Measurements of Flavour Dependent Fragmentation Functions in Z^0 -> qq(bar) Events

TL;DR

This study presents flavour-dependent fragmentation functions in Z0 → qq¯ events using OPAL data, separating the results for uds, c, and b quarks via D* meson and secondary-vertex tagging. A joint fit to momentum spectra in flavour-tagged hemispheres yields $x_p$ and $\xi_p$ distributions for each flavour, with the first measurement of flavour-specific $\xi_p$ maxima. The results show significantly softer fragmentation for b quarks compared with light quarks, while the inclusive fragmentation matches Jetset 7.4 predictions and confirms expected scaling violations. Systematic checks and cross-comparisons with other experiments support the robustness of the flavour-dependent conclusions. Overall, the work demonstrates the viability and physical relevance of flavour-resolved fragmentation studies at the Z0 resonance.

Abstract

Fragmentation functions for charged particles in Z -> qq(bar) events have been measured for bottom (b), charm (c) and light (uds) quarks as well as for all flavours together. The results are based on data recorded between 1990 and 1995 using the OPAL detector at LEP. Event samples with different flavour compositions were formed using reconstructed D* mesons and secondary vertices. The ξ_p = ln(1/x_E) distributions and the position of their maxima ξ_max are also presented separately for uds, c and b quark events. The fragmentation function for b quarks is significantly softer than for uds quarks.

Figures (5)

• Figure 1: (a): The decay length significance distribution in data (symbols) and Monte Carlo (solid curve). The contribution from uds and from c quarks in the Monte Carlo distribution has been shaded. The boundaries of the three decay length significance bins used in this analysis: $-10<\hbox{$L/\sigma_{L}$}<1$, $1<\hbox{$L/\sigma_{L}$}<5$ and $5<\hbox{$L/\sigma_{L}$}<50$ are indicated by vertical lines. (b) to (d): The distribution of the mass difference of the ${\rm D}^{*\pm}$ candidate and ${\rm D}^{0}$ candidate in the three different $x_{D^{*}}$ bins. The symbols show the data while the solid lines are the results of the fits described in the text.
• Figure 2: The upper plot shows the measured fragmentation functions for uds events (filled symbols), c events (open squares) and b events (open triangles) as well as the inclusive fragmentation function (solid line). The lower plot shows the ratio of the flavour dependent fragmentation functions to the inclusive fragmentation function. The error bars include statistical and systematic uncertainties. The systematic uncertainties are correlated between bins as well as between flavours.
• Figure 3: $\hbox{$\xi_{p}$} = \ln(1/\hbox{$x_{p}$})$ distribution for (a) all events, (b) uds events, (c) c events and (d) b events. The solid lines show the results of the skewed Gaussian fitted to the distributions in the indicated fit range and the dashed lines show the results of a normal Gaussian fit. The error bars include statistical and systematic uncertainties.
• Figure 4: Comparison of the results for the inclusive fragmentation function for this analysis (O) with results from ALEPH (A), DELPHI (D) and MARK II (M) at $\sqrt{s}=m_{{\rm Z}^0}$sameE1sameE2 and of the flavour dependent fragmentation function with the results from DELPHI delphi2. The error bars include statistical and systematic uncertainties. The Jetset 7.4 predictions for the fragmentation function are shown as full horizontal lines and the Herwig 5.9 predictions as dotted horizontal lines.
• Figure 5: Comparison of the results for the inclusive fragmentation function with results at different lower lowE and higher centre-of-mass energies highEopal133. The error bars include statistical and systematic uncertainties. The solid lines show the Jetset 7.4 prediction, assuming the centre-of-mass energy dependence of the flavour composition as predicted by the electroweak theory. The dotted lines show the Jetset 7.4 prediction assuming for all energies the same flavour mix as at $\sqrt{s} = 18$ GeV. The dotted line is almost entirely hidden behind the full line and even at $\sqrt{s} = 91.2$ GeV, only a negligible difference between the two curves can be seen because the effect of an increased b contribution is compensated largely by the effect of a decreased c contribution.