Author:

Pedro A. Movilla Fernandez

Title:

Studies of Quantum Chromodynamics and Determinations of the Strong Coupling Constant alpha_S at sqrt{s} = 14 - 44 GeV with the JADE Detector

Abstract:

A reanalysis of hadronic final states of e+e- annihilation data collected with the JADE detector at the PETRA storage ring (1979-1986) is presented. This thesis focuses on perturbative and non-perturbative aspects of Quantum Chromodynamics (QCD), in particular measurements of the coupling constant alpha_S of the strong interaction and the experimental investigation of power corrections to event shapes.

Major parts of the original detector simulation and offline event reconstruction software (which are partially older than 20 years) were successfully adapted and reactivated on current computer platforms. Due to the good performance of the programs, reasonable measurements of event shape observables are now possible at all PETRA energies.

The differential cross sections of the variables thrust T, heavy jet mass M_H, the wide and total jet broadening B_W and B_T, respectively, the C parameter and the differential 2-jet rate in the Durham scheme were determined at centre-of-mass energies sqrt{S}= 14-44GeV. The distributions were compared with the predictions of corresponding phenomenological hadronisation models previously tuned to LEP data at the Z0 mass scale, sqrt{S}=M_Z. The parton shower and string fragmentation model implemented in PYTHIA/JETSET is found to be well capable of describing event shapes of hadronic final states down to sqrt{S}= 14 GeV. The performance of the models ARIADNE (colour dipole scheme) and HERWIG (cluster fragmentation), however, is more moderate, and COJETS (independent fragmentation) is clearly disfavoured by the data. Obviously the model parameters of these generators need a retune at lower centre-of-mass energies.

From the measured event shapes, the strong coupling constant alpha_S was extracted by fits of the most complete perturbative calculations for the differential distributions available so far, namely a combination of O(alpha_S**2) and resummed NLLA calculations. The predictions which were folded with the hadronisation models agree well with the data in most cases and allow consistent determinations of alpha_S at all PETRA energies. The final results with total errors,

alpha_S(14.0 GeV)= 0.1704+0.0206-0.0171,
alpha_S(22.0 GeV)= 0.1513+0.0144-0.0121,
alpha_S(34.8 GeV)= 0.1431+0.0118-0.0096,
alpha_S(38.3 GeV)= 0.1397+0.0108-0.0087,
and
alpha_S(43.8 GeV)= 0.1306+0.0096-0.0080,

are substantially more precise than former PETRA measurements and are also in good agreement with the theoretical prediction for alpha_S(sqrt{s}) based on the world average value for alpha_S{M_Z}. The errors are dominated by renormalisation scale and hadronisation uncertainties. Using the new PETRA values and corresponding LEP results at sqrt{S}>=M_Z, the QCD expectation for the energy evolution of alpha_S can be verified more significantly on the basis of a homogeneous analysis technique within a wide energy range of the e+e- continuum.

Power corrections proportional to 1/sqrt{s} based on an analytical ansatz by Dokshitzer, Marchesini and Webber (DMW) were investigated as a promising alternative approach to describe non-perturbative effects in event shapes. Besides alpha_S, the model depends only on one additional free parameter, the 0'th moment alpha_0(muI) of the physical strong coupling constant alpha_S(mu) in the energy region mu=0...muI around the Landau pole. In addition to the JADE data, also measurements of other experiments up to sqrt{S}= 189 GeV were considered for experimental tests. Except for some discrepancies observed for the less inclusive quantities (B_W and M_H) in particular at lower centre-of-mass energies, the major parts of the overall event shape spectra are reproduced well. From global fits to differential distributions and mean values, the free parameters alpha_S and alpha_0 were simultaneously determined. The combined results (with total errors) are

alpha_S(M_Z)= 0.1126+0.0058-0.0048,
alpha_0(2 GeV)= 0.542+0.090-0.068
(distributions),

alpha_S(M_Z)= 0.1187+0.0031-0.0021,
alpha_0(2 GeV)= 0.485+0.066-0.045
(means).

Moreover, the predicted universality of alpha_0 which is an essential ingredient of the DMW model has been established experimentally within the systematic uncertainties.