SLAC-PUB-14763 BABAR-PUB-10/019
Observation and study of the baryonic B-meson decays B→D(∗)p(p(π) π)
P. del Amo Sanchez, J. P. Lees, V. Poireau, E. Prencipe, and V. Tisserand Laboratoire d’Annecy-le-Vieux de Physique des Particules (LAPP), Universit´ede Savoie, CNRS/IN2P3, F-74941 Annecy-Le-Vieux, France
J. Garra Tico and E. Grauges Universitat de Barcelona, Facultat de Fisica, Departament ECM, E-08028 Barcelona, Spain
M. Martinelliab, A. Palanoab, and M. Pappagalloab INFN Sezione di Baria; Dipartimento di Fisica, Universit`adi Barib, I-70126 Bari, Italy
G. Eigen, B. Stugu, and L. Sun University of Bergen, Institute of Physics, N-5007 Bergen, Norway
M. Battaglia, D. N. Brown, B. Hooberman, L. T. Kerth, Yu. G. Kolomensky, G. Lynch, I. L. Osipenkov, and T. Tanabe Lawrence Berkeley National Laboratory and University of California, Berkeley, California 94720, USA
C. M. Hawkes and A. T. Watson University of Birmingham, Birmingham, B15 2TT, United Kingdom
H. Koch and T. Schroeder Ruhr Universit¨at Bochum, Institut f¨ur Experimentalphysik 1, D-44780 Bochum, Germany
D. J. Asgeirsson, C. Hearty, T. S. Mattison, and J. A. McKenna University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z1
A. Khan and A. Randle-Conde Brunel University, Uxbridge, Middlesex UB8 3PH, United Kingdom
V. E. Blinov, A. R. Buzykaev, V. P. Druzhinin, V. B. Golubev, A. P. Onuchin, S. I. Serednyakov, Yu. I. Skovpen, E. P. Solodov, K. Yu. Todyshev, and A. N. Yushkov Budker Institute of Nuclear Physics, Novosibirsk 630090, Russia
M. Bondioli, S. Curry, D. Kirkby, A. J. Lankford, M. Mandelkern, E. C. Martin, and D. P. Stoker University of California at Irvine, Irvine, California 92697, USA
H. Atmacan, J. W. Gary, F. Liu, O. Long, and G. M. Vitug University of California at Riverside, Riverside, California 92521, USA
C. Campagnari, J. M. Flanigan, T. M. Hong, D. Kovalskyi, J. D. Richman, and C. West University of California at Santa Barbara, Santa Barbara, California 93106, USA
A. M. Eisner, C. A. Heusch, J. Kroseberg, W. S. Lockman, A. J. Martinez, T. Schalk, B. A. Schumm, A. Seiden, and L. O. Winstrom University of California at Santa Cruz, Institute for Particle Physics, Santa Cruz, California 95064, USA
C. H. Cheng, D. A. Doll, B. Echenard, D. G. Hitlin, P. Ongmongkolkul, F. C. Porter, and A. Y. Rakitin California Institute of Technology, Pasadena, California 91125, USA
R. Andreassen, M. S. Dubrovin, G. Mancinelli, B. T. Meadows, and M. D. Sokoloff University of Cincinnati, Cincinnati, Ohio 45221, USA
P. C. Bloom, W. T. Ford, A. Gaz, M. Nagel, U. Nauenberg, J. G. Smith, and S. R. Wagner
Work supported in part by US Department of Energy contract DE-AC02-76SF00515. SLAC National Accelerator Laboratory, Menlo Park, CA 94025 2
University of Colorado, Boulder, Colorado 80309, USA
R. Ayad∗ and W. H. Toki Colorado State University, Fort Collins, Colorado 80523, USA
H. Jasper, T. M. Karbach, J. Merkel, A. Petzold, B. Spaan, and K. Wacker Technische Universit¨at Dortmund, Fakult¨at Physik, D-44221 Dortmund, Germany
M. J. Kobel, K. R. Schubert, and R. Schwierz Technische Universit¨at Dresden, Institut f¨ur Kern- und Teilchenphysik, D-01062 Dresden, Germany
D. Bernard and M. Verderi Laboratoire Leprince-Ringuet, CNRS/IN2P3, Ecole Polytechnique, F-91128 Palaiseau, France
P. J. Clark, S. Playfer, and J. E. Watson University of Edinburgh, Edinburgh EH9 3JZ, United Kingdom
M. Andreottiab, D. Bettonia, C. Bozzia, R. Calabreseab, A. Cecchiab, G. Cibinettoab, E. Fioravantiab, P. Franchiniab, E. Luppiab, M. Muneratoab, M. Negriniab, A. Petrellaab, and L. Piemontesea INFN Sezione di Ferraraa; Dipartimento di Fisica, Universit`adi Ferrarab, I-44100 Ferrara, Italy
R. Baldini-Ferroli, A. Calcaterra, R. de Sangro, G. Finocchiaro, M. Nicolaci, S. Pacetti, P. Patteri, I. M. Peruzzi,† M. Piccolo, M. Rama, and A. Zallo INFN Laboratori Nazionali di Frascati, I-00044 Frascati, Italy
R. Contriab, E. Guidoab, M. Lo Vetereab, M. R. Mongeab, S. Passaggioa, C. Patrignaniab, E. Robuttia, and S. Tosiab INFN Sezione di Genovaa; Dipartimento di Fisica, Universit`adi Genovab, I-16146 Genova, Italy
B. Bhuyan and V. Prasad Indian Institute of Technology Guwahati, Guwahati, Assam, 781 039, India
C. L. Lee and M. Morii Harvard University, Cambridge, Massachusetts 02138, USA
A. Adametz, J. Marks, and U. Uwer Universit¨at Heidelberg, Physikalisches Institut, Philosophenweg 12, D-69120 Heidelberg, Germany
F. U. Bernlochner, M. Ebert, H. M. Lacker, T. Lueck, and A. Volk Humboldt-Universit¨at zu Berlin, Institut f¨ur Physik, Newtonstr. 15, D-12489 Berlin, Germany
P. D. Dauncey and M. Tibbetts Imperial College London, London, SW7 2AZ, United Kingdom
P. K. Behera and U. Mallik University of Iowa, Iowa City, Iowa 52242, USA
C. Chen, J. Cochran, H. B. Crawley, L. Dong, W. T. Meyer, S. Prell, E. I. Rosenberg, and A. E. Rubin Iowa State University, Ames, Iowa 50011-3160, USA
A. V. Gritsan and Z. J. Guo Johns Hopkins University, Baltimore, Maryland 21218, USA
N. Arnaud, M. Davier, D. Derkach, J. Firmino da Costa, G. Grosdidier, F. Le Diberder, A. M. Lutz, B. Malaescu, A. Perez, P. Roudeau, M. H. Schune, J. Serrano, V. Sordini,‡ A. Stocchi, L. Wang, and G. Wormser Laboratoire de l’Acc´el´erateur Lin´eaire, IN2P3/CNRS et Universit´eParis-Sud 11, Centre Scientifique d’Orsay, B. P. 34, F-91898 Orsay Cedex, France
D. J. Lange and D. M. Wright 3
Lawrence Livermore National Laboratory, Livermore, California 94550, USA
I. Bingham, C. A. Chavez, J. P. Coleman, J. R. Fry, E. Gabathuler, R. Gamet, D. E. Hutchcroft, D. J. Payne, and C. Touramanis University of Liverpool, Liverpool L69 7ZE, United Kingdom
A. J. Bevan, F. Di Lodovico, R. Sacco, and M. Sigamani Queen Mary, University of London, London, E1 4NS, United Kingdom
G. Cowan, S. Paramesvaran, and A. C. Wren University of London, Royal Holloway and Bedford New College, Egham, Surrey TW20 0EX, United Kingdom
D. N. Brown and C. L. Davis University of Louisville, Louisville, Kentucky 40292, USA
A. G. Denig, M. Fritsch, W. Gradl, and A. Hafner Johannes Gutenberg-Universit¨at Mainz, Institut f¨ur Kernphysik, D-55099 Mainz, Germany
K. E. Alwyn, D. Bailey, R. J. Barlow, G. Jackson, and G. D. Lafferty University of Manchester, Manchester M13 9PL, United Kingdom
J. Anderson, R. Cenci, A. Jawahery, D. A. Roberts, G. Simi, and J. M. Tuggle University of Maryland, College Park, Maryland 20742, USA
C. Dallapiccola and E. Salvati University of Massachusetts, Amherst, Massachusetts 01003, USA
R. Cowan, D. Dujmic, G. Sciolla, and M. Zhao Massachusetts Institute of Technology, Laboratory for Nuclear Science, Cambridge, Massachusetts 02139, USA
D. Lindemann, P. M. Patel, S. H. Robertson, and M. Schram McGill University, Montr´eal, Qu´ebec, Canada H3A 2T8
P. Biassoniab, A. Lazzaroab, V. Lombardoa, F. Palomboab, and S. Strackaab INFN Sezione di Milanoa; Dipartimento di Fisica, Universit`adi Milanob, I-20133 Milano, Italy
L. Cremaldi, R. Godang,§ R. Kroeger, P. Sonnek, and D. J. Summers University of Mississippi, University, Mississippi 38677, USA
X. Nguyen, M. Simard, and P. Taras Universit´ede Montr´eal, Physique des Particules, Montr´eal, Qu´ebec, Canada H3C 3J7
G. De Nardoab, D. Monorchioab, G. Onoratoab, and C. Sciaccaab INFN Sezione di Napolia; Dipartimento di Scienze Fisiche, Universit`adi Napoli Federico IIb, I-80126 Napoli, Italy
G. Raven and H. L. Snoek NIKHEF, National Institute for Nuclear Physics and High Energy Physics, NL-1009 DB Amsterdam, The Netherlands
C. P. Jessop, K. J. Knoepfel, J. M. LoSecco, and W. F. Wang University of Notre Dame, Notre Dame, Indiana 46556, USA
L. A. Corwin, K. Honscheid, R. Kass, and J. P. Morris Ohio State University, Columbus, Ohio 43210, USA
N. L. Blount, J. Brau, R. Frey, O. Igonkina, J. A. Kolb, R. Rahmat, N. B. Sinev, D. Strom, J. Strube, and E. Torrence University of Oregon, Eugene, Oregon 97403, USA 4
G. Castelliab, E. Feltresiab, N. Gagliardiab, M. Margoniab, M. Morandina, M. Posoccoa, M. Rotondoa, F. Simonettoab, and R. Stroiliab INFN Sezione di Padovaa; Dipartimento di Fisica, Universit`adi Padovab, I-35131 Padova, Italy
E. Ben-Haim, G. R. Bonneaud, H. Briand, G. Calderini, J. Chauveau, O. Hamon, Ph. Leruste, G. Marchiori, J. Ocariz, J. Prendki, and S. Sitt Laboratoire de Physique Nucl´eaire et de Hautes Energies, IN2P3/CNRS, Universit´ePierre et Marie Curie-Paris6, Universit´eDenis Diderot-Paris7, F-75252 Paris, France
M. Biasiniab, E. Manoniab, and A. Rossiab INFN Sezione di Perugiaa; Dipartimento di Fisica, Universit`adi Perugiab, I-06100 Perugia, Italy
C. Angeliniab, G. Batignaniab, S. Bettariniab, M. Carpinelliab,¶ G. Casarosaab, A. Cervelliab, F. Fortiab, M. A. Giorgiab, A. Lusianiac, N. Neriab, E. Paoloniab, G. Rizzoab, and J. J. Walsha INFN Sezione di Pisaa; Dipartimento di Fisica, Universit`adi Pisab; Scuola Normale Superiore di Pisac, I-56127 Pisa, Italy
D. Lopes Pegna, C. Lu, J. Olsen, A. J. S. Smith, and A. V. Telnov Princeton University, Princeton, New Jersey 08544, USA
F. Anullia, E. Baracchiniab, G. Cavotoa, R. Facciniab, F. Ferrarottoa, F. Ferroniab, M. Gasperoab, L. Li Gioia, M. A. Mazzonia, G. Pireddaa, and F. Rengaab INFN Sezione di Romaa; Dipartimento di Fisica, Universit`adi Roma La Sapienzab, I-00185 Roma, Italy
T. Hartmann, T. Leddig, H. Schr¨oder, and R. Waldi Universit¨at Rostock, D-18051 Rostock, Germany
T. Adye, B. Franek, E. O. Olaiya, and F. F. Wilson Rutherford Appleton Laboratory, Chilton, Didcot, Oxon, OX11 0QX, United Kingdom
S. Emery, G. Hamel de Monchenault, G. Vasseur, Ch. Y`eche, and M. Zito CEA, Irfu, SPP, Centre de Saclay, F-91191 Gif-sur-Yvette, France
M. T. Allen, D. Aston, D. J. Bard, R. Bartoldus, J. F. Benitez, C. Cartaro, M. R. Convery, J. Dorfan, G. P. Dubois-Felsmann, W. Dunwoodie, R. C. Field, M. Franco Sevilla, B. G. Fulsom, A. M. Gabareen, M. T. Graham, P. Grenier, C. Hast, W. R. Innes, M. H. Kelsey, H. Kim, P. Kim, M. L. Kocian, D. W. G. S. Leith, S. Li, B. Lindquist, S. Luitz, V. Luth, H. L. Lynch, D. B. MacFarlane, H. Marsiske, D. R. Muller, H. Neal, S. Nelson, C. P. O’Grady, I. Ofte, M. Perl, T. Pulliam, B. N. Ratcliff, A. Roodman, A. A. Salnikov, V. Santoro, R. H. Schindler, J. Schwiening, A. Snyder, D. Su, M. K. Sullivan, S. Sun, K. Suzuki, J. M. Thompson, J. Va’vra, A. P. Wagner, M. Weaver, C. A. West, W. J. Wisniewski, M. Wittgen, D. H. Wright, H. W. Wulsin, A. K. Yarritu, C. C. Young, and V. Ziegler SLAC National Accelerator Laboratory, Stanford, California 94309 USA
X. R. Chen, W. Park, M. V. Purohit, R. M. White, and J. R. Wilson University of South Carolina, Columbia, South Carolina 29208, USA
S. J. Sekula Southern Methodist University, Dallas, Texas 75275, USA
M. Bellis, P. R. Burchat, A. J. Edwards, and T. S. Miyashita Stanford University, Stanford, California 94305-4060, USA
S. Ahmed, M. S. Alam, J. A. Ernst, B. Pan, M. A. Saeed, and S. B. Zain State University of New York, Albany, New York 12222, USA
N. Guttman and A. Soffer 5
Tel Aviv University, School of Physics and Astronomy, Tel Aviv, 69978, Israel
P. Lund and S. M. Spanier University of Tennessee, Knoxville, Tennessee 37996, USA
R. Eckmann, J. L. Ritchie, A. M. Ruland, C. J. Schilling, R. F. Schwitters, and B. C. Wray University of Texas at Austin, Austin, Texas 78712, USA
J. M. Izen and X. C. Lou University of Texas at Dallas, Richardson, Texas 75083, USA
F. Bianchiab, D. Gambaab, and M. Pelliccioniab INFN Sezione di Torinoa; Dipartimento di Fisica Sperimentale, Universit`adi Torinob, I-10125 Torino, Italy
M. Bombenab, L. Lanceriab, and L. Vitaleab INFN Sezione di Triestea; Dipartimento di Fisica, Universit`adi Triesteb, I-34127 Trieste, Italy
N. Lopez-March, F. Martinez-Vidal, D. A. Milanes, and A. Oyanguren IFIC, Universitat de Valencia-CSIC, E-46071 Valencia, Spain
J. Albert, Sw. Banerjee, H. H. F. Choi, K. Hamano, G. J. King, R. Kowalewski, M. J. Lewczuk, I. M. Nugent, J. M. Roney, and R. J. Sobie University of Victoria, Victoria, British Columbia, Canada V8W 3P6
T. J. Gershon, P. F. Harrison, T. E. Latham, and E. M. T. Puccio Department of Physics, University of Warwick, Coventry CV4 7AL, United Kingdom
H. R. Band, S. Dasu, K. T. Flood, Y. Pan, R. Prepost, C. O. Vuosalo, and S. L. Wu University of Wisconsin, Madison, Wisconsin 53706, USA
We present results for B-meson decay modes involving a charm meson, protons, and pions using 455×106 BB pairs recorded by the BABAR detector at the SLAC PEP-II asymmetric-energy e+e− collider. The branching fractions are measured for the following ten decays: B0→D0pp, B0→D∗0pp, B0→D+ppπ−, B0→D∗+ppπ−, B−→D0ppπ−, B−→D∗0ppπ−, B0→D0ppπ−π+, B0→D∗0ppπ−π+, B−→D+ppπ−π−, and B−→D∗+ppπ−π−. The four B− and the two five-body B0 modes are ob- served for the first time. The four-body modes are enhanced compared to the three- and the five-body modes. In the three-body modes, the M(pp) and M(D(∗)0p) invariant mass distributions show enhancements near threshold values. In the four-body mode B0→D+ppπ−, the M(pπ−) dis- tribution shows a narrow structure of unknown origin near 1.5 GeV/c2. The distributions for the five-body modes, in contrast to the others, are similar to the expectations from uniform phase-space predictions.
PACS numbers: 13.25.Hw,12.38.Qk,12.39.Mk,14.20.Gk,14.40.Nd
+ I. INTRODUCTION cay modes were the CLEO observations of B Λc pπ + 0 ∗+ →− and B Λc pππ [1] and, later, of B D ppπ and B0 D→∗−pn [2]. These measurements supported→ the pre- B-meson decays to final states with baryons have been diction→ [3] that the final states with Λ baryons are not explored much less systematically than decays to meson- c the only sizable contributions to the baryonic B-meson only final states. The first exclusively reconstructed de- decay rate, and that the charm-meson modes of the form B D(∗)NN ′+anything, where the N (′) represent nucleon states,→ are also significant. Previous measurements show ∗Now at Temple University, Philadelphia, Pennsylvania 19122, a trend that the branching fractions increase with the USA number of final-state particles. The branching fractions †Also with Universit`adi Perugia, Dipartimento di Fisica, Perugia, for the four-body modes B0 D(∗)+ppπ− [2, 4] are ap- Italy proximately four times larger→ than those for the three- ‡Also with Universit`adi Roma La Sapienza, I-00185 Roma, Italy 0 (∗)0 § body modes B D pp [5], which, in turn, is approx- Now at University of South Alabama, Mobile, Alabama 36688, imately five times→ larger than those for the two-body USA 0 + ¶ modes B Λ p [6]. Also with Universit`adi Sassari, Sassari, Italy → c 6
We expand the scope of baryonic B-decay studies with b c 0 0 or ∗0 measurements of the branching fractions and the kine- B d W − u D D matic distributions of the following ten modes [7, 8]: d u 0 0 0 ∗0 p Three-body B D pp and B D pp, u → − → ∗ − Four-body B0 D+ppπ and B0 D +ppπ , (a) u ′′ B−→ D0ppπ− and B−→ D∗0ppπ−, u → → p Five-body B0 D0ppπ−π+ and B0 D∗0ppπ−π+, d ′′ B−→ D+ppπ−π− and B−→ D∗+ppπ−π−. → → u − 0 − Six of the modes—the four B and the two five-body B d π W − modes—are observed for the first time. b c We reconstruct the modes through twenty-six decay B− D0 or D∗0 chains consisting of all-hadronic final states (the list is u u given later with the results in Table I), e. g., u u p B+ D∗−ppπ+π+ d → (b) d D0π− → u p K+π−π+π−. u →
A D0 meson, as in the above example, is produced in FIG. 1: Typical quark-line diagrams representing (a) eight of the B modes and a D+ is produced in the remain- B0→D(∗)0pp and (b) B−→D(∗)0ppπ− modes. The gluon lines ing two. The D0-meson candidates are reconstructed are omitted. through decays to K−π+, K−π+π0, and K−π+π−π+; + − + + ∗0 and the D to K π π . The D -meson candidates − are reconstructed through decays to D0π0 and the D∗+ investigate a narrow structure in the M(pπ ) distribu- 2 as D0π+. tion near 1.5 GeV/c . Section VII states the conclusions. Typical quark-line diagrams for the three- and four- body modes with a D(∗)0 meson are shown in Fig. 1. The BABAR three-body modes involve internal emissions of the W − II. DETECTOR AND DATA SAMPLE boson, whereas the four- and five-body modes involve internal and external emission diagrams. We use a data sample with integrated luminosity of Baryonic B decays have a distinctive phenomenology 414 fb−1 (455 106 BB) recorded at the center-of-mass × whose features contrast with the patterns observed in energy √s=10.58 GeV with the BABAR detector at the meson-only final states. Experimentally, the overall rate PEP-II e+e− collider. The e+ and e− beams circu- enhancement of multi-body decays and the low-mass en- late in the storage rings at energies of 3.1 GeV and hancement in the baryon-antibaryon subsystem are ob- 9 GeV, respectively. The value of √s corresponds served [9–14]. Theoretically, these modes are used to to the Υ(4S) mass, maximizing the cross section for investigate a wide range of topics [15–25]. Among them e+e− bb Υ(4S) BB events. The BB production ac- are the predictions of the relative branching fractions, counts→ for→ approximately→ a quarter of the total hadronic the decay dynamics, and the hypotheses involving exotic cross section; the continuum processes e+e− uu, dd, QCD phenomena, such as tetra-, penta-, or septa-quark ss, and cc constitute the rest. → bound states. In particular, there have been discussions The main components of the BABAR detector [30] of pp peaks near threshold values and penta-quark in- are the tracking system, the Detector of Internally- (∗)+ termediate resonance decays Θc D p with respect to Reflected Cherenkov radiation (DIRC), the electromag- our modes [26–29]. → netic calorimeter, and the instrumented flux return. The paper is organized as follows: Section II describes The two-part charged particle tracking system mea- the data sample and the BABAR detector. Section III sures the momentum. The silicon vertex tracker, with presents the analysis method, introducing the key vari- five layers of double-sided silicon micro-strips, is closest ables MES and ∆E. Section IV shows the fits to the to the interaction point. The tracker is followed by a wire joint MES-∆E distributions. The fit yields and the cor- drift chamber filled with a helium-isobutane (80:20) gas responding branching fractions are given. Section V dis- mixture, which was chosen to minimize multiple scatter- cusses the systematic uncertainties. Section VI presents ing. The superconducting coil creates a 1.5 T solenoidal the kinematic distributions. For the three-body modes, field. the Dalitz plots of M 2(D(∗)0p) vs. M 2(pp) are given as The DIRC measures the opening angle of the well as the invariant mass plots of the variables. For the Cherenkov light cone, θC, produced by a charged particle four- and five-body modes, the two-body subsystem in- traversing one of the 144 radiator bars of fused silica. The variant mass plots are given. In the four-body modes, we light propagates in the bar by total internal reflection and 7
A. Monte Carlo-simulated event samples
+ K Monte Carlo (MC) event samples are produced and used
K+ to evaluate the analysis method. Two types of samples— π− signal and generic—are described below. The particle decays are generated using a combination of Evtgen [31] and Jetset 7.4 [32]. The interactions of the decay products traversing the detector are mod- Geant 4 − eled by [33]. The simulation takes into account π varying detector conditions and beam backgrounds dur- p p ing the data-taking periods. The signal MC sample is generated to characterize events with a B meson that decays to one of the signal modes (the accompanying B decays generically). The p typical size of 3 105 events per decay chain is two orders p of magnitude larger× than the expected signal in data. The generic MC sample is generated to characterize the entire data sample. The size is approximately twice that of the BABAR data sample. FIG. 2: Event display for the candidate decay B0→D0pp, D0→K+π−. The labeled tracks in the tracking system and DIRC rings at the perimeter correspond to the particles in the B. Event selection reconstructed decay chain. The remaining unlabeled tracks and rings are due to the decay of the other B0 meson in the The e+e− events are filtered for a signal B-meson candi- event. The beam axis is perpendicular to the image. date through the pre- and the final selections. The pre-selection requires the presence of proton- antiproton pair and a D0- or a D+-meson candidate is projected onto an array of photomultiplier tubes sur- (written as D without a charge designation) in one of rounding a water-filled box mounted at the back end of the 26 decay chains listed in Sec. I. the tracking system. The DIRC’s ability to distinguish Protons are identified with a likelihood-based algo- pions, kaons, and protons complements the energy loss rithm using the dE/dx and the θC measurements as de- measurements, dE/dx, in the tracking volume. scribed in Sec II. For a 1.0 GeV/c proton in the lab frame (typical of those produced in a signal mode), the selection The calorimeter measures the energies and positions efficiency is 98% and the kaon fake rate is 1%. of electron-photon showers with an array of 6580 finely- The D-meson candidates are selected using the invari- segmented Tl-doped CsI crystals. ant mass [34], M(D), and a kaon identification algorithm The flux return is instrumented with a combination of similar to that used for protons. The M(D) is required resistive plate chambers and limited streamer tubes for to be within seven times its resolution around the PDG the detection of muons and neutral hadrons. value [35] (superseded later during final selection). For a 0.9 GeV/c kaon in the lab frame (typical of those pro- A data event display is given in Fig. 2 for the candidate duced in a signal mode), the selection efficiency is 85% decay B0 D0pp, D0 K+π−. → → and the pion fake rate is 2%. For the D0 K−π+π0 and D∗0 D0π0 sub-decay modes, the π0 →γγ candidates are formed→ from two well- separated photons→ with 115 0 ∗+ − and B D ppπ [2, 5]; for the rest of the modes the C. Definitions of MES and ∆E latter value→ is used. The quantity z is computed for each discriminating variable for each decay chain. For The B meson beam-energy-substituted mass, MES, and the variables with a broad maximum in z, the cut values the difference between its energy and the beam energy, are chosen to be consistent across similar modes. ∆E, are defined with the quantities in the lab frame: In order to select D-meson candidates, M(D) is P P 2 required to be within 3 σ of the PDG value (s +4 B 0) 2 M(D) MES = · (PB) 0 − + − + 0 4 (E )2 − [35]. The resolutions σM(D) for D K π , K π π , s 0 (2) → − + − + + − + + QB Q √s K π π π , and D K π π are approximately 6, ∆E = · 0 . 10, 5, and 5 MeV/c2, respectively.→ For the modes involv- √s − 2 ing D0 K−π+π0 decays, the combinatoric background P events→ due to fake π0 candidates are suppressed using a The four-momentum vectors QB=(EB , B) and Q0=(E0, P 0) represent the B-meson candidate and model [36] that parameterizes the amplitude of the Dalitz + − plot distribution M 2(K−π+) vs. M 2(π+π0). The model the e e system, respectively. The two variables, accounts for the amplitudes and the interferences of de- when expressed in terms of center-of-mass quantities cays of K∗0 K−π+, K∗− K−π0, and ρ+ π+π0. The (denoted by asterisks), take the more familiar form, → → → ∗ 2 ∗ normalized magnitude of the decay amplitude is used to MES= s/4 (PB ) and ∆E=EB √s/2. The M -∆−E distributions for the− events passing the suppress the background events by requiring the quan- p ES tity to be greater than a value ranging from 1% to 5%, final selection are given for the six newly observed modes depending on the mode. in Fig. 3. Each point represents a candidate in an event. For many of the modes, a dense concentration of events is ∗ ∗ 2 In order to select D -meson candidates, the D -D visible near MES=5.28 GeV/c , the PDG B-meson mass mass difference, ∆M=M(D0π) M(D0), is required to − [35], and ∆E=0, as expected for signal events. The uni- be within 3 σ∆M of the PDG value [35]. The resolution form distribution of events over the entire plane away 2 ∗0 0 0 σ∆M is approximately 0.8 MeV/c for both D D π from the signal area is indicative of the general smooth- ∗+ 0 + 0 0 − →+ and D D π . For the mode B D ppπ π , the ness of the background event distribution. → 2 → requirement of ∆M>160 MeV/c excludes the contami- The MES-∆E plots are given in a box region of 0 ∗+ − ∗+ 0 + nation from B D ppπ , D D π decays. 5.22 2 2 ˆ N (M(D) M(D)PDG) (∆M ∆MPDG) e−N δ = − 2 + − 2 , (1) L(N,ˆ Ω)ˆ = P (yi; N,ˆ Ω)ˆ , (3) (σM(D)) (σ∆M ) N! i=1 Y is a function of the yield estimate Nˆ and the set of pa- where the PDG values [35] are labeled as such. The latter rameters Ω.ˆ The yi is the pair of MES and ∆E values term in the sum is included only if a D∗ is present in the for the B-meson candidate in the ith event and P is de- decay chain. If more than one candidate has the same δ scribed below. The quantity L is maximized [40–42] with value, we choose one randomly. respect to its arguments. 9 (a) (b) (c) 400 100 100 200 0 100 0 100 0 200 400 50 0 50 0 50 0 25 25 25 E (MeV) 0 E (MeV) 0 E (MeV) 0 ∆ ∆ ∆ -25 -25 -25 -50 -50 -50 5.24 5.28 5.24 5.28 5.24 5.28 2 2 2 MES (GeV/c ) MES (GeV/c ) MES (GeV/c ) (d) (e) (f) 50 20 10 0 50 0 20 0 10 50 0 50 0 50 0 25 25 25 E (MeV) 0 E (MeV) 0 E (MeV) 0 ∆ ∆ ∆ -25 -25 -25 -50 -50 -50 5.24 5.28 5.24 5.28 5.24 5.28 2 2 2 MES (GeV/c ) MES (GeV/c ) MES (GeV/c ) − 0 − 0 0 − + FIG. 3: Scatter plots of MES-∆E for the six newly observed B-meson decay modes: (a) B →D ppπ , (b) B →D ppπ π , (c) B−→D+ppπ−π−, (d) B−→D∗0ppπ−, (e) B0→D∗0ppπ−π+, and (f) B−→D∗+ppπ−π−. The first row of plots is related to the second by the exchange of the charm meson D↔D∗. The decay chain involving D0→K−π+ or D+→K−π+π+ is shown. ∗0 0 0 ∗+ 0 + 2 For (d, e, f), the decay chain involves D →D π or D →D π . The MES projection, in 1 MeV/c bins, is given above the scatter plot; the ∆E, in 1 MeV bins, on the right. For the projection plots, no selection is made on the complementary variable. The fit function is the sum of two terms polynomial for ∆E. The following function parameters are fixed to the val- P (y ; N,ˆ Ω)=ˆ N P (y ;Ω )+N P (y ;Ω ), (4) i sig sig i sig bgd bgd i bgd ues found by fitting the signal MC distributions: the ∆E which correspond to the signal and the background com- Gaussian width for Psig, the MES Gaussian width for Psig, ponent, respectively. For each component function, Pα is the MES power-law tail parameters for Psig, and the MES the two-dimensional function, Nα the yield, and Ωα the end-point parameter for Pbgd. Two exceptions are given parameters. The arguments of the function components after the detailed fit example. ˆ are related to the quantities in Eq. (4) by N= β Nβ A detailed example of the fit results is given in Fig. 4 ˆ for the decay chain B− D+ppπ−π−, D+ K−π+π+. and Ω= β Ωβ. P → → Each function component Pα is written as the prod- The plots in Figs. 4a and 4b show the MES distributions S uct of functions in MES and ∆E since the variables are for the ∆E signal and the ∆E sideband region, respec- largely uncorrelated. (The signal bias due to the small tively. Likewise, Figs. 4c and 4d show the respective correlation is treated as a systematic uncertainty.) The ∆E distributions for the analogous MES regions. The distributions for signal events peak in each variable, so fit function projections describe the distributions in the Psig is the product of functions composed of a Gaussian sideband regions well (Figs. 4b, 4d), which gives us con- core and a power-law tail [43]. The background event fidence that the background event distribution inside the distribution varies smoothly, so Pbgd is the product of signal box are also modeled well. a threshold function [38] for MES that vanishes at ap- The first exception to the fit procedure described above proximately 5.29 GeV/c2 and a second-order Chebyshev applies to the mode B− D∗0ppπ−. A term is added to → 10 2 Eq. (4) to account for the sizable contamination from the (a) mode B0 D∗+ppπ−. The fit function P is the same → peak 300 form as Psig with its parameters fixed to the values found by fitting the MC sample. The normalization Npeak is based on the branching fraction measured in this paper. 200 The second exception applies to four decay chains whose fits do not converge: B0 D0ppπ−π+, D0 K−π+π0; B0 D0ppπ−π+, D0 →K−π+π−π+; 0 → ∗0 − + →∗0 0 0 0 →− + 0 Events / 2 MeV/c 100 B D ppπ π , D D π , D K π π ; and 0→ ∗0 − + ∗0 → 0 0 0 →− + − + −18.8<∆E<14.7 MeV B D ppπ π , D D π , D K π π π . Two changes→ are made: the→ Gaussian→ parameters are fixed to the values found in the D0 K−π+ measurement, → 2 5.24 5.28 and the MES end-point parameter is floated. The fits 400 (b) converge after the changes. 300 IV. BRANCHING FRACTIONS 200 This section presents the B-meson branching fractions . B Section IV A shows the fits to the MES-∆E distributions. Events / 2 MeV/c 100 Sections IVB and IVC gives the values and their ra- ∆E<−28.9 or >24.8 MeV tios, respectively. Throughout thisB section, we simply state and use the systematic uncertainties of Sec. V. 0 5.24 5.28 M (GeV/c2) ES A. Fits of MES-∆E distributions (c) The MES distributions for the events in the ∆E signal 200 region for three-, four-, and five-body modes are given in Figs. 5–7, respectively. For all B-meson decay modes, the decay chains involving D0 K−π+ or D+ K−π+π+ show a peak. → → The fit function projection in each plot describes the 100 data well, except for the four decay chains corresponding Events / 4 MeV to Figs. 7b, 7c, 7d, and 7f, which had difficulties with fit 2 convergence as noted in the previous section. As we will 5273.6 400 B. Branching-fraction calculation Events / 4 MeV The B-meson branching fraction for each of the twenty- 200 2 MES<5270.3 MeV/c six decay chains, given in Table I, is given by 0 1 1 1 = Nsig Npeak , (5) -50 -25 0 25 50 ∗ B 2 NBB Υ D D ǫ − ∆E (MeV) B B B whose ingredients are as follows: the number of BB pairs, FIG. 4: Fit details for B−→D+ppπ−π−, D+→K−π+π+: 6 0 + − NBB=455 10 , the assumed Υ(4S) BB or B B MES and ∆E distributions in regions as noted on the plots. × → → branching fraction, Υ=1/2; the D-meson branching For (a, c) the top curve is the sum of Psig and Pbgd and the B ∗ fraction, D [35]; the D -meson branching fraction, bottom curve is the latter; for (b, d) the curve is Pbgd. B D∗ [35]; the reconstruction efficiency, ǫ; the signal B 11 2 (a) (b) 200 (c) 150 100 150 100 100 50 Events / 2 MeV/c 50 50 2 40 (d) (e) (f) 40 30 40 30 20 20 20 Events / 2 MeV/c 10 10 0 0 0 5.24 5.28 5.24 5.28 5.24 5.28 2 2 2 MES (GeV/c ) MES (GeV/c ) MES (GeV/c ) 0 0 0 ∗0 FIG. 5: MES fit projections for the three-body modes: (a–c) B →D pp and (d–f) B →D pp, where (a, d) are reconstructed via D0→K−π+, (b, e) D0→K−π+π0, and (c, f) D0→K−π+π−π+; and (d–f) D∗0→D0π0. Events with ∆E within 2.5σ of the mean value of the Gaussian function are shown. The top curve is the sum of Psig and Pbgd and the bottom curve is the latter. yield, N ; and the measured contamination, N , us- the charge of the charm mesons, D(∗)+ D(∗)0, sig peak ↔ ing the M(D)-sideband data sample. The D∗ is in- (iii) The ratio for the four-body mode to that of the cluded only when a D∗ decay is present inB the decay corresponding three-body mode with one fewer pion is chain. The efficiency ǫ is determined using the signal about four. MC sample and decreases with the particle multiplic- (iv) The ratio for the five-body mode to that of the cor- ity. The mode B0 D0pp, D0 K−π+ has the highest responding four-body mode with one fewer pion is about → 0 ∗0 → − + ∗0 0 0 value of ǫ at 19% and B D ppπ π , D D π and one-half. 0 − + − + → → D K π π π has the lowest at 1%. The patterns (iii, iv) imply 3-body< 5-body< 4-body. The→ values, given in Table II, are the combinations B B B [44] of theB above measurements using the statistical and the systematic uncertainties. All values are significant V. SYSTEMATIC UNCERTAINTIES with respect to their uncertainties.B For the previously observed modes, the results are consistent with earlier This section describes the systematic uncertainties for measurements. the B-meson branching fraction measurement. Sec- tion VA lists the sources, and Sec. VB gives the error matrices. C. Branching-fraction ratios Table III gives the ratio of the branching fractions for A. Sources modes related by D D∗, D(∗)0 D(∗)+, and the additionB ↔ ↔ of π. These ratios show four patterns: The sources of systematic uncertainties, which are listed (i) The ratios are roughly unity for the modes related in Table IV, can be organized as follows: by the spin of the charm mesons, D D∗. This result ↔ suggests that the additional degrees of freedom due to (i) Counting of the number of BB pairs, the D∗ polarization vector do not significantly modify • (ii–iv) Assumed branching fractions, the production rate. • (ii) The ratio is roughly unity for the modes related by (v–xi) Reconstruction efficiencies, • 12 2 800 (a) 600 400 Events / 2 MeV/c 200 2 200 (b) (c) (d) 100 150 100 100 50 50 Events / 2 MeV/c 50 2 400 800 (e) (f) (g) 600 300 600 400 200 400 200 Events / 2 MeV/c 100 200 2 (h) (i) (j) 100 200 200 150 50 100 100 Events / 2 MeV/c 50 0 0 0 5.24 5.28 5.24 5.28 5.24 5.28 2 2 2 MES (GeV/c ) MES (GeV/c ) MES (GeV/c ) 0 + − 0 ∗+ − − 0 − FIG. 6: MES fit projections for the four-body modes: (a) B →D ppπ , (b–d) B →D ppπ , (e–g) B →D ppπ , and (h–j) B−→D∗0ppπ−, where (a) is reconstructed via D+→K−π+π+, (b, e, h) D0→K−π+, (c, f, i) D0→K−π+π0, and (d, g, j) D0→K−π+π−π+; and (b–d) D∗+→D0π+ and (h–j) D∗0→D0π0. Events with ∆E within 2.5σ of the Gaussian mean value are shown. For (a–g) the top curve is the sum of Psig and Pbgd and the bottom curve is the latter; for (h–j) the middle curve is the sum of Ppeak and Pbgd. 13 2 (a) (b) 1500 (c) 200 1000 1000 100 500 500 Events / 2 MeV/c 2 400 300 40 (d) (e) (f) 300 30 200 20 200 100 Events / 2 MeV/c 10 100 0 0 2 (g) 300 200 Events / 2 MeV/c 100 2 5.24 5.28 (h) (i) 40 (j) 20 40 30 20 10 20 Events / 2 MeV/c 10 0 0 0 5.24 5.28 5.24 5.28 5.24 5.28 2 2 2 MES (GeV/c ) MES (GeV/c ) MES (GeV/c ) 0 0 − + 0 ∗0 − + − + − − FIG. 7: MES fit projections for the five-body modes: (a–c) B →D ppπ π , (d–f) B →D ppπ π , (g) B →D ppπ π , and (h–j) B−→D∗+ppπ−π−, where (a, d, h) are reconstructed via D0→K−π+, (b, e, i) D0→K−π+π0, (c, f, j) D0→K−π+π−π+, and (g) D+→K−π+π+; and (b–d) D∗+→D0π+ and (h–j) D∗0→D0π0. Events with ∆E within 2.5σ of the Gaussian mean value are shown. The top curve is the sum of Psig and the Pbgd and the bottom curve is the latter. We note that the plots in (b, c, e, f) had difficulties with fit convergence; see text. 14 (v) The charged track reconstruction efficiency is eval- TABLE I: Intermediate values for Table II: B-meson branch- uated using e+e− τ +τ − events, where one tau decays ing fractions for the decay chains. Nsig is the yield, Npeak is → the measured contamination (item xvii in Table IV), and ǫ is leptonically and the other hadronically. The uncertainty the reconstruction efficiency. The uncertainties are statistical. of 0.5% is due to the difference between the detection The rows marked by a dagger † have large systematic uncer- efficiency in the data and the MC samples. tainties; see text. The charges of the pions are implied as well (vi) The reconstruction efficiency of low-energy as the D∗0→D0π0 and D∗+→D0π+ decays, when applicable. charged pion from D∗+ D0π+ decays is sufficiently dif- ficult, in comparison to→ other tracks, that item (v) cannot B modes, D modes Nsig±σsig Npeak ǫ B±σstat −4 account for its uncertainty. Such a pion is often found (%) (10 ) using only the silicon vertex tracker because its momen- B0→D0pp, Kπ 351±20 7.6 19.0 1.02±0.06 tum is relatively low. The momentum dependence of B0→D0pp, Kππ0 431±28 24 7.0 0.95±0.06 pion identification is evaluated using the helicity angle 0 0 B →D pp, Kπππ 448±27 10 9.9 1.21±0.07 θhel distribution—the angle between the pion direction B0→D∗0pp, Kπ 110±12 −1.4 9.4 1.08±0.12 in the D∗+ rest frame and the D∗+ boost direction— B0→D∗0pp, Kππ0 148±15 3.9 3.2 1.17±0.12 because the two quantities are highly correlated. Since 0 ∗0 B →D pp, Kπππ 95±14 5.5 5.2 0.76±0.12 the pions are produced symmetrically in cos θhel, the ob- B0→D+ppπ−, Kππ 1816±53 55 12.6 3.32±0.10 served asymmetry in the distribution is indicative of the B0→D∗+ppπ−, Kπ 392±21 2.3 6.8 4.79±0.26 momentum dependence of the efficiency. The uncertainty B0→D∗+ppπ−, Kππ0 601±28 21 3.1 4.53±0.22 of 3.1% is due to the difference in the momentum depen- B0→D∗+ppπ−, Kπππ 378±22 20 3.7 3.92±0.24 dence in the data and the MC samples. − 0 − B →D ppπ , Kπ 1078±38 13 15.9 3.79±0.14 (vii) The π0 reconstruction efficiency is evaluated using − 0 − 0 B →D ppπ , Kππ 1176±54 41 5.5 3.34±0.16 τ +τ − events as in item (v) with an uncertainty of 3.0%. −→ 0 − ± ± B D ppπ , Kπππ 1296 57 33 7.8 4.38 0.20 (viii) The signal B-candidate reconstruction efficiency B−→D∗0ppπ−, Kπ 328±22 2.1 7.7 3.86±0.26 is evaluated using the MC samples. Since these sam- B−→D∗0ppπ−, Kππ0 482±35 47 2.9 3.99±0.32 B−→D∗0ppπ−, Kπππ 343±31 32 4.0 3.37±0.34 ples use the uniform phase-space decay model while the B0→D0ppπ−π+, Kπ 438±32 7.7 8.2 2.97±0.22 reported baryonic decay dynamics [2, 4, 5, 9–14, this pa- B0→D0ppπ−π+, Kππ0 663±65 160 2.9 2.83±0.36† per] are far from uniform, corrections are made in the B0→D0ppπ−π+, Kπππ 770±68 40 3.8 5.28±0.48† variables where the strongest variation are seen—in bins 2 2 (∗) B0→D∗0ppπ−π+, Kπ 61±12 1.8 2.9 1.87±0.38 of M (pp) vs. M (D p)—using the data and the MC B0→D∗0ppπ−π+, Kππ0 142±32 37 1.3 2.19±0.66† samples. The uncertainties ranging from 0.8% to 9.7% B0→D∗0ppπ−π+, Kπππ 163±30 13 1.3 4.93±0.99† are due to the limited statistics of the samples. B−→D+ppπ−π−, Kππ 475±37 6.6 6.7 1.66±0.13 (ix) The particle identification efficiencies for kaons B−→D∗+ppπ−π−, Kπ 57± 9 −12 2.9 1.98±0.26 and protons are evaluated using the MC samples, which B−→D∗+ppπ−π−, Kππ0 94±14 −0.6 1.3 1.82±0.27 are then corrected using a data sample rich in these B−→D∗+ppπ−π−, Kπππ 66±12 4.8 1.5 1.61±0.32 hadrons. The uncertainties ranging from 1.5% to 2.5% are due to the sample statistics associated with the cor- rection procedure. The sample, however, is dominated by the continuum events whose event topology is different (xii–xv) Fit functions and its parameters, and • from BB events. Items (x, xi) account for the differences. (xvi–xvii) Backgrounds peaking in MES or ∆E. (x, xi) The kaon and proton identification efficiencies • in the BB environment are evaluated using a data sample These contributions are described below. of D∗+ D0π+, D0 K−π+ and Λ pπ− decays, respec- (i) The number of BB pairs used in the analysis is the tively.→ The uncertainties→ of 0.5% and→ 1.0%, respectively, difference of the observed number of hadronic events and are due to the differences in the event topologies. the expected contribution from continuum events. The (xii) A subset of the fit function parameters is fixed latter is estimated using a separate data sample taken when fitting the MES-∆E distributions in the data sam- 40 MeV below the Υ(4S) peak. The uncertainty of 1.1% ple. Such parameter values are obtained by fitting the is mostly due to the difference in the detection efficiencies MC distributions, and they are assigned an uncertainty for hadronic events in the data and the MC samples. from this fit. The effect on the signal yield is eval- (ii) The Υ(4S) branching fraction is assumed to be uated by fitting the data sample with the parameter equal for B0B0 and B+B−. The uncertainty of 3.2% is value shifted by 1σ. The procedure is repeated for the difference of 1/2 and the PDG value [35]. each parameter in the set. The uncertainties of 1.3%, (iii, iv) The D- and D∗-meson branching fractions 2.8%, 5.7%, and 3.4% for the modes with D0 K−π+, − + 0 − + − + − + + + →− + + assume the PDG values [35]. The uncertainties of K π π , K π π π , K π π , and D K π π , → 1.3%, 3.7%, 2.5%, and 2.3% are the PDG uncer- respectively, are the quadrature sum of the fractional tainties for D0 K−π+, K−π+π0, K−π+π−π+, and yield changes. D+ K−π+π+,→ respectively; and 4.7% and 0.7% for (xiii) The choice of the signal fit function is evaluated D∗0→ D0π0 and D∗+ D0π+, respectively. using an alternate function, a fourth-order polynomial. → → 15 TABLE II: Main results of this paper: B-meson branching fractions for the ten modes. Also given are the values of χ2, the degrees of freedom (DOF), and the χ2 probabilities for the averaging of the results from Table I. The measurements are consistent with the previous results. 2 2 N-body B-meson decay mode B ±σstat±σsyst χ /DOF Prob(χ ) B from Refs. 2, 5 B from Ref. 4 (10−4) (%) (10−4) (10−4) Three-body B0 → D0pp 1.02±0.04 ±0.06 4.3/2 12 1.18±0.15±0.16 [5] 1.13±0.06±0.08 ′′ 0 ∗0 0.33 B → D pp 0.97±0.07 ±0.09 4.1/2 13 1.20±0.29 ±0.21 [5] 1.01±0.10±0.09 Four-body B0 → D+ppπ− 3.32±0.10 ±0.29 - - - 3.38±0.14±0.29 ′′ 0 ∗+ − 1.3 B → D ppπ 4.55±0.16 ±0.39 1.2/2 54 6.5 ±1.2 ±1.0 [2] 4.81±0.22±0.44 ′′ B− → D0ppπ− 3.72±0.11 ±0.25 3.4/219 - - ′′ B− → D∗0ppπ− 3.73±0.17 ±0.27 0.5/279 - - Five-body B0 → D0ppπ−π+ 2.99±0.21 ±0.45 0.3/285 - - ′′ B0 → D∗0ppπ−π+ 1.91±0.36 ±0.29 0.5/278 - - ′′ B− → D+ppπ−π− 1.66±0.13 ±0.27-- - - ′′ B− → D∗+ppπ−π− 1.86±0.16 ±0.19 0.2/291 - - TABLE III: Ratios of B-meson branching fractions of the TABLE IV: Systematic uncertainty list for B-meson branch- modes related by D↔D∗, D(∗)0↔D(∗)+, and the addition of ing fractions. The “D modes” represents D0→K−π+, π. The uncertainties are statistical. K−π+π0, and K−π+π−π+; and D+→K−π+π+. Ratio of the modes R ± σR Item Description Uncertainty (%) Related by spin of charm meson i Number of BB pairs 1.1 0 ∗0 0 0 B(B →D pp)/B(B →D pp) 0.95±0.08 ii B(Υ(4S): for Υ(4S)→BB 3.2 0 ∗+ − 0 + − B(B →D ppπ )/B(B →D ppπ ) 1.37±0.06 iii B(D): for D modes 1.8, 4.4, 3.2, 3.6 − ∗0 − − 0 − B(B →D ppπ )/B(B →D ppπ ) 1.00±0.05 iv B(D∗): for D∗→D0π0, D∗+→D0π+ 4.7, 0.7 B(B0→D∗0ppπ−π+)/B(B0→D0ppπ−π+) 0.64±0.13 v Charged particle reconstruction 0.5 B(B−→D∗+ppπ−π−)/B(B−→D+ppπ−π−) 1.12±0.13 vi π+ from D∗+→D0π+ 3.1 Related by charge of charm meson vii π0 reconstruction 3.0 B(B−→D0ppπ−)/B(B0→D+ppπ−) 1.12±0.05 viii Signal mode decay dynamics 0.8–9.7 B(B−→D∗0ppπ−)/B(B0→D∗+ppπ−) 0.82±0.05 ix Kaon and proton id using data 1.5–2.5 B(B0→D0ppπ−π+)/B(B−→D+ppπ−π−) 1.80±0.19 x Kaon id in BB event topology 0.5 B(B0→D∗0ppπ−π+)/B(B−→D∗+ppπ−π−) 1.03±0.21 xi Proton id in BB event topology 1.0 Related by addition of pion to three-body modes xii Fit function params: for D modes 1.3, 2.8, 5.7, 3.4 B(B−→D∗0ppπ−)/B(B0→D∗0pp) 3.84±0.33 xiii Signal fit function 0.6 B(B−→D0ppπ−)/B(B0→D0pp) 3.64±0.18 xiv Backgrnd. fit function: for D modes 0.8, 4.5, 1.3, 2.0 Related by addition of pion to four-body modes xv MES-∆E correlation 0.4–2.2 B(B−→D+ppπ−π−)/B(B0→D+ppπ−) 0.50±0.04 xvi Backgrnd. peaking in MES or ∆E for 0–5.5 (77–85) B(B−→D∗+ppπ−π−)/B(B0→D∗+ppπ−) 0.41±0.04 all modes (marked † in Table I) B(B0→D0ppπ−π+)/B(B0→D0ppπ−) 0.80±0.06 xvii Backgrnd. from baryonic modes 0.5–13.5 B(B0→D∗0ppπ−π+)/B(B0→D∗0ppπ−) 0.51±0.10 and ∆E distributions are produced according to Pbgd, The uncertainty of 0.6% is due to the yield difference and a signal MC sample from the full detector simula- with respect to the original fit function. tion. The uncertainties ranging from 0.1% to 1.8% are (xiv) The choice of the background fit function from the deviation of Nsig to the mean of the signal-yield is evaluated using a more general fit function with distribution. the addition of another such component. The (xvi) Background events whose distributions peak ei- 2 uncertainties—0.8%, 4.5%, 1.3%, and 2.0% for the ther at MES=5.28 GeV/c or ∆E=0 can alter the signal modes with D0 K−π+, K−π+π0, K−π+π−π+, and yield. For the B− D∗0ppπ− measurement, the vari- + − + + → D K π π , respectively—are due to the yield dif- ation of the normalization→ of the fit function for the → ferences with respect to the original fit function. B0 D∗+ppπ− contribution within the experimental un- → (xv) The small correlation between the MES and ∆E certainties has a negligible effect on the signal yield. For distributions introduces a bias in the signal yield. This other B decay modes, no such sources are found. How- effect is quantified by fitting pseudo-experiments. Each ever, the MES distributions for a few cases feature a broad 2 experiment contains a background sample whose MES hump with a width around 20 MeV/c spanning nearly 16 VI. KINEMATIC DISTRIBUTIONS TABLE V: Systematic uncertainties (%) combined for the B modes. For each D mode, two columns are given. The uncorrelated values are given on the left columns and the This section presents the kinematic distributions [45]. correlated on the right columns. The right columns exclude Sections VI A, VI B, and VI C give the plots for three-, items (iii, iv) of Table IV. four-, and five-body modes, respectively. Additional dis- − P cussion is devoted to the M(pπ ) feature in Sec. VI D. PP D→ K−π+ K−π+π0 K−π+π−π+ K−π+π+ PP We briefly describe the background-subtraction and B mode PP unc cor unc cor unc cor unc cor efficiency-correction methods used to obtain the differ- B0→D0pp 2.7 3.5 5.5 6.9 4.7 7.0 -- ential branching fraction plots (Figs. 9–11) as a function B0→D∗0pp 2.2 4.6 8.6 8.6 8.7 7.7 -- of two-body invariant mass variables. The differential B0→D+ppπ− ------5.7 5.7 branching fraction, in bins j of the plotted variable, is B0→D∗+ppπ− 4.2 6.3 7.6 8.6 9.9 8.9 -- the ratio of the number of signal events and the product B0→D0ppπ−π+ 14.5 3.9 81.2 7.0 77.3 7.0 -- of the correction factors as given in Eq. 5. The quantity B0→D∗0ppπ−π+ 13.8 4.9 86.3 8.8 85.4 7.7 -- in the numerator is the sum of the background-subtracted B−→D0ppπ− 4.4 4.2 8.8 7.2 11.6 7.3 -- event weights for events in bin j; the formulae are given B−→D∗0ppπ− 6.6 5.2 8.4 8.9 10.5 7.9 -- below. The efficiency-correction part of the denominator B−→D+ppπ−π− ------15.0 5.9 is found for bin j and is applied to each event weight. B−→D∗+ppπ−π− 5.9 6.8 14.9 9.0 19.0 9.2 -- The S-Plot method is used [46] to find the event weight, ρsig,sigPsig(yi)+ ρsig,bgdPbgd(yi) W (yi)= , (6) NsigPsig(yi)+ NbgdPbgd(yi) half of the signal box. The effect of the presence of such a source is quantified by adding a component Ppeak to where the yi is the pair of MES and ∆E values for the can- th the fit function whose parameters are fixed except for didate in the i event; the fit functions Pα were defined the normalization. Except for four decay chains—those in Eq. (4). In general, the weight W is approximately corresponding to Figs. 7b, 7c, 7e, and 7f—uncertainties 0 for a background event and 1 for a signal event. The ranging from zero to 5.5% are obtained from the changes ρsig,bgd quantifies the correlation between the signal and in yield when the additional component is included. For the background yields, the mentioned exceptions, the uncertainties range from N ′ 77% to 85%. As a consequence of the large uncertain- −1 Pλ(yi)Pλ (yi) (ρ ′ ) = . (7) ties, these four modes do not contribute significantly to λ,λ 2 i=1 NsigPsig(yi)+NbgdPbgd(yi) the final results. X (xvii) Background events from baryonic modes with- ∗ out a D meson are evaluated using the data sample. A. Three-body modes B→D( )pp An example case where the final states are identical is 0 − + 0 0 0 − + 0 B Λc pπ ,Λc pK π and B D pp, D K π π . For→ such a source,→ the M(D) distribution→ does→ not peak For the three-body modes, plots are given for Dalitz vari- ables and two-body invariant masses. at the D mass, so the contamination can be quantified by 2 (∗)0 2 repeating the analysis with the M(D)-sideband region. The Dalitz plots of M (D p) vs. M (pp) for the events in the MES-∆E signal box are given (Fig. 8a, 8b). Npeak is an additive correction factor for Nsig with un- certainties ranging from 0.5% to 13.5% due to the sample The allowed kinematic region is the shaded contour. statistics. The background events present in Figs. 8a and 8b are represented by Figs. 8c and 8d, respectively. The latter plots show the events in the MES-sideband regions with their normalizations determined from the background B. Error matrices yield in the signal box. The two-body invariant mass plots are given in Fig. 9. The error matrix, V , spanning the D modes of a given Differential branching fractions are plotted as a func- B mode is the sum of the statistical and systematic com- tion of M(D(∗)0p) and M(pp) for events in different re- ponents V =Vstat+Vsyst. gions of the complementary variable. The two low-mass 2 (∗)0 The Vstat is diagonal with elements (σstat,α) (Table I). enhancements near threshold values in M(D p) and The Vsyst is the sum of a diagonal part and an off- M(pp) correspond to the dense regions in the Dalitz diagonal part Vsyst=Vunc+Vcor (Table V). The Vunc is plots. The broad enhancement in Fig. 9(d) and 9(h) 2 diagonal with (σunc,α) . The Vcor is the sum of a diag- does not have a substantial contribution from J/ψ de- 2 onal part with (σcor,α) and an off-diagonal part with cays due to its width and current experimental limits on 0 0 ραβσcor,ασcor,β. The correlation coefficient ραβ is be- B D J/ψ,J/ψ pp [47]. tween two D0 modes α and β. The correlations among In→ general, we→ observe a strong similarity between the D0-meson branching fractions are the PDG values [35]; shapes of the corresponding distributions for B0 D0pp all others are assumed to be unity. and B0 D∗0pp. → → 17 (a) (b) (c) (d) FIG. 8: Dalitz plots M 2(pp) vs. M 2(D(∗)0p) for the three-body modes. Plots in the first column (a, c) correspond to B0→D0pp; 0 ∗0 the second column (b, d) B →D pp. Plots in the first row (a, b) are the events in the MES-∆E signal box; the second row (c, d) the events in the MES-sideband region normalized to the amount of background present in the respective plots in the first row. In the first row, near-threshold enhancements are seen compared to the respective sideband plots in the second row. The lines drawn at M 2(pp)=5, M 2(D0p)=9, and M 2(D∗0p)=10.5 GeV2/c4 are visual aides to show that the enhancements are mostly non-overlapping. The events are contained in the shaded contour representing the allowed kinematic region except for one outlier in (d), which failed the fit. The points are made larger for the plots in the second column for better visibility. B. Four-body modes B→D(∗)ppπ number of features. The M(pp) distributions show a threshold enhancement with respect to the expectations For the four-body modes, plots are given for two-body in- from the uniform phase-space decay model (Figs. 10a, 10e, 10i, 10m). The M(D(∗)p) distributions show no in- variant masses in Fig. 10. Differential branching fractions 2 (∗) (∗) dication of a penta-quark resonance at 3.1 GeV/c [48] are plotted as a function of M(pp), M(D p), M(D p), (∗) and M(pπ−). (Figs. 10b, 10f, 10j, 10n). The M(D p) distribution in one of the modes (Fig. 10k) suggests a threshold enhance- The two-body invariant-mass distributions show a 18 2 2 2 4 0.6 2 2 4 2 0 2 4 2 0 2 4 (a) M (pp)>5 GeV /c (b) M (pp)<5 GeV /c (c) M (D p)>9 GeV /c 0.3 (d) M (D p)<9 GeV /c 1 .4 0.4 0.2 ) / 50MeV/c 0.5 -5 .2 0.2 0.1 BF (10 0 0 0 0 3 3.5 4 3 3.5 4 2 2.5 3 3.5 2 2.5 3 3.5 M(D0p) (GeV/c2) M(D0p) (GeV/c2) M(pp) (GeV/c2) M(pp) (GeV/c2) 2 (e) M 2(pp)>5 GeV2/c4 (f) M 2(pp)<5 GeV2/c4 (g)M 2(D∗0p)>10.5 GeV2/c4 (h)M 2(D∗0p)<10.5 GeV2/c4 1 2 1 0.5 0.5 ) / 100MeV/c .5 1 -5 0 0 0 0 BF (10 3 3.5 4 3 3.5 4 2 2.5 3 2 2.5 3 M(D*0p) (GeV/c2) M(D*0p) (GeV/c2) M(pp) (GeV/c2) M(pp) (GeV/c2) FIG. 9: Differential branching fraction plots for the three-body B-meson modes: (a–d) B0→D0pp and (e–h) B0→D∗0pp. The captions give the various phase-space regions. The shaded region represents the uniform phase-space model with its area normalized to the data. The bin width for each row of plots is given on the left-most plot. ment, as was observed in the three-body modes, but the smaller at 10 MeV/c2 and the y-axis is the unweighted- distributions in the other modes show no such features uncorrected number of events. The events from the MES- (Figs. 10c, 10g, 10o). The M(pπ−) distribution in one sideband region is superimposed with its normalization of the modes (Fig. 10d) shows a narrow structure near determined from the background yield in the MES-∆E 1.5 GeV/c2, but it is less prominent in the distributions signal box. of the other modes (Figs. 10h, 10l, 10p). In order to measure the properties of the peak, the 2 The peak near 1.5 GeV/c does not correspond to a fit formalism of Eq. (4) is used. The signal component known state. The peak is discussed in detail in Sec. VI D. Psig is assumed to be a Breit-Wigner line shape. The background component Pbgd is taken from the same- sign M(pπ−) distribution. The distribution for the (∗) C. Five-body modes B→D ppππ B0 D+ppπ− mode is relatively smooth (Fig. 13a), and it describes→ the rise and fall of the opposite-sign distribu- For the five-body modes, plots are given for two-body in- tion well (Fig. 12a), whereas the same-sign distributions variant masses in Fig. 11. Branching fractions are plot- in the other modes show a more rapidly falling behavior ted as a function of M(pp), M(D(∗)p), M(D(∗)p), and around 1.5 GeV/c2 (Figs. 13b–13d). − M(pπ ). We note, however, that the use of the shape for Pbgd In contrast to the distributions for the three- and four- has limitations. Since the formation of the p or p is not body modes, the five-body distributions are generally necessarily symmetric with respect to the π− in these de- more consistent with the expectations from the uniform cays, the same-sign M(pπ−) combination may not pre- phase-space decay model. dict the true shape for the non-resonant component in the A notable absence, again, is the signal of a penta-quark opposite-sign M(pπ−) distribution. As a consequence, resonance at 3.1 GeV/c2 [48] (Figs. 11b, 11f, 11j, 11n). we cannot precisely quantify the systematic uncertainty associated with the lack of knowledge of the true back- ground shape. − 2 D. Narrow M(pπ ) peak at 1.5 GeV/c For the two neutral B modes, the fits of the opposite- sign distributions describe the entire kinematic range well The narrow peak in the M(pπ−) [49] distribution at (Figs. 12a, 12b). We note a small excess of events above 2 2 1.5 GeV/c , which we refer to as X, is discussed in this 1.65 GeV/c with respect to Pbgd, but no peak compo- section. nent is included in the fit at this mass. The fitted The opposite-sign M(pπ−) distributions corresponding X mass is 1494.4 4.1 MeV/c2 and 1500.8 4.4 MeV/c2, to Figs. 10d, 10h, 10l, 10p are shown in more detail in where the uncertainties± are statistical, for B0± D∗+ppπ− Fig. 12. In the detailed plots, the x-axis bin width is and B0 D+ppπ−, respectively. We measure→ the full → 19 2 c 3 3 / (a) (b) (c) (d) 4 4 50 MeV 2 2 / ) 5 − 2 2 1 1 10 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 2.5 M(pp) (GeV/c2) M(D+p) (GeV/c2) M(D+p) (GeV/c2) M(pπ-) (GeV/c2) 2 c 6 / (e) (f) (g) 6 (h) 4 50 MeV 5 4 / 4 ) 5 − 2 2 10 2 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 M(pp) (GeV/c2) M(D*+p) (GeV/c2) M(D*+p) (GeV/c2) M(pπ-) (GeV/c2) 2 c / (i) 3 (j) (k) 4 (l) 4 50 MeV 2 / 2 ) 5 2 − 2 10 1 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 2.5 M(pp) (GeV/c2) M(D0p) (GeV/c2) M(D0p) (GeV/c2) M(pπ-) (GeV/c2) 2 c / (m) 4 (n) 4 (o) (p) 4 5 50 MeV / ) 5 2 2 − 2 10 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 M(pp) (GeV/c2) M(D*0p) (GeV/c2) M(D*0p) (GeV/c2) M(pπ-) (GeV/c2) FIG. 10: Differential branching fraction plots as functions of M(pp), M(D(∗)p), M(D(∗)p), and M(pπ−) for the four-body B- meson modes: (a, b, c, d) B0→D+ppπ−, (e, f, g, h) B0→D∗+ppπ−, (i, j, k, l) B−→D0ppπ−, and (m, n, o, p) B−→D∗0ppπ−, respectively. The shaded region represents the uniform phase-space model normalized to the data. The possible presence of a narrow peak near 1.5 GeV/c2 in plots in (d, h, j, p) are shown in detail in Figs. 12a, 12b, 12c, and 12d, respectively, and discussed in Sec. VI D. The bin width for each row of plots is given on the left. 20 widths to be 51 18MeV and 43 17 MeV, respectively. B− D+ppπ−π−, and B− D∗+ppπ−π−—are observed The widths are significantly± wider± than detector resolu- for→ the first time (Figs. 6e–6g,→ 6h–6j, 7a, 7d, 7g, 7h–7j, tion, which is less than 4 MeV for a simulated X pπ− respectively). decay with a mass of 1.5 GeV/c2 and negligible width.→ The B-meson branching fraction measurements range In contrast to the neutral B modes, the opposite-sign from 0.97 10−4 to 4.55 10−4 with the hierarchy × × distributions for the two charged B modes exhibit a less 3-body< 5-body< 4-body (Table II). These results su- 2 B B B peaking behavior at 1.5 GeV/c . As a result, the param- persede the previous BABAR publication of B0 D0pp, eter for the width in the B− D0ppπ− mode is fixed to D∗0pp, D+ppπ−, and D∗+ppπ− [4]. The branching→ frac- 0 →+ − the value found in the B D ppπ mode; the results tions related by changes in the charge or the spin of the → of this fit are not used in the average. D meson are found to be similar (Table III). The known nucleon resonances N ∗ with the masses The kinematic distributions show a number of notable 1440, 1520, 1535, and 1650MeV/c2 are used in an at- features. For the three-body modes, threshold enhance- tempt to describe the X. The distribution is fit with ments are present in M(pp) and M(D(∗)0p) (Figs. 8, 9). the N ∗ fit function components each parameterized as For the four-body modes, a threshold enhancement is ob- a Breit-Wigner line shape. The normalization for each served in M(pp) and a narrow peak is seen in M(pπ−) component is allowed to vary independently. However, (Fig. 10). For the five-body modes, in contrast to the the fit does not describe the peak because the X is much other modes, the distributions are similar to the ex- narrower than any of the N ∗ resonances (Fig. 14a). pectations from the uniform phase-space decay model The overall significance of the X is difficult to mea- (Fig. 11). sure, due to our lack of knowledge of the true background The M(pπ−) distributions in the neutral B-meson de- shape, as discussed earlier, as well as further statistical cay mode show the most prominent peak near 1.5 GeV/c2. issues. We caution that the X analysis is not blind, the We obtained a mass of 1497.4 3.0 0.9 MeV/c2 and a full parameters are not chosen a priori, and the distribution width of 47 12 4 MeV, where± the± first uncertainties are under the no-X hypothesis may be only approximately statistical and± ± the second are systematic, respectively normal. Furthermore, even under the normal assump- (Figs. 12–14). Determining the significance and inter- tion, the presence of the mass and width nuisance pa- preting the origin of the peak are complicated by the rameters under the alternative hypothesis means that the fact that the background fit function is parameterized distributions of the S statistic is not likely to be pure χ2. by the distribution from the same-sign charge combina- We provide a measure of the statistical significance tions pπ−, a procedure which may not provide the true S= 2(ln L1 ln L0) of the X in the two neutral B background shape. − modes, where L1 is the likelihood value with Psig and Despite the relatively small branching fractions for p 0 + − −4 L0 is without Psig. The value is S=8.6 for B D ppπ these modes of order 10 , with product branching frac- → and S=6.9 for B0 D∗+ppπ−. tions of order 10−5 to 10−6 (including the D and D∗ The systematic→ uncertainties are mainly due to the modes), the large size of the BABAR data sample allowed Pbgd. We fit using an alternate fit function by adding us to observe signals containing hundreds of events in a component derived from the same-sign distribution of many of the modes. We are, therefore, able to probe their a different mode (Fig. 14b). The result is a mass shift of kinematic distributions that reflect the complex dynam- 0.8 MeV/c2 and a full width change of 4 MeV. An addi- ics of the multi-body final states. tional contribution of 0.5 MeV/c2 is added for the mass measurement due to the absolute uncertainty of the mag- netic field and the amount of detector material [50]. VIII. ACKNOWLEDGMENTS In summary, the unknown structure X can be charac- terized by a Breit-Wigner line shape: We are grateful for the extraordinary contributions of our PEP-II colleagues in achieving the excellent luminosity M(X) = 1497.4 3.0 0.9 MeV/c2 (8) ± ± and machine conditions that have made this work possi- Γ(X)= 47 12 4 MeV, ble. The success of this project also relies critically on the ± ± expertise and dedication of the computing organizations where the uncertainties are statistical and systematic, re- that support BABAR. The collaborating institutions wish spectively. to thank SLAC for its support and the kind hospitality extended to them. This work is supported by the US Department of Energy and National Science Foundation, VII. CONCLUSIONS the Natural Sciences and Engineering Research Council (Canada), the Commissariat `al’Energie Atomique and We have presented a study of ten baryonic B-meson Institut National de Physique Nucl´eaire et de Physique decay modes of the form B D(∗)pp(π)(π) using a data des Particules (France), the Bundesministerium f¨ur Bil- sample of 455 106 BB pairs.→ Significant signals are dung und Forschung and Deutsche Forschungsgemein- observed (Table× I). Six of the modes—B− D0ppπ−, schaft (Germany), the Istituto Nazionale di Fisica Nu- B− D∗0ppπ−, B0 D0ppπ−π+, B0 D→∗0ppπ−π+, cleare (Italy), the Foundation for Fundamental Research → → → 21 2 c / (a) 4 (b) (c) 4 (d) 4 4 50 MeV / ) 5 2 2 − 2 2 10 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 M(pp) (GeV/c2) M(D0p) (GeV/c2) M(D0p) (GeV/c2) M(pπ-) (GeV/c2) 2 c 6 / (e) (f) (g) (h) 4 75 MeV 5 4 5 / ) 5 − 2 2 10 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 M(pp) (GeV/c2) M(D*0p) (GeV/c2) M(D*0p) (GeV/c2) M(pπ-) (GeV/c2) 2 c / 4 (i) (j) (k) 4 (l) 4 4 100 MeV / ) 2 5 2 − 2 2 10 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 2.5 M(pp) (GeV/c2) M(D+p) (GeV/c2) M(D+p) (GeV/c2) M(pπ-) (GeV/c2) 2 c / 6 (m) (n) (o) (p) 6 4 4 100 MeV 4 / 4 ) 5 2 − 2 2 10 2 ( BF 0 0 0 0 2.0 2.5 3.0 3.0 3.5 4.0 3.0 3.5 4.0 1.0 1.5 2.0 M(pp) (GeV/c2) M(D*+p) (GeV/c2) M(D*+p) (GeV/c2) M(pπ-) (GeV/c2) FIG. 11: Differential branching fraction plots as functions of M(pp), M(D(∗)p), M(D(∗)p), and M(pπ−) for five-body B- meson modes: (a, b, c, d) B0→D0ppπ−π+, (e, f, g, h) B0→D∗0ppπ−π+, (i, j, k, l) B−→D+ppπ−π−, and (m, n, o, p) B−→D∗+ppπ−π−, respectively. The shaded region represents the uniform phase-space model normalized to the data. Each B-meson candidate for the plots in (d, h) contributes two entries for both pπ− and pπ+ combinations, so they are scaled accordingly. The bin width for each row of plots is given on the left. 22 2 2 50 70 (a) (b) 60 40 60 40 40 20 50 20 30 40 0 1.4 1.6 0 1.4 1.6 Events / 10 MeV/c 30 Events / 10 MeV/c 20 20 10 10 0 0 1 1.5 2 2.5 1 1.5 2 - 2 - 2 2 M(pπ ) (GeV/c ) 2 M(pπ ) (GeV/c ) 150 60 60 140 (c) (d) 100 40 120 50 50 20 100 40 0 1.4 1.6 0 1.4 1.6 80 30 Events / 10 MeV/c 60 Events / 10 MeV/c 20 40 20 10 0 0 1 1.5 2 2.5 1 1.5 2 M(pπ-) (GeV/c2) M(pπ-) (GeV/c2) FIG. 12: Fits of the opposite-sign M(pπ−) distribution for (a) B0→D+ppπ−, (b) B0→D∗+ppπ−, (c) B−→D0ppπ−, and (d) − ∗0 − B →D ppπ for events in the signal box of MES-∆E. The top curve is the sum of Psig and Pbgd while the bottom curve is Pbgd. The Pbgd is from the corresponding plot in Fig. 13. The shaded histograms are scaled MES sidebands. A small inset plot is a close-up of the region around 1.5 GeV/c2; its bin width is the same as in the larger plot. on Matter (The Netherlands), the Research Council of cil (United Kingdom). Individuals have received support Norway, the Ministry of Education and Science of the from the Marie-Curie IEF program (European Union), Russian Federation, Ministerio de Ciencia e Innovaci´on the A. P. Sloan Foundation (USA) and the Binational (Spain), and the Science and Technology Facilities Coun- Science Foundation (USA-Israel). [1] X. Fu et al. (CLEO Collaboration), Phys. Rev. Lett. 79, are beyond our sensitivity due to the λ2 suppression 3125 (1997). with respect to B0→D(∗)0pp. For the quantity λ2, see [2] S. Anderson et al. (CLEO Collaboration), Phys. Rev. L. Wolfenstein, Phys. Rev. Lett. 51, 1945 (1983). Lett. 86, 2732 (2001). [8] We use the convention that charge conjugation of parti- [3] I. Dunietz, Phys. Rev. D 58, 094010 (1998). cles and their decays is implied unless otherwise specified. [4] B. Aubert et al. (BABAR Collaboration), Phys. Rev. D [9] Y. J. Lee et al. (Belle Collaboration), Phys. Rev. Lett. 74, 051101 (2006). 93, 211801 (2004). [5] K. Abe et al. (Belle Collaboration), Phys. Rev. Lett. 89, [10] M. Z. Wang et al. (Belle Collaboration), Phys. Rev. D 151802 (2002). 76, 052004 (2007). [6] B. Aubert et al. (BABAR Collaboration), Phys. Rev. D [11] T. Medvedeva et al. (Belle Collaboration), Phys. Rev. D 78, 112003 (2008). 76, 051102 (2007). [7] We do not include B−→D(∗)−pp in the list because they [12] J. T. Wei et al. (Belle Collaboration), Phys. Lett. B 659, 23 2 2 60 (a) (b) 40 40 Events / 10 MeV/c Events / 10 MeV/c 20 20 0 0 1.0 1.5 2.0 2.5 1.0 1.5 2.0 - 2 - 2 2 M(pπ ) (GeV/c ) 2 M(pπ ) (GeV/c ) 60 (c) (d) 120 40 80 Events / 10 MeV/c Events / 10 MeV/c 20 40 0 0 1.0 1.5 2.0 2.5 1.0 1.5 2.0 M(pπ-) (GeV/c2) M(pπ-) (GeV/c2) FIG. 13: Fits of the same-sign M(pπ−) distribution for (a) B0→D+ppπ−, (b) B0→D∗+ppπ−, (c) B−→D0ppπ−, and (d) − ∗0 − B →D ppπ for events in the signal box of MES-∆E. The curve is the smoothed histogram that is used in the corresponding plot in Fig. 12 as Pbgd. The shaded histograms are scaled MES sidebands. 2 2 (a) (b) 60 60 60 40 60 40 20 20 40 0 1.4 1.6 40 0 1.4 1.6 Events / 10 MeV/c Events / 10 MeV/c 20 20 0 0 1.0 1.5 2.0 2.5 1.0 1.5 2.0 M(pπ-) (GeV/c2) M(pπ-) (GeV/c2) FIG. 14: Alternate fits of the opposite-sign M(pπ−) distribution for B0→D+ppπ− with (a) various N ∗ resonances and (b) an 0 ∗+ − additional Pbgd obtained from the B →D ppπ sample. The shaded histograms are the scaled MES-∆E sidebands. A small inset plot is a close-up of the region around 1.5 GeV/c2; its bin width is the same as the larger plot. 24 80 (2008). strum. Meth. A 506, 250 (2003). [13] J. H. Chen et al. (Belle Collaboration), Phys. Rev. Lett. [34] We follow the notation where M(D) is the invariant mass 100, 251801 (2008). of the reconstructed daughters of the D-meson candidate. [14] B. Aubert et al. (BABAR Collaboration), Phys. Rev. D [35] C. Amsler et al. (Particle Data Group), Phys. Lett. B 72, 051101 (2005) and Ibid. 76, 092004 (2007). 667, 1 (2008). [15] W. S. Hou and A. Soni, Phys. Rev. Lett. 86, 4247 (2001). [36] P. L. Frabetti et al. (E687 Collaboration), Phys. Lett. B [16] C. K. Chua, W. S. Hou, and S. Y. Tsai, Phys. Rev. D 65, 331, 217 (1994). 034003 (2002); Ibid. 66, 054004 (2002); and Phys. Lett. [37] W. D. Hulsbergen, Nucl. Instrum. Meth. A 552, 566 B 544, 139 (2002). (2005). [17] H. Y. Cheng and K. C. Yang, Phys. Rev. D 65, 054028 [38] B. Aubert et al. (BABAR Collaboration), Phys. Rev. D 65, (2002) (Erratum: Ibid. 65, 099901 (2002)); Ibid. 66, 032001 (2002) contains the definitions of the background 014020 (2002); Ibid. 66, 094009 (2002); Ibid. 67, 034008 fit function and the thrust quantities. (2003); Phys. Lett. B 533, 271 (2002); and Ibid. 633, [39] R. J. Barlow, Nucl. Instrum. Meth. A 297, 496 (1990). 533 (2006). [40] F. James and M. Roos, Comput. Phys. Commun. 10, 343 [18] I. I. Bigi, Eur. Phys. J. C 24, 271 (2002). (1975). [19] C. H. Chang and W. S. Hou, Eur. Phys. J. C 23, 691 [41] W. Verkerke and D. Kirkby, In Computing in High En- (2002). ergy and Nuclear Physics 2003 Conference Proceedings, [20] H. Y. Cheng and K. C. Yang, Phys. Rev. D 66, 014020 La Jolla, California. (2002); Ibid. 65, 054028 (2002); and Ibid. 67, 034008 [42] R. Brun and F. Rademakers, Nucl. Instrum. Meth. A (2003). 389, 81 (1997). [21] Z. Luo and J. L. Rosner, Phys. Rev. D 67, 094017 (2003). [43] M. J. Oreglia, PhD thesis, Stanford Univ., SLAC Report [22] H. Y. Cheng, J. Korean Phys. Soc. 45, S245 (2004); Int. 236, Appendix D (1980) and J. E. Gaiser, PhD thesis, J. Mod. Phys. A 21, 4209 (2006); and Nucl. Phys. Proc. Stanford Univ., Ibid. 255, Appendix F (1982). Suppl. 163, 68 (2007). [44] L. Lyons, D. Gibaut, and P. Clifford, Nucl. Instrum. [23] Y. K. Hsiao, Int. J. Mod. Phys. A 24, 3638 (2009). Meth. A 270, 110 (1988). [24] M. Suzuki, J. Phys. G 31, 755 (2005) and Ibid. 34, 283 [45] For Sec. VI, we use a fit strategy where the decay prod- (2007). ucts’ momenta are fit while constraining the B-meson [25] T. M. Hong, PhD thesis, Univ. of California, Santa Bar- candidate’s M(B) to the PDG value [35]. In contrast, we bara, SLAC Report 940, Works cited (2010) contains a do not impose this constraint for the branching-fraction more comprehensive list of the literature. measurements. This strategy ensures that the values of [26] J. L. Rosner, Phys. Rev. Lett. 21, 950 (1968); Phys. Rev. the kinematic variables for all the events lie within the D 68, 014004 (2003); Ibid. 69, 094014 (2004); and private allowed limits. communications (2007). [46] M. Pivk and F. R. Le Diberder, Nucl. Instrum. Meth. A [27] A. Datta and P. J. O’Donnell, Phys. Lett. B 567, 273 555, 356 (2005). (2003). [47] B. Aubert et al. (BABAR Collaboration), Phys. Rev. D [28] B. Kerbikov, A. Stavinsky, and V. Fedotov, Phys. Rev. 71, 091103 (2005). C 69, 055205 (2004). [48] A. Aktas et al. (H1 Collaboration), Phys. Lett. B 588, [29] C. H. Chang and H. R. Pang, Commun. Theor. Phys. 43, 17 (2004). 275 (2005). [49] We refer to both pπ− and pπ+ as opposite sign and to [30] B. Aubert et al. (BABAR Collaboration), Nucl. Instrum. both pπ− and pπ+ as same sign. Meth. A 479, 1 (2002). [50] B. Aubert et al. (BABAR Collaboration), Phys. Rev. D [31] D. J. Lange, Nucl. Instrum. Meth. A 462, 152 (2001). 72, 052006 (2005). [32] T. Sjostrand, Comput. Phys. Commun. 82, 74 (1994). [33] S. Agostinelli et al. (Geant 4 Collaboration), Nucl. In-