Fill and Sign the Mcu Bca Syllabus PDF Form

P-1022 v2.9 7 Nov. 2011 Revised Proposal: Intensity-Frontier Antiproton Physics with The Antiproton Annihilation Spectrometer (TAPAS∗) at Fermilab Giorgio Apollinari,4 David M. Asner,23 Wander Baldini,5 Larry Bartoszek,1 Daniel R. Broemmelsiek,4 Charles N. Brown,4 Alak Chakravorty,24 Paul Colas,2 Paul Derwent,4 Alexey Drutskoy,12 Michael Fortner,20 Ioannis Giomataris,2 Anjan Giri,11 Keith Gollwitzer,4 H. Richard Gustafson,18 Alan Hahn,4 Timothy Holmstrom,16 Gerald P. Jackson,7 Tord Johansson,25 David E. Johnson,4 Porter W. Johnson,10 Daniel M. Kaplan,10,∗ Penelope Kasper,4 ByeongRok Ko,13 Kwong Lau,9 Jonathan Lewis,4 Mario Macri,6 Mauro Marinelli,6 Michael Merkin,19 Sandip Pakvasa,8 Vaia Papadimitriou,4 HyangKyu Park,14 Todd K. Pedlar,17 Thomas J. Phillips,3 Olga Piskunova,15 Jerome Rosen,21 Giulio Stancari,4 Michelle Stancari,4 Ray Stefanski,4 Yagmur Torun,10 James T. Volk,4 Mitchell Wayne,22 Steven Werkema,4 William Wester,4 Christopher G. White,10 Herman B. White,4 G. P. Yeh4 1 Bartoszek Engineering, Aurora, IL 60506, USA 2 CEA Saclay, Gif-sur-Yvette, France 3 Duke University, Durham, North Carolina 27708, USA 4 Fermilab, Batavia, Illinois 60510, USA 5 INFN, Sezione di Ferrara, Ferrara, Italy 6 INFN, Sezione di Genoa, Genoa, Italy 7 Hbar Technologies, LLC, West Chicago, Illinois 60185, USA 8 University of Hawaii, Honolulu, Hawaii 96822, USA 9 University of Houston, Houston, TX 77004, USA 10 Illinois Institute of Technology, Chicago, Illinois 60616, USA 11 Indian Institute of Technology, Hyderabad, India 12 Institute for Theoretical and Experimental Physics, RU-117259 Moscow, Russia 13 Korea University, Seoul, 136-701, Korea 14 KyungPook National University, DaeGu, Korea 15 Lebedev Physical Institute, RU-117924 Moscow, Russia 16 Longwood University, Farmville, Virginia 23909, USA 17 Luther College, Decorah, Iowa 52101, USA 18 University of Michigan, Ann Arbor, Michigan 48109, USA 19 Moscow State University, Moscow, Russia 20 Northern Illinois University, DeKalb, Illinois 60115, USA 21 Northwestern University, Evanston, Illinois 60208, USA 22 Notre Dame University, Notre Dame, Indiana 46556, USA 23 Pacific Northwest National Laboratory, Richland, Washington 99352, USA 24 St. Xavier University, Chicago, Illinois 60655, USA 25 Uppsala University, SE-751 05 Uppsala, Sweden ∗ Spokesperson. E-mail address: kaplan@iit.edu Summary The Fermilab Antiproton Source is the world’s most intense source of antimatter. With the Tevatron program now behind us, this unique facility can help make the case for Fermilab’s continued accelerator operations. The Antiproton Source can be used for unique, dedicated antimatter studies, including medium-energy p-annihilation experiments. We propose to assemble a powerful, yet cost-effective, solenoidal magnetic spectrometer for antiproton-annihilation events, and to use it at the Fermilab Antiproton Accumulator to measure the charm production cross section, study rare hyperon decays, search for hyperon CP asymmetry, precisely measure the properties of several charmonium and nearby states, and make the first measurements of the Drell–Yan continuum in medium-energy antiproton annihilation. Should the charm production cross section be as large as some have proposed, we will also be able to measure D0 –D0 mixing with high precision and discover (or sensitively limit) charm CP violation. The observation of charm or hyperon CP violation would be evidence for physics beyond the Standard Model, with possible implications for the origin of the baryon asymmetry of the universe — the question of what happened to all the antimatter that must have been produced in the Big Bang. The experiment will be carried out by an international collaboration and will require some four years of running time. As possibly the sole hadron experiment in progress at Fermilab during that time, it will play an important role in maintaining a broad particle physics program at Fermilab and in the U.S. It will thus help us to continue attracting creative and capable young people into science and technology, and introducing them to the important technologies of accelerators, detectors, and data acquisition and analysis — key roles in society that accelerator-based particle physics has historically played. i Contents Summary i List of Figures iv List of Tables vi 1 Introduction 1 2 TAPAS Physics Overview 2.1 Hyperons . . . . . . . . . . . . . . . 2.2 Charmonium and X(3872) . . . . . . 2.3 Antiproton Drell–Yan Studies . . . . 2.4 Charm-Meson Mixing, CP Violation, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . and Rare Decays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Experimental Approach 6 4 Measurement Program 4.1 Hyperon CP Violation and Rare Decays . . . . . . . 4.1.1 Hyperon sensitivity estimates . . . . . . . . . 4.2 Precision Measurements in the Charmonium Region 4.3 Antiproton Drell–Yan . . . . . . . . . . . . . . . . . 4.4 Charm Mixing, CP Violation, and Rare Decays . . . 4.4.1 D0 mixing . . . . . . . . . . . . . . . . . . . . 4.4.2 Direct CP violation . . . . . . . . . . . . . . 4.4.3 Rare charm decays . . . . . . . . . . . . . . . 4.4.4 Charm cross-section and sensitivity estimates 4.4.5 Charm Monte Carlo and background studies 4.5 Additional Physics . . . . . . . . . . . . . . . . . . . 4.6 Competition . . . . . . . . . . . . . . . . . . . . . . . 5 Experiment Description 5.1 Beam . . . . . . . . . . . . . . . 5.2 Targets . . . . . . . . . . . . . . 5.2.1 Cluster-jet target . . . . . 5.2.2 Wire target . . . . . . . . 5.2.3 Frozen-hydrogen target . 5.3 Luminosity Monitor . . . . . . . 5.4 Magnetic Spectrometer . . . . . . 5.4.1 Superconducting solenoid 5.4.2 Silicon vertex detectors . 5.4.3 Scintillating-fiber tracking 5.4.4 TPC tracking . . . . . . . 5.5 Particle Identification . . . . . . 5.6 Calorimeter . . . . . . . . . . . . 5.7 Triggers . . . . . . . . . . . . . . 5.8 Data Acquisition System . . . . . 2 3 4 5 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 8 10 12 13 16 18 19 20 21 25 29 31 . . . . . . . . . . . . . . . 32 32 33 34 34 35 37 37 38 38 39 41 42 44 44 46 6 Budget and Schedule 6.1 Summary of Recuperated Equipment . . . . . . . . . . . . . . . . . . . . . . 6.2 Budget Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 47 48 51 7 Collaboration 51 8 Competition for the Facility 51 iii List of Figures 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Sketch of “upgraded E835” apparatus . . . . . . . . . . . . . . . . . . . . . Mass spectrum for 3-track final states in HyperCP, and dimuon mass spectrum of the three HyperCP Σ+ → pµ+ µ− candidate events. . . . . . . . . . √ Cross sections for various pp processes vs. momentum and s . . . . . . . . Examples of expected X(3872) lineshapes in J/ψπ + π − and D0 D0 π 0 final states in the molecular hypothesis . . . . . . . . . . . . . . . . . . . . . . . Doubly differential NLO Drell–Yan cross sections in pp and pp fixed-target collisions at 8 GeV p or p kinetic energy . . . . . . . . . . . . . . . . . . . . Mass spectrum of Drell–Yan `+ `− pairs in fixed-target pp collisions with 8 GeV p kinetic energy as calculated to NLO . . . . . . . . . . . . . . . . . CTEQ NLO Drell–Yan signal with simulated π + π − -mis-ID and cc-doublesemileptonic backgrounds superimposed . . . . . . . . . . . . . . . . . . . . World average of D0 –D0 mixing parameters . . . . . . . . . . . . . . . . . . Total cross sections vs. antiproton momentum for pp → D0 D∗0 and pp → D+ D∗− from Braaten formula and Regge calculation of Titov and Kämpfer. Histogram of Drell–Yan continuum after D∗ –D0 mass-difference and D0 decay-vertex cuts as described in text, with exponential fit plus Gaussian representing D0 → e+ e− . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Comparison of leading Feynman diagrams for pp → D∗0 D0 and pp → K ∗+ K − Some leading Feynman diagrams for pp → K ∗ Kπ, pp → D∗ Dπ, and pn → D∗ Dπ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total cross sections for pp → cc . . . . . . . . . . . . . . . . . . . . . . . . . Magnetic-field dependence of a) number of events accepted, b) decay-distance resolution, c) D0 mass resolution, and d) D∗ –D0 mass-difference resolution Transverse-momentum histograms for charged pions from accepted taggedD0 events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Histograms of D∗ and D0 mass, and D∗ -D0 mass difference . . . . . . . . . Monte Carlo simulation of D0 decay-vertex distribution as reconstructed in 272-µm-pitch scintillating-fiber detectors, compared with that of random hadron pairs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Histogram of reconstructed K ∓ π ± mass from MIPP analysis; histogram of reconstructed D∗ -D mass difference for MIPP events with K ∓ π ± mass within 2σ of D0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Data on Λ pt spectra from WA89, ISR, STAR, UA1, and CDF. . . . . . . . Integrated luminosity vs time projected for Belle II at SuperKEKB . . . . . Tagged D0 decays from the sequence D∗+ → D0 π + , D0 → K + K − , as reconstructed by LHCb in 195 pb−1 of data . . . . . . . . . . . . . . . . . . . . . Illustration of luminosity leveling achieved in E835. . . . . . . . . . . . . . . Schematic of solid-hydrogen target built at KEK for TRIUMF experiment . Schematic of solid-hydrogen target proposed for our experiment . . . . . . . Schematic of BESS solenoid . . . . . . . . . . . . . . . . . . . . . . . . . . . Sketch of possible Silicon Vertex Detector geometry. . . . . . . . . . . . . . Layout of MICE scintillating-fiber detectors . . . . . . . . . . . . . . . . . . Observed photoelectron yield in MICE tracker cosmic-ray tests. . . . . . . . CAD drawing and photo of MICE tracker support frame . . . . . . . . . . . Schematic and photo of a MICE scintillating-fiber ribbon . . . . . . . . . . iv 7 9 10 12 14 15 15 17 17 21 22 23 24 26 26 27 27 28 30 31 32 34 36 36 38 40 40 41 41 42 31 32 33 Momentum-vs.-time-difference plot for hadrons from simulated D0 decays. . Cross-sectional schematic diagram of TOF Barrel detector. . . . . . . . . . Transverse-momentum histogram of Geant4 minimum-bias events. . . . . . v 43 44 45 List of Tables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Antiproton energies and intensities at existing and future facilities. . . . . . Experimental observations of X(3872). . . . . . . . . . . . . . . . . . . . . . Construction and Installation Budget Summary. . . . . . . . . . . . . . . . Measured and estimated pp → hyperon-antihyperon cross sections just above threshold. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Example sensitivity estimate for D∗ -tagged D0 → Kπ decays. . . . . . . . . Various exclusive pp cross sections to final states containing K ∗0 . . . . . . . Key parameters of simulated detectors. . . . . . . . . . . . . . . . . . . . . . Detector positions used in simulations. . . . . . . . . . . . . . . . . . . . . . Illustrative signal-to-background estimate for D∗ -tagged D0 → Kπ decays. . BESS solenoid parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . Event length estimate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Target Budget Estimate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Luminosity Monitor Budget Estimate. . . . . . . . . . . . . . . . . . . . . . SciFi Budget Estimate. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Time-of-Flight Budget Estimate. . . . . . . . . . . . . . . . . . . . . . . . . Trigger Systems Budget Estimate. . . . . . . . . . . . . . . . . . . . . . . . Data Acquisition Budget Estimate. . . . . . . . . . . . . . . . . . . . . . . . Infrastructure Budget Estimate. . . . . . . . . . . . . . . . . . . . . . . . . . Illustrative Schedule Estimate . . . . . . . . . . . . . . . . . . . . . . . . . . vi 1 4 7 11 22 23 25 25 28 39 47 49 49 49 49 50 50 50 51 1 Introduction We propose to assemble a simple, cost-effective, yet powerful magnetic spectrometer at the AP-50 experimental area of the Fermi National Accelerator Laboratory Antiproton Source, by integrating, and strategically augmenting, existing equipment from previous experiments. This will capitalize on Fermilab’s very substantial investment in the Antiproton Source, by far the world’s best facility for producing antiprotons. The TAPAS apparatus will allow uniquely sensitive investigations of • hyperons, • charmonium and nearby states, • antiproton-induced Drell–Yan lepton pairs, and • charm mesons. We will use it to study and search for rare decays and symmetry-violating effects with world-leading sensitivities, and to make unique measurements of valence quarks in nucleons and nuclei at high x and low Q2 . Some of these measurements will be made for the first time, and others will improve on existing measurements by an order of magnitude or more. This program of measurements, which could be completed by about 2018, may well constitute the only hadron physics carried out at Fermilab for many years prior to Project X turn-on. As such, it will substantially broaden the Lab’s physics program and multiply the number of available thesis topics severalfold. It will thus help in attracting talented U.S. physics students into our field, and provide them with the valuable skills and experience that medium-scale accelerator-based experiments traditionally have done. Table 1 compares the parameters of current and future antiproton sources. The highestenergy and highest-intensity antiproton source is at Fermilab. Even while stacking only 15% of the time (so as to allow fixed-target collisions the rest of the time), it can support a luminosity of 2 × 1032 cm−2 s−1 on an internal target. The CERN Antiproton Decelerator (AD) Table 1: Antiproton energies and intensities at existing and future facilities. p Stacking: Operation: Facility Kinetic Energy Rate Duty Hours p/yr (GeV) (1010 /hr) Factor /yr (1013 ) 0.005 CERN AD – – 3800 0.4 0.047 Fermilab Accumulator: current operation 8 > 25 90% 5550 > 150 proposed here ≈ 3.5–8 20 15% 5550 17 > 2018*) FAIR (∼ 1–14 3.5 15%* 2780* 1.5 ∗ The lower number of operating hours at FAIR compared with that at other facilities arises from the collection ring being shared between the antiproton and radioactive-beam programs. Due to the modular staging of the FAIR facility, stacking of antiprotons will initially be done in the experiment ring, leading to the small duty factor shown here. FAIR’s stacking ring is planned for installation several years after initial operation. 1 provides low-energy antiproton beams at a tiny fraction of the intensity available at Fermilab. Germany’s billion-Euro plan for the Facility for Antiproton and Ion Research (FAIR) at Darmstadt includes construction — only just beginning — of 30 and 90 GeV rapid-cycling synchrotrons, and of low- and medium-energy antiproton and ion storage rings [1]. Antiproton operation at FAIR is not anticipated before 2018. The Fermilab Antiproton Source has previously served medium-energy antiproton fixed-target experiments, including the charmonium experiments E760 and E835. With the completion of the Tevatron program, it is once again potentially available for dedicated antiproton experiments. For the selected topics we propose, it is the world’s most sensitive facility, and will remain so for many years. It is worth noting in Table 1 that an additional order of magnitude in intensity becomes available if the Antiproton Source can stack full-time. This offers the possibility of a luminosity upgrade (to ∼ 1033 cm−2 s−1 ) by adding a small third ring to the complex. The rate capability of the detector would need to be upgraded accordingly. One would then have a sensitivity for small cross sections and for high-statistics measurements with which FAIR would be unable to compete for the foreseeable future. The detector capabilities at L ∼ 1032 cm−2 s−1 could also be upgraded, by (for example) building a new calorimeter, with photodetectors that are insensitive to magnetic fields, and a new spectrometer solenoid to go around it. We believe these potential upgrades are worth considering, but that they should be designed based on measured cross sections for signals and backgrounds, rather than on the models and simulations that are currently available, which are likely to be unreliable due to the paucity of data on which they are based. Thus a relatively quick experiment based largely on available equipment is the logical next step in pursuing antiproton physics at Fermilab, provided such an experiment has enough capability to be compelling on its own. We argue in the following that it does. 2 TAPAS Physics Overview In the flavor problem, nature has handed us a supremely challenging puzzle. Forty years since the Standard Model’s founding, our failure to discern what deeper theory underlies it bespeaks the difficulty of the challenge. The clues being few, we can ill afford to ignore any area where new ones might lie. The progress of technology now enables unprecedented sensitivities to rare effects, giving access to new-physics signatures previously thought too difficult to pursue. Several important issues can be studied in a medium-energy antiproton-beam fixedtarget experiment. These include the possible contributions of new physics to hyperon decay and charm mixing and decay, and the mechanism(s) underlying the mysterious X, Y , and Z states in the charmonium region [2]. Without knowing the nature of the new physics we seek, it is difficult to rank these by impact and importance. But should new physics be discovered in any one of them, it would immediately become the most interesting particlephysics topic of the day. Despite much effort on B and K mixing and CP violation (CPV), evidence for physics beyond the Standard Model in those sectors has proved elusive [3].1 We should therefore look elsewhere as well. The key questions — whether new physics contributes appreciably to hyperon decay, charm decay, or charm mixing — hinge on the degree to which these phenomena violate CP symmetry [5]. 1 The evidence for anomalous CP violation in Bs mixing observed by the DØ collaboration [4], if confirmed, may indicate that new physics does indeed contribute to CP asymmetries at detectable levels. 2 New sources of CP violation are expected by many. The baryon asymmetry of the universe can in principle be understood in terms of CP violation [6], but the CKM contribution to CPV is too small by many orders of magnitude [7], suggesting that additional contributions from new physics were at play in the early universe. Proposed Standard Model extensions (e.g., non-minimal SUSY [8], multi-Higgs models [9], left–right-symmetric models [10, 11], and the SME of Kostelecký et al. [12]) abound in possible new, CP-violating phases, which could account for the baryon asymmetry and could show up in sensitive heavy-quark experiments. These issues have motivated substantial, world-wide efforts seeking to discover physics beyond the Standard Model in neutral-meson mixing and CPV. We are proposing to take the next step in sensitivity in hyperons and charm. At the same time we will have the opportunity to shed light on an intriguing, current mystery — the nature of the X(3872) and its cousins — and perform unique and valuable measurements of parton distribution functions via the Drell–Yan process. As recently emphasized at PANIC11 by G. Perez [13], both the K and B sectors appear consistent with minimal flavor violation, suggesting that the non-SM CPV that baryogenesis requires must reside elsewhere. He therefore stressed the importance of the search for new physics in charm, where SM CPV is strongly suppressed. In an alternate view, it is now fashionable to hope that the “action” will be in the neutrino sector — baryogenesis via leptogenesis — a prospect whose thorough testing will take decades and giga-dollars. The importance of the issue impels us toward a thorough, near-term investigation in charm and hyperons — the more so since the opportunity to do so is so cost-effective, and so synergistic with the current situations of Fermilab and of U.S. HEP. We stress that in a healthy research enterprise, experiments at widely differing funding levels ought not to compete with one another. For example, TAPAS will not materially affect the construction or running schedule of LBNE. The lack of signals to date at LHC for supersymmetry or other new physics serves to remind us that virtual effects accessible in intensity-frontier experiments cover a much wider mass range than does the LHC. Perez [13] and many previous authors [14, 15, 16] have shown that D0 mixing and charm CPV and rare decays (for example) already provide some of the strongest bounds on new-physics mass scales and couplings, with masses in the 103 TeV region or higher if unit coupling strengths are assumed. We must not “put all our eggs in the LHC basket.” To the contrary, we must exploit these intensity-frontier opportunities to learn about virtual effects due to new physics at scales that are inaccessible via direct production. 2.1 Hyperons While CPV is by now well studied in the K 0 and B-meson sectors, CPV in hyperon decay has yet to be established. Its relative suppression in the Standard Model creates an observational window in which new physics could play a dominant role. Such is also the case for rare hyperon decays. Two potentially interesting hyperon signals may already have been glimpsed in the Fermilab HyperCP experiment, albeit with low statistical significance: ( ) evidence for CP violation observed via the AΞΛ asymmetry parameter in Ξ ∓ decay [17], and for flavor-changing neutral currents in Σ+ decay [18]. While a dedicated experiment to follow up each of these < 3σ effects would be hard to justify, the opportunity for substantial increases in hyperon statistics using the same apparatus that can make the other measurements described here is highly appealing. Some R&D will be required to assess the feasibility of improving on the Ξ−/Ξ+ sensitivity achieved by HyperCP. In any case, we will 3 be able to study Σ+ and Ω− decays and search for Ω−/Ω+ CP violation with unprecedented ( ) sensitivities,2 and world-leading Ξ ∓ studies may also be possible. These hyperon measurements offer a window into new physics different from, and complementary to, those of K, B, and D mesons. As detailed below, we aim to achieve sensitivity to both the AΞΛ and ∆Ω CP asymmetries at the ≈ 1 × 10−4 level. This is three times better than the HyperCP AΞΛ result. There is no previous measurement of ∆Ω . It might be argued that such signals will be difficult to interpret theoretically, even if they are (say) an order of magnitude larger than expected in the Standard Model. However, in our experience, such physics is invariably led, and driven, by experiment. A significant observation of an apparent new-physics CPV signal will be a major breakthrough, and will stimulate the ingenuity of theorists in ways that current phenomenological speculation cannot. Furthermore, by pushing sensitivities for more than just a single hyperon CP asymmetry, we will create the opportunity to observe multiple interesting signals, whose pattern could tell us more than would a single observation. 2.2 Charmonium and X(3872) The X(3872) has been observed by several groups (see Table 2) and is a well established state [19]. Despite its proximity in mass to various charmonium levels, it does not appear to be a charmonium state itself [2]. As we will see, pp annihilation has the potential to make uniquely incisive measurements of its properties and thereby reveal its true nature. Table 2: Experimental observations of X(3872). Expt. Belle BABAR CDF DØ Belle Belle Belle BABAR BABAR Year 2003 2004 2004 2004 2004 2005 2006 2008 2008 Mode π + π − J/ψ π + π − J/ψ π + π − J/ψ π + π − J/ψ ω(π + π − π 0 )J/ψ γJ/ψ D0 D0 π0 γψ, γψ 0 D0 D0 π0 Events 35.7 ± 6.8 25.4 ± 8.7 730 ± 90 522 ± 100 10.6 ± 3.6 13.6 ± 4.4 23.4 ± 5.6 23.0 ± 6.4, 25.4 ± 7.3 33 ± 7 Ref. [20] [21] [22] [23] [24] [25] [26] [27] [28] By scanning the Antiproton Accumulator beam energy across the resonance, Fermilab experiments E760 and E835 made the world’s most precise measurements of charmonium masses and widths [29, 30]. This technique can be used to measure the X(3872) mass, width, and lineshape with a precision unobtainable by any other means. At present [19], the upper limit on the X(3872) width is 2.3 MeV. With ∼ 1000 events, we aim to measure its mass and its width with to ∼ 100 keV. The other key advantage of pp annihilation is its ability to directly produce charmonium states of all quantum numbers, in contrast to e+ e− machines which produce primarily 1−− states and the few states that couple directly to them, or (with lower statistics) states accessible in B decay or 2γ production. In addition to studying the X(3872), now that the 2 For convenience, inclusion of charge-conjugate decays is hereinafter implied where not otherwise stated. 4 masses of the ηc0 and hc are reasonably well determined [19], detailed scans dedicated to the spectroscopy of both states are possible, and should be performed. The various X, Y , and Z states appear to pose intriguing and important questions [31]. Some have been interpreted as charmonium states, others as coupled-channel effects, tetraquarks, or meson-antimeson molecules. There may be a new spectroscopy coming into view — an exciting prospect. So far they have been observed only in e+ e− annihilation or in B decays. The opportunity to bring a new methodology — pp formation — to bear on the problem could be extremely valuable. 2.3 Antiproton Drell–Yan Studies Measurements of Drell–Yan lepton-pair production have played a key role in determining sea-quark and gluon distributions in the nucleon. Global structure-function fits to deepinelastic lepton-scattering and Drell–Yan data form the foundation for our understanding of important Tevatron and LHC cross sections. This involves extrapolation from high-x measurements obtained at low Q2 to the low-x and high-Q2 collisions (mainly of gluons) at the LHC that can produce (for example) Higgs bosons or supersymmetric particles. However, data at high x and low Q2 are very sparse. Antiproton collisions at the Antiproton Source offer the opportunity for unprecedented studies of Drell–Yan production at high x and low Q2 . Moreover, the global structure-function fits suffer from poor χ2 values, reflecting tension among the various data sets. In these fits, valence-quark distributions are determined predominantly from deep-inelastic scattering data, while sea distributions depend predominantly on Drell–Yan data. These two distinct categories of experimental data have differing biases, nuclear effects, and systematic uncertainties, which appear not to be fully understood. In contrast, antiproton-beam Drell–Yan data directly measure valence distributions, offering the opportunity to determine both valence and sea distributions for the first time from just one type of measurement: Drell–Yan. 2.4 Charm-Meson Mixing, CP Violation, and Rare Decays As pointed out by many authors (see for example [3, 14, 15, 16, 32, 33]), charm presents an excellent venue in which to search for new physics. Not only is it the only up-type quark for which mixing is observable, but Standard Model backgrounds to new physics are suppressed in charm: the CKM factors are small, and the most massive quark participating in loop diagrams is the b. There are thus many potential signatures in charm (such as CPV in D0 –D0 mixing) that would be direct indications of new physics. Furthermore, compared to beauty, charm has both a large hadroproduction cross section and large branching ratios to decay modes of interest. In the past, the largest charm samples have been obtained in high-energy hadroproduction experiments (e.g., Fermilab E791 and, now, CDF) and at the B factories. We argue that over the next several years, a medium-energy charm experiment at the Fermilab Antiproton Source might be the world’s most sensitive: 1) hadroproduction has an enormous charm-production advantage over e+ e− colliders: charm hadroproduction cross sections are typically ∼ µb, vs. 1 nb for e+ e− ; and 5 2) the low charged-particle multiplicity in medium-energy p annihilation offers a substantial signal-to-background advantage over hadroproduction at high energy. Of course, luminosity favors e+ e− (by factors of 10 to 102 ), and until now, backgrounds have as well. Moreover, high-energy hadroproduction has the advantage of longer decay > 10) [19] in high-energy interdistances. But the higher charged-particle multiplicity (hnch i ∼ actions is responsible for the dominant background to charm in high-energy experiments — combinatorics — whose suppression has required tight vertex cuts. The much lower chargedparticle multiplicity (hnch i ≈ 3) [19] in p collisions near open-charm threshold should lead to charm samples with cleanliness comparable to that at the B factories, with the application of only modest cuts, and hence, high efficiency. As we will see, antiproton annihilation at the Fermilab Accumulator may thus enable the reconstruction of clean charm-meson decay samples a factor of ten or more larger than those of the B factories. The competition to medium-energy p collisions is LHCb (discussed further below), which may have significant systematic biases, due e.g. to trigger effects, production and detection asymmetries, and large rates of b → c decays, and a possible “super-B factory”; when such a facility will be in full-luminosity operation is at present uncertain. The immediate question for us is whether Fermilab should seek to continue to compete in this area. We believe the answer is a clear yes. 3 Experimental Approach We have proposed [34] to assemble an “upgraded E835” apparatus (Fig. 1), including a magnetic spectrometer, with precision vertexing and particle-identification capabilities. Since the E835 apparatus did not include a magnet, various cross sections needed to assess experiment performance and reach remain unmeasured; however, they can be estimated with some degree of confidence. If these cross sections are of the expected magnitudes, it should be possible with this apparatus to make the world’s best measurements of hyperon rare decays and CPV, charm mixing and CPV, as well as of the other states mentioned above. At a minimum, besides precision charmonium measurements, the experiment will measure several cross sections for the first time. Because much of the equipment and infrastructure needed for this experiment are already available, needing only to be integrated rather than built from scratch, we are offered a remarkable and unusual opportunity to do valuable physics quickly and at modest cost. The E760/835 barrel calorimeter, in storage at Fermilab, can easily be reinstalled in the AP-50 pit. A spectrometer solenoid that fits inside the calorimeter is available at KEK. Charged-particle tracking can be performed with scintillating fibers, taking advantage of the very capable scintillating-fiber readout system from the Fermilab DØ experiment [35] which, with the end of the Tevatron program, now becomes available. Precision (δt < 10 ps r.m.s.), cost-effective time-of-flight (TOF) counters under development [36] are likely to be available by the time they are needed for this experiment. High-bandwidth triggering and data-acquisition systems will be needed, and again can exploit hardware available from DØ and CDF. The estimated assembly and installation costs, summarized in Table 3, total less than $10M. We estimate the time from start of funding to initial shakedown at about 2 years. Apparatus details are presented in Sec. 5, and the budget and schedule are detailed in Sec. 6. We assume pp or pN luminosity of 2 × 1032 cm−2 s−1 , one order of magnitude beyond that of E835, which can be accomplished by use of a denser internal target than the E835 6 Return Yoke TPC TOF SciFi SciFi TOF SciFi SciFi TOF TOF Figure 6: E835 apparatus layout (from [67]). Figure 6: E835 apparatus layout (from [67]). Figure 1: Sketch of (left) “upgraded E835” apparatus as simulated: a 1 T solenoid surrounds fine-pitch scintillating-fiber detectors, and is surrounded by precision TOF counters, all within the existing E760/835 Central Calorimeter (return yoke not shown); (right) current apparatus baseline, including small Time Projection Chamber and return yoke needed for proper functioning of calorimeter phototubes. Table 3: Construction and Installation Budget Summary; see Sec. 6.2 for details. Item Cost (k$) Contingency (k$) Targets 430 160 Luminosity monitor 60 20 Scintillating-fiber tracking system 1,820 610 Time-of-Flight system 500* 500 Triggering 1,390 460drawn to the same scale as Fig. 6, 7: The and central tracking system, Figure 7: The DØ solenoid and central tracking system, drawn to theFigure same scale asDØ Fig.solenoid 6, shown as currently installed within the DØ calorimeters shown as currently installed the DØ calorimeters (from [68]). Data within acquisition system 490 153(from [68]). Infrastructure 1,350 550 TOTALS 6,040 15 2,450 15 ∗ Detailed TOF cost estimates based on the University of Chicago “Large-Area Picosecond Photo-Detectors” project are not yet available. This is a preliminary estimate from H. Frisch, for which we assume 100% contingency. hydrogen cluster-jet [37]. This could be a cryogenic, frozen-hydrogen target (already under development, as discussed in Sec. 5.2) or a thin metal wire or pellet; these would be operated in the halo of the antiproton beam.3 3 A denser cluster-jet target may also be a possibility and is under development by the PANDA collaboration [38]. 7 4 Measurement Program Of the suite of measurements we propose, some can be performed simultaneously, while others must be carried out sequentially due to their particular beam-energy requirements. For example, precision measurements of the X(3872) require scanning of the beam energy across the resonance in small steps, while the Ω CP-violation study requires running somewhat above the 3345 MeV threshold for pp → Ω+ Ω− , and Drell–Yan and charm-meson studies are likely to be optimized by running at the highest available Accumulator beam kinetic energy, 8 GeV. We are thus proposing a measurement program which will take approximately 4 years; the schedule is discussed further in Sec. 6. It should also be noted that the X(3872) running requires a hydrogen target, and the charm running may benefit from one but could also use a metal target, while the hyperon running could be done with a hydrogen or a metal target, and Drell–Yan data should be taken with both hydrogen and nuclear targets. We next discuss in greater detail each of the major proposed measurements. 4.1 Hyperon CP Violation and Rare Decays Hyperon CP asymmetries probe parity-conserving currents, hence they can potentially provide information about new physics that is complementary to that from the B and K systems.4 Hyperon CPV measurements have yet to reach Standard Model (CKM) sensitivity levels, but possible new-physics contributions can stand out, and be detected, against the small CKM background. The world’s largest hyperon samples are from the Fermilab HyperCP Experiment [39], including 2.1 × 109 reconstructed Ξ− and 0.5 × 109 Ξ+ decays, and 1010 produced Σ+ . The main HyperCP goal was to substantially advance sensitivity to ( ) ( ) the Ξ ∓ → Λ π ∓ decay-angle CP asymmetry, A ≡ (α + α)/(α − α) [40], where α (α) is the hyperon (antihyperon) longitudinal parity-violation parameter [41]. In this it succeeded, extending sensitivities by some two orders of magnitude over previous results. HyperCP observed unexpected possible signals at the > 2σ level for new physics in the rare hyperon decay Σ+ → pµ+ µ− [18] (Fig. 2) as well as in the CP asymmetry AΞΛ ≈ AΞ + AΛ = [−6.0 ± 2.1 (stat) ± 2.0 (syst)] × 10−4 [17]. It also set the world’s first limit on CPV in Ω− decay: AΩΛ = [−0.4 ± 9.1 (stat) ± 8.5 (syst)] × 10−2 [42]. Since the pp → Ω+ Ω− threshold lies in the same mass region as charmonium, the proposed experiment can further ( ) test these observations using Ω− → Ξ− µ+ µ− decays and potential Ω ∓ CPV, signaled e.g. ( ) ( ) by possible small Ω–Ω decay-rate differences in Λ K ∓ or Ξ 0 π ∓ final states [43]. It may also be possible to run just above the pp → Ξ+ Ξ− threshold and improve on the statistics and systematics of the HyperCP AΞΛ measurement. Note that while the HyperCP evidence is suggestive of the range of possible new-physics effects in hyperon decay, high-sensitivity hyperon studies are well motivated more generally, irrespective of the HyperCP signals. While CPT symmetry requires the lifetimes of particle and antiparticle to be identical, partial-rate asymmetries violate only CP. For most hyperon decays, partial-rate asymmetries are expected to be undetectably small. However, this need not be the case for Ω− decays to ΛK − and Ξ0 π − , for which the particle/antiparticle partial-rate asymmetries could be as large as 2 × 10−5 in the Standard Model and one to two orders of magnitude larger if non-SM contributions are appreciable [43]. These are quantities for which there are no 4 While K is also sensitive to parity-conserving currents, it is consistent with arising entirely from the SM, hence offers little direct information about new physics. 8 a b Figure 2: (Left) Mass spectrum for 3-track final states consistent with being single-vertex pµ+ µ− events in HyperCP positive-beam data sample: (a) wide mass range (semilog scale); (b) narrow range around Σ+ mass; (c) after application of additional cuts as described in Ref. [18]. (Arrows indicate mass of Σ+ .) (Right) Dimuon mass spectrum of the three HyperCP Σ+ → pµ+ µ− candidate events compared with Monte Carlo spectrum assuming (a) SM virtual-photon form factor (solid) or isotropic decay (dashed), or (b) decay via a narrow resonance X 0 . The mass distribution is consistent with a new narrow resonance, with 2.4σ statistical significance. previous measurements. The quantities to be measured are ∆ΛK Γ(Ω− → ΛK − ) − Γ(Ω+ → ΛK + ) Γ(Ω− → Ξ0 π − ) − Γ(Ω+ → Ξ0 π + ) , ∆ ≡ Ξπ Γ(Ω− → ΛK − ) + Γ(Ω+ → ΛK + ) Γ(Ω− → Ξ0 π − ) + Γ(Ω+ → Ξ0 π + ) (1) 1 ≈ (Γ − Γ) (2) 2Γ (3) ≈ 0.5 (1 − N/N ) , ≡ where in the last step we have assumed equal numbers (N ) of Ω and (N ) of Ω events, as would be the case in the experiment proposed here. As a benchmark, sensitivity at the 10−4 level thus requires O(107 ) reconstructed events. Measuring such a small branchingratio difference reliably will require the clean, exclusive event sample produced less than a π 0 mass above threshold, or 4.94 < pp < 5.44 GeV/c. As detailed below, running at + this momentum, we expect to obtain 108 exclusive pp → Ω Ω− events. We have begun simulation studies of the limiting systematics in such measurements, with promising results, indicating that 10−4 sensitivity is likely achievable. Besides partial-rate differences, other possible new-physics signals in Ω decay include decay-angle asymmetries [42], T -odd asymmetries in e.g. Ω− → Ξ− π + π − [44], and confirmation of the HyperCP Σ+ → pµ+ µ− signal in Ω− → Ξ− µ+ µ− , where, due to the greater Q value, the branching ratio is expected to be of O(10−6 ) if the X 0 possibly observed in HyperCP is real [45].5 The experiment we propose will extend sensitivities in all such sig5 Such a particle, if confirmed, could be evidence for nonminimal supersymmetry [46] or other new physics [47, 48]. 9 Figure 3: Cross sections (in mb) for various pp processes vs. momentum and √ s (from [53]). natures. (The recent negative evidence for X 0 → µ+ µ− in K [49] and B decays [50] does not entirely rule out the possibility that the “HyperCP particle” is real — although it does significantly restrict its parameter space — since the B 0 and K 0 decay modes in question probe couplings of the X 0 different from those in hyperon decay [51].) 4.1.1 Hyperon sensitivity estimates There have been numerous measurements of hyperon production by low-energy antiprotons. Johansson et al. [52] report cross sections measured by PS185 at LEAR, but the maximum LEAR p momentum (2 GeV/c) was insufficient to produce Ξ’s or Ω’s. Chien et al. [53] report measurements of a variety of hyperon final states, performed with the BNL 80-inch liquid-hydrogen bubble chamber, in a 6.935 GeV/c electrostatically separated antiproton beam at the AGS; Baltay et al. [54] summarize data taken at lower momenta. In 80,000 pictures Chien et al. observed some 1,868 hyperon or antihyperon events, corresponding to a total hyperon-production cross section of 1.310 ± 0.105 mb [53]. The corresponding cross section measured at 3.7 GeV/c was 720 ± 30 µb, and 438 ± 52 µb at 3.25 GeV/c [54] (see Fig. 3). The inclusive hyperon-production cross section at 5.4 GeV/c is thus about 1 mb. At L = 2 × 1032 cm−2 s−1 this amounts to some 2 × 105 hyperon events produced per second, or 4 × 1012 per year. To estimate the exclusive pp → ΩΩ cross section requires some extrapolation, since it has yet to be measured (moreover, even for pp → Ξ+ Ξ− only a few events have been seen). A rule of thumb is that each strange quark costs between one and two orders of magnitude in cross section, reflecting the effect of the strange-quark mass on the hadronization process. This is borne out by e.g. HyperCP, in which 2.1 × 109 Ξ− → Λπ − and 1.5 × 107 Ω− → ΛK − decays were reconstructed [39]; given the 160 GeV/c hyperon momentum and 6.3 m distance from HyperCP target to decay pipe, this corresponds to ≈ 30 Ξ− ’s per Ω− produced at the target. A similar ratio is observed in HERA-B [55]. In exclusive pp → hyperon-antihyperon production there could be additional effects, since as one proceeds from Λ to Ξ to Ω fewer 10 Table 4: Measured and estimated pp → hyperon-antihyperon cross sections just above threshold. Reaction pp → ΛΛ pp → Ξ− Ξ+ pp → Ω− Ω+ p momentum 1.642 3.0 5.4 Cross section ≈ 65 µb ≈ 2 µb∗ ≈ 60 nb Ref. [52] [54, 56] – *While the cross section at 3.0 GeV/c p momentum has not been measured, that at 3.5 GeV/c has been and is shown here. and fewer valence quarks are in common between the initial and final states. Nevertheless, the cross section for Ξ+ Ξ− somewhat above threshold (pp ≈ 3.5 GeV/c) is ≈ 2 µb [54, 56, 57], or about 1/30 of the corresponding cross section for ΛΛ. Thus the ≈ 65 µb cross section measured for pp → ΛΛ at pp = 1.642 GeV/c at LEAR [52] implies σ(pp → ΩΩ) ≈ 60 nb at 5.4 GeV/c (Table 4). The forgoing extrapolation implies ≈ 2 × 108 ΩΩ events produced per year. For detector acceptance times efficiency of 50% and given the various branching ratios, an estimated 2.1 × 107 decays each in Ω− → Ξ0 π − and Ω+ → Ξ0 π + are observed, and 6.1 × 107 each in Ω− → ΛK − and Ω+ → ΛK + , giving the following statistical sensitivities for partial-rate asymmetries: 0.5 δ∆Ξπ ≈ √ ≈ 1.1 × 10−4 , NΞπ 0.5 ≈ 6.4 × 10−5 . δ∆ΛK ≈ √ NΛK (4) (5) At these sensitivies, if the CP asymmetry in Ξ− → Λπ − is as large as suggested by HyperCP, one might expect to see signals in one or both of these Ω decay modes. Note that in pp → ΩΩ, no valence quarks are in common between the initial and final states, thus the Ω and Ω should have similar kinematics, thereby minimizing systematic uncertainties. (Further in the future — but beyond the scope of this Proposal — an additional, dedicated p storage ring could decelerate antiprotons to the ΛΛ, Σ+ Σ− , and Ξ− Ξ+ thresholds, where an experiment at 1033 luminosity might amass the clean, > 1010 -event samples needed to definitively confirm or refute the HyperCP evidence [17] for CP asymmetry in the Ξ− → Λπ − decay sequence; alternatively, efficient deceleration to the Ξ− Ξ+ threshold region might be possible in the existing Accumulator, but its feasibility is yet to be established.6 The feasibility of such a precise CP-asymmetry measurement has been argued in [57].) In addition, the measured ≈ 1 mb cross section for associated production of inclusive hyperons [53] would mean ∼ 1012 Σ+ events produced per year, which could directly confront the HyperCP evidence (at ≈ 2.4σ significance) for a possible new particle of mass 214.3 MeV/c2 in the three observed Σ+ → pµ+ µ− events (Fig. 2). Whether the the existing Accumulator suffices to amass the needed statistics at Ξ− Ξ+ threshold depends on the efficiency of decelerating in the Accumulator to ≈ 3.0 GeV/c, which in turn depends on how the lattice is manipulated as the beam crosses transition, or so as to avoid transition crossing altogether [59] — a complex set of questions, requiring R&D that can only be performed with dedicated use of the Antiproton Source. 6 11 -2 -1 0 Energy (MeV) 1 2 -2 -1 0 Energy (MeV) 1 2 FIG. 3: Line shapes of X(3872) for γre + iγim = 47.5 MeV. The curves the shapes line shape in FIG. 4:areLine of X(3872) for γre + iγim = (38.4 + 12.0i) MeV. The curves are the line + π − in(solid-blue 0 (dashed J/ψ π + π − (solid line), the shape in D0 D̄0 πof line), andX(3872) theshape D ∗0 D̄in0 energy Figure 4:lineExamples expected lineshapes in J/ψπ J/ψ π +distribution π − (solid line), the line shape D0 D̄ 0 π 0 (dashedcurve) line), andand the D∗0 D̄ 0 energy distribution (dash-dotted line). The two line shapes have been normalized so the resonances below (dash-dotted line). The two line shapes have been normalized so the resonances below the threshold 0 0 0 D D π (dashed-red) final states forthe various parameter choices in the molecular hypothesis threshold have the same peak height. have the same peak height. (from [58]). D0 experimental resolution. IV. 4.2 D0 Precision Measurements in the Charmonium Region D ENERGY DISTRIBUTIONS FOR THE D 0 D̄0 π 0 DECAY CHANNEL ∗0 π0 Usingwethe Fermilab Antiproton experiments E760 and E835 made the world’s most In this section, summarize the essential aspects ofSource, the line shape of the X(3872) in the D 0 D̄ 0precise π 0 channel. We also determineof thecharmonium energy distribution that follows from the [29, 30]. This precision (< 100 D̄∗0 keV) measurements masses and widths identification of D 0 D̄ 0 π 0 events with energy near the D ∗0 D̄ 0 threshold with D ∗0 D̄ 0 and was enabled by the small energy spread of the stochastically cooled antiproton beam and the 0 D 0 D̄ ∗0 events above the threshold. D̄ D̄0 + In the decay B+ → K D 0 D̄ 0 π 0motion , the momentum distributions for D 0 D̄ 0 π 0loss near in the the H cluster-jet absence of +Fermi and negligible energy target. Although 2 X(3872) resonance can be calculated from the sum of the two diagrams in Fig. 5. The open FIG. 5: Diagrams for the production of D0 D̄ 0 π 0 . The open dot represents the B → K transition ∗0 charmonium has by now been studied, dot represents the B + → K + transition which creates a D ∗0extensively D̄ 0 orthat D 0 D̄creates at short distances. 0 a∗0 number of questions remain, most D ∗0 D̄0 or D ∗0 D̄0 at a short-distance scale. The double line represents the propagation The double notably line represents the exact propagator for the resonant superposition of D D̄[2] and improved measurement of h the nature of the mysteriousof the X(3872) state resonant linear combination of the pair of charm mesons. The two diagramsc involve either and D 0 D̄ ∗0 , whose dependence on the total energy E of D 0 D̄ 0 π 0 is given by∗0 the scattering 0 a∗0virtual D (left diagram) a virtual D̄ ∗0 (right diagram). ∗0X may wellor be parameters [31]. The width of the small compared to 1 MeV [58]. andin ηEq. amplitude f (E) (2). In the propagators for the virtual D and D̄ , the width Γ ∗0 c 0 must be taken into account. The coupling of the π to the charm mesons is linear in the The unique precision of the pp energy-scan technique is ideally suited to making the precise E and theinthree pion momentum. The differential distribution in the total energy E and the momenta: momenta 0 hypothesis that the pD , pD̄ , andmass, pπ of thelineshape, D 0 , D̄ 0 , and πand has width the form measurements needed to test the intriguing p2 p2 p2D 0 molecule [60]. As + D̄ hypothesis, + π , (14) = −δ !2 shown in Fig.E4, D ∗the Dπ + molecular X(3872) is!! a D∗0 D in the ! 2MD0 2MD0 2mπ0 1 1 √ ! dΦDD̄π dE . dΓ ∝ |f (E)|2 p2π !! 2 + 2 (13) ! lineshape of X(3872) be −distinctive and dependent on decay mode. For optimal s pD the − 2µE − iµΓ∗0 pwill − 2µE iµΓ∗0 D̄ where δD∗ Dπ is the energy released in the decay of D ∗0 to D 0 π 0 : π0 resolution, these measurements will require the use of a hydrogen target: either an improved version of the E835 gas jet or a windowless, frozen-hydrogen target [61] (see below). The formation cross not been measured, but 9 section of X(3872) in pp annihilation has 10 it has been estimated to be similar in magnitude to that of the χc states [62, 63]. In E760, the χc1 and χc2 were detected in pp → χc → γJ/ψ (branching ratios of 36% and 20%, respectively [19]) with acceptance times efficiency of 44 ± 2%, giving about 500 observed events each for an integrated luminosity of 1 pb−1 taken at each resonance; at the mass peak, 1 event was observed per nb−1 [64]. The lower limit B[X(3872) → π + π − J/ψ] > 0.042 at 90% C.L. [65] implies that in a day at the peak of the X(3872) (8 pb−1 × [1000 events/pb−1 ] × 0.04/0.36 × acceptance-efficiency ratio of final states of ≈ 50%), about 500 events would be observed. Even if the formation cross section is an order of magnitude less than those of the χc states, the tens of events per day of running at the peak will be greater than the background observed by E835.7 By way of comparison, Table 2 shows current sample sizes, which are likely to increase by not much more than an order of magnitude as the respective analyses are completed. (Although CDF and DØ could amass samples of order 104 X(3872) decays, the large backgrounds in the CDF and DØ observations, reflected The differential 3-body phase space dΦDD̄π includes a delta function that relates the energy δD∗ Dπ ≡ MD∗0 − MD0 − mπ0 = 7.14 ± 0.07 MeV. 7 This pp → X(3872) sensitivity will be competitive even with that of the SuperKEKB [66] upgrade. 12 (15) in the uncertainties on the numbers of events listed in Table 2, limit their incisiveness.) We have concentrated here on one decay mode of the X(3872): X(3872) → π + π − J/ψ. Large samples will of course also be obtained in other modes as well, increasing the statistics and allowing knowledge of X(3872) branching ratios to be improved. Given the uncertainties in the cross section and branching ratios, the above may well be an under- or overestimate of the pp formation and observation rates, perhaps by as much as an order of magnitude. Nevertheless, it appears that a new experiment at the Antiproton Accumulator could obtain the world’s largest clean samples of X(3872), in perhaps as little as a month of running. In a few months of running, hundreds to thousands of observed events can be expected in all of the known decay modes, and many more, as-yet-unknown, modes should be seen as well. We will also have the opportunity to study the angular distributions of both the known and unknown modes. The high statistics, event cleanliness, and unique precision available in the pp formation technique could enable the world’s smallest systematics. This experiment could thus provide a definitive test of the nature of the X(3872). Although others of the X, Y , and Z particles are not as narrow as the X(3872), their pp formation and observation in a variety of decay modes could nevertheless shed light on whether a new spectroscopy of meson-antimeson molecules, multiquark states, gluonic hybrids — or something else entirely — is being glimpsed. 4.3 Antiproton Drell–Yan Figure 5 compares the NLO Drell–Yan lepton-pair production cross sections, computed using a next-to-leading order (NLO) CTEQ code [67], for 8 GeV (kinetic) pp and pp collisions. We see that the presence of valence antiquarks in the antiproton amplifies the Drell–Yan cross section by about a factor of 20 at 1.25 GeV/c2 mass in pp as compared to pp collisions. Moreover, the “amplification” factor increases with mass. The ability to measure this cross section depends on the sizes of the dominant backgrounds, which are expected to be lepton pairs from the independent semileptonic decays of charm-anticharm pairs, and pion pairs misidentified as lepton pairs. We have simulated each of these backgrounds based on plausible models and find them to be one to two orders of magnitude below the signal for masses in the range 2.0 < m < 3.0 GeV/c2 (beyond which the J/ψ should dominate the signal). In the corresponding range of lepton momentum, electrons can be distinguished from pions more readily than can muons. We thus propose to use e+ e− pairs to measure the Drell–Yan cross section. The particle identification criteria available include comparison of the calorimetric energy to the magnetically measured momentum (E/p), calorimeter transverse shower shape, time-of-flight (TOF) measurement, and ionization-rate (dE/dx) measurement in the TPC. E/p cuts alone typically provide π/e rejection of 10−3 per track. Combining this rejection with that from shower shape, dE/dx and TOF, we expect to suppress the random-pion-pair background by a factor of 10−10 . Furthermore, the main backgrounds will each be measured with high statistics, allowing a precision background subtraction to be performed. Figure 6 shows the pp Drell–Yan cross section integrated over Feynman-x, and Fig. 7 compares it with the simulated background spectra. We have estimated the π/emisidentification background by generating 8 GeV fixed-target pp events in Geant4 [68], forming all possible π + π − pairs in each event, and scaling the resulting mass spectrum down by 10−10 . We have estimated the charm background by generating D∗ D pairs and allowing them to decay semileptonically, assuming a pp → D∗ D cross section of 3 µb. We see that while charm is the dominant background over most of the useful mass range, the 13 Figure 5: Doubly differential NLO Drell–Yan cross sections in (top) pp and (bottom) pp fixed-target collisions at 8 GeV p or p kinetic energy, as calculated by P. Reimer using CTEQ code [67]. Units of x and z axes are GeV/c2 and nb/GeV, respectively. Drell–Yan signal does appear large enough to be measured. Studies of pp-produced Drell–Yan dileptons at 8 GeV can also address the following outstanding physics issues: 1. The Lam–Tung relation [69], derived in the 1980s as a consequence of the spin-1/2 nature of the quarks, was found to be significantly violated in pion-induced Drell–Yan data [70, 71]. More recent Drell–Yan results from Fermilab E866 [72, 73] in pp and pd, and from CDF pp at large dilepton pt and mass [74], showed that the violation is much less pronounced than in the pion data. The origin of the violation of the Lam–Tung relation is still poorly understood. Drell–Yan data from TAPAS, measured in a very different kinematic regime compared to previous results, could provide unique new information. 14 nb/GeV Figure 6: Mass spectrum of Drell–Yan `+ `− pairs in fixed-target pp collisions with 8 GeV p kinetic energy as calculated to NLO, on linear and semilogarithmic scales. (Based on calculation by Reimer [67].) e+ e− mass (GeV/c2 ) Figure 7: CTEQ NLO Drell–Yan signal (solid histogram) [67] with simulated π + π − -mis-ID (dashed) and cc-double-semileptonic (dotted) backgrounds superimposed, assuming 10−5 background rejection per pion track. Additional rejection of the charm background may be possible using vertex information. 15 2. It has been pointed out [75] that measurement of the azimuthal angular distribution (the “cos 2φ” term) of the Drell–Yan process can be used to extract a novel transverse-momentum–dependent parton distribution function known as the Boer– Mulders function. The Boer–Mulders function corresponds to the correlation between the quark transverse spin and the quark transverse momentum in an unpolarized hadron. Pronounced cos 2φ azimuthal angular dependence has been observed in pioninduced [70, 71], but not in proton-induced [72, 73], Drell–Yan production. This difference is interpreted as a consequence of the valence-like nature of the Boer–Mulders function probed using pion beams, since proton-induced Drell–Yan involves the seaquark Boer–Mulders function. TAPAS could further test this since the valence-valence nature of pp collisions implies that the azimuthal cos 2φ angular dependence should be large. These data could also provide additional input for extracting these novel Boer–Mulders functions. 3. There is considerable interest in accurate measurements of the Q2 dependence of the Weinberg mixing angle using parity-violating reactions [76]. The NuTeV anomaly remains to be understood, and major new experiments have been proposed at the Jefferson Laboratory to measure parity violation in electron deep-inelastic scattering and in Møller scattering. TAPAS Drell–Yan data would provide a unique opportunity to study parity violation at the Q2 scale of a few GeV2 . The interference of the virtual-photon and Z 0 exchange would show up as a forward-backward asymmetry in the Drell–Yan data [77]. A simulation study is currently underway to determine the feasibility of such a measurement and will be reported at the PAC meeting. 4. It has been predicted that the Drell–Yan cross sections could be significantly higher than indicated by the NLO calculation when “threshold resummation” is taken into account [78]. The kinematics of the TAPAS Drell–Yan events would be ideal for a stringent test of this prediction. This information is also crucial for future Drell–Yan experiments being planned at the FAIR antiproton and the J-PARC proton facilities where relatively low energy antiproton and proton beams will be utilized. 4.4 Charm Mixing, CP Violation, and Rare Decays After a > 20-year search, D0 –D0 mixing is now established at > 10 standard deviations [79, 80] (Fig. 8), thanks to the B factories and CDF. The level of mixing (∼ 1%) is consistent with the wide range of Standard Model predictions [5]; however, this does not preclude a significant and potentially detectable contribution from new physics [32, 81]. Since some new-physics models predict differing effects in the charge-2/3 (“up-type”) and –1/3 quark sectors [32, 81], it is important to carry out such studies not only with s and b hadrons, but with charm mesons as well — the only up-type system for which meson mixing can be measured. While the total charm-production cross section for ≈ 8 GeV antiprotons incident on proton or nucleon targets is challenging to compute from first principles, recent phenomenological estimates imply values in the 1–10 µb range [82, 83, 84, 85, 86]. This is sufficiently large that the experiment we propose could amass a sample ten or more times larger than those of the B factories. For example, model-dependent calculations of the exclusive cross √ section σ(pp → D∗0 D0 ) peak at about 1 µb at s ≈ 4.2 GeV [84, 85, 86] (Fig. 9; see further discussion in Sec. 4.4.4). This corresponds to antiprotons of 8 GeV kinetic energy (the Antiproton Source design energy) impinging on a fixed target and, at L = 2 × 1032 cm−2 s−1 , 16 y (%) 2 HFAG-charm FPCP 2010 CPV allowed 1.5 1 0.5 0 1m 2m 3m 4m 5m ï0.5 ï1 ï1 ï0.5 0 0.5 1 1.5 2 x (%) 0 0 mixing parameters x ≡ ∆m/Γ, y ≡ ∆Γ/2Γ: best-fit 0 average of D –D Figure 8: World 10 * values are x = (0.59 ±pp−>DD 0.20)%, y = (0.80 ± 0.13)%, and no mixing (x = y = 0) is disfavored by 10.2σ [79]. 10 −1 ! (µb) 10 10 pp−>DD −2 10 0 10 o * *o 10 * −1 −3 σ (µb) −1 10 D+D*− 10 −4 ! (µb) σ (μb) pp−>DD DD 10 0 10 6 10 −2 −3 −2 o *o DD 8 10 12 14 16 18 pLab + o 10 −3 + *− DD *o (GeV/c)D D 10 *− DD −4 6 8 10 12 14 16 18 pLab (GeV/c) pLab (GeV/c) 6 8 10 12 14 16 18 Figure 9: Total cross sections vs. antiproton momentum for pp → D0 D∗0 (solid) and pp → pLab (GeV/c) D+ D∗− (dashed) from (left) Braaten formula (Eq. 11) [63] and (right) Regge calculation of Titov and Kämpfer [85, 86]. Given their uncertainties, these estimates are in agreement as to the order of magnitude of the cross section. 10 −4 represents some 4 × 109 events produced per year. Since there will also be D∗± D∓ , D∗ D∗ , DD, DDπ,... events, the total charm sample will be even larger, and with the use of a target nucleus heavier than hydrogen, the charm-production A-dependence [87, 88] should enhance statistics by a further factor of a few. The total sample could thus substantially exceed the 109 events produced at the B factories. Indeed, we project below in excess of 1010 tagged-D0 events produced per year of running. By localizing the primary interactions to ∼ 10 µm along the beam (z) direction, a thin wire or frozen-hydrogen target (or perhaps a small metallic pellet suspended on a low-mass stem) can allow the D-meson decay distance to be resolved. The low charged multiplicity 17 at these energies [19] implies small combinatorial background, so that clean samples can be amassed using only modest vertex cuts, and thus, with high efficiency. Medium energy pp or pN annihilation may thus be the optimal way to study charm mixing, and to search for possible new-physics contributions via the clean signature [15, 3, 32] of charm CPV. 4.4.1 D0 mixing Several signatures for charm mixing have been observed and indicate that charm mixing is at the upper end of the range expected in the SM [19]. These involve differing time-dependences of “right-sign” (RS) Cabibbo-favored and “wrong-sign” (WS) D0 decays (arising both from doubly Cabibbo-suppressed decay and from mixing), differing lifetimes of decays to CP-even and mixed-CP final states, and Dalitz-plot analyses of 3-body D0 decays. These processes are sensitive to various combinations of the reduced mixing parameters x ≡ ∆m/Γ, y ≡ ∆Γ/2Γ. As already mentioned, mixing at the observed level could be due to SM physics, but there could also be an appreciable or even dominant contribution from new physics, which could be indicated by CP violation. The first publications of statistically significant signals for D0 –D0 mixing were from BABAR [89] and Belle [90] and employed D0 → K ± π ∓ decays. Neglecting CP violation, for small mixing the ratio of the WS to RS decay rates is given by R(t) = RD + p RD y 0 Γt + x02 + y 02 (Γt)2 . 4 (6) Here, RD is the rate of the doubly-Cabibbo-suppressed (DCS) D0 → K + π − decay, and x0 and y 0 are “rotated” mixing parameters: x0 = x cos δKπ + y sin δKπ , y 0 = −x sin δKπ + y cos δKπ , (7) (8) where δKπ is the str