6, K.Johnston 13, M.Jones 7, P.Kammel 5, R.Kazimi 7, P.M.King 6,5, A.Kolarkar 11, E.Korkmaz 15, W.Korsch 11, S.Kox 4, J.Kuhn 9, J.Lachniet 9, L.Lee 8, J.Lenoble 2, E.Liatard 4, J.Liu 6, B.Loupias 2,7, A.Lung 7, G.A.MacLachlan 16, D.Marchand 2, J.W.Martin 10,17, K.W.McFarlane 18, D.W.McKee 16, R.D.McKeown 10, F.Merchez 4, H.Mkrtchyan 3, B.Moffit 1, M.Morlet 2, I.Nakagawa 11, K.Nakahara 5, M.Nakos 16, R.Neveling 5, S.Niccolai 2, S.Ong 2, S.Page 8, V.Papavassiliou 16, S.F.Pate 16, S.K.Phillips 1, M.L.Pitt 14, M.Poelker 7, T.A.Porcelli 15,8, G.Quéméner 4, B.Quinn 9, W.D.Ramsay 8, A.W.Rauf 8, J.-S.Real 4, J.Roche 7,1, P.Roos 6, G.A.Rutledge 8, J.Secrest 1, N.Simicevic 13, G.R.Smith 7, D.T.Spayde 5,19, S.Stepanyan 3, M.Stutzman 7, V.Sulkosky 1, V.Tadevosyan 3, R.Tieulent 4, J.van de Wiele 2, W.van Oers 8, E.Voutier 4, W.Vulcan 7, G.Warren 7, S.P.Wells 13, S.E.Williamson 5, S.A.Wood 7, C.Yan 7, J.Yun 14 1 College of William and Mary, 2 Institut de Physique Nucléaire d'Orsay, 3 Yerevan Physics Institute, 4 Laboratoire de Physique Subatomique et de Cosmologie- Grenoble, 5 University of Illinois, 6 University of Maryland, 7 Thomas Jefferson National Accelerator Facility, 8 University of Manitoba, 9 Carnegie Mellon University, 10 California Institute of Technology, 11 University of Kentucky, 12 TRIUMF, 13 Louisiana Tech University, 14 Virginia Tech, 15 University of Northern British Columbia, 16 New Mexico State University, 17 University of Winnipeg, 18 Hampton University, 19 Grinnell College

15. Measure forward & backward asymmetries  16 scintillator rings  recoil protons for forward measurement  electrons for backward measurements  elastic/inelastic for 1 H,  quasi-elastic for 2 H  Q 2 = 0.12~1.0 for forward  Q 2 = 0.3, 0.5, 0.8 for backward Forward measurements complete (101 C electrons). 10 13 protons per detector ring! E beam = 3.03 GeV, 0.33 - 0.93 GeV I beam = 40 uA, 80 uA P beam = 75%, 80%  = 52 – 76 0, 104 – 116 0  = 0.9 sr, 0.5 sr l target = 20 cm L = 2.1, 4.2 x 10 38 cm -2 s -1 A ~ -1 to -50 ppm, -12 to -70 ppm Overview of the G 0 experiment

16. G 0 in Hall C beam monitoring girder superconducting magnet (SMS) scintillation detectors cryogenic supply cryogenic target service module electron beamline Lumi monitors

17. Beam Properties New Tiger laser system for G 0  40uA, 32 ns time structure, much higher bunch charge.  P beam = 73.7  1.0%.  Need to minimize the helicity correlated beam properties to avoid false asymmetry. Accelerator “IA” Pockels Cell (charge) and PZT mirrors (position) feedback system to minimize helicity correlated beam properties

18. Spectrometer  Toroidal magnet, elastic protons dispersed in Q 2 along focal surface  Acceptance 0.12<Q 2 <1.0 GeV for 3 GeV incident beam  16 scintillator rings at the focal plane. 8 octants.  Azimuthally symmetric!  Detector 15 acceptance: 0.44-0.88 GeV 2  Detector 14: Q 2 = 0.41, 1.0 GeV 2  Detector 16: “super- elastic”, crucial to measure the background lead collimators elastic protons detectors target beam

19. Timing in the Experiment Accelerator pulse structure t 32 ns I beam Beam Helicity +1 ON DAQ OFF t 1/30 s ~500  s “Macropulse” Measurement timing Typical t.o.f. spectrum

20. Electronics High rate counting experiment, coinc. rate ~1MHz per scintillator pair. Electronics deadtime well-understood. Fast time encoding (ToF histogramming electronics. beam pick-off signal  T=0) DAQ rate, every helicity flip (30Hz) PMT Left PMT Right PMT Left PMT Right Mean Timer Mean Timer Coinc TDC / LTD Front Back Scalers: Histogramming

21. A phys Blinding Factor Analysis Overview Raw Asymmetries, A meas Beam and instrumental corrections: Deadtime Helicity-correlated beam properties Leakage beam Beam polarization Background correction Q2Q2 nucleon form factors Unblinding GEs+GMsGEs+GMs

22. Electronics Deadtime Corrections To the first order This shows up as 1)a decrease of normalized yield with I beam (almost like a target density reduction) 2)a correlation between A m and A Q. The DT effect is largely corrected based on the model of the electronics The residual A Q correlation is removed by the linear regression The final residual A phys dependence is corrected based on A phys and f residual (~0.05  0.05 ppm).

23. 499MHz Leakage Beam Use “cut0” region in actual data to measure leakage current and asymmetry throughout run. Worked out nicely.  ~ 50 nA 499 MHz beam leaks into G 0 beam (~ 40 uA)  Leakage current has large, varying asymmetry (A ~ 600 ppm).  A leak = -0.71  0.14 ppm (global uncertainty!) “G 0 ” A Q = 1000 ppm “Leakage” A Q = -1000 ppm 0 16 32 ns Leakage demo, extreme case BCM integrates charge, no sensitivity to the micro-structure!

24. Positive background asymmetry? Where do they come from ????

25. Physics Origin of the Positive Background Asymmetries GEANT simulation Weak decay-particles rescatter inside the spectrometer that make into the detector. Low rate, large asymmetry! Hyperons being produced in the scattering is highly polarized. Y → N+π is a PV decay with very large asymmetry (~1) However the decay-nucleon is highly supressed by the acceptance.

26. Results of the Hyperon Simulation Source explained; used measured data in the correction.

27. Background Correction  Yield & Asymmetry “2-step” Fit  Extract bin-by-bin dilution factor f b (t) by fitting time-of-flight spectra (gaussian signal + 4nd order polynomial background)  Use results to perform asymmetry fit: A e = const, A b = quadratic  2 / =31.1/40  2 / =37.5/44 constantquadratic detector 8

28. Det. 15 Background Corrections Elastic peak broadened (~6 ns) because of increased Q 2 acceptance. Smooth variation of the background yield and asymmetry over detector range 12-14, 16. So make linear interpretation over detector number to determine f b (t) and A b (t).

29. Detector 15 Asymmetry Compare interpolated background asymmetry and data Assume elastic asymmetry in each bin is a constant. Take interpolated f b and A b, fit three A e.

30. Elastic Asymmetries “No vector strange” asymmetry, A NVS, is A(G E s,G M s = 0) Nucleon EM form factors: Kelly PRC 70 (2004) 068202 Inner error bars: stat; outer: stat & pt-pt sys Primary contributors to the global error band:  Leakage correction (low Q 2 )  background correction (high Q 2 )

31. G E s +  G M s, Q 2 = 0.12-1.0 GeV 2 A  2 test based on the random and correlated errors: the non-vector-strangeness hypothesis is disfavored at 89%!  “Model” uncertainty is the EW rad. corr. unc. Dominated by the uncertainty of G A e.  3 nucleon form factor fits; spread indicate uncertainties.  |Kelly-FW| heavily driven by the difference in G E n. Are the data consistent with zero?

32. World Data with G 0 Q 2 =0.48 GeV 2 Strange quark contributes to  p at -10% level.

33. “Complete” picture with G 0 data Emerging picture: At low Q 2 end, despite the small , the data are positive, consistent with a large and positive G M s. Data go toward zero around Q 2 ~0.2, which suggests G E s might have a negative bump there. Data curve back up again for Q 2 >0.3 (note the growing  ), indicating that G M s stays positive. For G 0,  ~ 0.94Q 2

34. Global Fits to World Data à la Kelly Toy model, minimal physics input dipole form  Kelly form ensures G E s (0)=0, G E s  1/Q 4 when Q 2 large  Fixed b 3 = 1, all other variables float  Fit all 24 data points: 18 G 0, 3 HAPPEX, 2 A4, 1 SAMPLE

35. Results of the fit Excellent fit  2 = 14.9 19 d.o.f. Correlation Coeff.

36. Compare G E s with G E n, and G M s with G M p Recall the factor of -1/3 G E s and G M s Separately -1/3  s /  p = -18% -1/3G E s (0.2)/G E n (0.2)~40%

37. Very Naïve Interpretation (Conclusion?) Hannelius, Riska + Glozman, Nucl. Phys. A 665 (2000) 353 Interpret G E s and G M s results from the momentum space into spatial distribution. So the G E s and G M s results are both favor the picture that nucleon has an s quark “core” and s quark “skin” Naïve kaon cloud (s quark spatially outside) leads to a negative  s. Disfavored by the data  “s quark skin”. R. Jaffe, PLB 229 (1989) 275 Recall the well-known charge distribution in the neutron: positive bump in G E n  neutron has a positive core and negative “skin”. -1/3G E s has a positive bump  s quark spatially outside on average!

38. Strange FF in the Near Future G 0 backward: detect electrons at  = 108° Q 2 = 0.3, 0.5, 0.8 GeV 2 both LH 2 and LD 2 targets PV-A4 backward:  = 145° Q 2 = 0.23, 0.47 GeV 2 (underway) HAPPEX (H and He4) running now high precision at Q 2 = 0.1 GeV 2 high precision at Q 2 = 0.6 GeV 2 (proposed July 2005)

39. Prospective G 0 Data @ Q 2 = 0.8, 0.23 GeV 2 Run in Spring 06 at Q 2 = 0.79 GeV 2 (H and D targets) Possible run at Q 2 = 0.23 GeV 2 next (H alone?)

40. Summary The successful G 0 forward angle experiment yield the first measurement of parity-violating asymmetries over broad Q 2 range. PRL 95, 092001(2005) Emerging picture: G M s and G E s are both likely nonzero oG M s positive at low Q 2 and stays positive up to Q 2 =1.0 (GeV/c) 2 oG E s might have a negative bump at Q 2 ~0.2 (GeV/c) 2 os quark skin of the nucleon??? Stay tuned for G 0 backward results

41. Target 20 cm LH 2, aluminum target cell longitudinal flow, v ~ 8 m/s, P > 1000 W! negligible density change < 0.5% measured small boiling contribution: 260 ppm : 1200 ppm (stat. width)

42. Formalism Including EW Rad. Corr. Where and  Each asymmetry measurement can be cast into a linear combination of G E s and G M s, assuming everything else is known.  In forward angle, use theoretical value and uncertainty of G A e. Uncertainty dominated by the “anapole” term. At tree level, R’s are zeros. M.J. Mosolf et al, Phys Rep. 239, No. 1(1994) S.L Zhu et al, PRD 62,033008(2000)

43. Feedback Performance All parameters coverage to zero! A Q (ppm)-0.14  x (nm) 3 ± 4  y (nm) 4 ± 4

44. Acceptance Large and continuous acceptance for protons.

45. Helicity Correlated Beam Properties and Their Corrections So require Small ΔP i  Small sensitivity to P i  Azimuthal symmetry  large reduction of detector sensitivity to beam positions  Response of spectrometer to beam changes well understood  False asymmetries (and the uncertainty) due to helicity- correlated beam parameters very small (~ -0.02 ppm)

46. Measured Asymmetry upon Beam Spin Reversal

47. Detector 1-14 Background Uncertainty  Allowed background yield varied within “lozenge”. Similar approach for asymmetry.  Separated point-to-point (pt-pt) uncertainties in background correction from global uncertainties. E.g. linear  quadratic model of A b move A e downward for detectors 1-14

48. Det. 15 Asymmetry Compare interpolated background asymmetry and data In detector 15, the global uncertainties larger because bins are continuous. Assume elastic asymmetry in each bin is a constant. Take interpolated f b and A b, fit three A e.

49. Different Nucleon EM FF Parametrizations

50. Interpolate G 0 Data Three overlapping Q 2 with other experiments: Q 2 = 0.1(HAPPEX, SAMPLE, A4), Q 2 = 0.23 (A4), Q 2 =0.48 (HAPPEX)  Q 2 = 0.1 extrapolate G 0 using A i /Q 2 i for first 3 Q 2 points Q 2 = {0.122, 0.128, 0.136}  Q 2 = 0.23 (PVA4-I), 0.477 (HAPPEX-I) GeV 2 Interpolate A i /Q 2 i for Q 2 = {0.210, 0.232, 0.262},{0.410, 0.511, 0.631}  Average the results of flat and linear interpolation. Use the ½ difference as an additional “model” uncertainty.

51. Combine World Data I.Start from the experimental asymmetries and uncertainties II.Use a common set of form factor and EW parameters III.Calculate G E s +  G M s IV.Separate G E s and G M s V.The sensitivity to nucleon FF and EW parameters are evaluated separately by changing the model input globally and repeat I-IV General procedure:

52. G 0 Backward Angle Measurements  Match forward angle range with measurements at 3 momentum transfers  New detectors and electronics:  Cryostat Exit Detectors (CEDs): separate elastic and inelastic electrons by trajectory  Cerenkov Detectors: pion detectors  Counting electronics only (no ToF separation)  Trigger change to run with standard beam (499 MHz) Q2Q2 Beam EnergyTargetRateAsymmetry (GeV 2 )(GeV)(MHz)(ppm) 0.30.424H2H2 2.03-18 D2D2 2.80-25 0.50.576H2H2 0.718-32 D2D2 1.10-43 0.80.799H2H2 0.190-54 D2D2 0.274-72 Scheduled: Mar 06 – May 06 resume at Sep 06