PRODUCTION of HIGH MASS Cv and E'e' PAIRS in the UA2
Total Page:16
File Type:pdf, Size:1020Kb
142 PRODUCTION OF HIGH MASS cv AND e'e' PAIRS IN THE UA2 EXPERIMENT AT THE CERN pp COLLIDER The UA2 Collaboration (Berne, CERN, CopenhagenT Orsay;" Pavía, Sacïay) Dî 84100S6163 presented by J. SCHACHER University of Berne, Switzerland ABSTRACT We present new results on intermediate vector boson production at the CERN pp collider. A comparison is made with the predictions of the standard model of the unified electroweak Glashow-Salam-Weinberg theory. - 143 - 1, INTRODUCTION We report here the results from a search for electrons with > 15 GeV/c produced at the CERN pp collider (Vs = 540 GeV) during its 1982 and 1983 periods of operation. Following a general discussion of the topology of the events containing an electron candidate, we shall compare the data with expectations in the framework of the electroweak standard model [1] for the reactions p + p - W~ + anything (1) - e" • v (?) p • p - Z° + anything (2) -* e* + e" or e* + e" + y where W~ and Z° are the postulated charged and neutral Intermediate Vector Bosonc (IVB), respectively, According to the amount of data collected we are now in the position to study some details of the IVB production, e.g. the influence of emission of gluon radiation on the distribution of the W transverse momentum (Fig.1). Fig.l Typical diagram for W and Z production, taking into account emission of gluon (g) radiation. 7T : parton in p or p. - 144 - Preliminary results from the study reported here have already been presented elsewhere [2] and a more complete discussion can be found in a recent publication [3]. 2. THE DETECTOR The experimental apparatus, shown in Fig.2, has been described in detail elsewhere [4]. At the centre of the apparatus a system of cylindrical chambers (the vertex detector [5]) measures charged particle trajectories in a region without magnetic field. The vertex detector consists of : a) four multi-wire proportional chambers, (CI to C4), having cathode strips with pulse height read-out at ±45° to the wires ; b) two drift chambers with measurement of the charge division on a total of 12 wires per track. The drift chambers are used to obtain both tracking information and to evaluate the most likely ionisation 10 associated with each track. From the reconstructed tracks the position of the event vertex is determined with a precision of ±1 mm in all directions. Fig.2 A view of the UA2 detector in a plane containing the beam line. - 145 - The vertex detector is surrounded by an electromagnetic and hadronic calorimeter (central calorimeter [6]), which covers the full azimuth and a polar angle interval 40° < 6 < 140°. The calorimeter is segmented into 240 independent cells, each covering 10° in 6 and 15° in 4> an¿ built in a tower structure pointing to the centre of the interaction region. The celfs are segmented longitudinally into a 17 radiation lengths thick electromagnetic compartment (lead-scintillator) followed by two hadronic compartments (iron-scintillator) of = 2 absorption lengths each. In the angular region covered by the central calorimeter a cylindrical tungsten converter, 1.5 radiation lengths thick, followed by a cylindrical proportional chamber (C5), is located just after the vertex detector. This device localises electromagnetic showers initiated in the tungsten with a precision of ±3 mm, as verified using test-beam electrons. For the first 15 nb-1 of integrated luminosity, collected during the Autumn of 1982, the azimuthal coverage of the central calorimeter was only 300°. The remaining interval (±30° around the horizontal plane) was cohered by a magnetic spectrometer which included a lead-glass array to measure charged and neutral particle production [7]. The two forward regions (20° < 8 < 37.5° and 142.5° < 9 < 160°) are each equipped with twelve toroidal magnet sectors with an average bending power of .38 Tm. Each sector is instrumented with a) three drift chambers [8] located after the magnetic field region. Each chamber contains three planes with wires at -7°, 0° and *7° with respect to the magnetic fie'd direction. b) a 1.4 radiation lengths thick lead-iron converter, followed by a preshower counter which consists of two pairs of layers of 20 mm diameter proportional tubes (MTPC), staggered by a tube radius and equipped with pulse height measurement [9]. This device localises electromagnetic showers initiated in the converter with a precision of ±6 mm. c) an electromagnetic calorimeter consisting of lead-scintiilator counters assembled in ten independent cells, each covering 15° in $ and 3.5° in 6. Each cell is subdivided into two independent longitudinal sections, 24 and 6 radiation lengths thick, respectively, the latter providing rejection against hadrons. - 146 - The systematic uncertainty in the energy calibration of the electromagnetic calorimeters for the data presented here amounts to an average value of ±1.5%. The cell-to-cell calibration has a distribution with a r.m.s. of 2.2%. The energy resolution for electrons is measured to be c^/E = 0.14/VE [6] in the central calorimeter and 0.17/VE in the forward ones (E in GeV). 3. DATA TAKING AND DATA REDUCTION In order to implement a trigger sensitive to electrons of high transverse momentum, the photomultiplier gains ¡n all calorimeters were adjusted so that their signals were proportional to the transverse energy. Because of the cell dimensions, electromagnetic showers initiated by electrons may be shared among adjacent cells. Trigger thresholds were applied, therefore, to linear sums of signals from matrices of 2 x 2 cells, rather than to individual cells. In the central calorimeter, all possible 2x2 matrices were considered; in the two forward ones, we included only those consisting of cells belonging to the same sector. A W trigger signal (for Z° trigger: see Ref.12) was generated whenever the linear sum from at least one such matrix exceeded a threshold which was typically set at 8 GeV. To suppress background from sources other than pp collisions, we required a coincidence with two signals obtained from scintillator hodoscopes covering the polar angle interval 0.47° - 2.84° with respect to the beams on both sides of the collision region. These hodoscopes, which were part of an experiment to measure the pp total cross-section £ï0], gave a coincidence signal in more than 98% of all non-diffractive pp collisions. Approximately 7 x 10s W triggers were recorded during the 1982 and 1983 runs, corresponding to an integrated luminosity SÛ - 131 nb"1, A first data reduction is made by requiring the presence of an energy cluster with a transverse energy greater than 15 GeV. In the central calorimeter, clusters are obtained by joining all electromagnetic cells which share a side and contain at least 0.5 GeV. A halo contribution from the cells having at least one side in common with a cluster is also added. The forward calorimeter clusters consist of at most two adjacent cells having the same azimuth (the cells are far from the interaction point and are much larger than 147 - the lateral extension of an electromagnetic shower - the dead region between cells at different azimuth does noi allow clustering across it). In the surviving events, a search is .made for configurations consistent with the presence of a high-p-j. electron among the collision products. An electron is identified from the observation of a) the presence of a cluster of energy deposition in the first compartment (electromagnetic) of the calorimeters with a small lateral size and only a small energy leakage in the hadronic compartment. b) the presence of a reconstructed charged particle track which peints to the energy cluster. The pattern of energy deposition must agree with that expected from an isolated electron incident along the track direction. c) the presence of a hit in the preshower counter with an associated pulse height larger than that of a minimum ionising particle (m.i.p.). The distance of the hit from the track must be consistent with the space resolution of the counter itself. A set of cuts has been defined according these requirements. A detailed description of these cuts can be found in Ref.3. The efficiencies of the simultaneous application of these cuts are 76% and 80% in the central and forward regions, respectively. 4. TOPOLOGY OF EVENTS CONTAINING AN ELECTRON CANDIDATE. After application of the electron cuts the sample is reduced to 225 events, containing genuine electrons and still fake electrons resulting from misidentification of hadrons. The p^. distribution is shown in Fig.3. The background of fake electrons can be shown to fall mainly into two categories. In approximately 70% of the cases we are dealing with "overlaps", i.e. jets fragmenting into a hard TT° with a charged pion nearby in angle. The rest of the background results from n°s undergoing Dalitz decays or conversions in the beam-pipe. From studies of hadron jets [11] we expect the fake electrons to be accompagnied by other high-p^. jets at approximately opposite azimuth. We shall th- refore search for high-p^. jets by grouping together adjacent cells with energy into clusters using an algorithm described elsewhere [11 ]. Clusters with more than 3 GeV of transverse energy are retained and called jets. For minimum bias triggers such clusters occur :n only 15% of the events. - 148 - Electron camMates/GeV/c Threshold SO 30 1, 20 - 10 S 10 15 35 ¡.0 45 SO 55 PT ICeV/c) Fig.3 Transverse momentum distribution of the 225 electron candidates satisfying the electron cuts. We find that 45 events contain no jet, the electron candidate being the only high-pT particle observed. Their Pj distribution is shown in Fig.4a. Such events contain either a neutrino, as in the case of W — ev decay, or other high-p.j- particles having escaped detection.