University of Groningen Search for decays of neutral beauty mesons into four muons LHCb Collaboration Published in: Journal of High Energy Physics DOI: 10.1007/JHEP03(2017)001 IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2017 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): LHCb Collaboration (2017). Search for decays of neutral beauty mesons into four muons. Journal of High Energy Physics, 2017(3), [001]. https://doi.org/10.1007/JHEP03(2017)001 Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date: 06-04-2021 https://doi.org/10.1007/JHEP03(2017)001 https://research.rug.nl/en/publications/search-for-decays-of-neutral-beauty-mesons-into-four-muons(411758db-eef4-4647-9c7f-a1d6099ccbee).html https://doi.org/10.1007/JHEP03(2017)001 J H E P 0 3 ( 2 0 1 7 ) 0 0 1 Published for SISSA by Springer Received: November 24, 2016 Revised: February 2, 2017 Accepted: February 18, 2017 Published: March 1, 2017 Search for decays of neutral beauty mesons into four muons The LHCb collaboration E-mail: tobias.tekampe@cern.ch Abstract: A search for the non-resonant decays B0s → µ+µ−µ+µ− and B0 → µ+µ−µ+µ− is presented. The measurement is performed using the full Run 1 data set collected in proton-proton collisions by the LHCb experiment at the LHC. The data correspond to integrated luminosities of 1 and 2 fb−1 collected at centre-of-mass energies of 7 and 8 TeV, respectively. No signal is observed and upper limits on the branching fractions of the non-resonant decays at 95% confidence level are determined to be B(B0s → µ+µ−µ+µ−) < 2.5 × 10−9, B(B0 → µ+µ−µ+µ−) < 6.9 × 10−10. Keywords: B physics, Flavour Changing Neutral Currents, Hadron-Hadron scattering (experiments), Rare decay, Supersymmetry ArXiv ePrint: 1611.07704 Open Access, Copyright CERN, for the benefit of the LHCb Collaboration. Article funded by SCOAP3. doi:10.1007/JHEP03(2017)001 mailto:tobias.tekampe@cern.ch https://arxiv.org/abs/1611.07704 http://dx.doi.org/10.1007/JHEP03(2017)001 J H E P 0 3 ( 2 0 1 7 ) 0 0 1 Contents 1 Introduction 1 2 Detector and simulation 3 3 Event selection 3 4 Selection efficiencies and systematic uncertainties 4 5 Normalisation 6 6 Results 8 7 Conclusion 9 The LHCb collaboration 14 1 Introduction The rare decays B0 (s) → µ+µ−µ+µ− proceed through b→ d(s) flavour-changing neutral- current processes, which are strongly suppressed in the Standard Model (SM).1 In the main non-resonant SM amplitude, one muon pair is produced via amplitudes described by electroweak loop diagrams and the other is created by a virtual photon as shown in figure 1(a). The branching fraction of the non-resonant B0s → µ+µ−µ+µ− decay is expected to be 3.5 × 10−11 [1]. Theories extending the SM can significantly enhance the B0 (s) → µ+µ−µ+µ− decay rate by contributions of new particles. For example, in minimal supersymmetric models (MSSM), the decay can proceed via new scalar S and pseudoscalar P sgoldstino particles, which both decay into a dimuon final state as shown in figure 1(b). There are two types of couplings between sgoldstinos and SM fermions. Type-I couplings describe interactions between a sgoldstino and two fermions, where the coupling strength is proportional to the fermion mass. Type-II couplings describe a four-particle vertex, where a scalar and a pseudoscalar sgoldstino interact with two fermions. Branching fractions up to B(B0s → SP) ≈ 10−4 and B(B0→ SP) ≈ 10−7 are possible [2]. Sgoldstinos can decay into a pair of photons or a pair of charged leptons [3]. In this analysis the muonic decay is considered, as the coupling to electrons is smaller and the large τ-lepton mass limits the available phase space. The branching fractions of the sgoldstino decays strongly depend on the model parameters such as the sgoldstino mass and the supersymmetry breaking scale. In the search for Σ+→ pµ+µ− decays the HyperCP collaboration found an excess of events, 1The inclusion of charge-conjugate processes is implied throughout. – 1 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 W t γ W γ, Z0 d, s b̄ µ− µ+ µ− µ+ (a) S P s, d b̄ µ− µ+ µ− µ+ (b) s̄ W c̄ c s b̄ µ− µ+ µ− µ+ φ J/ψ (c) Figure 1. Feynman diagrams for (a) the non-resonant B0 (s) → µ+µ−µ+µ− decay, (b) a supersym- metric B0 (s) → S(→ µ+µ−)P(→ µ+µ−) decay and (c) the resonant B0s → J/ψ (→ µ+µ−)φ(→ µ+µ−) decay (see text). which is consistent with the decay Σ+→ Pp with P → µ+µ− and a pseudoscalar mass of m(P) = 214.3 ± 0.5 MeV [4]. So far only limits on the SM and MSSM branching fractions at 95% confidence level have been measured by LHCb based on the data recorded in 2011 [5] to be B(B0s → µ+µ−µ+µ−) < 1.6 × 10−8, B(B0→ µ+µ−µ+µ−) < 6.6 × 10−9, B(B0s → S(→ µ+µ−)P(→ µ+µ−)) < 1.6 × 10−8, B(B0→ S(→ µ+µ−)P(→ µ+µ−)) < 6.3 × 10−9. These limits are based on assumed sgoldstino masses of m(S) = 2.5 GeV/c2, which is ap- proximately the central value of the allowed mass range, and m(P) = 214.3 MeV/c2. The dominant SM decays of neutral B mesons into four muons proceed through res- onances like φ, J/ψ and ψ(2S). The most frequent of these decays is B0s → J/ψφ, where both the J/ψ and the φ mesons decay into a pair of muons, as shown in figure 1(c). In the following, this decay is referred to as the resonant decay mode and treated as a back- ground. From the product of the measured branching fractions of the underlying decays B(B0s → J/ψφ), B(J/ψ → µ+µ−), and B(φ→ µ+µ−) [6] its branching fraction is calculated to be (1.83 ± 0.18) × 10−8. In this paper a search for the non-resonant SM process, and for the MSSM-induced B0 (s) → µ+µ−µ+µ− decays is presented, using pp collision data recorded by the LHCb detec- tor during LHC Run 1. Potentially contributing sgoldstinos are assumed to be short lived, such that they do not form a displaced vertex. The analysed data correspond to integrated luminosities of 1 and 2 fb−1 collected at centre-of-mass energies of 7 and 8 TeV, respec- tively. The branching fraction is measured relative to the decay B+→ J/ψ (→ µ+µ−)K+, which gives a clean signal with a precisely measured branching fraction [6]. This yields a significant improvement compared to the previous measurement, where the use of the B0 → J/ψK∗0 decay as normalisation mode resulted in a large systematic uncertainty originating from the S-wave fraction and the less precisely measured branching fraction. The advantage of normalising to a well-known B meson decay is that dominant systematic uncertainties originating mainly from the bb cross-section cancel in the ratio. – 2 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 2 Detector and simulation The LHCb detector [7, 8] is a single-arm forward spectrometer covering the pseudorapidity range 2 < η < 5, designed for the study of particles containing b or c quarks. The detector includes a high-precision tracking system consisting of a silicon-strip vertex detector sur- rounding the pp interaction region, a large-area silicon-strip detector located upstream of a dipole magnet with a bending power of about 4 Tm, and three stations of silicon-strip detectors and straw drift tubes placed downstream of the magnet. The tracking system provides a measurement of momentum, p, of charged particles with a relative uncertainty that varies from 0.5% at low momentum to 1.0% at 200 GeV/c. The minimum distance of a track to a primary vertex (PV), the impact parameter, is measured with a resolution of (15 + 29/pT) µm, where pT is the component of the momentum transverse to the beam, in GeV/c. Different types of charged hadrons are distinguished using information from two ring-imaging Cherenkov (RICH) detectors. Photons, electrons and hadrons are identified by a calorimeter system consisting of scintillating-pad and preshower detectors, an elec- tromagnetic calorimeter and a hadronic calorimeter. Muons are identified by a system composed of alternating layers of iron and multiwire proportional chambers. In the simulation, pp collisions are generated using Pythia [9, 10] with a specific LHCb configuration [11]. Decays of hadronic particles are described by EvtGen [12], in which final-state radiation is generated using Photos [13]. The interaction of the generated particles with the detector, and its response, are implemented using the Geant4 toolkit [14, 15] as described in ref. [16]. 3 Event selection The online event selection is performed by a trigger, which consists of a hardware stage, based on information from the calorimeter and muon systems, followed by a software stage, which applies a full event reconstruction. In this analysis candidate events are first required to pass the hardware trigger, which for 7 TeV (8 TeV) data selects events with at least one muon with a transverse momentum of pT > 1.48 GeV/c (pT > 1.76 GeV/c) or at least one pair of muons with the product of the transverse momenta larger than (1.296)2 GeV2/c2 ((1.6)2 GeV2/c2). In the subsequent software trigger, at least one of the final-state particles is required to have pT > 1 GeV/c and an impact parameter larger than 100 µm with respect to all PVs in the event. In the offline selection, the B0 (s) decay vertex is constructed from four good quality muon candidates that form a common vertex and have a total charge of zero. The vertex is required to be significantly displaced from any PV. Among the four final-state muons, there are four possible dimuon combinations with zero charge. In all four combinations, the mass windows corresponding to the φ (950–1090 MeV/c2), J/ψ (3000–3200 MeV/c2) and ψ(2S) (3600–3800 MeV/c2) resonances are vetoed. This efficiently suppresses any background from any of the three mentioned resonances to a negligible level. Contributions of other charmonium states are found to be negligible. – 3 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 The MatrixNet (MN) [17], a multivariate classifier based on a Boosted Decision Tree [18, 19], is applied in order to remove combinatorial background, where a candi- date B0 (s) vertex is constructed from four muons that do not originate from a single B meson decay. The input variables are the following properties of the B0 (s) candidate: the decay time, the vertex quality, the momentum and transverse momentum, the cosine of the direction angle (DIRA), and the smallest impact parameter chisquare (χ2IP) with respect to any PV, where χ2IP is defined as the difference between the vertex-fit χ 2 of a PV recon- structed with and without the B0 (s) candidate. The DIRA is defined as the angle between the momentum of the reconstructed B0 (s) candidate and the vector from the PV with the smallest χ2IP to the B 0 (s) decay vertex. As training samples, simulated B0s → µ+µ−µ+µ− and B0→ µ+µ−µ+µ− events, generated with a uniform probability across the decay phase space, are used as a signal proxy. Before training, the signal simulation is weighted to correct for known discrepancies between data and simulation as described later. The back- ground sample is taken from the far and the near sidebands in data as defined in table 1. In order to verify that the classification of each event is unbiased, 10-fold cross-validation [20] is employed. Background arising from misidentifying one or more particles is suppressed by apply- ing particle identification (PID) requirements. Information from the RICH system, the calorimeters and the muon system is used to calculate the difference in log-likelihood be- tween the hypothesis of a final-state particle being a pion or a muon, DLLµπ. Events in the signal region are not examined until the analysis is finalised. Events outside the signal region are split into the far sidebands, used to calculate the expected background yield, and the near sidebands, used to optimise the cuts on the MN response and the minimum DLLµπ values of the four muon candidates in the final state. The optimization of the cuts is performed on a two-dimensional grid maximising the figure of merit [21] FoM = εsignal σ/2 + √ N expected bkg ×εbkg . The intended significance in terms of standard deviations (σ) is set to three. Very similar selection criteria are found when using five. The expected background yield before applying the MN and PID selection, N expected bkg , is determined from a fit to the events in the near sidebands using an exponential function. For each grid point the background efficiency, εbkg, is measured using events from the near sidebands. The signal efficiency, εsignal, is measured for each grid point using simulated B0 (s) → µ+µ−µ+µ− decays. Lacking a model for non-resonant B0 (s) → µ+µ−µ+µ− simulation, the selection of the preceding measurement was developed on B0s → J/ψ (→ µ+µ−)φ(→ µ+µ−) data. Now that a suitable simulation model is available, significant improvements in terms of signal efficiency and background rejection are made by employing a multivariate classifier and being able to measure the selection efficiency from simulation. 4 Selection efficiencies and systematic uncertainties The optimal working point corresponds to signal efficiencies of (0.580 ± 0.003)% and – 4 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 Mass interval ( MeV/c2) Near sidebands [5020, 5220] and [5426, 5626] Far sidebands [4360, 5020] and [5626, 6360] Signal region [m(B0) − 60,m(B0s ) + 60] B0s search region [m(B 0 s ) − 40,m(B0s ) + 40] B0 search region [m(B0) − 40,m(B0) + 40] Table 1. Definitions of intervals in the B0 and B0s reconstructed invariant mass distributions. (0.568 ± 0.003)% for the B0s → µ+µ−µ+µ− and B0→ µ+µ−µ+µ− decay modes, respec- tively. Sources of peaking background such as B0→ K∗0µ+µ−, in which the kaon and the pion originating from the K∗ decay are misidentified as muons, are reduced to a negligible level by the optimised selection. The efficiencies for the MSSM processes are measured using simulated samples of the B0 (s) → S(→ µ+µ−)P(→ µ+µ−) decays, where the B0 (s) me- son decays into a pseudoscalar sgoldstino with a mass of 214.3 MeV/c2 [4] and a scalar sgoldstino with a mass of 2.5 GeV/c2. Both the P and S particles are simulated with a decay width of Γ = 0.1 MeV/c2, which corresponds to a prompt decay. The measured efficiencies are the same for the B0s and the B 0 decays and amount to (0.648 ± 0.003)%. The difference between the SM and the MSSM efficiencies originates from the fact that in the case of the decay proceeding via P and S sgoldstinos, the decay products are more likely to be within the acceptance of the LHCb detector. In order to test the dependence of the measured B0 (s) → S(→ µ+µ−)P(→ µ+µ−) branching fractions on the mass of the scalar sgoldstino, the selection efficiency is measured in bins of dimuon invariant mass while requiring the corresponding other dimuon mass to be between 200 and 950 MeV/c2. An efficiency variation of O(20%) is observed. The selection applied to the normalisation mode B+→ J/ψ (→ µ+µ−)K+ differs from that applied to the signal modes in the PID criteria and that no multivariate analysis technique is applied. The total efficiency is (1.495 ± 0.006)%. The uncertainties on the ef- ficiencies are driven by the limited number of simulated events and are treated as systematic uncertainties of 0.4–0.5%. The total efficiency is calculated as the product of the efficiencies of the different stages of the selection. As an alternative to the trigger efficiency calculated on simulation, the value is calculated on B+ → J/ψ (→ µ+µ−)K+ data [22] and a systematic uncertainty of 3% is assigned corresponding to the relative difference. The efficiency of the MN classifier to select the more frequent decay B0s → J/ψ (→ µ+µ−)φ(→ K+K−) is compared between data and simulation. The relative difference of 0.3% is assigned as a systematic uncertainty. Another source of systematic uncertainty arises from the track finding efficiency. Again, values obtained from data [23] and simulation are compared and the deviation is treated as a correction factor for the efficiency, while the uncertainty on the deviation, 1.7%, is assigned as a systematic uncertainty. In general the agreement in the observables used in the selection between data and simulation is very good, although there are some distributions that are known to deviate. – 5 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 Therefore, the gradient boosting reweighting technique [24] is used to calculate weights that correct for differences between data and simulation in B0s → J/ψ (→ µ+µ−)φ(→ K+K−). The weighting is performed in the track multiplicity, the B transverse momentum, the χ2 of the decay vertex fit and the χ2IP. The first two are chosen because they are correlated with the PID variables and the latter two dominate the feature ranking obtained from the MN training. These weights are applied to the B0 (s) → µ+µ−µ+µ− and B+→ J/ψ (→ µ+µ−)K+ simulation samples, and are used to calculate the MN and the PID efficiencies. In order to account for inaccuracies of this method resulting from the kinematic and topological differences between the decay modes, systematic uncertainties of 3.6% are assigned based on the difference of the MN efficiency on B0 (s) → µ+µ−µ+µ− and B0s → J/ψ (→ µ+µ−)φ(→ K+K−). For the B+ → J/ψ (→ µ+µ−)K+ decay mode, the efficiencies are measured with and without weights and the observed difference of 2.3% is assigned as systematic uncertainty. In order to determine accurate efficiencies of the applied PID requirements, calibra- tion samples of muons from J/ψ → µ+µ− and φ → µ+µ− decays and of kaons from D∗+ → D0(→ K−π+)π+ decays are used. The relative frequency for kaons and muons to pass the PID criteria is calculated in bins of track multiplicity, particle momentum and pseudorapidity. Different binning schemes are tested and the observed differences in the efficiencies of 1% for B+→ J/ψ (→ µ+µ−)K+ and 0.5% for B0 (s) → µ+µ−µ+µ− are assigned as systematic uncertainties. Additionally, 3% of the simulated B0 (s) → µ+µ−µ+µ− decays contain muons with low transverse momentum outside the kinematic region covered by the calibration data. This fraction is assigned as a systematic uncertainty. Candidates that have a reconstructed invariant mass within ±40 MeV/c2 around the known B0 (s) mass, which corresponds to ±2σ of the mass resolution, are treated as signal candidates. The ac- curacy of the efficiency of this cut is evaluated on B0s → J/ψ (→ µ+µ−)φ(→ K+K−) data. A systematic uncertainty of 0.5% corresponding to the relative difference of the efficiency measured on data and simulation is assigned. Systematic uncertainties of 0.9% and 0.5% in the case of B0 (s) → µ+µ−µ+µ− and B+ → J/ψ (→ µ+µ−)K+ originate from the imper- fections of the efficiency of the event reconstruction due to soft photon radiation and 0.6% from mismatching of track segments between different tracking stations in the detector, which is measured using simulated events. All relevant sources of systematic uncertainty along with the total values are summarised in table 2. The most significant improvements with regard to the preceding measurement are the larger available data sample, and the choice of the B+→ J/ψ (→ µ+µ−)K+ decay as normalisation mode, which has the advan- tage of a precisely measured branching fraction and the absence of an additional systematic uncertainty originating from the S-wave correction. 5 Normalisation The B+ → J/ψ (→ µ+µ−)K+ signal yield is determined by performing an unbinned ex- tended maximum likelihood fit to the K+µ+µ− invariant mass distribution. In this fit the J/ψ mass is constrained [25] to the world average [6]. The normalisation yield is found to be N(B+ → J/ψ (→ µ+µ−)K+) = 687890 ± 920. The J/ψK+ mass spectrum along – 6 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 )2c) (MeV/ + Kψ(J/m 5200 5400 5600 ) 2 c E v e n ts / ( 3 .5 M e V / 2 10 3 10 4 10 5 10 LHCb + Kψ J/→ + B Combinatorial X + Kψ J/→B Figure 2. Fit to the B+→ J/ψ (→ µ+µ−)K+ invariant mass distribution. The signal contribution is modelled by a Hypatia2 [26] function (blue dotted line), the combinatorial background by an exponential function (green dash-dotted line). Partially reconstructed decays, such as B0→ J/ψK∗0 where one pion is not reconstructed, are modelled by a Gaussian function with an exponential tail towards the lower mass side (red dashed line). Data are shown by black dots. with the fit result is shown in figure 2. A systematic uncertainty of 0.3% is assigned to the determined B+→ J/ψ (→ µ+µ−)K+ yield by using an alternative fit model and performing a binned extended maximum likelihood fit. The B0 (s) → µ+µ−µ+µ− branching fraction is calculated as B(B0(s)→ µ +µ−µ+µ−) = N(B0(s)→ µ +µ−µ+µ−) ×ηd,s, with ηd,s = ε(B+→ J/ψ (→ µ+µ−)K+) ×B(B+→ J/ψ (→ µ+µ−)K+) ε(B0 (s) → µ+µ−µ+µ−) ×N(B+→ J/ψ (→ µ+µ−)K+) × fu fd,s , where N(B+→ J/ψ (→ µ+µ−)K+) and N(B0 (s) → µ+µ−µ+µ−) are the observed yields of the normalisation and the signal channel, respectively. The ratio between the production rates of B0s and B 0 was measured by LHCb to be fs/fd = 0.259±0.015 [27]. The measure- ment was performed using pp collision data at √ s = 7 TeV, but found to be stable between√ s = 7 TeV and 8 TeV by a previous LHCb measurement [28]. The ratio between the B+ and B0 production rates is assumed to be unity. As a consequence fs/fu is equal to fs/fd. The single event sensitivities, ηd,s, amount ηSMs = (8.65 ± 0.80) × 10−10, ηSMd = (2.29 ± 0.16) × 10−10, ηMSSMs = (7.75 ± 0.72) × 10−10, ηMSSMd = (2.01 ± 0.14) × 10−10, – 7 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 for the B0s and the B 0 decay modes in the SM and in the MSSM scenario. Here, the uncertainties are the combined values of the statistical uncertainty on the B+ → J/ψ (→ µ+µ−)K+ yield and the systematic uncertainty. In the case of ηs the systematic uncertainty is dominated by the ratio of fu/fs and in the case of ηd by the weighting procedure applied to correct for the difference between data and simulation. The individual sources of systematic uncertainties given in table 2 are assumed to be uncorrelated and are combined quadratically. The total systematic uncertainty is 9.2% for the B0s decay and 7.2% for the B 0 decay. These values are small compared to the statistical uncertainty on the expected number of background events in the B0 and B0s search regions. The whole analysis strategy is cross-checked by measuring the B0s → J/ψ (→ µ+µ−)φ(→ µ+µ−) branching fraction. The obtained value has a precision of 20% and is compatible with the product of the branching fractions of the underlying decays taken from ref. [6]. The number of expected background events is determined by fitting an exponential function to the far sidebands of m(µ+µ−µ+µ−). Extrapolating and integrating the fitted function in ±40 MeV/c2 wide windows around the B0 (s) meson masses yields the number of expected background events, N expected bkg (B 0) = 0.55+0.24−0.19 (stat) ± 0.20 (syst) and N expected bkg (B 0 s ) = 0.47 +0.23 −0.18 (stat) ± 0.18 (syst). The statistical uncertainty is the combination of the Poissonian uncertainty originating from the limited size of the data sample and the uncertainty on the fit parameters. As an alternative fit model a second-order polynomial is used and the difference between these background expectations is assigned as a systematic uncertainty. 6 Results The final distribution of the reconstructed mass of the four muon system is shown in figure 3. No candidates are observed in either the B0 or the B0s search region, which is consistent with the expected background yield. The Hybrid CLs procedure [29–31], with log-normal priors to account for uncertainties of both background and efficiency estimations, is used to convert the observations into upper limits on the corresponding branching fractions. The exclusion at 95% confidence level assuming the SM single event sensitivities is shown in figure 4. The result for the corresponding MSSM values is presented in figure 5. The limits on the branching fractions of the B0s and B 0 decays are anti-correlated. Replacing the log-normal priors by gamma distributions yields the same results. Assuming negligible cross-feed between the B0s and the B 0 search regions, the observed upper limits on the branching fractions at 95% confidence level are found to be B(B0s → µ+µ−µ+µ−) < 2.5 × 10−9, B(B0→ µ+µ−µ+µ−) < 6.9 × 10−10, B(B0s → S(→ µ+µ−)P(→ µ+µ−)) < 2.2 × 10−9, B(B0→ S(→ µ+µ−)P(→ µ+µ−)) < 6.0 × 10−10. – 8 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 Source Value [%] Selection efficiency 0.4 − 0.5 Trigger efficiency 3.0 MN efficiency 0.3 Track finding efficiency 1.7 Weighting B0 (s) → µ+µ−µ+µ− 3.6 Weighting B+→ J/ψ (→ µ+µ−)K+ 2.3 PID binning B+→ J/ψ (→ µ+µ−)K+ 1.0 PID binning B0 (s) → µ+µ−µ+µ− 0.5 Kinematic coverage of PID calibration data 3.0 ±40 MeV/c2 search region efficiency 0.5 Soft photon radiation B0 (s) → µ+µ−µ+µ− 0.9 Soft photon radiation B+→ J/ψ (→ µ+µ−)K+ 0.5 Track segments mismatching 0.6 Normalisation fit 0.3 fu/fs 5.8 B(B+→ J/ψK+) 3.0 B(J/ψ → µ+µ−) 0.1 Combined ηs SM 9.2 Combined ηd SM 7.2 Combined ηs MSSM 9.2 Combined ηd MSSM 7.2 Table 2. Summary of systematic uncertainties affecting the single event sensitivities along with the total systematic uncertainty calculated by adding up the individual components in quadrature. The dominating uncertainty arising from fu/fs only contributes to ηs. The uncertainty of the stated selection efficiencies arising from the limited number of simulated events is 0.5% for B0 → µ+µ−µ+µ− and 0.4% for all other considered decay modes. 7 Conclusion In summary, a search for non-resonant B0 (s) → µ+µ−µ+µ− decays has been presented. In addition, the sensitivity to a specific MSSM scenario has been probed. The applied selection focuses on finding four muon tracks that form a common vertex. For the SM scenario and the MSSM decay through short-lived scalar and pseudoscalar new particles, the limits set by the previous measurement performed by LHCb on a subset of the present data, are improved by a factor of 6.4 (7.3) for the SM (MSSM) mode in the case of the B0s decay and by a factor of 9.5 (10.5) in the case of the B0 decay. – 9 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 Figure 3. Mass distribution of selected B0 (s) → µ+µ−µ+µ− events observed in 3 fb−1 of data in all considered B mass regions. Background (red line) is modelled by an exponential function. Signal subregions for B0 and B0s searches are also shown. The error bars on the individual points with n entries are ± √ n. ] 9− 10×) [ − µ+ µ− µ+ µ→ 0 s (BΒ 0 2 4 6 ] 9 − 1 0 × ) [ − µ + µ − µ + µ → 0 (B Β 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 LHCb 95% CL Exclusion observed σ 1 ±expected σ 2 ±expected Figure 4. Expected and observed 95% CL exclusion in B(B0→ µ+µ−µ+µ−) vs. B(B0s → µ+µ−µ+µ−) parameters plane. Acknowledgments We express our gratitude to our colleagues in the CERN accelerator departments for the excellent performance of the LHC. We thank the technical and administrative staff at the LHCb institutes. We acknowledge support from CERN and from the national – 10 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 ] 9− 10×) [ − µ+ µ→(PΒ ×) − µ+ µ →(SΒ × S P) → 0 s (BΒ 0 2 4 6 ] 9 − 1 0 × ) [ − µ + µ → (P Β × ) − µ + µ → (S Β × S P ) → 0 (B Β 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 LHCb S P→ 0 s S P; B→ 0 B − µ+ µ →; P − µ+ µ →S = 214 MeV P = 2.5 GeV; mSm 95% CL Exclusion observed σ 1 ±expected σ 2 ±expected Figure 5. Expected and observed 95% CL exclusion in B(B0 → S(→ µ+µ−)P(→ µ+µ−)) vs. B(B0s → S(→ µ+µ−)P(→ µ+µ−)) parameters plane with scalar and pseudoscalar S and P as described in section 3. agencies: CAPES, CNPq, FAPERJ and FINEP (Brazil); NSFC (China); CNRS/IN2P3 (France); BMBF, DFG and MPG (Germany); INFN (Italy); FOM and NWO (The Nether- lands); MNiSW and NCN (Poland); MEN/IFA (Romania); MinES and FASO (Russia); MinECo (Spain); SNSF and SER (Switzerland); NASU (Ukraine); STFC (United King- dom); NSF (U.S.A.). We acknowledge the computing resources that are provided by CERN, IN2P3 (France), KIT and DESY (Germany), INFN (Italy), SURF (The Netherlands), PIC (Spain), GridPP (United Kingdom), RRCKI and Yandex LLC (Russia), CSCS (Switzer- land), IFIN-HH (Romania), CBPF (Brazil), PL-GRID (Poland) and OSC (U.S.A.). We are indebted to the communities behind the multiple open source software packages on which we depend. Individual groups or members have received support from AvH Foundation (Ger- many), EPLANET, Marie Sk lodowska-Curie Actions and ERC (European Union), Conseil Général de Haute-Savoie, Labex ENIGMASS and OCEVU, Région Auvergne (France), RFBR and Yandex LLC (Russia), GVA, XuntaGal and GENCAT (Spain), Herchel Smith Fund, The Royal Society, Royal Commission for the Exhibition of 1851 and the Leverhulme Trust (United Kingdom). Open Access. This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits any use, distribution and reproduction in any medium, provided the original author(s) and source are credited. References [1] Y. Dincer and L.M. Sehgal, Electroweak effects in the double Dalitz decay Bs → `+`−`′+`′−, Phys. Lett. B 556 (2003) 169 [hep-ph/0301056] [INSPIRE]. – 11 – http://creativecommons.org/licenses/by/4.0/ http://dx.doi.org/10.1016/S0370-2693(03)00131-X https://arxiv.org/abs/hep-ph/0301056 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0301056 J H E P 0 3 ( 2 0 1 7 ) 0 0 1 [2] S.V. Demidov and D.S. Gorbunov, Flavor violating processes with sgoldstino pair production, Phys. Rev. D 85 (2012) 077701 [arXiv:1112.5230] [INSPIRE]. [3] S.V. Demidov and D.S. Gorbunov, More about sgoldstino interpretation of HyperCP events, JETP Lett. 84 (2007) 479 [hep-ph/0610066] [INSPIRE]. [4] HyperCP collaboration, H. Park et al., Evidence for the decay Σ+ → pµ+µ−, Phys. Rev. Lett. 94 (2005) 021801 [hep-ex/0501014] [INSPIRE]. [5] LHCb collaboration, Search for rare B0 (s) → µ+µ−µ+µ− decays, Phys. Rev. Lett. 110 (2013) 211801 [arXiv:1303.1092] [INSPIRE]. [6] Particle Data Group collaboration, C. Patrignani et al., Review of particle physics, Chin. Phys. C 40 (2016) 100001 [INSPIRE]. [7] LHCb collaboration, The LHCb detector at the LHC, 2008 JINST 3 S08005 [INSPIRE]. [8] LHCb collaboration, LHCb detector performance, Int. J. Mod. Phys. A 30 (2015) 1530022 [arXiv:1412.6352] [INSPIRE]. [9] T. Sjöstrand, S. Mrenna and P.Z. Skands, PYTHIA 6.4 physics and manual, JHEP 05 (2006) 026 [hep-ph/0603175] [INSPIRE]. [10] T. Sjöstrand, S. Mrenna and P.Z. Skands, A brief introduction to PYTHIA 8.1, Comput. Phys. Commun. 178 (2008) 852 [arXiv:0710.3820] [INSPIRE]. [11] LHCb collaboration, Handling of the generation of primary events in Gauss, the LHCb simulation framework, J. Phys. Conf. Ser. 331 (2011) 032047 [INSPIRE]. [12] D.J. Lange, The EvtGen particle decay simulation package, Nucl. Instrum. Meth. A 462 (2001) 152 [INSPIRE]. [13] P. Golonka and Z. Was, PHOTOS Monte Carlo: a precision tool for QED corrections in Z and W decays, Eur. Phys. J. C 45 (2006) 97 [hep-ph/0506026] [INSPIRE]. [14] Geant4 collaboration, J. Allison et al., GEANT4 developments and applications, IEEE Trans. Nucl. Sci. 53 (2006) 270. [15] GEANT4 collaboration, S. Agostinelli et al., GEANT4: a simulation toolkit, Nucl. Instrum. Meth. A 506 (2003) 250 [INSPIRE]. [16] LHCb collaboration, The LHCb simulation application, Gauss: design, evolution and experience, J. Phys. Conf. Ser. 331 (2011) 032023 [INSPIRE]. [17] A. Gulin, I. Kuralenok and D. Pavlov, Winning the transfer learning track of Yahoo!’s learning to rank challenge with YetiRank, JMLR Proc. 14 (2011) 63. [18] L. Breiman, J. H. Friedman, R. A. Olshen and C.J. Stone, Classification and regression trees, Wadsworth international group, Belmont U.S.A. (1984). [19] R.E. Schapire and Y. Freund, A decision-theoretic generalization of on-line learning and an application to boosting, J. Comput. Syst. Sci. 55 (1997) 119. [20] A. Blum, A. Kalai and J. Langford, Beating the hold-out: bounds for k-fold and progressive cross-validation, in the proceedings of the Twelfth Annual Conference on Computational Learning Theory (COLT’99), July 7–9, New York, U.S.A. (1999). [21] G. Punzi, Sensitivity of searches for new signals and its optimization, eConf C 030908 (2003) MODT002 [physics/0308063] [INSPIRE]. – 12 – http://dx.doi.org/10.1103/PhysRevD.85.077701 https://arxiv.org/abs/1112.5230 http://inspirehep.net/search?p=find+EPRINT+arXiv:1112.5230 http://dx.doi.org/10.1134/S0021364006210028 https://arxiv.org/abs/hep-ph/0610066 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0610066 http://dx.doi.org/10.1103/PhysRevLett.94.021801 http://dx.doi.org/10.1103/PhysRevLett.94.021801 https://arxiv.org/abs/hep-ex/0501014 http://inspirehep.net/search?p=find+EPRINT+hep-ex/0501014 http://dx.doi.org/10.1103/PhysRevLett.110.211801 http://dx.doi.org/10.1103/PhysRevLett.110.211801 https://arxiv.org/abs/1303.1092 http://inspirehep.net/search?p=find+EPRINT+arXiv:1303.1092 http://dx.doi.org/10.1088/1674-1137/40/10/100001 http://dx.doi.org/10.1088/1674-1137/40/10/100001 http://inspirehep.net/search?p=find+J+%22Chin.Phys.,C40,100001%22 http://dx.doi.org/10.1088/1748-0221/3/08/S08005 http://inspirehep.net/search?p=find+J+%22JINST,3,S08005%22 http://dx.doi.org/10.1142/S0217751X15300227 https://arxiv.org/abs/1412.6352 http://inspirehep.net/search?p=find+EPRINT+arXiv:1412.6352 http://dx.doi.org/10.1088/1126-6708/2006/05/026 http://dx.doi.org/10.1088/1126-6708/2006/05/026 https://arxiv.org/abs/hep-ph/0603175 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0603175 http://dx.doi.org/10.1016/j.cpc.2008.01.036 http://dx.doi.org/10.1016/j.cpc.2008.01.036 https://arxiv.org/abs/0710.3820 http://inspirehep.net/search?p=find+EPRINT+arXiv:0710.3820 http://dx.doi.org/10.1088/1742-6596/331/3/032047 http://inspirehep.net/search?p=find+J+%22J.Phys.Conf.Ser.,331,032047%22 http://dx.doi.org/10.1016/S0168-9002(01)00089-4 http://dx.doi.org/10.1016/S0168-9002(01)00089-4 http://inspirehep.net/search?p=find+J+%22Nucl.Instrum.Meth.,A462,152%22 http://dx.doi.org/10.1140/epjc/s2005-02396-4 https://arxiv.org/abs/hep-ph/0506026 http://inspirehep.net/search?p=find+EPRINT+hep-ph/0506026 http://dx.doi.org/10.1109/TNS.2006.869826 http://dx.doi.org/10.1109/TNS.2006.869826 http://dx.doi.org/10.1016/S0168-9002(03)01368-8 http://dx.doi.org/10.1016/S0168-9002(03)01368-8 http://inspirehep.net/search?p=find+J+%22Nucl.Instrum.Meth.,A506,250%22 http://dx.doi.org/10.1088/1742-6596/331/3/032023 http://inspirehep.net/search?p=find+J+%22J.Phys.Conf.Ser.,331,032023%22 http://www.jmlr.org/proceedings/papers/v14/gulin11a/gulin11a.pdf http://dx.doi.org/10.1006/jcss.1997.1504 https://arxiv.org/abs/physics/0308063 http://inspirehep.net/search?p=find+EPRINT+physics/0308063 J H E P 0 3 ( 2 0 1 7 ) 0 0 1 [22] S. Tolk, J. Albrecht, F. Dettori and A. Pellegrino, Data driven trigger efficiency determination at LHCb, LHCb-PUB-2014-039 (2014). [23] LHCb collaboration, Measurement of the track reconstruction efficiency at LHCb, 2015 JINST 10 P02007 [arXiv:1408.1251] [INSPIRE]. [24] A. Rogozhnikov, Reweighting with boosted decision trees, J. Phys. Conf. Ser. 762 (2016) 012036 [arXiv:1608.05806] [INSPIRE]. [25] W.D. Hulsbergen, Decay chain fitting with a Kalman filter, Nucl. Instrum. Meth. A 552 (2005) 566 [physics/0503191] [INSPIRE]. [26] D. Mart́ınez Santos and F. Dupertuis, Mass distributions marginalized over per-event errors, Nucl. Instrum. Meth. A 764 (2014) 150 [arXiv:1312.5000] [INSPIRE]. [27] LHCb collaboration, Updated average fs/fd b-hadron production fraction ratio for 7 TeV pp collisions, LHCb-CONF-2013-011 (2013). [28] LHCb collaboration, Measurement of the B0s → µ+µ− branching fraction and search for B0 → µ+µ− decays at the LHCb experiment, Phys. Rev. Lett. 111 (2013) 101805 [arXiv:1307.5024] [INSPIRE]. [29] R.D. Cousins and V.L. Highland, Incorporating systematic uncertainties into an upper limit, Nucl. Instrum. Meth. A 320 (1992) 331 [INSPIRE]. [30] A.L. Read, Presentation of search results: the CL(s) technique, J. Phys. G 28 (2002) 2693. [31] T. Junk, Confidence level computation for combining searches with small statistics, Nucl. Instrum. Meth. A 434 (1999) 435 [hep-ex/9902006] [INSPIRE]. – 13 – http://cds.cern.ch/record/1701134 http://dx.doi.org/10.1088/1748-0221/10/02/P02007 http://dx.doi.org/10.1088/1748-0221/10/02/P02007 https://arxiv.org/abs/1408.1251 http://inspirehep.net/search?p=find+EPRINT+arXiv:1408.1251 http://dx.doi.org/10.1088/1742-6596/762/1/012036 http://dx.doi.org/10.1088/1742-6596/762/1/012036 https://arxiv.org/abs/1608.05806 http://inspirehep.net/search?p=find+EPRINT+arXiv:1608.05806 http://dx.doi.org/10.1016/j.nima.2005.06.078 http://dx.doi.org/10.1016/j.nima.2005.06.078 https://arxiv.org/abs/physics/0503191 http://inspirehep.net/search?p=find+EPRINT+physics/0503191 http://dx.doi.org/10.1016/j.nima.2014.06.081 https://arxiv.org/abs/1312.5000 http://inspirehep.net/search?p=find+EPRINT+arXiv:1312.5000 http://cds.cern.ch/record/1559262 http://dx.doi.org/10.1103/PhysRevLett.111.101805 https://arxiv.org/abs/1307.5024 http://inspirehep.net/search?p=find+EPRINT+arXiv:1307.5024 http://dx.doi.org/10.1016/0168-9002(92)90794-5 http://inspirehep.net/search?p=find+J+%22Nucl.Instrum.Meth.,A320,331%22 http://dx.doi.org/10.1088/0954-3899/28/10/313 http://dx.doi.org/10.1016/S0168-9002(99)00498-2 http://dx.doi.org/10.1016/S0168-9002(99)00498-2 https://arxiv.org/abs/hep-ex/9902006 http://inspirehep.net/search?p=find+EPRINT+hep-ex/9902006 J H E P 0 3 ( 2 0 1 7 ) 0 0 1 The LHCb collaboration R. Aaij40, B. Adeva39, M. Adinolfi48, Z. Ajaltouni5, S. Akar6, J. Albrecht10, F. Alessio40, M. Alexander53, S. Ali43, G. Alkhazov31, P. Alvarez Cartelle55, A.A. Alves Jr59, S. Amato2, S. Amerio23, Y. Amhis7, L. An41, L. Anderlini18, G. Andreassi41, M. Andreotti17,g, J.E. Andrews60, R.B. Appleby56, F. Archilli43, P. d’Argent12, J. Arnau Romeu6, A. Artamonov37, M. Artuso61, E. Aslanides6, G. Auriemma26, M. Baalouch5, I. Babuschkin56, S. Bachmann12, J.J. Back50, A. Badalov38, C. Baesso62, S. Baker55, W. Baldini17, A. Baranov35, R.J. Barlow56, C. Barschel40, S. Barsuk7, W. Barter40, M. Baszczyk27, V. Batozskaya29, B. Batsukh61, V. Battista41, A. Bay41, L. Beaucourt4, J. Beddow53, F. Bedeschi24, I. Bediaga1, L.J. Bel43, V. Bellee41, N. Belloli21,i, K. Belous37, I. Belyaev32, E. Ben-Haim8, G. Bencivenni19, S. Benson43, J. Benton48, A. Berezhnoy33, R. Bernet42, A. Bertolin23, C. Betancourt42, F. Betti15, M.-O. Bettler40, M. van Beuzekom43, Ia. Bezshyiko42, S. Bifani47, P. Billoir8, T. Bird56, A. Birnkraut10, A. Bitadze56, A. Bizzeti18,u, T. Blake50, F. Blanc41, J. Blouw11,†, S. Blusk61, V. Bocci26, T. Boettcher58, A. Bondar36,w, N. Bondar31,40, W. Bonivento16, I. Bordyuzhin32, A. Borgheresi21,i, S. Borghi56, M. Borisyak35, M. Borsato39, F. Bossu7, M. Boubdir9, T.J.V. Bowcock54, E. Bowen42, C. Bozzi17,40, S. Braun12, M. Britsch12, T. Britton61, J. Brodzicka56, E. Buchanan48, C. Burr56, A. Bursche2, J. Buytaert40, S. Cadeddu16, R. Calabrese17,g, M. Calvi21,i, M. Calvo Gomez38,m, A. Camboni38, P. Campana19, D.H. Campora Perez40, L. Capriotti56, A. Carbone15,e, G. Carboni25,j, R. Cardinale20,h, A. Cardini16, P. Carniti21,i, L. Carson52, K. Carvalho Akiba2, G. Casse54, L. Cassina21,i, L. Castillo Garcia41, M. Cattaneo40, Ch. Cauet10, G. Cavallero20, R. Cenci24,t, D. Chamont7, M. Charles8, Ph. Charpentier40, G. Chatzikonstantinidis47, M. Chefdeville4, S. Chen56, S.-F. Cheung57, V. Chobanova39, M. Chrzaszcz42,27, X. Cid Vidal39, G. Ciezarek43, P.E.L. Clarke52, M. Clemencic40, H.V. Cliff49, J. Closier40, V. Coco59, J. Cogan6, E. Cogneras5, V. Cogoni16,40,f , L. Cojocariu30, G. Collazuol23,o, P. Collins40, A. Comerma-Montells12, A. Contu40, A. Cook48, G. Coombs40, S. Coquereau38, G. Corti40, M. Corvo17,g, C.M. Costa Sobral50, B. Couturier40, G.A. Cowan52, D.C. Craik52, A. Crocombe50, M. Cruz Torres62, S. Cunliffe55, R. Currie55, C. D’Ambrosio40, F. Da Cunha Marinho2, E. Dall’Occo43, J. Dalseno48, P.N.Y. David43, A. Davis59, O. De Aguiar Francisco2, K. De Bruyn6, S. De Capua56, M. De Cian12, J.M. De Miranda1, L. De Paula2, M. De Serio14,d, P. De Simone19, C.-T. Dean53, D. Decamp4, M. Deckenhoff10, L. Del Buono8, M. Demmer10, A. Dendek28, D. Derkach35, O. Deschamps5, F. Dettori40, B. Dey22, A. Di Canto40, H. Dijkstra40, F. Dordei40, M. Dorigo41, A. Dosil Suárez39, A. Dovbnya45, K. Dreimanis54, L. Dufour43, G. Dujany56, K. Dungs40, P. Durante40, R. Dzhelyadin37, A. Dziurda40, A. Dzyuba31, N. Déléage4, S. Easo51, M. Ebert52, U. Egede55, V. Egorychev32, S. Eidelman36,w, S. Eisenhardt52, U. Eitschberger10, R. Ekelhof10, L. Eklund53, S. Ely61, S. Esen12, H.M. Evans49, T. Evans57, A. Falabella15, N. Farley47, S. Farry54, R. Fay54, D. Fazzini21,i, D. Ferguson52, A. Fernandez Prieto39, F. Ferrari15,40, F. Ferreira Rodrigues2, M. Ferro-Luzzi40, S. Filippov34, R.A. Fini14, M. Fiore17,g, M. Fiorini17,g, M. Firlej28, C. Fitzpatrick41, T. Fiutowski28, F. Fleuret7,b, K. Fohl40, M. Fontana16,40, F. Fontanelli20,h, D.C. Forshaw61, R. Forty40, V. Franco Lima54, M. Frank40, C. Frei40, J. Fu22,q, E. Furfaro25,j, C. Färber40, A. Gallas Torreira39, D. Galli15,e, S. Gallorini23, S. Gambetta52, M. Gandelman2, P. Gandini57, Y. Gao3, L.M. Garcia Martin69, J. Garćıa Pardiñas39, J. Garra Tico49, L. Garrido38, P.J. Garsed49, D. Gascon38, C. Gaspar40, L. Gavardi10, G. Gazzoni5, D. Gerick12, E. Gersabeck12, M. Gersabeck56, T. Gershon50, Ph. Ghez4, S. Giaǹı41, V. Gibson49, O.G. Girard41, L. Giubega30, K. Gizdov52, V.V. Gligorov8, D. Golubkov32, A. Golutvin55,40, A. Gomes1,a, I.V. Gorelov33, C. Gotti21,i, M. Grabalosa Gándara5, R. Graciani Diaz38, L.A. Granado Cardoso40, E. Graugés38, – 14 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 E. Graverini42, G. Graziani18, A. Grecu30, P. Griffith47, L. Grillo21,40,i, B.R. Gruberg Cazon57, O. Grünberg67, E. Gushchin34, Yu. Guz37, T. Gys40, C. Göbel62, T. Hadavizadeh57, C. Hadjivasiliou5, G. Haefeli41, C. Haen40, S.C. Haines49, S. Hall55, B. Hamilton60, X. Han12, S. Hansmann-Menzemer12, N. Harnew57, S.T. Harnew48, J. Harrison56, M. Hatch40, J. He63, T. Head41, A. Heister9, K. Hennessy54, P. Henrard5, L. Henry8, J.A. Hernando Morata39, E. van Herwijnen40, M. Heß67, A. Hicheur2, D. Hill57, C. Hombach56, H. Hopchev41, W. Hulsbergen43, T. Humair55, M. Hushchyn35, N. Hussain57, D. Hutchcroft54, M. Idzik28, P. Ilten58, R. Jacobsson40, A. Jaeger12, J. Jalocha57, E. Jans43, A. Jawahery60, F. Jiang3, M. John57, D. Johnson40, C.R. Jones49, C. Joram40, B. Jost40, N. Jurik61, S. Kandybei45, W. Kanso6, M. Karacson40, J.M. Kariuki48, S. Karodia53, M. Kecke12, M. Kelsey61, I.R. Kenyon47, M. Kenzie49, T. Ketel44, E. Khairullin35, B. Khanji12, C. Khurewathanakul41, T. Kirn9, S. Klaver56, K. Klimaszewski29, S. Koliiev46, M. Kolpin12, I. Komarov41, R.F. Koopman44, P. Koppenburg43, A. Kosmyntseva32, A. Kozachuk33, M. Kozeiha5, L. Kravchuk34, K. Kreplin12, M. Kreps50, P. Krokovny36,w, F. Kruse10, W. Krzemien29, W. Kucewicz27,l, M. Kucharczyk27, V. Kudryavtsev36,w, A.K. Kuonen41, K. Kurek29, T. Kvaratskheliya32,40, D. Lacarrere40, G. Lafferty56, A. Lai16, G. Lanfranchi19, C. Langenbruch9, T. Latham50, C. Lazzeroni47, R. Le Gac6, J. van Leerdam43, J.-P. Lees4, A. Leflat33,40, J. Lefrançois7, R. Lefèvre5, F. Lemaitre40, E. Lemos Cid39, O. Leroy6, T. Lesiak27, B. Leverington12, Y. Li7, T. Likhomanenko35,68, R. Lindner40, C. Linn40, F. Lionetto42, B. Liu16, X. Liu3, D. Loh50, I. Longstaff53, J.H. Lopes2, D. Lucchesi23,o, M. Lucio Martinez39, H. Luo52, A. Lupato23, E. Luppi17,g, O. Lupton57, A. Lusiani24, X. Lyu63, F. Machefert7, F. Maciuc30, O. Maev31, K. Maguire56, S. Malde57, A. Malinin68, T. Maltsev36, G. Manca7, G. Mancinelli6, P. Manning61, J. Maratas5,v, J.F. Marchand4, U. Marconi15, C. Marin Benito38, P. Marino24,t, J. Marks12, G. Martellotti26, M. Martin6, M. Martinelli41, D. Martinez Santos39, F. Martinez Vidal69, D. Martins Tostes2, L.M. Massacrier7, A. Massafferri1, R. Matev40, A. Mathad50, Z. Mathe40, C. Matteuzzi21, A. Mauri42, B. Maurin41, A. Mazurov47, M. McCann55, J. McCarthy47, A. McNab56, R. McNulty13, B. Meadows59, F. Meier10, M. Meissner12, D. Melnychuk29, M. Merk43, A. Merli22,q, E. Michielin23, D.A. Milanes66, M.-N. Minard4, D.S. Mitzel12, A. Mogini8, J. Molina Rodriguez1, I.A. Monroy66, S. Monteil5, M. Morandin23, P. Morawski28, A. Mordà6, M.J. Morello24,t, J. Moron28, A.B. Morris52, R. Mountain61, F. Muheim52, M. Mulder43, M. Mussini15, D. Müller56, J. Müller10, K. Müller42, V. Müller10, P. Naik48, T. Nakada41, R. Nandakumar51, A. Nandi57, I. Nasteva2, M. Needham52, N. Neri22, S. Neubert12, N. Neufeld40, M. Neuner12, A.D. Nguyen41, T.D. Nguyen41, C. Nguyen-Mau41,n, S. Nieswand9, R. Niet10, N. Nikitin33, T. Nikodem12, A. Novoselov37, D.P. O’Hanlon50, A. Oblakowska-Mucha28, V. Obraztsov37, S. Ogilvy19, R. Oldeman49, C.J.G. Onderwater70, J.M. Otalora Goicochea2, A. Otto40, P. Owen42, A. Oyanguren69,40, P.R. Pais41, A. Palano14,d, F. Palombo22,q, M. Palutan19, J. Panman40, A. Papanestis51, M. Pappagallo14,d, L.L. Pappalardo17,g, W. Parker60, C. Parkes56, G. Passaleva18, A. Pastore14,d, G.D. Patel54, M. Patel55, C. Patrignani15,e, A. Pearce56,51, A. Pellegrino43, G. Penso26, M. Pepe Altarelli40, S. Perazzini40, P. Perret5, L. Pescatore47, K. Petridis48, A. Petrolini20,h, A. Petrov68, M. Petruzzo22,q, E. Picatoste Olloqui38, B. Pietrzyk4, M. Pikies27, D. Pinci26, A. Pistone20, A. Piucci12, S. Playfer52, M. Plo Casasus39, T. Poikela40, F. Polci8, A. Poluektov50,36, I. Polyakov61, E. Polycarpo2, G.J. Pomery48, A. Popov37, D. Popov11,40, B. Popovici30, S. Poslavskii37, C. Potterat2, E. Price48, J.D. Price54, J. Prisciandaro39, A. Pritchard54, C. Prouve48, V. Pugatch46, A. Puig Navarro41, G. Punzi24,p, W. Qian57, R. Quagliani7,48, B. Rachwal27, J.H. Rademacker48, M. Rama24, M. Ramos Pernas39, M.S. Rangel2, I. Raniuk45, F. Ratnikov35, G. Raven44, F. Redi55, S. Reichert10, A.C. dos Reis1, C. Remon Alepuz69, V. Renaudin7, S. Ricciardi51, S. Richards48, M. Rihl40, K. Rinnert54, – 15 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 V. Rives Molina38, P. Robbe7,40, A.B. Rodrigues1, E. Rodrigues59, J.A. Rodriguez Lopez66, P. Rodriguez Perez56,†, A. Rogozhnikov35, S. Roiser40, A. Rollings57, V. Romanovskiy37, A. Romero Vidal39, J.W. Ronayne13, M. Rotondo19, M.S. Rudolph61, T. Ruf40, P. Ruiz Valls69, J.J. Saborido Silva39, E. Sadykhov32, N. Sagidova31, B. Saitta16,f , V. Salustino Guimaraes2, C. Sanchez Mayordomo69, B. Sanmartin Sedes39, R. Santacesaria26, C. Santamarina Rios39, M. Santimaria19, E. Santovetti25,j, A. Sarti19,k, C. Satriano26,s, A. Satta25, D.M. Saunders48, D. Savrina32,33, S. Schael9, M. Schellenberg10, M. Schiller40, H. Schindler40, M. Schlupp10, M. Schmelling11, T. Schmelzer10, B. Schmidt40, O. Schneider41, A. Schopper40, K. Schubert10, M. Schubiger41, M.-H. Schune7, R. Schwemmer40, B. Sciascia19, A. Sciubba26,k, A. Semennikov32, A. Sergi47, N. Serra42, J. Serrano6, L. Sestini23, P. Seyfert21, M. Shapkin37, I. Shapoval45, Y. Shcheglov31, T. Shears54, L. Shekhtman36,w, V. Shevchenko68, B.G. Siddi17,40, R. Silva Coutinho42, L. Silva de Oliveira2, G. Simi23,o, S. Simone14,d, M. Sirendi49, N. Skidmore48, T. Skwarnicki61, E. Smith55, I.T. Smith52, J. Smith49, M. Smith55, H. Snoek43, M.D. Sokoloff59, F.J.P. Soler53, B. Souza De Paula2, B. Spaan10, P. Spradlin53, S. Sridharan40, F. Stagni40, M. Stahl12, S. Stahl40, P. Stefko41, S. Stefkova55, O. Steinkamp42, S. Stemmle12, O. Stenyakin37, S. Stevenson57, S. Stoica30, S. Stone61, B. Storaci42, S. Stracka24,p, M. Straticiuc30, U. Straumann42, L. Sun64, W. Sutcliffe55, K. Swientek28, V. Syropoulos44, M. Szczekowski29, T. Szumlak28, S. T’Jampens4, A. Tayduganov6, T. Tekampe10, M. Teklishyn7, G. Tellarini17,g, F. Teubert40, E. Thomas40, J. van Tilburg43, M.J. Tilley55, V. Tisserand4, M. Tobin41, S. Tolk49, L. Tomassetti17,g, D. Tonelli40, S. Topp-Joergensen57, F. Toriello61, E. Tournefier4, S. Tourneur41, K. Trabelsi41, M. Traill53, M.T. Tran41, M. Tresch42, A. Trisovic40, A. Tsaregorodtsev6, P. Tsopelas43, A. Tully49, N. Tuning43, A. Ukleja29, A. Ustyuzhanin35,x, U. Uwer12, C. Vacca16,f , V. Vagnoni15,40, A. Valassi40, S. Valat40, G. Valenti15, A. Vallier7, R. Vazquez Gomez19, P. Vazquez Regueiro39, S. Vecchi17, M. van Veghel43, J.J. Velthuis48, M. Veltri18,r, G. Veneziano57, A. Venkateswaran61, M. Vernet5, M. Vesterinen12, B. Viaud7, D. Vieira1, M. Vieites Diaz39, H. Viemann67, X. Vilasis-Cardona38,m, M. Vitti49, V. Volkov33, A. Vollhardt42, B. Voneki40, A. Vorobyev31, V. Vorobyev36,w, C. Voß67, J.A. de Vries43, C. Vázquez Sierra39, R. Waldi67, C. Wallace50, R. Wallace13, J. Walsh24, J. Wang61, D.R. Ward49, H.M. Wark54, N.K. Watson47, D. Websdale55, A. Weiden42, M. Whitehead40, J. Wicht50, G. Wilkinson57,40, M. Wilkinson61, M. Williams40, M.P. Williams47, M. Williams58, T. Williams47, F.F. Wilson51, J. Wimberley60, J. Wishahi10, W. Wislicki29, M. Witek27, G. Wormser7, S.A. Wotton49, K. Wraight53, K. Wyllie40, Y. Xie65, Z. Xing61, Z. Xu41, Z. Yang3, Y. Yao61, H. Yin65, J. Yu65, X. Yuan36,w, O. Yushchenko37, K.A. Zarebski47, M. Zavertyaev11,c, L. Zhang3, Y. Zhang7, Y. Zhang63, A. Zhelezov12, Y. Zheng63, A. Zhokhov32, X. Zhu3, V. Zhukov9, S. Zucchelli15 1 Centro Brasileiro de Pesquisas F́ısicas (CBPF), Rio de Janeiro, Brazil 2 Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, Brazil 3 Center for High Energy Physics, Tsinghua University, Beijing, China 4 LAPP, Université Savoie Mont-Blanc, CNRS/IN2P3, Annecy-Le-Vieux, France 5 Clermont Université, Université Blaise Pascal, CNRS/IN2P3, LPC, Clermont-Ferrand, France 6 CPPM, Aix-Marseille Université, CNRS/IN2P3, Marseille, France 7 LAL, Université Paris-Sud, CNRS/IN2P3, Orsay, France 8 LPNHE, Université Pierre et Marie Curie, Université Paris Diderot, CNRS/IN2P3, Paris, France 9 I. Physikalisches Institut, RWTH Aachen University, Aachen, Germany 10 Fakultät Physik, Technische Universität Dortmund, Dortmund, Germany 11 Max-Planck-Institut für Kernphysik (MPIK), Heidelberg, Germany 12 Physikalisches Institut, Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany 13 School of Physics, University College Dublin, Dublin, Ireland 14 Sezione INFN di Bari, Bari, Italy – 16 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 15 Sezione INFN di Bologna, Bologna, Italy 16 Sezione INFN di Cagliari, Cagliari, Italy 17 Sezione INFN di Ferrara, Ferrara, Italy 18 Sezione INFN di Firenze, Firenze, Italy 19 Laboratori Nazionali dell’INFN di Frascati, Frascati, Italy 20 Sezione INFN di Genova, Genova, Italy 21 Sezione INFN di Milano Bicocca, Milano, Italy 22 Sezione INFN di Milano, Milano, Italy 23 Sezione INFN di Padova, Padova, Italy 24 Sezione INFN di Pisa, Pisa, Italy 25 Sezione INFN di Roma Tor Vergata, Roma, Italy 26 Sezione INFN di Roma La Sapienza, Roma, Italy 27 Henryk Niewodniczanski Institute of Nuclear Physics Polish Academy of Sciences, Kraków, Poland 28 AGH — University of Science and Technology, Faculty of Physics and Applied Computer Science, Kraków, Poland 29 National Center for Nuclear Research (NCBJ), Warsaw, Poland 30 Horia Hulubei National Institute of Physics and Nuclear Engineering, Bucharest-Magurele, Romania 31 Petersburg Nuclear Physics Institute (PNPI), Gatchina, Russia 32 Institute of Theoretical and Experimental Physics (ITEP), Moscow, Russia 33 Institute of Nuclear Physics, Moscow State University (SINP MSU), Moscow, Russia 34 Institute for Nuclear Research of the Russian Academy of Sciences (INR RAN), Moscow, Russia 35 Yandex School of Data Analysis, Moscow, Russia 36 Budker Institute of Nuclear Physics (SB RAS), Novosibirsk, Russia 37 Institute for High Energy Physics (IHEP), Protvino, Russia 38 ICCUB, Universitat de Barcelona, Barcelona, Spain 39 Universidad de Santiago de Compostela, Santiago de Compostela, Spain 40 European Organization for Nuclear Research (CERN), Geneva, Switzerland 41 Insitute of Physics, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland 42 Physik-Institut, Universität Zürich, Zürich, Switzerland 43 Nikhef National Institute for Subatomic Physics, Amsterdam, The Netherlands 44 Nikhef National Institute for Subatomic Physics and VU University Amsterdam, Amsterdam, The Netherlands 45 NSC Kharkiv Institute of Physics and Technology (NSC KIPT), Kharkiv, Ukraine 46 Institute for Nuclear Research of the National Academy of Sciences (KINR), Kyiv, Ukraine 47 University of Birmingham, Birmingham, United Kingdom 48 H.H. Wills Physics Laboratory, University of Bristol, Bristol, United Kingdom 49 Cavendish Laboratory, University of Cambridge, Cambridge, United Kingdom 50 Department of Physics, University of Warwick, Coventry, United Kingdom 51 STFC Rutherford Appleton Laboratory, Didcot, United Kingdom 52 School of Physics and Astronomy, University of Edinburgh, Edinburgh, United Kingdom 53 School of Physics and Astronomy, University of Glasgow, Glasgow, United Kingdom 54 Oliver Lodge Laboratory, University of Liverpool, Liverpool, United Kingdom 55 Imperial College London, London, United Kingdom 56 School of Physics and Astronomy, University of Manchester, Manchester, United Kingdom 57 Department of Physics, University of Oxford, Oxford, United Kingdom 58 Massachusetts Institute of Technology, Cambridge, MA, United States 59 University of Cincinnati, Cincinnati, OH, United States 60 University of Maryland, College Park, MD, United States 61 Syracuse University, Syracuse, NY, United States 62 Pontif́ıcia Universidade Católica do Rio de Janeiro (PUC-Rio), Rio de Janeiro, Brazil, associated to2 – 17 – J H E P 0 3 ( 2 0 1 7 ) 0 0 1 63 University of Chinese Academy of Sciences, Beijing, China, associated to3 64 School of Physics and Technology, Wuhan University, Wuhan, China, associated to3 65 Institute of Particle Physics, Central China Normal University, Wuhan, Hubei, China, associated to3 66 Departamento de Fisica , Universidad Nacional de Colombia, Bogota, Colombia, associated to8 67 Institut für Physik, Universität Rostock, Rostock, Germany, associated to12 68 National Research Centre Kurchatov Institute, Moscow, Russia, associated to32 69 Instituto de Fisica Corpuscular (IFIC), Universitat de Valencia-CSIC, Valencia, Spain, associated to38 70 Van Swinderen Institute, University of Groningen, Groningen, The Netherlands, associated to43 a Universidade Federal do Triângulo Mineiro (UFTM), Uberaba-MG, Brazil b Laboratoire Leprince-Ringuet, Palaiseau, France c P.N. Lebedev Physical Institute, Russian Academy of Science (LPI RAS), Moscow, Russia d Università di Bari, Bari, Italy e Università di Bologna, Bologna, Italy f Università di Cagliari, Cagliari, Italy g Università di Ferrara, Ferrara, Italy h Università di Genova, Genova, Italy i Università di Milano Bicocca, Milano, Italy j Università di Roma Tor Vergata, Roma, Italy k Università di Roma La Sapienza, Roma, Italy l AGH — University of Science and Technology, Faculty of Computer Science, Electronics and Telecommunications, Kraków, Poland m LIFAELS, La Salle, Universitat Ramon Llull, Barcelona, Spain n Hanoi University of Science, Hanoi, Viet Nam o Università di Padova, Padova, Italy p Università di Pisa, Pisa, Italy q Università degli Studi di Milano, Milano, Italy r Università di Urbino, Urbino, Italy s Università della Basilicata, Potenza, Italy t Scuola Normale Superiore, Pisa, Italy u Università di Modena e Reggio Emilia, Modena, Italy v Iligan Institute of Technology (IIT), Iligan, Philippines w Novosibirsk State University, Novosibirsk, Russia x Moscow Institute of Physics and Technology, Moscow, Russia †Deceased – 18 – Introduction Detector and simulation Event selection Selection efficiencies and systematic uncertainties Normalisation Results Conclusion The LHCb collaboration