Power corrections to event shapes with mass-dependent operators

Vicent Mateu, Iain W. Stewart, Jesse ThalerPhysical 2013 · DOI · arXiv

Abstract

We introduce an operator depending on the "transverse velocity" $r$ that describes the effect of hadron masses on the leading $1/Q$ power correction to event-shape observables. Here, $Q$ is the scale of the hard collision. This work builds on earlier studies of mass effects by Salam and Wicke and of operators by Lee and Sterman. Despite the fact that different event shapes have different hadron mass dependence, we provide a simple method to identify universality classes of event shapes whose power corrections depend on a common nonperturbative parameter. We also develop an operator basis to show that at a fixed value of $Q$, the power corrections for many classic observables can be determined by two independent nonperturbative matrix elements at the 10% level. We compute the anomalous dimension of the transverse velocity operator, which is multiplicative in $r$ and causes the power correction to exhibit nontrivial dependence on $Q$. The existence of universality classes and the relevance of anomalous dimensions are reproduced by the hadronization models in Pythia 8 and Herwig++, though the two programs differ in the values of their low-energy matrix elements.

Introduction

In high-energy particle physics, "event shapes" such as thrust or the C-parameter are essential tools for studying the geometry of particle jets produced in collisions. While perturbative Quantum Chromodynamics (QCD) provides accurate predictions at high energy scales ($Q$), these calculations eventually encounter the non-perturbative regime of "hadronization," where quarks and gluons transform into massive hadrons. Historically, power corrections—shifts in distributions suppressed by $1/Q$—were often modeled assuming massless hadrons. However, real hadron masses break the universality of these corrections. This paper introduces a rigorous field-theoretic framework using a new "transverse velocity" operator to account for these mass effects, bridging the gap between high-energy theory and experimental detector data.

The Transverse Velocity Operator and Universality Classes

The core of this framework is the definition of a new operator, $\hat{\mathcal{E}}_T(r, y)$, where $r = p_\perp / m_\perp$ represents the "transverse velocity." This operator allows for tracking how a particle's mass affects its contribution to an event shape. By generalizing the standard energy flow operator, the authors can categorize event shapes into "universality classes" defined by a function $g(r)$. Shapes within the same class share a common non-perturbative shift parameter, simplifying the description of diverse observables.

The transverse velocity operator $\hat{\mathcal{E}}_T(r, y)$ (bottom) generalizes the standard energy flow operator $\hat{\mathcal{E}}_T(\eta)$ (top) to account for the mass-dependent transverse velocity $r$ of particles within a jet.

Different experimental measurement schemes (e.g., E-scheme, P-scheme) handle particle masses differently. The authors demonstrate that the choice of scheme fundamentally alters the functional form of the power corrections. For instance, the "E-scheme," which measures only energies and angles, is shown to be a special case where universality is more easily maintained.

Comparison of different event-shape schemes (Original, P, E, R, J) and their corresponding functional dependencies for various non-perturbative parameters.

Renormalization Group Evolution and Results

A significant finding of this work is that non-perturbative power corrections are not constant but evolve, or "run," with the energy scale $Q$. By calculating the one-loop anomalous dimension of the transverse velocity operator, the authors prove that the power correction exhibits logarithmic dependence on $Q$: $$\mu \frac{d}{d\mu} \Omega_1(r, \mu) = -\frac{\alpha_s C_A}{\pi} \ln(1-r^2) \Omega_1(r, \mu)$$ This "running" explains why constant-shift models often fail when comparing data across different collision energies, such as those from the Large Electron-Positron (LEP) collider.

Validation using Monte Carlo generators (Pythia 8 and Herwig++) confirms these theoretical predictions. Despite their different internal fragmentation models, both programs reproduce the predicted universality classes and logarithmic running. Furthermore, the authors show that a simple two-parameter basis can describe mass-dependent power corrections for most classic event shapes within 10% accuracy. Re-analyzing thrust data with this running correction shifts the extracted strong coupling constant $\alpha_s(m_Z)$ by approximately $\pm 0.0005$, a significant portion of the current experimental uncertainty.

Conclusion

This work provides a rigorous foundation for including hadron mass effects in QCD power corrections. By identifying universality classes and calculating the scale evolution of these corrections, it offers a more precise path for extracting fundamental parameters like $\alpha_s$ from experimental data. Future work will involve extending this framework to multi-jet events and higher-order loop calculations to further refine the precision of QCD predictions.

Summary updated

Abstract

We introduce an operator depending on the ``transverse velocity'' $r$ that describes the effect of hadron masses on the leading $1/Q$ power correction to event-shape observables. Here, $Q$ is the scale of the hard collision. This work builds on earlier studies of mass effects by Salam and Wicke [J. High Energy Phys. 05 (2001) 061] and of operators by Lee and Sterman [Phys. Rev. D 75, 014022 (2007)]. Despite the fact that different event shapes have different hadron mass dependence, we provide a simple method to identify universality classes of event shapes whose power corrections depend on a common nonperturbative parameter. We also develop an operator basis to show that at a fixed value of $Q$, the power corrections for many classic observables can be determined by two independent nonperturbative matrix elements at the 10% level. We compute the anomalous dimension of the transverse velocity operator, which is multiplicative in $r$ and causes the power correction to exhibit nontrivial dependence on $Q$. The existence of universality classes and the relevance of anomalous dimensions are reproduced by the hadronization models in Pythia 8 and $\mathrm{Herwig}\text{\hskip-0.22em}+\text{\hskip-0.22em}+$, though the two programs differ in the values of their low-energy matrix elements.
BibTeX
@inproceedings{mateu2013power,
  author = {Vicent Mateu and Iain W. Stewart and Jesse Thaler},
  title = {{Power corrections to event shapes with mass-dependent operators}},
  booktitle = {Physical},
  year = {2013},
  doi = {10.1103/physrevd.87.014025},
}
Figure