1. Progress and opportunities of unpolarised Drell-­Yan program Wen-Chen Chang Institute of Physics, Academia Sinica, Taiwan COMPASS beyond 2020 Workshop March 21-22, 2016 1

2. Outline 2

3. Progress of Drell-Yan Experiments in the past 3

4. Deep Inelastic Scattering (DIS) and Drell-Yan Processes 4

5. Parton Distribution Function (PDF) of Proton: MMHT 2014 PDFs 5 L. A. Harland-Lang, A. D. Martin, P. Motylinski, R.S. Thorne, arXiv:1412.3989

6. Naïve Assumption: NMC (Gottfried Sum Rule): NA51 (Drell-Yan, 1994): E866/NuSea (Drell-Yan, 1998): Light Antiquark Flavor Asymmetry: Drell-Yan Experiments 6

7. 7 E. Pereza and E. Rizvib, arXiv:1208.1178

8. 8

9. Dimuon Invariant-mass Distributions (2014 COMPASS DY) 9

10. Dimuon x 1,x 2 and p T Distributions (2014 COMPASS DY) 10

11. Projections: COMPASS polarized DY ● 2015: ~80 000 DY events M>4 GeV/c 2 from NH3 target ● 2018: ~116 000 events (assuming same conditions, but 160 effective days) Sivers asymmetry statistical error:~1.8% Catarina Marques Quintans 11

12. Opportunity (1): Nucleon (+ ̅ ) & Kaon PDFs 12

13. Large Uncertainty of (+ ̅ ) at valence-quark region 13 arXiv:1412.3989

14. Charged Kaon SIDIS HERMES Collaboration, PRD 89, 097101 (2014) 14

15. (+ ̅ ) from SIDIS of kaons HERMES Collaboration, PRD 89, 097101 (2014) 15

16. Large Uncertainty in the extraction of (+ ̅ ) W.C. Chang and J.C. Peng, PRD 92, 054020 (2015) 16

17. Kaon Partonic Structure NA3 Collaboration, PLB 93, 354 (1980) 17

18. Valence-quark distribution functions in the kaon and pion Chen, et al., arXiv:1602.01502 18 Algebraic formulae for the dressed-quark propagators and pion and kaon Bethe-Salpeter amplitudes where SU(3) symmetry breaking is implemented.

19. Valence-quark distribution functions in the kaon and pion Chen, et al., arXiv:1602.01502 19

20. Kaon-induced Drell-Yan Process: Avoiding Fragmentation Functions Uncertainty 20 Kaon-induced Drell-Yan cross sections will determine nucleon strange quark structure kaon PDFs Kaon-induced Drell-Yan cross sections will determine nucleon strange quark structure kaon PDFs Kaon beam and LH 2 target

21. Opportunity (2): Violation of Lam-Tung relation & Boer-Mulders Functions 21

22. 22 Drell-Yan decay angular distributions Collins-Soper frame  and  are the decay polar and azimuthal angles of the μ + in the dilepton rest-frame

23. NA10 @ CERN: Violation of Lam-Tung Relation Z. Phys. 37 (1988) 545 23   +W 140 GeV   +W 194 GeV   +W 286 GeV

24. E615 @ FNAL: Violation of Lam-Tung Relation PRD 39, 92 (1989) 24 252-GeV   +W

25. Azimuthal Asymmetries Require Nontrivial Spin Correlation 25 What will be the mechanism? Brandenburg, Nachtmann & Mirkes, ZPC 60 (1993) 697

26. Brandenburg, Nachtmann & Mirkes [Z. Phy. C 60 (1993) 697]: QCD Vacuum Effect 26 O. Nachtmann, ECT DY workshop, 2012 spin-flip synchrotron gluons 0.2-0.3 fm O. Nachtmann / Annals of Physics 350 (2014) 347–378

27. Boer [PRD 60, 014012 (1999)]: Hadronic Effect, Boer-Mulders Functions 27 Spin-orbit correlation of transversely polarized noncollinear partons inside an unpolarized hadron

28. Chen & Li [PLB 726 (2013) 262] Breaking of Factorization by Glauber Gluons 28 S e is large only for the pion as Nambu-Goldstone boson.

29. Theoretical Interpretations of Lam- Tung Violation in pion-induced DY 29 Boer-Mulders Function QCD chromo- magnetic effect Glauber gluon Origin of effectHadronQCD vacuumPion specific Quark-flavor dependenceYesNo Hadron dependenceYesNoYes Large P T limit0Nonzero0 Yes (valence quarks involved) Yes Yes (valence quarks involved) YesYes/No Yes (valence quarks involved) YesNo No (sea quarks involved) YesNo ReferencesPRD 60, 014012 (1999)Z. Phy. C 60,697 (1993)PLB 726, 262 (2013) Kaon and antiproton beams

30. Consistency of LT relation in p+p and p+d DY: E866 (PRL 99 (2007) 082301; PRL 102 (2009) 182001) 30 ν(π - W  µ + µ - X)~ [valence h 1 ┴ (π)] * [valence h 1 ┴ (p)] ν(pd  µ+µ-X) ~ [valence h 1 ┴ (p)] * [sea h 1 ┴ (p)] Sea-quark BM functions are much smaller than valence quarks.

31. Boer-Mulders functions from unpolarized pD and pp Drell-Yan data 31 Z. Lu and I. Schmidt, PRD 81, 034023 (2010) V. Barone et al., PRD 82, 114025 (2010) Sign of BM functions and flavor dependence?

32. Lu, Ma & Schimdt [PLB 639 (2006) 494]: Flavor separation of the Boer–Mulders functions 32 MIT Bag Model Spectator Model Large Nc limit LD 2 target

33. Opportunity (3): Distribution amplitudes 33

34. CIP (PRL 43, 1219, (1979)) : Longitudinally Polarized Photon at large x 1 34

35. 35

36. Brandenburg et al. (PRL 73, 939 (1994)) Higher-twist Effect & Pion Distribution Amplitude Pion distribution amplitude: distribution of LC momentum fractions in the lowest-particle number valence Fock state. 36

37. Brandenburg et al. (PRL 73, 939 (1994)) Pion Distribution Amplitude  :E615 37

38. Brandenburg et al. (PRL 73, 939 (1994)): Pion/Kaon/Antiproton Distribution Amplitude 38 Kaon and antiproton beams

39. 39

40. Opportunity (4): Flavor-dependent EMC effect 40

41. Cloet et. al (PRL 102, 252301, 2009): Flavor dependence of the EMC effect? 41 A (Z=N)

42. Dutta et al. (PRC 83, 04220, 2011): Pion-induced Drell-Yan and the flavor-dependent EMC effect 42

43. Dutta et al. (PRC 83, 04220, 2011): Pion-induced Drell-Yan and the flavor-dependent EMC effect 43 LD 2 and nuclear targets

44. Feasibility for COMPASS (Beyond 2020) 44

45. Wish List of COMPASS from Unpolarized Drell-Yan Program 45 Very Long LH2/LD2 targets Many nuclei targets High-intensity pion beam High-intensity kaon beam High-intensity antiproton beam

46. Dimuon Vertex Distributions (2014 DY) 46

47. Fraction of particles in the positive or negative M2-Hadron-beam at COMPASS target 47 http://www.staff.uni-mainz.de/jasinsk/index.htm ~1% ~97% ~2.5% 1x10 8 /sec

48. Beam PID: CEDAR (Cerenkov Differential Counters with Achromatic Ring Focus) 48 Improvement of CEDARs system performance for higher rate capability: “CEDARs for DY run ”, Ivan Gnesi in 2014 June COMPASS TB MeetingCEDARs for DY run

49. Z eff A eff Z eff /A eff Density ρ (g/cm 3 ) pion interaction length λ int (g/cm 2 ) Thickness L (cm) Effective Thickness L eff (cm) NH 3 5.9411.710.507 0.85 x 0.5 (F f ) 114.8111090.36 1x10 8 Vertex Detector: Polystyrene [C 6 H 5 CHCH 2 ] n 5.6211.160.5041.06113.71.51.496.66x10 7 Al 13 26.980.482 2.70136.7 7.0 6.54 6.56 x10 7 W (Beam Plug) 74 183.80.403 19.25218.712011.36 5.72x10 7 Pion interaction length (Aluminum) = 50.64 cm 7.0 cm = 13.8 % of pion interaction length Primary π - Beam Flux Al W 6.56 x 10 7 sec -1 5.72 x 10 7 sec -1 VTX NH 3 I0I0 1 x 10 8 sec -1 Al (7cm) + W Estimation of #DY 49 Attenuation of hadron beams in target materials Takahiro Sawada Final configuration: 7-cm Al target, 267 mm from the beam plug.

50. interaction length λ int (g/cm 2 ) Thickness L (cm) Effective Thickness L eff (cm) NH 3 127 110 92.0 1x10 8 *(0.025*0.5) = 1,250,000 Vertex Detector: Polystyrene [C 6 H 5 CHCH 2 ] n 125 1.5 1.49865,000 Al149 7.0 6.57854,000 W (Beam Plug)234 120 12.2752,000 Estimation of #DY Attenuation of beam hadrons in material interaction length λ int (g/cm 2 ) Thickness L (cm) Effective Thickness L eff (cm) NH 3 82.5 110 84.0 1x10 8 *(0.01*0.8) = 800,000 Vertex Detector: Polystyrene [C 6 H 5 CHCH 2 ] n 81.3 1.5 1.49454,000 Al103 7.0 6.40445,000 W (Beam Plug)183 120 9.51371,000 Assumption: K- : 2.5 % x 0.5 (PID efficiency) anti-p : 1% x 0.8 (PID efficiency) ( 1 ) (0.692) (0.683) (0.602) (0.567) (0.556) (0.463) ( 1 ) ( scale ) Takahiro Sawada 50

51. Expected Statistical Precision of Dimuon Angular Distributions 51 NA10 λ μ ν λ μ ν λ μ ν p T (GeV/c) COMPASS, DY on W target, 140-day data taking 4 < M μμ < 9 GeV/c 2

52. Expected Statistical Precision of Dimuon Angular Distributions 52 NA10 λ μ ν λ μ ν λ μ ν COMPASS, DY on W target, 140-day data taking 4 < M μμ < 9 GeV/c 2 x1x1 x1x1 x1x1 λ μ ν

53. Improved CEDAR: Expected Statistics of Unpolarized Drell-Yan Events 53 NH 3 Al (7cm)W 140,00027,100270,000 1,7503503,700 1,2602201,800 140-day data taking, with the efficiencies of 2015 DY run. DY ( 4 < M μμ < 9 GeV/c 2 ) NA3 21,220 700 E615 27,977 E537 387 NA10 284,200

54. 54 WHAT ABOUT A RF SEPARATED  p BEAM ??? RF2 RF1 DUMP Choose e.g.   p  = 2  (L f / c) (  1 -1 –  2 -1 ) with  1 -1 –  2 -1 = (m 1 2 -m 2 2 )/2p 2 L DUMP Momentum selection First and very preliminary thoughts, guided by recent studies for P326 CKM studies by J.Doornbos/TRIUMF, e.g. http://trshare.triumf.ca/~trjd/rfbeam.ps.gz E.g. a system with two cavities:

55. 55 Base rough estimates on acceptance values for RF separated K + beam (as provided by J.Doornbos) CKM K + beam  p beam Beam momentum [GeV/c]60100 Momentum spread [%]±2±1 Angular emittance H, V [mrad]±3.5, ±2.5 Solid angle [  sterad]10-12  % wanted particles lost on stopper3720 As the  p kick is more favourable than for K +, I assume that 80% of  p pass beyond the beam stopper. Acceptance 10   sterad, 2 GeV/c

56. 56 H.W.Atherton formula tells us : 0.42  p / int.proton / GeV Assume target efficiency of 40% Then for 10 13 ppp on target one obtains: 0.4. 10 13. 0.42. . 10 -5. 2. 0.8  ppp = 8 10 7  ppp for a total intensity probably not exceeding 10 13 ppp, knowing that e - and  are well filtered, but K + only partly. Due to 10 8 limit on total flux, max antiproton flux remains limited by purity (probably about 50%). Hence ≈ 5 10 7  ppp VERY PRELIMINARY CONCLUSION

57. RF/Separated Kaon/antiproton : Expected Statistics of Unpolarized Drell-Yan Events 57 NH 3 Al (7cm)W 140,00027,100270,000 14,0002,80029,600 15,7502,75022,500 140-day data taking, with the efficiencies of 2015 DY run. DY ( 4 < M μμ < 9 GeV/c 2 ) NA3 21,220 700 E615 27,977 E537 387 NA10 284,200 A flux of 1x10 7 /sec for kaon/antiproton is assumed.

58. Conclusions Drell-Yan process is a powerful tool to explore the partonic structures of nucleons and (unstable) mesons. Drell-Yan program with Improved CEDAR/RF- separated beam and LH 2 /LD 2 /nuclear targets will bring unique opportunities for COMPASS to address many important unresolved issues in understanding the flavor and TMD structures of proton, antiproton, pion, kaon and nuclei. 58