Improved measurement of branching fractions for $\pi \pi$ transitions among ${\rm Y}(nS)$ states

Using samples of $(5.93\pm 0.10)\times {10}^6$ ${\rm Y}(3S)$ decays and $(9.11\pm 0.14)\times {10}^6$ ${\rm Y}(2S)$ decays collected with the CLEO detector, we report improved measurements of the following branching fractions: $\mathcal{B}({\rm Y}(3S)\to {\rm Y}(1S){\pi }^{+}{\pi }^{-})=(4.46\pm 0.01\pm 0.13)\%$, $\mathcal{B}({\rm Y}(2S)\to {\rm Y}(1S){\pi }^{+}{\pi }^{-})=(18.02\pm 0.02\pm 0.61)\%$, $\mathcal{B}({\rm Y}(3S)\to {\rm Y}(1S){\pi }^0{\pi }^0)=(2.24\pm 0.09\pm 0.11)\%$, $\mathcal{B}({\rm Y}(2S)\to {\rm Y}(1S){\pi }^0{\pi }^0)=(8.43\pm 0.16\pm 0.42)\%$, and $\mathcal{B}({\rm Y}(3S)\to {\rm Y}(2S){\pi }^0{\pi }^0)=(1.82\pm 0.09\pm 0.12)\%$. In each case, the first uncertainty reported is statistical, while the second is systematic.

Figure 1

The spectrum of bottomonium states below $B\overline{B}$ threshold for different $J^{\mathrm{PC}}$ combinations. Established bottomonium states are indicated by the solid horizontal lines, while those indicated by dashed horizontal lines have never been observed. Arrows connecting various bottomonium states represent hadronic transitions that have been observed.

Figure 2

${\pi }^{+}{\pi }^{-}$ recoil mass distributions for the four charged dipion transition analyses, (a) ${\rm Y}(3S)\to {\pi }^{+}{\pi }^{-}{\rm Y}(1S)$ inclusive, (b) ${\rm Y}(2S)\to {\pi }^{+}{\pi }^{-}{\rm Y}(1S)$ inclusive, (c) ${\rm Y}(3S)\to {\pi }^{+}{\pi }^{-}{\rm Y}(1S)$ exclusive, and (d) ${\rm Y}(2S)\to {\pi }^{+}{\pi }^{-}{\rm Y}(1S)$ exclusive. The data are represented by symbols with uncertainty, while the histograms overlaid in each inclusive plot represents background-subtracted data. Note the logarithmic scale for the exclusive analyses.

Figure 3

${\pi }^0{\pi }^0$ recoil mass distributions in the neutral dipion transition analyses, (a) ${\rm Y}(3S)\to {\pi }^0{\pi }^0{\rm Y}(1S)$, (b) ${\rm Y}(3S)\to {\pi }^0{\pi }^0{\rm Y}(2S)$, and (c) ${\rm Y}(2S)\to {\pi }^0{\pi }^0{\rm Y}(1S)$. The data are represented by the symbols with errors, while the histograms represent the result of the fit using the MC shape and a linear background function. Note the logarithmic scale.