Recent Belle II Results on Hadronic $\boldsymbol{B}$ Decays

Hadronic $B$ decays provide precise constraints on the parameters of the Cabibbo-Kobayashi-Maskawa (CKM) quark-mixing matrix and are sensitive probes of physics beyond the standard model (SM). Measurement of new decay channels expands our knowledge of the flavour sector. Belle II [1], located at the SuperKEKB [2] asymmetric collider at KEK, is a hermetic solenoidal magnetic spectrometer surrounded by particle-identification detectors, an electromagnetic calorimeter, and muon detectors. Optimised for the reconstruction of bottom-antibottom pairs produced at the threshold in $\Upsilon(4S)$ decays, Belle II has competitive, unique, or world-leading reach in many key quantities associated with hadronic $B$ decays. The CKM angle $\phi_{3}/\gamma=\arg\left(-\frac{V_{ud}V^{*}_{ub}}{V_{cd}V_{cb}^{*}}\right)$ is a fundamental constraint for charge-parity (CP) violation in the SM and can be reliably determined in tree-level processes, with negligible loop-amplitude contributions. The angle $\phi_{2}/\alpha=\arg\left(-\frac{V_{td}V^{*}_{tb}}{V_{ud}V_{ub}^{*}}\right)$ is currently limiting the precision of non-SM tests based on global fits of the quark-mixing matrix unitarity and is best determined through measurement of $B\rightarrow\rho\rho$ and $B\rightarrow\pi\pi$ decays. For $B\rightarrow K\pi$ decays, an isospin sum rule [3] combines the branching fractions and CP asymmetries of the decays, providing a null test of the SM. The capability of investigating all the related final states jointly, under consistent experimental settings, is unique to Belle II.

The challenge in analysing these channels lies in the large amount of $e^{+}e^{-}\rightarrow q\bar{q}$ continuum background. Binary-decision-tree classifiers $C$ are used to discriminate between signal and continuum events using event topology, kinematic, and decay-length information. [4] The determination of the signal yields is mainly based on observables that exploit the specific properties of at-threshold production: the energy difference between the $B$ candidate and the beam energy, $\Delta E=E_{B}^{*}-E_{\text{beam}}^{*}$, and the beam-constrained mass $M_{\text{bc}}=\sqrt{E_{\text{beam}}^{*}/c^{2}-(\boldsymbol{p_{B}^{*}}/c)^{2}}$, where $E_{B}^{*}$ and $\boldsymbol{p_{B}^{*}}$ are the energy and momentum of the $B$ candidate, respectively, and $E_{\text{beam}}^{*}$ is the beam energy, all in the center-of-mass frame.

The composition of a large fraction of the $B$ hadronic width is unknown, which limits significantly our capability to model $B$ decays in simulation, impacting the precision of many measurements. Belle II is pursuing a systematic program of exploration of hadronic $B$ decays. We report a preliminary measurement of the branching fraction of $B^{-}\rightarrow D^{0}K^{-}K_{S}^{0}$ decay [5], with a precision that is three times better than the previous best results [6], and first observation of three new decay channels ($\bar{B}^{0}\rightarrow D^{+}K^{-}K_{S}^{0}$, $B^{-}\rightarrow D^{*0}K^{-}K_{S}^{0}$, $\bar{B}^{0}\rightarrow D^{*+}K^{-}K_{S}^{0}$) [5]. The branching fractions

are extracted from a $362\text{ fb}^{-1}$ Belle II sample using likelihood fits to the unbinned distributions of the energy difference $\Delta E$, where the first uncertainties are statistical and the second are systematic. The invariant mass $m(KK_{S}^{0})$ of the two kaons is investigated. For all four channels, the $m(KK_{S}^{0})$ distribution exhibits a peaking structure in the low-mass region, which departs from the expected three-body phase space distribution. Structures are also observed in the Dalitz distributions (Figure 1).

The angle $\phi_{3}$ is studied through the interference of $b\rightarrow c\bar{u}s$ and $b\rightarrow u\bar{c}s$ transition amplitudes in tree-level $B$ hadron decays. The current world average $\phi_{3}=\left(65.9^{+3.3}_{-3.5}\right)^{\circ}$ is dominated by LHCb measurements [7, 8]. $\phi_{3}$ can be determined through different approaches featuring different $D$ final states from $B$ decays into charmed $B$ final states. The most precise Belle II result is obtained with self-conjugate $D$ final states ($D\rightarrow K_{S}^{0}h^{+}h^{-}$, $h=K,\pi$) [9], where several intermediate resonances are involved in $D$ decays, resulting in variation of the CP asymmetry over the phase space. Two other approaches have also been pursued: one with Cabibbo-suppressed $D$ decays, and the other with the $D$ meson decaying to two-body CP eigenstates. In both analyses it is required that $M_{\text{bc}}>5.27\text{ GeV}/c^{2}$, $|\Delta E|<0.15\text{ GeV}$ ($\Delta E>-0.13\text{ GeV}$ for the latter), and a loose requirement on $C$ removes about 60% of the continuum background. The signal yields are extracted with fits of $\Delta E$ and the continuum suppression classifier.

The least precisely known CKM angle $\phi_{2}/\alpha$ is starting to limit the testing power of the CKM model. Belle II has the unique capability of measuring all $B\rightarrow\rho\rho$ and $B\rightarrow\pi\pi$ decays, from which $\phi_{2}$ can be determined. The combined information from these decays exploits isospin symmetry, reducing the effect of hadronic uncertainties. The $B$ candidates are required to have $M_{\text{bc}}>5.27\text{ GeV}/c^{2}$ and $|\Delta E|<0.15\text{ GeV}$ ($<0.30\text{ GeV}$ for final states involving neutral pions), followed by continuum suppression that removes 90-99% of continuum background.

The isospin sum rule $I_{K\pi}$ is defined by

where $\mathcal{B}_{K\pi}$ and $\mathcal{A}_{K\pi}$ are the branching fractions and the CP asymmetries, and $\tau_{B^{0}}/\tau_{B^{+}}=0.9273\pm 0.0033$ [7] is the ratio of $B^{0}$ and $B^{+}$ lifetimes. The SM prediction of the sum rule is zero, with a precision of better than 1%, in the limit of isospin symmetry and no electroweak penguins contributions. Any large deviation from the SM prediction is an indication of non-SM physics. The experimental precision of the sum rule is limited by $\mathcal{A}_{K^{0}\pi^{0}}$ [7].

We studied all the final states associated with the sum rule: $B^{0}\rightarrow K^{+}\pi^{-}$, $B^{+}\rightarrow K_{S}^{0}\pi^{+}$, $B^{+}\rightarrow K^{+}\pi^{0}$, and $B^{0}\rightarrow K_{S}^{0}\pi^{0}$ using $362\text{ fb}^{-1}$ of Belle II data. The analyses of the various decays follow a similar strategy, with common selections applied to the final states particles. $B$ candidates are required to satisfy $5.272<M_{\text{bc}}<5.288\text{ GeV}/c^{2}$, $|\Delta E|<0.3\text{ GeV}$, and a loose requirement of $C$ that suppresses 90-99% of continuum background. A fit is performed on the $\Delta E$-$C^{\prime}$ distribution, where the flavour tagging algorithm [26] is employed to determine the flavour of the $B$ candidate in $B^{0}\rightarrow K_{S}^{0}\pi^{0}$ decay due to absence of primary charged particles. The $\Delta E$ distributions are shown in Figure 2. The measured branching fractions and CP asymmetries, as well as the sum rule calculated using these measurements, are listed in Table 3. They agree with the world averages [7] and have competitive precisions. In particular, the time-integrated and time-dependent results of $B^{0}\rightarrow K_{S}^{0}\pi^{0}$ are combined to achieve the world’s best result for $\mathcal{A}_{K^{0}\pi^{0}}$, and consequentially for $I_{K\pi}$ a competitive precision that is limited by the statistical uncertainty.

In summary, we present five new results from Belle II: measurement of $B\rightarrow D^{(*)}KK_{S}^{0}$ decays, analyses of $\phi_{3}/\gamma$ with the GLS and GLW methods, precise measurements of two-body decays that contribute to the determination of $\phi_{2}/\alpha$, and the $K\pi$ isospin sum rule with a competitive precision to the world’s best result.