Prospects for X(3872) Detection at Panda Jens Sören Lange, Martin Galuska, Thomas Geßler, Wolfgang Kühn, Stephanie Künze, Yutie Liang, David Münchow, Björn Spruck, Matthias Ullrich, and Marcel Werner II. Physikalisches Institut, Justus-Liebig-Universität Gießen for the PANDA Collaboration 12 th International Conference on Meson-Nucleon Physics and the Structure of the Nucleon 05/31-06/04, 2010 College of William and Mary Williamsburg, Virginia This talk: only MC

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 2 The X(3872)  A charmonium(-like) state found in X(3872) → J/   +  −  Observation of decay into J/  → C=+1  interesting properties: breaks isospin in the decays J/( !  +  − ), J/( !  +  −  0 ) → is it charmonium?  Within  m < 1 MeV of the DD* threshold S-wave molecular state?  (to be) investigated at Belle, BaBar, CDF-II, D0 BES-III, Belle-II  However: width is unknown <2.3 MeV (Belle) Belle Belle, Phys. Rev. Lett.91(2003) CDF-II, Phys. Rev. Lett.93(2004) D0, Phys. Rev. Lett.93(2004) BaBar, Phys. Rev. D71(2005)071103

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 3 Charmonium Production in  pp collisions  X(3872) has C=+  In e + e − collisions only J PC =1–− can be produced directly inpp collisions any quantum number  Resonance scan with a cooled anti-proton beam: the width of the X(3872) could be measured ! Panda Experiment at FAIR 2 gluons: 0−+, 0++, gluons: 1−−, 1+− J=0,2 J=1

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 4 GSI today Future facility 2R=1.084 km

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 5 HESR (High Energy Storage Ring) PANDA High intensity mode (stochastic cooling) High resolution mode (e – cooling) High intensity mode (stochastic cooling) High resolution mode (e – cooling) Internal targets Pellets Cluster jet Nuclei: Be, C, Si, Al Internal targets Pellets Cluster jet Nuclei: Be, C, Si, Al ' 2 x 10 9 J/ per year For Anti-Protons

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 6 The Pellet Target Here: WASA Target d=25 m fall speed ¸ 60 m/s

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 7 Example of Charmonium Width Measurements (J/) = 99 § 12 § 6 keV  (´) = 306 § 36 § 16 keV by FERMILAB E760 and E835 beam momentum resolution p/p ‚ 2 £ 10 −4 → s FWHM resolution 0.5 MeV  £ BR (J/) ' 630 nb Events/Luminosity (nb) J/ ´´ Phys. Rev. D 47(1993)772

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 8 Panda Experiment MVD Micro Vertex Detector TPC Time Projection Chamber EMC Electromagnetic Calorimeter

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 9 PandaRoot Framework  Simulation, Digitization Reconstruction, Analysis  43,000 geometry volumes  ¸ 400,000 lines of C++ code  Transport engines: Geant3, Geant4, FLUKA  Event Generators: EvtGen, DPM, PYTHIA, UrQMD  ¸ 20 Linux platforms  Improvements since Panda Physics Book arXiv:  X(3872) simulations for the 1 st time  field maps  track finder and track fitter  final state radiation PHOTOS, Comp. Phys. Comm. 79(1994) For further information

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 10 PandaRoot Implementation of relevant Detectors MVD 120 pixel modules 100x100 µm 2 pixel size ~10 7 readout channels 400 strip modules ~0.5 m 2 active area 7x10 4 readout channels TPC 135 padrows 135,169 pads of 2x2 mm 2 EMC ~17,200 crystals PbWO4 (radiation hard, fast  decay ~6 ns) 28 X 0

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 11 Estimated Rates for X(3872) Formation at Panda  p beam = GeV/c  Baseline assumption:  peak = 50 nb  If X is a loosely bound DD* state Chen, Ma, Phys. Rev. D77(2008) cross section estimate · 443 nb  Branching fractions of decays of X(3872) into DD* : J/  +  − : J/  assumed as 9:1:0  J/ ! e + e − and  +  − BR ' 6% each  Reconstruction efficiency ' 50% (track reconstruction efficiency for low momentum  § ) ! 250 pb  high resolution mode p/p =10 −5 is low luminosity mode: L=2 £ cm −2 s −1  duty factor = 50%  L int = 0.86 pb −1 /day  20 points for resonance scan 2 days/point  Yield ' 215 events of X ! J/  +  − per day at peak

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 12 X(3872)! J/   +  – Event, PandaRoot Simulation Beampipe Targetpipe e§e§ e§e§ §§ §§

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 13 X(3872) Events in Panda MVD+TPC PandaRoot Simulation Beampipe Interaction XYZ coordinates / cm

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 14 Comparison of e + e − and  +  − final states Combinatorial Background only 2-particle invariant mass Radiative Tail due to Final state radiation (PHOTOS)

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 15 Particle Indentification (PID) Using E EMC /p track (requires track-cluster matching) E EMC is deposited cluster energy P track is reconstructed track momentum (TPC + MVD)  ´ → J/   +  −, no J/  decay

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 16 Example MC Results for X(3872)! J/   +  − 2-particle invariant mass4-particle invariant mass X(3872) J/ without PID with PID

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 17 Variation of Angular Distributions  Panda is fixed target experiment  Negligible effect on reconstruction efficiency (<1%)  ! Lorentz boost dominates acceptance phasespace vector → vector + pseudoscalar + pseudoscalar

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 18 Kinematic Fit: Vertex Constraint and Mass Constraint Example: reference mode ´ ! J/  +  – No fit Fit J/ ‘‘ ‘‘ Constraint  m ' 3 MeV

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 19 Background: p  p ! 2  + 2  − COMPILATION OF CROSS-SECTIONS 3. p AND anti-p INDUCED REACTIONS By High-Energy Reactions Analysis Group (V. Flaminio et al.). CERN-HERA-79-03, Oct pp. 50 b  total (pp ) ' 70 mb

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 20 Shape of Background Dual Parton Model (DPM) Kinematic situation favourable → background decreases just below J/ mass Capella, Sukhatme, Tan, Tran Thanh Van Phys. Rept. 236(1994)225 Kaidalov, Volkovitsky, Z. Phys. C63(1994)517 Uzhinsky, Galoyan hep-ph/ Shape varying across J/ signal region Fit by 1 st order Chebyshev polynom 2 days data taking

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 21 Scan of a Resonance R Convolution of resonance width and beam resolution

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 22 MC, X(3872) Scan, each 2 days,  p/p = mode MeV MeV MeV MeV MeV MeV MeV MeV MeV Tagged J/ yield, 2-particle invariant mass / GeV Mass(e + e − ) / GeV

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 23 X(3872) Excitation Function, MC  Exclusive final state pp → X(3872)  each data point 2 days data taking  full background  Preliminary result: fitted width is ~20% larger than input, but (s) still · 100 keV (this would be a syst. error on the width of X(3872))  Unfolding (integral equation) is ongoing work (Master Thesis, Univ. Gießen)  Preliminary estimate for Panda  ¸ 168 keV for p/p=10 −4 s / GeV X(3872) yield Reconstruction efficiency ' 72% Preliminary stat. errors only ' 79 keV

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 24 Summary  For the 1 st time, MC simulations for X(3872) production at Panda have been performed  Detailed simulation and reconstruction:  final state radiation  vector → vector angular distribution  track finder and track fitter  PID by track and cluster matching  background with dual parton model  Preliminary results are encouraging  Extraction of width is ongoing (integral equation)  For more information about FAIR see Talk by Ulrich Wiedner, Plenary, Friday

Backup

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 26 Details of MC Scan  Signal X(3872) Geant3.21 PHOTOS on, angular distribution V ! V PS PS  Background DPM with fast simulation (detector resolution parametrized)  input beam resolution not varying over scan range  Track finder and track fitter using TPC and MVD N Hits and  2 Cut  Background assumed not varying with  s  statistical errors only in fit  yield outside is assumed zero (but hadronic cross section is high)

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 27 Resonance Scan Note: the cross section depends on the spin J of the resonance ! a way to determine the quantum number of X(3872)  R is independant from BR x BR because it only enters into  peak

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 28 The Pellet Target Here: WASA Target d=25 m fall speed ¸ 60 m/s

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 29 MVD (Micro Vertex Detector)  4 barrels & 8 disks  inner layers pixels  outer layers strips  forward mixed  Silicon Pixel Detectors  120 modules  100x100 µm 2 pixel size  ~10 7 readout channels  maximum rate <10 MHits s – 1 per module  Radiation length ~1% X 0 per layer  Silicon Strip Detectors  400 modules  ~0.5 m 2 active area  7x10 4 readout channels p/GeV/c dE/dx (a.u.)  K p e MVD dE/dx

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 30 STT (Straw Tube Tracker)  ~4100 tubes, axial or skewed (15 double layers)  R= cm  L=1.5m  (only) m=50 kg  tube diameter 10mm wall Mylar, 30μm  anode wire W/Re, 20μm  spatial resolution σ rφ ≈150μm (axial layers) σ z ≈3-10mm (skewed layers)  gas 90%Ar,10%CO 2 over-pressure for stabilization p=2 bar  radiation length ~1-1.3% X 0  wire U=2 kV  p T /p T ' 1.2%  prototype under investigation

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 31 TPC (Time Projection Chamber)  R= cm  135 padrows  135,169 pads of 2x2 mm 2  Multi-GEM for amplification and ion backflow suppression  Gas: Ne/CO 2 (+CH 4 /CF 4 )  µs drift time ! 700 events pile-up  gating grid continously open  p/p ' 1%  prototype under investigation

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 32 DIRC (Detection of Internally Reflected Cerenkov Light) 16 Quartz (n=1.47) Bars d=1.7 cm, R=48 cm BaBar design electron 0.7 GeV/c

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 33 EMC (Electromagnetic Calorimeter) ~17,200 crystals PbWO 4 (radiation hard, fast  Decay ~6 ns) 28 X 0 dE/dx=13.0 MeV/cm

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 34 EMC (Electromagnetic Calorimeter) operated at T=–25 o C E/p

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 35 MUO (Muon Detector)  R = 1.3–1.8 m  tubes (for x coordinate) gold plated tungsten wire, d=0.05 mm, L=4 m, wire U=+3.6 kV, cathode U=-1.2 kV  copper strips (for z coordinate) U=+1.8 kV  ! pad size 1x1 cm 2 view from GeoManager x y z d=1 cm

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 36 Resonance Scan Note: the cross section depends on the spin J of the resonance ! a way to determine the quantum number of X(3872)  R is independant from BR x BR because it only enters into  peak

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 37 Chen and Ma Estimate Prediction for cross section with X as loosely bound DD* state Chen, Ma, Phys. Rev. D77(2008) width  X 136 keV MeV cross section nb

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda vs. 1-

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 39 Charmonium Production Cross Section in ppbar Barnes, Li, Phys.Rev.D75:054018,2007. hep-ph/ nb With one additional meson

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 40 Dual Parton Model A. Capella, U. Sukhatme, C.-I. Tan, and J. Tran Thanh Van, Phys. Rept. 236, 225 (1994). A. B. Kaidalov and P. E. Volkovitsky, Z. Phys. C63, 517 (1994). [12] V. V. Uzhinsky and A. S. Galoyan, hep-ph/ (2002).

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 41  (2S)! J/   +  − as reference channel  § from X(3872) have higher momentum Reconstruction efficiency is ' 21% higher

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 42 Variation of Angular Distributions  With PID Cut  PHOTOS ON (in X decay and J/Psi decay), e + e − only (no Muons)  Negigible effect on reconstruction efficiency (< 1%) ! Lorentz boost is dominating angular distribution Phasespace V ! V PS PS

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 43 Comparison of e + e − and  +  − final states Radiative Tail in J/ mass generates tail at higher masses in 4-particle invariant mass ' 5 days data taking 4-particle invariant mass

MENU10, Williamsburg, June 2, 2010Sören Lange | X(3872) at Panda 44 Resonance Scan Double scan technique Constant Orbit Constant Field (beam energy is changed by stochastic cooling pick-ups)