TL;DR
This paper presents the results of a search for lepton-flavor violating (LFV) decays of the tau lepton into three charged leptons, using a large dataset collected by the BABAR experiment. The study found no evidence for these rare decays, setting new stringent upper limits on their branching fractions, which are crucial for constraining theories beyond the Standard Model of particle physics.
What Problem Does This Paper Address?
The Standard Model (SM) of particle physics, while incredibly successful, has some limitations. One of these is its prediction that lepton flavor is conserved. This means that a muon should always behave like a muon, and a tau lepton like a tau lepton, never transforming into one another. However, the discovery of neutrino oscillations implies that neutrinos can change flavor, suggesting that lepton flavor might not be absolutely conserved in the charged sector either.
LFV decays, particularly those involving the tau lepton (which is much heavier than the electron or muon and thus more prone to decay into lighter leptons), are a key area to search for physics beyond the SM. Specifically, this paper focuses on neutrinoless tau decays into three charged leptons (e.g., $\tau^- \rightarrow e^- \mu^+ \mu^-$ or $\tau^- \rightarrow e^- e^+ e^-$). Observing such a decay would be a clear sign of new physics. The challenge is that these decays are predicted to be extremely rare, making them very difficult to detect amidst a large background of standard processes. This paper aims to push the experimental limits on the branching fractions of these LFV decays to unprecedented levels.
Methods
The BABAR experiment collected a substantial dataset of approximately 430 million tau-lepton pair events. The core of the analysis involves identifying events where a tau lepton decays into three charged leptons. This requires sophisticated particle identification (PID) techniques to distinguish between electrons and muons, as well as precise tracking of charged particles.
The analysis employs a multivariate algorithm for electron identification, utilizing information from the electromagnetic calorimeter and tracking system. For muon identification, a bagged decision tree (BDT) algorithm is used, incorporating data from the instrumented flux return (IFR) and inner trackers. The paper highlights improvements in these PID techniques compared to previous analyses, which are crucial for reducing misidentification probabilities and improving sensitivity.
Signal events are simulated using Monte Carlo (MC) methods, incorporating higher-order radiative corrections and using established tau decay simulation tools. The analysis defines a "signal region" in a two-dimensional plane of reconstructed invariant mass ($M_{rec}$) and energy difference ($\Delta E$) of the three-prong tau decay candidate. Events falling within this region, with specific particle identification criteria, are considered potential LFV signal events. Backgrounds from standard tau decays and other Standard Model processes (like Bhabha scattering and two-photon events) are carefully modeled using control samples and fitted to the data.
Datasets
The primary dataset used in this analysis consists of integrated luminosity of 468 fb⁻¹ collected by the BABAR detector at the PEP-II collider. This data was collected at two center-of-mass energies: $\sqrt{s} = 10.58$ GeV (426 fb⁻¹) and $\sqrt{s} = 10.54$ GeV (42 fb⁻¹).
Monte Carlo simulations were extensively used to model signal events and estimate background processes. Control samples, derived from the collected data, were used to validate the PID efficiencies and to model specific background contributions.
Key Insights and Ideas
- Improved PID is Key: The paper emphasizes that advancements in electron and muon identification techniques were critical for achieving higher signal efficiencies and reducing background, directly leading to improved limits.
- Blind Analysis: A "blind analysis" procedure was employed, meaning the signal region was kept hidden until the selection criteria were finalized. This is a standard but essential technique to prevent experimenter bias.
- Comprehensive Background Modeling: The analysis meticulously models and accounts for various background sources, including standard tau decays and other Standard Model processes. This thoroughness is vital for setting reliable upper limits on rare decays.
- Constraining New Physics: The stringent limits set on LFV tau decays provide valuable constraints for theoretical models proposing new particles or interactions that could mediate such processes.
Key Results
The analysis considered six possible LFV decay modes: $\tau^- \rightarrow e^- e^+ e^-$, $\tau^- \rightarrow \mu^- e^+ e^-$, $\tau^- \rightarrow e^- \mu^+ \mu^-$, $\tau^- \rightarrow \mu^- \mu^+ \mu^-$, $\tau^- \rightarrow e^- e^+ \mu^-$, and $\tau^- \rightarrow \mu^- e^+ \mu^-$.
In all six decay modes, the number of observed events in the signal region was found to be compatible with the expected background. Consequently, the study sets new upper limits on the branching fractions for these LFV decays. At the 90% confidence level, these limits range from $1.8 \times 10^{-8}$ to $3.3 \times 10^{-8}$. These results supersede previous BABAR analyses and are consistent with the latest findings from the Belle experiment.
Limitations and Future Work
The authors acknowledge that the limits are primarily determined by the statistical uncertainties in the background estimation and the signal efficiency. Systematic uncertainties, particularly those related to particle identification, are the dominant contributors to the error on the signal efficiency.
The paper does not explicitly detail future work, but the established methodology and the pursuit of higher luminosity datasets at future colliders would naturally lead to even more stringent limits on these rare LFV processes. The continued search for LFV decays remains a crucial avenue for exploring physics beyond the Standard Model.
Summary updated