Test of lepton flavor universality with measurements of $R(D^{+})$ and $R(D^{*+})$ using semileptonic $B$ tagging at the Belle II experiment

Abstract

We report measurements of the ratios of branching fractions ${\cal R}(D^{(*)+}) = \frac{{\cal B}(\overline{B}{}^0 \to D^{(*)+} \,τ^- \, \overlineν_τ)}{{\cal B}(\overline{B}{}^0 \to D^{(*)+} \, \ell^- \, \overlineν_\ell)}$, where $\ell$ denotes either an electron or a muon. These ratios test the universality of the charged-current weak interaction. The results are based on a $365\, \mathrm{fb}^{-1}$ data sample collected with the Belle II detector at the SuperKEKB $e^+e^-$ collider, which operates at a center-of-mass energy corresponding to the $Υ(4S)$ resonance, just above the threshold for $B\overline{B}{}$ production. Signal candidates are reconstructed by selecting events in which the companion $B$ meson from the $Υ(4S) \to B\overline{B}{}$ decay is identified in semileptonic modes. The $τ$ lepton is reconstructed via its leptonic decays. We obtain ${\cal R}(D^+) = 0.418 \pm 0.074 ~({\mathrm{stat}}) \pm 0.051 ~({\mathrm{syst}})$ and ${\cal R}(D^{*+}) = 0.306 \pm 0.034 ~({\mathrm{stat}}) \pm 0.018 ~({\mathrm{syst}})$, which are consistent with world average values. Accounting for the correlation between them, these values differ from the Standard Model expectation by a collective significance of $1.7$ standard deviations.

Figures (14)

• Figure 1: The five input variables for the multiclassification BDT are shown for electrons and muons, and the $D^+\ell$ and $D^{*+} \ell$ categories combined.
• Figure 2: The densities of the four fit categories in the $z_\mathrm{diff} = z_{\ell} - z_\mathrm{bkg}$ vs. $z_\tau$ plane and the chosen binning are shown. The four components, shown from top left to bottom right, are semitauonic, semileptonic, $D^{**}$ and gap-mode backgrounds, and other backgrounds. The binned histogram shows the density of data events.
• Figure 3: The fitted classifier distributions for the $D^{+}e^-$, $D^{+}\mu^-$, $D^{*+}e^-$, and $D^{*+}\mu^-$ categories are shown. The hatched regions of the histograms correspond to the systematic uncertainties. The black lines demarcate the first two bins from the remaining bins, which contain significantly fewer events, and are thus displayed with two different $y$-axis scales on the left and right sides.
• Figure 4: Distributions of the signal enriched bins (5,6,10--20) for $E_\mathrm{extra}$ with the results of the fit superimposed.
• Figure 5: Distributions of the signal enriched bins (5,6,10--20) for $p_{\ell}^{*}$ with the results of the fit superimposed.