HREPHAH 0H3HKA, 2004, moM 67, № 1, c. 128-137
VERY HIGH MULTIPLICITY PHYSICS
SOFT AND HARD INTERACTIONS IN pp COLLISIONS
AT -y/i = 1800 AND 630 GeV
© 2004 F. Rimondi* (for the CDF Collaboration)
Dipartimento di Fisica and Istituto Nazionale di Fizica Nucleare, University di Bologna, Italy
Received May 21, 2003
We present a study of pp collisions at a/s = 1800 and 630 GeV collected using a minimum bias trigger by the CDF experiment in which the data set is divided into two classes corresponding to "soft" and "hard" interactions. For each subsample, the analysis includes measurements of the multiplicity, transverse momentum (pT) spectrum, and the average pT and event-by-event pT dispersion as a function of multiplicity. A comparison of results shows distinct differences in the behavior of the two samples as a function of the center-of-mass (CM) energy. We find evidence that the properties of the soft sample are invariant as a function of CM energy.
1. INTRODUCTION
Hadron interactions are often classified as either "hard" or "soft" [1, 2]. Although there is no formal definition for either, the term "hard interactions" is typically understood to mean high transverse energy (ET) parton—parton interactions associated with such phenomena as high-ET jets, while the soft component consists of everything else. Whereas perturbative QCD provides a reasonable description of high-ET jet production, there is no equivalent theory for the low-ET multiparticle production processes that dominate the inelastic cross section. Some QCD inspired models [2] attempt to describe these processes by the superposition of many parton interactions extrapolated to very low momentum transfers. It is not known, however, if these or other collective multiparton process is at work.
The study of low-ET interactions usually involves collecting data using minimum bias (MB) triggers, which, ideally, sample events in fixed proportion to the production rate — in other words, in their "natural" distribution. Lacking a comprehensive description of the microscopic processes [3] involved in low-ET interactions, our knowledge of the details of low transverse momentum (pT) particle production rests largely upon empirical connections between phenomenological models and data collected with MB triggers at many center-of-mass (CM) energies. Such comparisons are further complicated by the difficulty in isolating events of a purely soft or purely hard nature.
E-mail: Franco.Rimondi@bo.infn.it
This paper adopts a novel approach in addressing this issue using samples of pp collisions at y/s = = 1800 and 630 GeV collected with a MB trigger. The analysis first divides the full MB samples into two subsamples, one highly enriched in soft interactions, the other relatively depleted of soft interactions. We then compare inclusive distributions and final state correlations between the subsamples and as a function of CM energy in order to gain insights into the mechanisms of particle production in soft interactions. The results in the isolated soft sample exhibit some interesting properties, in particular, an unpredicted invariance with CM energy. The results presented in this paper published in [4].
2. DATA SET AND EVENT SELECTION
Data samples have been collected with the CDF detector at the Fermilab Tevatron Collider. The CDF apparatus has been described elsewhere [5]; here only the parts of the detector utilized for the present analysis are discussed.
Data at 1800 GeV were collected with a MB trigger during runs 1A and 1B, and at 1800 and 630 GeV during run 1C. This trigger requires coincident hits in scintillator counters located at 5.8 m on either side of the nominal interaction point and covering the pseudorapidity (n = — log(tan(6/2)), where 6 is an angle with respect to the proton direction) interval 3.2 < \n\ < 5.9, in coincidence with a beam-crossing signal.
The analysis uses charged tracks reconstructed within the Central Tracking Chamber (CTC). The CTC is a cylindrical drift chamber covering a n
interval of about three units with full efficiency for \n\ < 1 and pt> 0.4 GeV/c.
Inside the CTC inner radius, a set of time projection chambers (VTX) [6] provides r—z tracking information out to a radius of 22 cm for \n\ < 3.25. The VTX is used in this analysis to find the z position of event vertices, defined as a set of tracks converging to the same point along the z-axis. Reconstructed vertices are classified as either "primary" or "secondary" based upon a combination of the number of tracks pointing to the vertex and the forward—backward symmetry of these tracks. High-multiplicity vertices with highly symmetric topologies are considered to be primaries; low-multiplicity, highly asymmetric vertices are classified as secondaries.
The transverse energy flux was measured by a calorimeter system [7] covering from —4.2 to 4.2 in n.
The 1800-GeV data sample consists of subsam-ples collected during three different time periods. Approximately 1 700 000 events were collected in run 1A at an average luminosity of 3.3 x 1030 s-1 cm-2, 1 500 000 in run 1B at an average luminosity of 9.1 x 1030 s-1 cm-2 and 106000 in run 1C at an average luminosity of 9.0 x 1030 s-1 cm-2. The 630-GeV data set consists of about 2 600 000 events recorded during run 1C at an average luminosity of 1.3 x 1030 s-1 cm-2.
Additional event selection conducted off-line removed the following events: (i) events identified as containing cosmic-ray particles as determined by time-of-flight measurements using scintillator counters in the central calorimeter; (ii) events with no reconstructed tracks; (iii) events exhibiting symptoms of known calorimeter problems; (iv) events with at least one charged particle reconstructed in the CTC to have pT > 400 MeV/c, but no central calorimeter tower with energy deposition above 100 MeV; (v) events with more than one primary vertex; (vi) events with a primary vertex more than 60 cm away from the center of the detector (in order to keep full tracking efficiency in the CTC and avoid energy leakage through exposed cracks in the calorimeter); (vii) events with no primary vertices.
After all event selection cuts, 2 079558 events remain in the full minimum bias sample at y/s = = 1800 GeV (runs 1A + IB + 1C) and 1 963 157 in that at y/s = 630 GeV (run 1C).
The vast majority of rejected events failed the vertex selection. About 0.01% of selected events contain background tracks from cosmic rays that are coincident in time with the beam crossing and pass near the event vertex. The residual beam gas contamination is about 0.02%.
The systematic uncertainties that arise from the event selection criteria and other sources, are discussed in Section 6.
3. TRACK SELECTION
Reconstructed tracks within each event must pass selection criteria designed to remove the main sources of background. Tracks must pass through a miniumum number of layers in the CTC, and have a minimum number of hits in each superlayer in order to reduce the number of tracks with reconstruction errors. Fake and secondary particle tracks are removed by requiring that tracks pass within 0.5 cm of the beam axis, and within 5 cm along the z-axis of the primary event vertex. Accepting only tracks with pT > 0.4 GeV/c and within \n\ < 1.0 ensures full efficiency and acceptance.
We define the charged track multiplicity in an event, N*h, as the number of selected CTC tracks in the event. The mean pT of the event is defined as
1 Nh ch i
unless stated otherwise.
4. SELECTION OF SOFT AND HARD INTERACTIONS
The identification of soft and hard interactions is largely a matter of definition [8]. In this analysis we use a jet reconstruction algorithm to distinguish between the two classes. The algorithm employs a cone with radius R = (An2 + A^2)1/2 = 0.7 to define "clusters" of calorimeter towers belonging to the jet. To be considered, a cluster must have a transverse energy (ET) of at least 1 GeV in a seed tower, plus at least 0.1 GeV in an adjacent tower.
In the regions \n\ < 0.02 and 1.1 < \n\ < 1.2, a track clustering algorithm is used instead of the calorimeter algorithm in order to compensate for energy lost in calorimeter cracks. A track cluster is defined as one track with pT > 0.7 GeV/c and at least one other track with pT > 0.4 GeV/c in a cone of radius R = 0.7.
We define a soft event as one that contains no cluster with ET > 1.1 GeV. All other events are classified as hard.
5. EFFICIENCY CORRECTIONS
The track reconstruction efficiency for the CTC has been investigated for several different analyses and under various conditions at CDF [9—11]. For this analysis, we have calculated a full-event track reconstruction efficiency using a parametric MC sample. Version 5.7 of the Pythia generator was used with the MB configuration tuned to match the inclusive multiplicity and pT distributions of the 1800-GeV
sample. For each inclusive distribution, a track finding efficiency correction was computed by taking the ratio of the Pythia generated distribution to the corresponding distribution from tracks traced through the apparatus. The efficiency for reconstructing the correct event charged multiplicity is about 95% up to a multiplicity of about 20, falling to about 85% at multiplicities above about 20.
The same Pythia MC sample was used to evaluate the background from gamma ray conversions, charged and neutral particle decays. Correction factors due to these effects have been computed as a function of track pT and the event multiplicity.
There exists a small contamination from diffrac-tive events even in the restricted region of phase space examined in this study. We have evaluated this contamination with a special Pythia MC run in which only the diffractive generation algorithm was switched on. The data were then subjected to the full event and track selection procedure. The correction for this effect is estimated to be about 5% in the zero multiplicity bin, decreasing rapidly to zero for N*h ~
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.