Data accumulated by the TASSO detector across the whole range of energies spanned at PETRA, 12⩽ s ⩽46.8 GeV , have been analysed in terms of cluster algorithms. Using parameters optimised at 35 GeV CM energy, three perturbative QCD+fragmentation models were compared with the data. The O( α s 2 ) model gives too few 4,5- cluster events, implying that higher order QCD contributions are required to describe the data. The parton cascade model, incorporating many orders in perturbation theory, gives a better description of the rates of ⩾ 4 clusters, but shows a lack of hard gluon emission by giving too few 3-, and too many 2-cluster events. When hard gluon emission is taken into account, by the cascade model incorporating the O( α s ) matrix element, all cluster rates are reproduced well. All the models describe the trend of the evolution of the cluster rates between 〈 s 〉 = 14 and 43.8 GeV. We find that the rate of 3-jet events seen in the data decreases as s increases in a manner consistent with the Q 2 dependence of α s as predicted by QCD.
No description provided.
No description provided.
Corrected 3 jet rate with YCUT=0.08.
The topology of hadronic e + e − annihilation events has been analysed using the sphericity tensor and a cluster method. Comparison with quark models including gluon bremsstrahlung yields good agreement with the data. The strong-coupling constant is determined in 1st order QCD to be α S =0.19±0.04 (stat) ± 0.04 (syst.) at 22 GeV and α S =0.16 ±0.02± 0.03 at 34 GeV. The differential cross section with respect to the energy fraction carried by the most energetic parton agrees with the prediction of QCD, but cannot be reproduced by a scalar gluon model. These results are stable against variations of the transverse momentum distribution of the fragmentation function within the quoted errors.
No description provided.
Hadronic events obtained with the CELLO detector at PETRA were compared with first-order QCD predictions using two different models for the fragmentation of quarks and gluons, the Hoyer model and the Lund model. Both models are in reasonable agreement with the data, although they do not completely reproduce the details of many distributions. Several methods have been applied to determine the strong coupling constant α S . Although within one model the value of α S varies by 20% among the different methods, the values determined using the Lund model are 30% or more larger (depending on the method used) than the values determined with the Hoyer model. Our results using the Hoyer model are in agreement with previous results based on this approach.
DATA CORRECTED WITH HOYER MODEL (ALPHA-S=0.15).
DATA CORRECTED WITH LUND MODEL (ALPHA-S=0.25).
No description provided.
Forum