Rare B−Decays: an Update from the ATLAS Experiment Ann-Kathrin Perrevoort

1 Rare B−Decays: An Update from the ATLAS Experiment a;∗ 2 Ann-Kathrin Perrevoort, on behalf of the ATLAS Collaboration a 3 Nikhef, 4 Science Park 105, 1098 XG Amsterdam, The Netherlands 5 E-mail: [email protected] Rare flavour-changing neutral current processes are sensitive probes for physics beyond the Stan- dard Model. The ATLAS experiment at the Large Hadron Collider has performed a measurement 0 + − 0 + − of the branching fraction of Bs ! µ µ decays and a search for B ! µ µ decays with data collected in 2015 and 2016, in combination with a previous result by ATLAS on data collected in 6 2011 and 2012. The results are combined with recent results by the CMS and LHCb experiments. The angular decay distributions of B0 ! K∗0 µ+ µ− decays are measured with 20:3 fb−1 of data collected in 2012. For both analyses, the expected improvement in sensitivity and precision is estimated for the High-Luminosity LHC. ATL-PHYS-PROC-2021-010 26 February 2021 BEAUTY2020 21-24 September 2020 Kashiwa, Japan (online) ∗Speaker © Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). https://pos.sissa.it/ Rare B−Decays: An Update from the ATLAS Experiment Ann-Kathrin Perrevoort 7 1. Introduction + − 8 The flavour-changing neutral current (FCNC) processes, b ! sµ µ , are mediated in the 9 Standard Model (SM) via loop or box diagrams and thus occur at low rates. This turns rare B 10 meson decays into sensitive probes of physics beyond the SM, as such processes could enhance the 11 branching fractions and modify angular decay distributions. 12 The ATLAS experiment [1] at the Large Hadron Collider (LHC) measures the branching 0 + − 0 + − 13 fractions of the rare decays B ! µ µ and Bs ! µ µ as well as angular decay distributions 0 ∗0 + − 14 of B ! K µ µ . The former measurement is combined with recent results from the CMS and 15 LHCb experiments. For both measurements, the expected sensitivities for the High-Luminosity 16 LHC upgrade (HL-LHC) [2] are estimated. 0 + − 17 ! 2. Measurement of the branching fractions of B¹sº µ µ 0 18 The ATLAS experiment has performed a measurement of the branching fractions B¹B ! + − 0 + − −1 19 µ µ º and B¹B ! µ µ º using 36:2 fb of pp collision data at a centre-of-mass energy of p s 20 s =13 TeV collected in 2015 and 2016 (early LHC Run 2) [3], and has combined the results with p −1 21 the Run 1 measurement based on 25 fb data at a s of 7 and 8 TeV [4]. 22 The Run 2 data is collected using a two-level trigger system. In the hardware-based first-level µ1 µ2 23 ¹ º trigger system, dimuon events passing transverse momentum thresholds pT ; pT of (4 GeV, 6 GeV) 24 are selected. The software-based high-level trigger requires in addition a dimuon invariant mass mµµ 25 between 4 GeV and 8:5 GeV. Further selections are based on the quality of the reconstructed muons 26 and the reconstructed B¹sº vertex. Events with 4:766 GeV < mµµ < 5:966 GeV are considered in 27 the analysis, of which the range 5:166 GeV < mµµ < 5:526 GeV is defined as signal region. 28 The contribution of continuum combinatorial background originating from muons of uncorre- 29 lated hadron decays is reduced by the means of a boosted decision tree (BDT). The BDT takes 15 30 variables into account, and is trained and tested on data sidebands and simulated signal events. The + − 31 mµµ distributions of partially reconstructed b ! µ µ X decays have a tail reaching into the signal 32 region, and therefore, are taken into account in the fit. A further background component stems from 0 ¹0º ± ± 33 B¹sº ! hh decays with the hadrons h = π ; K being mis-identified as muon candidates. This 34 background peaks in the signal region and is reduced by a tight muon quality selection. 0 + − 35 ! The B¹sº µ µ branching fractions are measured relative to the abundant and well-measured ± + − ± 36 reference channel B ! J/ ¹! µ µ ºK . The yield in the reference channel is determined in an + − ± 37 unbinned maximum likelihood fit to the invariant mass mJ/ K± of the µ µ K system, while the 0 + − 38 ! efficiency relative to the signal decay B¹sº µ µ is taken from MC simulations. 39 The signal yield is extracted in an simultaneous unbinned maximum likelihood fit to the mµµ 40 distribution in four BDT intervals of constant signal efficiency. The fitted signal and background 41 contributions in the highest BDT interval are shown in Fig.1 on the left. Due to the limited mass 0 + − 0 + − 42 resolution, the peaks of B ! µ µ and Bs ! µ µ overlap and are statistically separated in the 0 + − 0 + − 43 fit. Using the Run 2 data set, 80 ± 22 Bs ! µ µ and −12 ± 20 B ! µ µ events are extracted. 0 + − 44 The result is in agreement with the SM predictions. The branching fraction of the Bs ! µ µ 0 + − +0:96 +0:49 −9 45 B¹ ! º ¹ º ¹ º × decay obtained in a Neyman construction is Bs µ µ = 3:21−0:91 stat. −0:30 syst. 10 . 0 + − 0 + − −10 46 For B ! µ µ , an upper limit is set to B¹B ! µ µ º < 4:3 × 10 at 95 % confidence level 2 Rare B−Decays: An Update from the ATLAS Experiment Ann-Kathrin Perrevoort B¹ 0 ! + −º Figure 1: Results from the measurement of B¹sº µ µ [3]. Left: Distributions of mµµ in the unblinded data in the highest BDT interval. Superimposed is the result of the maximum likelihood fit; Right: Likelihood 0 + − 0 + − contours for the simultaneous fit to B¹Bs ! µ µ º and B¹B ! µ µ º, for values of −2∆ ln¹Lº equal to 2.3, 6.2, 11.8. B¹ 0 ! + −º Figure 2: Likelihood contours for the results of B¹sº µ µ for the ATLAS, CMS, and LHCb experiments as well as their combination [6]. Left: Contours correspond to values of −2∆ ln¹Lº =2.3, 6.2, and 11.8; Right: Contours of the combination for values of −2∆ ln¹Lº =2.3, 6.2, 11.8, 19.3, and 30.2. 47 (CL). The likelihood functions from the Run 2 result are combined with the Run 1 results, yielding 0 + − +0:8 −9 0 + − 0 + − 48 B¹ ! º × ! B¹ ! º Bs µ µ = 2:8−0:7 10 , and an upper limit for B µ µ of B µ µ < −10 49 2:1 × 10 at 95 % CL. The likelihood contours are shown in Fig.1 on the right. The results differ 0 + − 0 + − 50 by 2.4 standard deviations σ from the SM [5] in the B¹B ! µ µ º-B¹Bs ! µ µ º plane. 51 2.1 Combination of ATLAS, CMS and LHCb results 0 + − 52 The results on B¹Bs ! µ µ º of the ATLAS experiment are consistent with results obtained 53 by the CMS and LHCb experiments with data collected between 2011 and 2016 [6] (see Fig.2). 0 + − 0 54 The results of the three experiments are combined. A Bs ! µ µ branching fraction of B¹Bs ! + − +0:37 −9 0 + − 55 µ µ º : × B ! µ µ = 2 69−0:35 10 is measured and an upper limit on the branching fraction 0 + − −10 56 is set at B¹Bs ! µ µ º < 1:9 × 10 at 95 % CL. This corresponds to a deviation from the SM 57 predictions by 2.1 σ. 0 ∗0 + − 58 3. Angular analysis of B ! K µ µ 0 ∗0 + − + − 59 An angular analysis of the decay B ! K ¹! K π ºµ µ is performed with the ATLAS p −1 60 experiment using 20:3 fb of pp collision data at s = 8 TeV collected in 2012 [7]. The angular 3 Rare B−Decays: An Update from the ATLAS Experiment Ann-Kathrin Perrevoort 0 0 Figure 3: The measured values of P4 and P5 compared with predictions from the theoretical calculations [9– 11]. Statistical and total uncertainties are shown for the data, i.e. the inner mark indicates the statistical uncertainty and the total error bar the total uncertainty. Figures taken from [7]. 2 61 differential decay rate is described by three angles and the invariant dimuon mass squared, q . 62 This analysis employs the same parametrisation as used by the LHCb experiment in terms of ¹0º 63 the optimised parameters Pi which minimise uncertainites from hadronic form factors [8]. A 64 trigonometric folding of the angular variables is performed due to limited statistics. 65 In the event selection, trigger chains with one, two, or at least three muons are combined 2 66 to ensure a high signal yield and a sensitivity down to low q . Events with two charged tracks 67 compatible with an invariant Kπ mass, mKπ, between 846 MeV and 946 MeV and two muons with 2 2 68 q 2 »0:04; 6:0¼ GeV , that have an invariant mass of the Kπµµ system, mKπµµ, in the range of 2 2 69 5110 to 5700 MeV are selected for analysis. Events with q > 6:0 GeV are excluded to suppress 0 ∗0 0 + 70 ! / the radiative tail from B K J decays. A veto on the D and D¹sº mass ranges is applied 0 0 + 71 ! / to diminish background contributions from B D D¹sº X decays.

Load more