Evidence for $B^{-}\to {\tau }^{-}{\overline{\nu }}_{\tau }$ with a semileptonic tagging method

We present a measurement of the decay $B^{-}\to {\tau }^{-}{\overline{\nu }}_{\tau }$ using a data sample containing $657\times {10}^6$ $B\overline{B}$ pairs collected at the ${\rm Y}(4S)$ resonance with the Belle detector at the KEKB asymmetric-energy $e^{+}{e}^{-}$ collider. A sample of $B^{+}{B}^{-}$ pairs are tagged by reconstructing one $B^{+}$ meson decaying semileptonically. We detect the $B^{-}\to {\tau }^{-}{\overline{\nu }}_{\tau }$ candidate in the recoil. We obtain a signal with a significance of 3.6 standard deviations including systematic uncertainties, and measure the branching fraction to be $\mathcal{B}(B^{-}\to {\tau }^{-}{\overline{\nu }}_{\tau })=[{1.54}_{-0.37}^{+0.38}(\mathrm{stat}{)}_{-0.31}^{+0.29}(\mathrm{syst})]\times {10}^{-4}$. This result confirms the evidence for $B^{-}\to {\tau }^{-}{\overline{\nu }}_{\tau }$ obtained in a previous Belle measurement that used a hadronic $B$ tagging method.

Figure 1

$\cos {\theta }_{B,D^{(*)}\ell }$ distribution for candidate events with $E_{\mathrm{ECL}}<1.2\mathrm{GeV}$ selected with all $B_{\mathrm{tag}}$ and $B_{\mathrm{sig}}$ requirements except for those on $\cos {\theta }_{B,D^{(*)}\ell }$. Leptonic and hadronic $\tau$ decay modes are combined. The points with error bars are data. The dot-dashed, dotted, and dashed histograms are the MC expectation for events without $B^{+}\to {\overline{D}}^{(*)0}{\ell }^{+}{\nu }_{\ell }$ decays for $B^{+}{B}^{-}$, sum of $B^{+}{B}^{-}$ and $B^0{\overline{B}}^0$, and sum of $B\overline{B}$ and non-$B\overline{B}$ events, respectively.

Figure 2

$E_{\mathrm{ECL}}$ distribution for double semileptonic tagged events. The points with error bars are data and the solid histogram is the MC expectation scaled to the luminosity of the data. The dashed histogram is the MC expectation multiplied by the correction function described in the text. The inset shows the ratio of data to the MC expectation and the correction function.

Figure 3

$E_{\mathrm{ECL}}$ distribution of semileptonic tagged events with the fit result for (a) all $\tau$ decay modes combined, (b) ${\tau }^{-}\to e^{-}{\overline{\nu }}_e{\nu }_{\tau }$, (c) ${\tau }^{-}\to {\mu }^{-}{\overline{\nu }}_{\mu }{\nu }_{\tau }$, and (d) ${\tau }^{-}\to {\pi }^{-}{\nu }_{\tau }$. The points with error bars are data. The hatched histogram and solid open histogram are the background and the signal contributions, respectively.