Transcript Bez nadpisu
Time-of-flight hadron identification from C+C at 2 A GeV measured by the HADES spectrometer Vladimír Pospíšil 1) within diploma thesis Experimental study of relativistic nuclear collisions with di-lepton spectrometer HADES supervisor of diploma work : Pavel Tlustý 2) DST production : Kaliopi Kanaki 3) G3 Simulation : Jehad Mousa 4) 1) 5. year student of Czech Technical Univerzity, Faculty of Nuclear Sciences and Physical Engineering, Prague 2) Czech Academy of Sciences, Nuclear Physics Institute, Řež 3) Institut für Kern– und Hadronenphysik, Forsungszentrum Rossendorf, Germany 4) University of Cyprus, Department of Physics, Nicosia, Cyprus The HADES spectrometer installed at GSI Darmstadt is devoted to study production of di– electron pairs from proton– and pion–induced reactions and nucleus-nucleus collisions. Extraction of rare lepton pairs in high hadron multiplicity events requires efficient particle identification (PID). Hadrons are identified mostly by momentum and time–of–fligh measurement. For all charged particles momentum is measured by a tracking system placed before and after the toroidal magnet, and the TOF detector provides time of flight. The C+C reaction at 2 A GeV has been recently studied, with the main aim to reconstruct the di– electron signal from the decay of hadrons. Data obtained from this experiment are analyzed at present. The particle identification method is based on testing of hypothesis, that the reconstructed track can be identified as a certain particle specie. In case of hadrons, the particle momentum and velocity are used for the PID decision. For the PID probability calculation the detector response has to be known. This is achieved by parametrization of the velocity distribution of particles in each momentum bin of a suitable size. Probability density function for each particle type is then counted and the Bayes theorem is applied for identification of individual particles. Hadron PID is main topic of diploma thesis of V. Pospíšil, 5. year student of Czech Technical Univerzity, Faculty of Nuclear Science and Physical Engineering. Main aim of this poster is to outline content of the diploma thesis and to show its progress achieved up to now. • Email contact : [email protected] PID method Each subdetector response on specific particle is given by the probability density function : Pk( h | p, xk ) , where h is type of the particle, xk is measured variable and k is index of the subdetector (k = TOF in this work). As an example let`s take deuterons and protons with p = 750 MeV/c. Their PDFs are gaussians with mean at β = p/( m02c2 + p2)1/2 and wideness determined by the detector accuracy. Uknown particle with p = 750 MeV/c is identified as a deuteron, if its measured x is less than xC. Probability of misidentification of deuteron as a proton is Xc Xc with momentum p = 750 MeV/c. PID process Distributions of velocities for deuterons and protons Experimental data by HADES • Obtaining PDF and relative abundance parameters Momentum vs. beta diagrams are taken from the DST files and projections on beta are made for each momentum bin. Pojection are then fitted by the sum of the gaussians and a background (simple quadratic function). h Ah ( p, h) l Al P ( x p, h) References : BABAR Barlow et al., www.slac.stanford.edu/BFROOT/www/Statistics/Report/report/pdf STAR Fisyak, www.usatlas.bnl.gov/~fisyak//d0/photons/pure.ps A.G.Frodesten, O. Skjeggestad, H. Tofe: Probability and Statistics in Particle Physics P( x p, proton ) ( x h )2 z h Ah exp( ) 2 2 h P( x p, h) (h) ~ P ( x p, h) P( x p, ) ( ) where φ(h) is particle relative abundance function. P( x p, deuteron ) prob. of confusion proton as a deutron is PDFs of deuterons and protons In experimental data there is not equal amounts of deuterons and protons in each event. This fact is not considered by normalized gauss functions, so also relative abundance functions have to be created for each particle. If both normalized PDFs and rel. abundance functions are known, particle the identification could be achieved using the Bayes theorem : zh 2 h Ah • C+C 2 AGeV exp. data was taken during november 2002 (200 mil. events, 56 mil. used in hadron PID). • For low res. tracking used inner MDCs and TOF/TOFino + SHOWER • For high res. tracking were used inner and outer MDCs. During nov02 experiment just two outer MDCs were installed (sect. 1 and 4). • Two metods of momentum reconstructing were used: Kickplane (mg. field is replaced by bending plane) and Spline (track is counted using Maxwell equations). • δp 10% low resolution • δp ??% high resolution Hadron PID progress PDF parameters are evaluated at the moment. High resolution parameters are half– completed (sectors 1. and 4. spline method are done). Using the gaussian (zh) parameters, the PDFs and rel. abundance are easily counted. This should be done for each HADES sector and for multiple azimutal angle regions (because momentum resolution depends on azimutal angle). This task is done by specialized ROOT macros. • Identification of individual particles Examples of taken PDF parameters. At these figures are μh, бh and number of particles (бhxAh) of negative pions in azimutal angle region 50° - 55°, sector 1. Particle identification is performed by HYDRA, the ROOT based collection of libraries for HADES simulation and analysis tasks. HYDRA is presently developed, the libraries for high momentum resolution analysis will be ready soon. Reference : www-hades.gsi.de Near future activity • PDF parameters for kickplane high resolution method have to be carried out, as same as Screenshot of the macro which fits projections and counts PDF parameters for low resolution. • All parameters have to be sorted into a file readable by HYDRA • Identification of individual hadrons have to be performed • Deadline of hadron PID is end of november 2004