Test of lepton family universality and search for lepton and baryon number violation at Belle
TTest of lepton family universality and search for leptonand baryon number violation at Belle
D. Sahoo π,π, β π Tata Institute of Fundamental Research, Mumbai, India π Utkal University, Bhubaneswar, IndiaOn behalf of the Belle Collaboration
E-mail: [email protected]
The electroweak penguin B decays mediated by π β π β + β β transitions are ο¬avour-changingneutral current processes, and are thus sensitive to new physics owing to potential contributionsof heavy particles in the quantum loop. Recently, LHCb has obtained interesting results, wherethe possible hints of lepton family universality violation (LFUV) could be seen. We report a newmeasurement of the LFUV observable π πΎ , the ratio of branching fractions of π΅ β πΎ π + π β to π΅ β πΎπ + π β , based on the full data sample recorded by Belle at the Ξ₯ ( π ) resonance from π + π β collisions produced by the KEKB collider. We also report results on lepton ο¬avor violating π΅ decays, π΅ + β πΎ + π Β± π β and π΅ β πΎ π π Β± π β . The π΅ -factory at KEK is also a π factory, creatinga copious amount of ππ pairs. We have used these data to look for lepton and baryon numberviolating π decays π β β ππ + π β , ππ β π β , ππ + π β , ππ β π + , ππ + π β , and ππ β π β . Results of thesearch are also reported. β Speaker Β© Copyright owned by the author(s) under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). https://pos.sissa.it/ a r X i v : . [ h e p - e x ] J a n est of lepton family universality and search for lepton and baryon number violation at Belle D. Sahoo
1. KEKB and Belle
The Belle detector [1] was placed at the interaction point of the KEKB asymmetric-energy π + π β collider [2]. It was a large-solid-angle magnetic spectrometer comprising six subdetectors:silicon vertex detector, central drift chamber, aerogel Cherenkov counter, time-of-ο¬ight counter,CsI(Tl) crystal electromagnetic calorimeter, and πΎ πΏ and muon detector. The data used in thestudies reported here were recorded by Belle at and near the Ξ₯ ( π ) and Ξ₯ ( π ) resonances.
2. Test of lepton family universality
In the standard model (SM), the electroweak couplings of gauge bosons to leptons are in-dependent of their ο¬avor and the property is known as lepton family universality. The decays π΅ β πΎβ + β β ( β = π, π ) , mediated by the π β π β + β β quark-level transition, constitute ο¬avor-changing neutral current processes. Such processes are forbidden at tree level in the SM but canproceed via suppressed loop-level diagrams, and are therefore sensitive to particles predicted ina number of new physics models [3, 4]. A robust observable to test the SM prediction is thelepton-family-universality (LFU) ratio π πΎ = β« π Ξ ππ [ π΅ β πΎ π + π β ] ππ β« π Ξ ππ [ π΅ β πΎπ + π β ] ππ , where the decay rate Ξ is integrated over the available range of the dilepton invariant-mass squared, π β‘ π ( β + β β ) . LHCb [5] has measured π πΎ with a diο¬erence of about 2.5 standard deviations ( π ) from the SM prediction in the π β ( . , . ) GeV bin. Previous measurement of the samequantity was performed by Belle in the whole π range using a data sample of 657 Γ π΅π΅ events.We present here the results [6] obtained from a multidimensional ο¬t performed on the full Ξ₯ ( π ) data sample of Belle, which replace our previous result [7]. Following four channels are studied: π΅ + β πΎ + π + π β , π΅ + β πΎ + π + π β , π΅ β πΎ π π + π β , and π΅ β πΎ π π + π β based on 711 fb β Ξ₯ ( π ) datacorresponding to 772 Γ π΅π΅ events. The signal yield is extracted with an extended unbinnedmaximum-likelihood ο¬t to the distributions of π bc β‘ βοΈ πΈ β | (cid:174) π π΅ | , Ξ πΈ β‘ πΈ π΅ β πΈ beam , and neuralnetwork output (background suppression variable). Here πΈ beam is the beam energy, and πΈ π΅ and (cid:174) π π΅ are the energy and momentum of the reconstructed π΅ meson candidates, all calculated in the center-of-mass (CM) frame. From the ο¬t we obtain 137 Β±
14 and 138 Β±
15 events in π΅ + β πΎ + π + π β and π΅ + β πΎ + π + π β decays, respectively. Similarly, the yields for the neutral channels π΅ β πΎ π π + π β and π΅ β πΎ π π + π β are 27 . + . β . and 21 . + . β . events. The ο¬t has also been performed in diο¬erent π bins and the π πΎ results are shown in Figure 1. The π πΎ values for diο¬erent π bins are consistentwith the SM predictions, and the value for the whole π range is 1 . + . β . Β± .
02. Our π πΎ + resultfor π β ( . , . ) GeV is higher by 1.6 π compared to the LHCb result.
3. Lepton ο¬avor violation in π΅ decays In many theoretical models, lepton ο¬avor violation (LFV) accompanies LFU violation [8].Extrapolating from the level of neutrino mixing, LFV is only possible at rates far below the currentexperimental sensitivity. In case of a signal, this will unequivocally constitute signature of physics2 est of lepton family universality and search for lepton and baryon number violation at Belle
D. Sahoo
Figure 1: π πΎ in bins of π for π΅ β πΎ + β + β β (top left), π΅ β πΎ π β + β β (top right), and both modes combined(bottom). The red marker represents the bin of 1 < π < , and blue markers are for 0 . < π < < π < .
12, 10 . < π < . π > .
18 GeV bins. The green marker is for the whole π regionexcluding the charmonium resonances. beyond the SM. The LFV in π΅ decays can be studied via π΅ β πΎ π Β± π β . The most stringent upperlimits on π΅ + β πΎ + π + π β and π΅ + β πΎ + π β π + set by LHCb [9] are 6 . Γ β and 7 . Γ β at 90%conο¬dence level (CL). Also, π΅ β πΎ π Β± π β decays were searched for by BaBar [10], setting a 90%CL upper limit on their branching fractions at 2 . Γ β .The signal yields for LFV decays in Belle [6] are obtained by performing extended maximum-likelihood ο¬ts, similar to those for π΅ β πΎβ + β β channels. The signal-enhanced projection plotsobtained from the ο¬t for LFV decays are shown in Figure 2. The ο¬tted yields are 11 . + . β . ,1 . + . β . , and β . + . β . for π΅ + β πΎ + π + π β , π΅ + β πΎ + π β π + , and π΅ β πΎ π π Β± π β , respectively.The 90% CL upper limits on the branching fractions are B ( π΅ + β πΎ + π + π β ) < . Γ β , B ( π΅ + β πΎ + π β π + ) < . Γ β , and B ( π΅ β πΎ π π Β± π β ) < . Γ β . As there is a 3 . π evidencefor signal in π΅ + β πΎ + π + π β , we also quote the branching fraction of ( . + . β . Β± . ) Γ β .The earlier limit on the neutral decay channel is improved by an order of magnitude.
4. Lepton-ο¬avor-, lepton-number- and baryon-number-violating π decays As lepton ο¬avor, lepton number ( L ) and baryon number ( B ) are accidental symmetries of theSM, there is no reason to expect them to be conserved in all possible particle interactions. In fact,lepton ο¬avor violation has already been observed in neutrino oscillations [11]. While B is presumedto have been violated in the early universe, its exact mechanism still remains unknown. It is one ofthe three criteria formulated by Sakharov [12] to explain the matter-antimatter asymmetry observed3 est of lepton family universality and search for lepton and baryon number violation at Belle D. Sahoo
Figure 2:
Signal-enhanced π bc projection of three-dimensional unbinned maximum likelihood ο¬ts to thedata for decays π΅ + β πΎ + π + π β (left), π΅ + β πΎ + π β π + (middle), and π΅ β πΎ π π Β± π β , respectively. Points witherror bars are the data, blue solid curves are the ο¬tted results for the signal-plus-background hypothesis, reddashed curves denote the signal component, while cyan long-dashed, green dash-dotted, and black dashedcurves represent continuum, π΅π΅ background, and charmless π΅ decays, respectively. in the universe. Any observation of processes involving B violation would be a clear signal ofnew physics. Such processes are studied in diο¬erent scenarios of physics beyond the SM, such assupersymmetry [13], grand uniο¬cation [14], and models with black holes [15].Based on 1 fb β of π π collision data, LHCb [16] has studied the last two channels, setting90% CL upper limits on their branching fractions: B ( π β β ππ + π β ) < . Γ β and B ( π β β ππ β π β ) < . Γ β . Using experimental bounds on proton decay, authors in Refs. [17β19] predicta branching fraction in the range of 10 β β10 β for these kinds of decays. We report herein theresults [20] on a search for six L - and B -violating decays: π β β ππ + π β , ππ β π β , ππ + π β , ππ β π + , ππ + π β , and ππ β π β [21] using 921 fb β of data, equivalent to ( Β± ) Γ π + π β events, recordedwith the Belle detector at the KEKB asymmetric-energy π + π β collider.The π lepton is reconstructed by combining a proton or an antiproton with two chargedlepton candidates. To identify the signal, we use two kinematic variables: the reconstructed mass π rec β‘ βοΈ πΈ πββ (cid:48) β (cid:174) π πββ (cid:48) and the energy diο¬erence Ξ πΈ β‘ πΈ CM πββ (cid:48) β πΈ CMbeam , where πΈ πββ (cid:48) and (cid:174) π πββ (cid:48) arethe sum of energies and momenta, respectively, of the π , β and β (cid:48) candidates. The beam energy πΈ CMbeam and πΈ CM πββ (cid:48) are calculated in the CM frame.To optimize the event selection and obtain signal detection eο¬ciency, we use Monte Carlo(MC) simulation samples. Various MC generators are used to generate background and signal MCsamples. Background samples include π + π β β π + π β ( πΎ ) , π + π β β ππ ( π’ππ π continuum and π΅π΅ ),Bhabha scattering, dimuon and two-photon mediated events.At the preliminary level, we try to retain as much generic π + π β β π + π β events as possiblein the sample while reducing obvious backgrounds by applying suitable selection requirements ondiο¬erent kinematic variables. At the next stage of selection, we apply dedicated selection criteriato pick up candidate events that are more signal-like. We perform a sideband study to identify thesources of background that are dominated by events with a misidentiο¬ed proton or antiproton, aswell as to verify the overall data-MC agreement. Since we follow the blind analysis method, beforelooking at data in the signal region, we estimate the background contribution in that region. Forthis we choose a Ξ πΈ strip by hiding the signal region to predict the background expected in thatregion as shown in Figure 3.We donβt ο¬nd any excess of events in the signal region as shown in Figure 3 and Table 1.4 est of lepton family universality and search for lepton and baryon number violation at Belle D. Sahoo (GeV) rec M β β β β E ( G e V ) β e + ep β Ο (GeV) rec M β β β β E ( G e V ) β e pe β Ο (GeV) rec M β β β β E ( G e V ) β Β΅ + ep β Ο (GeV) rec M β β β β E ( G e V ) β + Β΅ ep β Ο (GeV) rec M β β β β E ( G e V ) β Β΅ Β΅ p β Ο (GeV) rec M β β β β E ( G e V ) β + Β΅ Β΅ p β Ο Figure 3: Ξ πΈ β π rec distribution where the red box denotes the signal region and the green Ξ πΈ strip is usedto calculate the expected background. Black dots represent the data. As the number of events observed is consistent with the background prediction, we calculate anupper limit on the signal yield using a frequentist method based on a double-sided proο¬le likelihoodtest-statistic [22, 23] with CL s + b as the p-value. We have set 90% CL upper limits on the branchingfractions of these lepton ο¬avor, lepton and baryon number violating tau decays in the range of ( . . ) Γ β . In Table 1 we list results for all channels. In the case of π β β ππ β π β and ππ β π + ,our limits are improved by an order of magnitude compared to LHCb [16]. For the remaining fourchannels, we set limits for the ο¬rst time. Table 1:
Signal detection eο¬ciency ( π ), number of expected background events ( π bkg ), number of observeddata events ( π obs ), 90% CL upper limits on the signal yield and branching fraction for various decay channels. Channel π ( % ) π bkg π obs π ULsig
B (Γ β ) π β β ππ + π β . . Β± .
71 1 3 . < . π β β ππ β π β . . Β± .
46 1 3 . < . π β β ππ + π β . . Β± .
44 0 2 . < . π β β ππ β π + . . Β± .
63 0 2 . < . π β β ππ β π β . . Β± .
14 1 3 . < . π β β ππ β π + . . Β± .
07 0 2 . < .
5. Summary
We have presented recent Belle results on π πΎ and LFV searches. We also report preliminaryBelle results on L - and B -violating π decays. More precision and improved results on these studies5 est of lepton family universality and search for lepton and baryon number violation at Belle D. Sahooare expected with its upgrade, namely Belle II.
References [1] A. Abashian et al. (Belle Collaboration), Nucl. Instrum. Methods Phys. Res., Sec. A , 117(2002); also see Section 2 in J. Brodzicka et al. , Prog. Theor. Exp. Phys. , 04D001 (2012).[2] S. Kurokawa and E. Kikutani, Nucl. Instrum. Methods Phys. Res., Sec. A , 1 (2003), andother papers included in this volume; T. Abe et al. , Prog. Theor. Exp. Phys. , 03A001(2013) and following articles up to 03A011.[3] G. Hiller and F. Kruger, Phys. Rev. D , 074020 (2004).[4] C. Bobeth, G. Hiller, and G. Piranishvili, JHEP (2007) 040.[5] R. AaΔ³ et al. (LHCb Collaboration), JHEP (2017) 055.[6] S. Choudhury et al. (Belle Collaboration), arXiv:1908.01848v2 (2020).[7] J. T. Wei et al. (Belle Collaboration), Phys. Rev. Lett. , 171801 (2009).[8] S. L. Glashow, D. Guadagnoli, and K. Lane, Phys. Rev. Lett. , 091801 (2015).[9] R. AaΔ³ et al. (LHCb Collaboration), Phys. Rev. Lett. , 241802 (2019).[10] B. Aubert et al. (BaBar Collaboration), Phys. Rev. D , 092001 (2006).[11] Y. Fukuda et al. (Super-Kamiokande Collaboration), Phys. Rev. Lett. , 1562 (1998); Q. R.Ahmad et al. (SNO Collaboration), Phys. Rev. Lett. , 011301 (2002).[12] A. D. Sakharov, JETP Lett. , 24 (1967).[13] S. P. Martin, arXiv:hep-ph/9709356 (1997).[14] H. Georgi and S. L. Glashow, Phys. Rev. Lett. , 438 (1974).[15] J. D. Bekenstein, Phys. Rev. D , 1239 (1972).[16] R. AaΔ³ et al. (LHCb Collaboration), Phys. Lett. B , 36 (2013).[17] W. J. Marciano, Nucl. Phys. B, Proc. Suppl. , 3 (1995).[18] W.-S. Hou, M. Nagashima, and A. Soddu, Phys. Rev. D , 095001 (2005).[19] J. Fuentes-Martin, J. Portoles, and P. Ruiz-Femenia, JHEP , 134 (2015).[20] D. Sahoo et al. (Belle Collaboration), arXiv:2010.15361v1 (2020).[21] Inclusion of charge-conjugate processes is implied unless explicitly stated otherwise.[22] G. J. Feldman and R. D. Cousins, Phys. Rev. D , 3873 1998.[23] L. Moneta et al.et al.