The final 20 layer prototype for the AMS transition radiation detector : beamtests, data analysis, MC-studies

The AMS TRD prototype, as described in this thesis, was subjected to 2 high energy beamtests at CERN test facilities (X7,H6) in summer 2000, where more than 3 million events of p+,e-,pi- and mu- data with beam energies up to 250 GeV have been recorded. The data analysis is based solely on single track events. These are selected from several selection criteria, applied to all events with a successfully reconstructed single track from linear regression. The rejection analysis was mainly focused on the prototype's capability of rejecting protons against electrons. These rejection factors as functions of beam energy were determined in two different ways, namely using the Cluster Counting and a Likelihood method. The Cluster Counting (CC) method is based on the binary counting of hits on the track above a certain energy threshold, *Ecut*, and the requirement of a minimum number, *Hcut*, of such hits for an event to be selected "electron-like". This *Hcut*, referring to a defined *Ecut*, is chosen such that 90% of the electrons are accepted. The application of this same *Hcut* to all proton events is used to calculate the rejection factor, as the ratio of total number of proton events in the sample, divided by those events that are selected "electron-like". This analysis derived rejection factors ranging from (625 +/- 117) at 20 GeV down to (80 +/- 5) at 250 GeV beam energy. The indicated errors are calculated from the statistical uncertainty of the selected number of proton events. The likelihood (LH) method uses the full proton and electron energy spectra as single hit probability distributions, after normalisation to an integral of one. For each event, the single hit probabilities for the hits on the track are multiplied and the geometric mean is calculated separately for the electron (Pe) and proton (Pp) hypotheses. The electron likelihood is finally calculated as the ratio Le = -ln(Pe/(Pe+Pp)). In the *Le* distribution of electron events the *LHcut* is now chosen such, that 90% of the electron events are accepted and the proton rejection is derived as before. This analysis derived rejection factors ranging from (1429 +/- 408) at 20 GeV down to (143 +/- 12) at 250 GeV beam energy. Compared to the CC this denotes on average an improvement by more than a factor of 2. In the same way the rejection factors for pions against electrons are derived from likelihood analysis. They range from (1000 +/- 400) at 20 GeV beam energy down to (19.2 +- 0.7) at 100 GeV. This described rejection analysis was carried out, using the 20 GeV electron sample as reference. The repeated analysis, using a 40 GeV electron reference sample, reproduced the derived rejection factors, within the statistical errors. The GEANT 3.21 software package, improved in the simulation of dE/dx energy loss, and additional supplements to generate and detect transition radiation, first implemented by members of the HERA-B Collaboration, were used to reproduce the beamtest results. The detector characteristics, like the geometry and part of the readout electronics, as well as the beamtest conditions, have been introduced into the GEANT as precisely as possible. Supplementary to this, several adjustments to the above mentioned, original GEANT supplements, have been accomplished to give the best agreement to the measured energy spectra. The such adapted and optimised MC simulation is capable to reproduce the measured proton, pion and electron energy spectra over the full range of particle energies. The derived rejection factors of protons against electrons up to a particle energy of 160 GeV, and those of pions, over the full pion energy range, very well agree with those rejection factors derived from the measured data. Above 160 GeV proton energy a clear discrepancy appears in the comparison between data and MC rejection factors. An additionally introduced diffractive proton proton interaction process is capable to resolve this discrepancy up to the highest energetic measured data. All other attempts to resolve this disagreement failed to follow suit.

Grantor dc:publisher

Publikationsserver der RWTH Aachen University

Year dc:date

2003

Author dc:creator

• Orboeck, Jörg

Contributors dc:contributor

• Schael, Stefan

• info:eu-repo/classification/ddc/530
• Physik
• AMS
• TRD
• transition radiation
• rejection
• TR GEANT Simulation

Statement dc:rights

• info:eu-repo/semantics/openAccess

Language dc:language

eng

Identifier

https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-120793%22

OAI identifier oai:identifier

oai:publications.rwth-aachen.de:58970