1. Argonne National Laboratory is managed by The University of Chicago for the U.S. Department of Energy Probing Hadronic Structure with Baryonic Probes – Drell-Yan Measurements Donald Geesaman Physics Division Argonne National Laboratory

2. 2 How to make progress on hadron structure? Observables – –Spectroscopy Parton distributions fit into the first category as Deep inelastic scattering Flavor sensitivity from neutrinos or semi-inclusive flavor tagging Graphic stolen from JLab

3. 3 Recent Progress Spin and orbital angular momentum –sea carries little spin (HERMES) –anticipating gluon measurements at RHIC and Compass –Lots of interest on transverse spin distributions Strange quark content to Electric and Magnetic form factors.02<x bj <1

4. 4 Drell-Yan Experiments on the horizon FNAL E906 – 120 GeV proton induced Drell-Yan – in the proton – in nuclei –parton energy loss J-PARC – 50 GeV proton induced Drell-Yan – in the proton –parton energy loss RHIC –polarized proton Drell-Yan –W production at higher energies (s 1/2 ~ 500 GeV) GSI – FAIR antiproton-induced Drell-Yan (with polarization?) –Transversity –Sivers Function

5. 5 Proton-induced Drell-Yan scattering (Fixed Target): A laboratory for studying sea quark distributions Leading Order x target x beam proton }X}X }X}X -- ++ Detector acceptance chooses range in x target and x beam. x F = x beam – x target > 0 high-x Valence Beam quarks. Low/interm.-x sea Target quarks.

6. 6 Method to study flavor dependence of anti-quark distributions Assumptions Use full parton distributions in analysis. The approximate sensitivity is

7. 7 pQCD - Gluon splitting? Meson Cloud? Chiral Solitons? Instantons? Models describe well, but not —pQCD becoming dominant? LA-LP-98- 56 Soon lattice moment analysis may also weigh in. Structure of the nucleon: What produces the nucleon sea? Peng et al.

8. 8 The models all have close relations between antiquark flavor asymmetry and spin Statistical Parton Distributions

9. 9 Fermilab Accelerator Complex: Fixed Target Program Fixed Target Beam lines Tevatron 800 GeV Main Injector 120 GeV E866 vs. E906: 800 vs. 120 GeV Cross section scales as 1/s –7 x that of 800 GeV beam Backgrounds (J/  decay) scale as s –7 x Luminosity for same detector rate as 800 GeV beam 50 x statistics!!

10. 10 FNAL E906 Collaboration Abilene Christian University Donald Isenhower, Mike Sadler, Rusty Towell Argonne National Laboratory John Arrington, Don Geesaman *, Kawtar Hafidi, Roy Holt, Hal Jackson, Paul E. Reimer *, David Potterveld University of Colorado Ed Kinney Fermi National Accelerator Laboratory Chuck Brown University of Illinois Jen-Chieh Peng * Co-Spokespersons Los Alamos National Laboratory Gerry Garvey, Mike Leitch, Pat McGaughey, Joel Moss Rutgers University Ron Gilman, Charles Glashausser, Xiaodong Jaing, Ron Ransome Texas A & M University Carl Gagliardi, Bob Tribble, Maxim Vasiliev Thomas Jefferson National Accelerator Facility Dave Gaskell Valparaiso University Don Koetke

11. 11 Drell-Yan Acceptance Programmable trigger removes likely J/  events Transverse momentum acceptance to above 2 GeV Spectrometer could also be used for J/ ,  0 studies

12. 12 Projected errors on ratios of D to H

13. 13 Does deuterium structure affect the results at higher x Important to also extend nuclear results to higher x

14. 14 Projected results on ratio of d-bar/u-bar Parton Distributions PDF fits are and uncertainties completely dominated by E866. E906 will significantly extend these measurements and improve on uncertainty. Absolute cross sections on deuterium give d-bar+u-bar Impact Collider/LHC sensitivity for tests of the Standard Model—Background. Origins of the Proton Sea—Models explain d-bar > u-bar. No theory (model) expects the results seen for x > 0.3.

15. 15 Structure of nucleonic matter: How do sea quark distributions differ in a nucleus? Comparison with Deep Inelastic Scattering (DIS) Antishadowing not seen in Drell- Yan—Valence only effect?— better statistical precision needed—E906. sea Intermediate-x sea PDF’s set by -DIS on iron—unknown nuclear effects. What can the sea parton distributions tell us about nuclear binding?

16. 16 Structure of nucleonic matter: Where are the nuclear pions? Nucleon motion in the nucleus tends to reduce parton distributions – f(y) peaked below y=1. Rescaling effects also reduce parton distribution for x>0.15 Antiquark enhancement expected from Nuclear Pions.

17. 17 Radiative corrections calculations are now finished—small effect. Fermilab E906 will add much more precise high-x data. Drell-Yan Absolute Cross Sections: Proton Structure as x  1 MRST and CTEQ: d/u  0 as x  1 Reach high-x through beam proton—Large xF)large xbeam. Proton-Proton—no nuclear corrections—4u(x) + d(x) Proton-deuterium (cross check) agrees with proton-proton data. Parton distributions overestimate cross section. Working with CTEQ to incorporate data in global PDF fits. PRELIMINARY

18. 18 Effects beyond leading order Interpretability of Drell-Yan based on factorization theorem ~ 0.4 GeV 2 ~ 70 GeV on Pb, 23 GeV on Ca How low in mass can we interpret continuum spectrum as Drell-Yan? Parton Energy Loss –radiative or collisional Decay Angular Distribution of Drell-Yan pQCD = 1, , = 0... More generally Lam-Tung relation +2 = 1 not satisfied in pion-induced Drell-Yan at high x. Related to chiral odd quark transversity function h 1 t (x,p t )

19. 19 Parton Energy Loss Colored parton moving in strongly interacting media. Only initial state interactions are important—no final state strong interactions. E866 data are consistent with no energy loss Treatment of parton propagation length and shadowing are critical –Johnson et al. find 2.2 GeV/fm from the same data Energy loss  1/s—larger at 120 GeV Important to understand RHIC data.

20. 20 Role of J-PARC (Letter of Intent L15) Obviously can in principle reach higher x. Energy may restrict nuclear parton dependence to light nuclei. Other nuclear effects would be interesting – but would need higher energy data to separate effects If polarization possible, whole new ball game

21. 21 Energy Loss at 50 GeV Even a suggestion that nuclear spin-orbit interaction has significant effect

22. 22 GSI – FAIR Drell-Yan with anti-protons In anti-proton-proton collisions, the focus is on valence quark distributions Initial plans are ~15 GeV antiprotons –This really restricts di-muon mass range –Are di-lepton masses below 4 GeV interpretable as Drell-Yan? –Also important question for RHIC There are ideas for asymmetric collider to get some reach into “safe” Drell-Yan region--- s 1/2 = 5.5 GeV  s 1/2 =14.7 GeV POLARIZED ANTIPROTON-PROTON collisions –Novel polarization technique looks feasible –Emphasis on transverse distributions –double spin – Twist two transversity distribution –single spin – h 1 t (x,p t )

23. 23 GSI: Phase II (PAX@HESR) EXPERIMENT: 1. Asymmetric collider: polarized antiprotons in HESR (p=15 GeV/c) polarized protons in CSR (p=3.5 GeV/c) 2. Internal polarized target with 22 GeV/c polarized antiproton beam. Physics: Transversity Second IP with minor interference with PANDA

24. 24 Drell-Yan kinematics Drell-Yan kinematics Statistics concentrates at low-M 2 x~(P p P pbar ) -1/2 x 1 x 2 s = M 2 = Q 2 x 1 -x 2 = x F = 2 p L √s

25. 25 PAX-Precision of h 1 measurement 1 year of data taking 15 + 3.5 GeV/c Collider L = 2∙10 30 cm -2 s -1 hundreds of events/day 10 % precision on the h 1 u (x) in the valence region

26. 26 Summary Fixed-Target Drell- Yan is the ideal way to study the quark sea. What is the structure of the nucleon? –d-bar/u-bar at intermediate-x –Parton distributions as x  1 What is the structure of nucleonic matter? –Where are the nuclear pions? –Is antishadowing a valence effect? Do partons lose energy? Many of these problems have been with us for years, but they are still at the heart of hadron structure PRELIMINARY