Charmless Hadronic Beauty Decays at LHCb

A summary of six LHCb results on the topic of charmless hadronic b-hadron decays is presented. These are comprised of: a search for the decay Bs→ K0 S K+K− and updated branching fraction measurements of B(s) → K0 S h+h′− decays (h=K,π) [1]; the first observation of the decays B0→ ppπ+π−, Bs→ ppKK,Bs→ ppK+π− and strong evidence for the decay B0→ ppK+K− [2]; the first observation of the decay Bs→ pΛK− [3]; a search for the decay Bs→ φη′ [4]; the first observation of the decay Ξ− b → pK−K− [5] and evidence for CP-violation in Λb→ pπ−π+π− decays [6].


Introduction
Charmless b-hadron decays provide fertile ground for studying CP violation and searching for the effects of new physics.These decays can typically proceed via either b → u tree level diagrams or b → s, d gluonic loop level diagrams.The former of these transisitons is heavily suppressed by the CKM matrix element V ub , which means the tree and loop level diagrams often have similar amplitudes.The consequences of this are two fold; firstly the rates of these processes are sensitive to the presence of new physics entering into the loop diagrams and secondly CP-violation can arise from the interference of the tree and loop level diagrams.Therefore, by measuring branching fractions and CP-violation observables indirect searches for physics beyond the standard model can be performed.Furthermore, these measurements provide vital input for QCD calculations.
An overview of recent results from the LHCb experiment on the topic of charmless b-hadron decays is presented.These all make use of the Run I dataset collected by the LHCb detector during the years 2011 and 2012 which correspond to integrated luminosities of approximately 1 fb −1 and 2 fb −1 respectively.The LHCb detector is a single-arm forward spectrometer with a pseudorapidity coverage of 2 < η < 5, which has an acceptance for bb pairs of ∼ 25%.The detector is specialised for studying particles containing beauty and charm quarks.The unique sub-detectors relevant for the studies presented here are: the silicon strip vertex locator (VELO) surrounding the beam pipe which gives an impact parameter resolution of (15 ± 29 p T ) µm [7] and the two ring imaging cherenkov sub-detectors (RICH) which provide excellent particle identification ability [8].S particles that decay inside the VELO produce Long tracks and K 0 S particles that decay outside the VELO produce downstream tracks [12].
only enter via the loop diagrams which dominate the amplitudes of these charmless decays, the overall weak phase of b → ccs decays would remain unchanged.Therefore, deviations in the values of weak phase measurements between those determined from b → q qs (q=s,d,u) decays and from b → ccs decays could be an indication of new physics.Mixing induced CP violation has been measured in charmed decays through channels such as B 0 → J/ψ K 0 S [10], and a time-dependent flavour tagged analysis of the three body Dalitz plot for B 0 → K 0 S π + π − and B 0 → K 0 S K + K − decays would facilitate a comparison with charmless decays.
Unfortunately, the size of the data samples currently available at LHCb mean a flavour tagged time-dependent analysis of these decay modes is not feasible.In fact, the B 0 s → K 0 S K + K − is still unobserved.Therefore, the first aim is to observe all decay modes and develop selections that can later be used for Dalitz analyses.Previous LHCb results on these channels, making use of 1 fb −1 of 7 TeV data collected in 2011, made the first observation of the decay modes B 0 s → K 0 S π + π − , B 0 s → K 0 S K ± π ∓ and measured the branching fraction of all observed B 0 (s) → K 0 S h + h − decays relative to the well measured channel B 0 → K 0 S π + π − [11].Updated branching fraction results, which now include the 2 fb −1 of 8 TeV data collected in 2012, are presented below along with an updated search for the decay B 0 s → K 0 S K + K − .Due to the long lived nature of the neutral K 0 S particle, two separate reconstruction categories are used in this measurement.If the K 0 S decays inside the LHCb VELO the tracks from the daughter particles can be fully reconstructed.However, if the K 0 S decays outside of the VELO there is less precise tracking information available and the mass resolution of the reconstructed K 0 S is broader.Consequently, separate mass fits are required for each reconstruction category.As depicted in Figure 1, when the K 0 S decays outside of the VELO it is labelled a downstream event and when the K 0 S decays inside the VELO it is labelled as a long event.
A BDT classifier is used to separate signal and background events.The variables that enter into this classifier are deliberately chosen to avoid significant variations in efficiency over the phase space of the decay as such variations would be undesirable for future three body Dalitz plot analyses.There are two separate optimisations of the requirement on the BDT output for each final state; one for the favoured mode and one for the suppressed mode.It is possible for decays such as Λ 0 b → K 0 S pπ − to remain present as mis-ID backgrounds, therefore particle identification requirements are imposed to Table 1.Signal yield and branching fraction results for each of the B 0 (s) → K 0 S h + h − decay modes remove such backgrounds.There can be further backgrounds from B and Λ 0 b decays where the decay has proceeded through an intermediate charm (charmonium) resonance, for example B 0 → DK 0 S .These backgrounds are removed with explicit mass vetos of width 30(48) MeV around the world average mass of the charm (charmonium) resonance.
In order to extract the signal yields, the selected data samples are separated into 12 mutually exclusive categories; three separate final states, two different opitmisations and two different reconstruction categories.An unbinned simultaneous extended maximum likelihood fit to all categories is then performed.The signal decays are modelled with the sum of two Crystal Ball functions (CB) [13] and combinatorial background is modelled with a linear function.Cross-feed backgrounds, where events from one signal final state are mis-identified as another, are modelled with emperical shapes taken from simulated samples and Gaussian constraints are applied to the relative yield of these backgrounds.Partially reconstructed backgrounds from decays such as B 0 s → (K * 0 → K 0 S π 0 )(K * 0 → K − π + ) are modelled with ARGUS functions [14].The results of this fit for the favoured and suppressed optimisations are shown in Figure 2 and Figure 3 respectively.The resulting signal yields are shown in Table 1.
After correcting efficiencies for the variation over the phase space of the decay, the ratio of branching fractions (relative to the decay B 0 → K 0 S π + π − ) is determined separately for each reconstruction category.These are then combined for each decay mode by performing a minimum χ 2 fit where the convariance matrix includes the statistical and systematic uncertainties.The world average branching fraction measurement, with the previous LHCb result omitted, for the B 0 → K 0 S π + π − mode is then used to obtain the branching fraction results shown in Table 1.
The significance of the observed signal yield for the B 0 s → K 0 S K + K − mode is equivalent to 2.5 Gaussian standard deviations; this decay mode still remains unobserved.Therefore, a 90% confidence level interval is instead derived using the Feldman-Cousins method [15].
Despite not observing the B 0 s → K 0 S K + K − mode, these branching fraction measurements offer improved precision over, whilst still being consistent with, the previous LHCb results [11].Work is now underway to perform Daltiz plot analyses of the dominant decay modes -

First Observation of the Decay
The inclusive branching fraction of B decays to baryonic final states is ∼ 7% of the total decay width; many experimental studies of baryonic B decays have been carried out.At previous e + e − collider experiments the observation of many baryonic decays of B 0 and B + mesons have been reported [16,17].More recently, the first observation of a baryonic B + c decay was reported by the LHCb collaboration [18].However, the baryonic decay of a B 0 s meson has never previously been observed.The only summing the three periods of data taking, with the selection optimisation chosen for the favoured decay modes for (left) downstream and (right) long K 0 S reconstruction categories.In each plot, data are the black points with error bars and the total fit model is overlaid (solid blue line).The B 0 (B 0 s ) signal components are the pink (orange) short-dashed (dotted) lines, while fully reconstructed misidentified decays are the green and dark blue dashed lines close to the B 0 and B 0 s peaks.The sum of the partially reconstructed contributions from B to open charm decays, charmless hadronic decays, B 0 !⌘ 0 K 0 S and charmless radiative decays are the red dash triple-dotted lines.The combinatorial background contribution is the gray long-dash dotted line.
events in the fitted data sample.The average efficiency for each decay mode: .The B 0 (B 0 s ) signal candidates are shown by the magenta (orange) short dashed line, partially reconstructed backgrounds are shown by the red double dot dash line, the combinatorial background is shown by the grey dot dash line and the blue and green long dashed lines are mis-ID backgrounds.The full fit is shown by the solid blue line.[1] evidence for such a decay is reported by the Belle collaboration, where 4.4σ evidence for the decay B 0 s → Λ − c Λπ + was seen [19].One interesting aspect of baryonic B decays is the hierachy of branching fractions; multi-body baryonic B decays generally have higher branching fractions than two body baryonic decays.It is predicted that the decay B 0 s → pΛK − has a branching fraction of the order of 10 −6 which, based on similar charmless b decays, suggests it is likely to be accessible with the LHCb Run I dataset [20,21].It is also predicted that CP violating asymmetries in the decay B 0 → pΛπ − could be as high as ∼ 12% [22], which provides extra motivation for studying the B 0 (s) → pΛh family of decays.For these reasons, a search for the decay B 0 s → pΛK − has been performed using the LHCb Run I dataset.The branching fraction of the B 0 s → pΛK − is measured relative to the topologically very similar decay B 0 → pΛπ − in order to reduce systematic uncertainties.The branching fraction of this decay, B(B 0 → pΛπ − ) = (3.14 ± 0.29) × 10 −6 , has been previously measured by the BaBar and Belle collaborations [23].For the same reasons as described in Section 2, the data is separated into two reconstruction categories, depending on whether the Λ is reoncstructed inside or outside of the LHCb summing the three periods of data taking, with the selection optimisation chosen for the suppressed decay modes for (left) downstream and (right) long K 0 S reconstruction categories.Colours and line styles follow the same conventions as in Fig. 1 where N B 0 (s) !K 0 S h ± h 0⌥ is the fitted signal yield, is given for each signal decay in Table 1.They are presented for each data-taking period and K 0 S reconstruction category in Tables 3-5 in Appendix A. Their relative uncertainties due to the finite size of the simulated event samples vary from 1% to 20%.
The imperfections of the simulation are corrected for in several respects.Inaccuracies of the tracking efficiency in the simulation are mitigated by weighting the h + and h 0− tracks by a correction factor obtained from a data calibration sample [49].An analogous correction is applied for the K 0 VELO.After applying explicit mass vetoes to remove background events from the decay B 0 → Λ − c p, an MLP classifier is used to reduce combinatorial background.There are then further PID requirements applied to the p and K − / π − from the B meson, but no PID requirements are placed on the p from the Λ because no evidence for contamination from K 0 S → π + π − decays has been found.The signal and control channel yields are extracted by performing an extended maximum likelihood fit to the signal and control channel simultaneously.The results of these fits for both years of data taking and both reconstruction categories are shown in Figure 4.These mass fit models include components for: signal, combinatorial background and partially reconstructed background from the decays B 0 → pΣπ − and B 0 s → pΣK − .It is clear from Figure 4 that signal events are unambiguously present in both channels; the resulting total yields are N(B 0 s → pΛK − ) = 234 ± 29 and N(B 0 → pΛπ − ) = 519 ± 28.This is the first observation of a baryonic B 0 s decay, with a statistical significance of greater than 15 gaussian standard deviations.The branching fraction is subsequently measured to be,  B 0 s !p⇤K − decay.The combinatorial background yield and shape parameters are treated independently in each subsample and are allowed to vary in the fit.
Figure 1 presents the fit to the p⇤h − invariant mass distributions for all subsamples combined.Both B 0 !p⇤⇡ − and B 0 s !p⇤K − signals are prominent.In particular, the B 0 s !p⇤K − decay is observed with a statistical significance above 15 standard deviations, estimated from the change in log-likelihood between fits with and without the B 0 s !p⇤K − signal component [33].It constitutes the first observation of a baryonic B 0 s decay.The yields summed over all subsamples are N (B 0 !p⇤⇡ − ) = 519 ± 28 and , where the uncertainties are statistical only.
The sPlot technique is used to subtract the background and obtain the phase space distribution of signal candidates.Figure 2 shows the m(p⇤) invariant mass distributions for the B 0 !p⇤⇡ − and B 0 s !p⇤K − candidates after correcting for the distribution selection efficiencies.Both distributions show a pronounced enhancement at threshold in the baryon-antibaryon invariant mass, first suggested in Ref. [5] and observed in several baryonic B decay modes.
The sources of systematic uncertainty arise from the fit model, the knowledge of the selection efficiencies, and the uncertainties on the B 0 !p⇤⇡ − branching fraction and on the ratio of hadronization probabilities f s /f d .Uncertainties on the selection efficiencies arise from residual di↵erences between data and simulation in the trigger, reconstruction, selection and particle identification.Additional uncertainties arise due to the limited size of the simulation samples and the corresponding uncertainty on the distribution of the efficiencies across the decay phase space.As the efficiencies depend on the signal decay-time distribution, the e↵ect coming from the di↵erent lifetimes of the B 0 s mass eigenstates has been evaluated [34].Pseudoexperiments are used to estimate the e↵ect of using alternative shapes for the fit components, of including additional backgrounds in the fit such as partially reconstructed decays, and of excluding the B 0 !p⌃ 0 ⇡ − and B 0 s !p⌃ 0 K − decays that show no significant contribution.Intrinsic biases in the fitted signal yields are s → pΛK − ) channel candidate events.The grey shaded area is combinatorial background, partially reconstructed backgrounds are represented by the dot and dot dash green and pink lines and signal is represented by black and red dashed lines.The full fit is shown by the solid blue line.[3] where the first uncertainty is statistical, the second uncertainty is systematic, the third is due to the control channel branching fraction and the fourth is due to the knowledge of the ratio of B 0 s /B 0 b quark fragmentation fractions ( f s f d ).The sum of the branching fractions from both B 0 s and B 0 s is quoted because the identical final states mean it is not possible to determine the flavour of the B 0 s meson without a flavour tagged analysis.
Another area of interest in the study of baryonic B decays is threshold enhancement in the baryon anti-baryon system.This effect has previously been observed by the BaBar and Belle collaborations in several baryonic B decays [17].This effect is investigated in the B 0 → pΛπ − and B 0 s → pΛK − decays by using the sPlot technique to perform a background subtraction and correct for efficiency variations across the phase space of the decay [24].The efficiency corrected and background subtracted invariant mass distributions for the p Λ system are shown in Figure 5; clear threshold enhancement is observed.investigated with ensembles of simulated pseudoexperiments.A small bias is found and added to the systematic uncertainty on the fit model.The systematic uncertainty due to the knowledge of the efficiencies involved in the definition of fit constraints is negligible.The total systematic uncertainty on the B 0 s !p⇤K − branching fraction is given by the sum of all uncertainties added in quadrature and amounts to 10.5%; it is dominated by the systematic uncertainty on the fit model.
The uncertainty on the branching fraction of the normalization decay, [25], is taken as a systematic uncertainty from external inputs.The 5.8% uncertainty on the latest f s /f d combination from LHCb, f s /f d = 0.259 ± 0.015 [35], is taken as a second source of systematic uncertainty from external inputs.The B 0 s !p⇤K − branching fraction, determined relative to that of the B 0 !p⇤⇡ − normalization channel according to Eq. 1, is measured to be 4 Search for B 0 (s) → pph + h − decays As described in Section 3 the baryonic decay of a B 0 s meson has now been observed.However, a four body baryonic decay of the B 0 s meson is still to be observed and previous results for the B 0 meson only set an upper limit B(B 0 → ppπ + π − ) < 2.5 × 10 −4 at 90% confidence level [25].There is additional motivation in studying four body baryonic B meson decays because they are an ideal place to search for CP violation using triple-product correlations (TPCs); in contrast to the case of three body decays, they do not involve the spins of the final state particles.This motivation is strengthened by the first evidence of CP violation in baryonic B decays being seen in B + → ppK + decays [26] and predictions of CP violation as high as 20% in other baryonic B decays [27].
A search for the family of decays B 0 (s) → pph + h − is reported, where h is a kaon or pion.The branching fractions of these decays are measured relative to the decay B 0 → J/ψ K * 0 , where the resonances are reconstructed through the J/ψ → pp and K * 0 → K + π − decay channels.In order to remove charm and charmonium resonances in the signal channels the requirement m(pp) < 2.85 GeV is imposed on the signal channels and specfic mass vetoes are applied to remove Λ + c and D 0 decays.The selection further consists of a BDT, which makes use of 15 variables, to reduce combinatorial background and PID requirements remove mis-ID backgrounds.
The signal yields are extracted with an extended maximum likelihood fit which is simultaneous across the invariant masses of all 3 possible final states: m(ppK + π − ), m(ppK + K − ) and m(ppπ + π − ).A separate fit is performed for the B 0 → J/ψ K * 0 control channel which is multi-dimensional for the m(ppK + π − ), m(pp) and m(K + π − ) invariant masses.These fits are shown in Figure 6 and the resulting yields and branching fractions are shown in Table 2.This is the first observation of the decays B 0 → ppπ + π − and B 0 s → ppK + K − with significances of > 25σ.The decay B 0 s → ppK + π − is also observed with a significance of 6.5σ, there is strong evidence for the decay B 0 → ppK + K − at the level of 4.1σ and evidence at the level of 2.6σ is seen for the B 0 s → ppπ + π − decay.A limit is set on the B 0 s → ppπ + π − channel by integrating the likelihood with a uniform prior in the region of positive branching fraction.where the B 0 !J/ K ⇤ (892) 0 branching fraction is taken from Ref. [35] and the others from Ref. [11].Branching Fraction/10 −6 B 0 → ppK + K − 68 ± 17 0.113 ± 0.028 ± 0.011 ± 0.008 Signal yields and branching fraction results for the six B 0 (s) → pph + h − decays.The first uncertainty is statistical, the second systematic, the third is due to the uncertainty on the normalisation channel branching fraction and the fourth (in the case of B 0 s mesons) is due to the ratio of B 0 s /B 0 fragmentation fractions.
As described in Section 3, threshold enhancement has been observed in the baryon antibaryon system of several B meson decays.To investigate threshold enhancement in the B 0 → ppK + π − decay, the sPlot technique is again used to perform a background subtraction and correct for the efficiency variation across the phase space of the decay.The normalised, background subtracted and efficiency corrected distribution of m(pp) can be seen in Figure 7 [24].As with the decay B 0 s → pΛK − (see Figure 5), clear threshold enhancement is present.the efficiency of the hardware stage of the trigger.As the efficiencies gnal decay-time distribution, the e↵ect coming from the di↵erent lifetimes igenstates has been evaluated.The systematic uncertainties due to the hadrons are also included.a search for the four-body charmless baryonic decays B 0 (s) !pphh 0 has by the LHCb collaboration with a sample of proton-proton collision data an integrated luminosity of 3 fb −1 .First observations are obtained for pp⇡⇡, nonresonant B 0 !ppK⇡, B 0 s !ppKK and B 0 s !ppK⇡, while reported for the B 0 !ppKK mode and an upper limit is set on the ching fraction.In particular, four-body baryonic B 0 s decays are observed and a threshold enhancement in the baryon-antibaryon mass spectra is ryonic B 0 s decays [2]. 5 Search for the decay B 0 s → η φ Current experimental knowledge of B 0 s decays to a light pseudoscalar (P) meson and a vector meson (V) is very limited and theoretical predictions for the branching fractions of such decays have large uncertainties and cover a large range.For example, a QCD factorisation approach can yield the result B(B 0 s → η φ) = (0.05 1.18  −0.19 ) × 10 −6 [28], whereas a perturbative QCD approach yields the prediction B(B 0 s → η φ) = (20.0+16.3 −9.1 ) × 10 −6 [29] (see Table 1 in Ref. [4] for a summary of theoretical predictions).These large uncertainties are a consequence of the currently limited knowledge of penguin contributions, the ω-φ mixing angle, the s-quark mass and form factors. Therefore, a measurement of B(B 0 s → η φ) would be a useful input to the knowledge of B 0 s to φ form factors and would subsequently improve the accuracy of branching fraction predictions for B 0 s → Pφ decays.A search for the decay B 0 s → η φ is reported, where the η is reconstructed through the channel η → π + π − γ and the φ through the channel φ → K + K − .The well studied decay B + → K + η is used as a control channel due to its relatively large branching fraction B(B + → K + η ) = (70.6 ± 2.5) × 10 −6 and the presence of only combinatorial background [23,30].The selection makes use of a BDT with 9 variables to reject the majority of combinatorial background and particle identification requirements are imposed on the hadrons and the photon from the η .
The signal yields are extracted with an extended maximum likelihood fit to the B 0 s → η φ and B + → K + η channels simultaneously.The PDF used in the B 0 s → η φ (B + → K + η ) channel is two dimensional for the m(η φ) and m(π + π − γ) (m(η K + ) and m(π + π − γ)) variables.In the case of the B 0 s → η φ channel, there is an irreducible background from B 0 s → φφ decays where one of the φ mesons decays through φ → K + K − and the other through φ → π + π − π 0 but one of the photons from the decay of the π 0 is not reconstructed.This background is therefore modelled in the fit with a two-dimensional Gaussian kernel function [31].The fit results for the B 0 s → η φ channel are shown in Figure 8.  Sets of pseudoexperiments are used to evaluate possible fit biases.Fits on samples generated from the probability density function (PDF) with parameters obtained from the data are found to be unbiased.The procedure is then repeated using simulated B 0 s !φφ events instead of generating the corresponding background component from the PDF.Biases of −1.3 ± 0.3 on the signal yield and of (−1.16 ± 0.33) ⇥ 10 −4 on the ratio of yields are observed.The results obtained with data are corrected for these biases and systematic uncertainties computed as the quadratic sum of the statistical uncertainty on the bias and half of the bias value are assigned.
Additional systematic uncertainties a↵ect the signal yield and the yield ratio.The mass fit is repeated with di↵erent combinatorial background PDFs: linear functions are replaced with exponential functions, and the parabolic function is replaced with a third-order polynomial.The quadratic sum of the di↵erences between the values obtained in these alternative fits and the nominal result is assigned as a systematic uncertainty.The limited size of the simulated B 0 s !φφ sample leads to an uncertainty on the determination of the nonparametric PDF for the physics background, which is propagated as a systematic uncertainty.The e↵ect of fixing some of the model parameters in the fit is studied by performing a large number of fits on the data, with the fixed parameters sampled s → η φ channel.The solid blue line represents the result of the two-dimensional simultaneous fit to both channels.The red dashed line shows the signal component, the green dashed line shows the combinatorial background with a real η , the blue dot-dash line shows combinatorial background without a real η and the black dot-dash line is the irreducible B 0 s → φφ background described in the text.[4] The yields from this fit are N(B 0 s → η φ) = −3.2+5.0 −3.8 and N(B + → K + η ) = 11081 ± 127, therefore bayesian upper limits are set on B(B 0 s → η φ) using a uniform prior.The resulting limit is, B(B 0 s → η φ) < 0.82(1.01)× 10 −6 at 90% (95%) CL, which is the first upper limit to be set on B(B 0 s → η φ).This result is consistent with the lower end of the range of predictions available for this branching fraction; the central values of several predictions are inconsistent with this result, see for example Refs.[29,[32][33][34].This result therefore provides very useful constraints on the theoretical models used to predict branching fractions and CP asymmetries of charmless B 0 s decays to a pseudoscalar and vector meson final state.

Search for the decays Ξ
Prior to the collection of the Run I dataset by LHCb, opportunities to study charmless decays of bbaryons were very limited.Consequently, the only observed decays were Λ 0 b → pπ − and Λ 0 b → pK − by the CDF collaboration [35].More recently, LHCb has observed the 37] and Λ 0 b → Λφ [38].There has also been evidence at the level of 3σ seen for the decay Λ 0 b → Λη b and f Ω − b , are not known.This search uses the topologically similar and well measured decay B − → K + K − K − as a control channel, therefore the reported results are the product of branching fraction and ratio of fragmentation fractions, denoted R ph − h − .
The event selection firstly consists of a neural network, which makes use of 8 variables, to reduce combinatorial background.A tight particle identification requirement is then imposed on the proton to reject background from B − → K + h − h − decays.Further particle identification requirements on the other hadrons are optimised in a manner that ensures mutually exclusive samples for each possible final state.There are then vetoes applied to remove charm backgrounds from decays such as The signal yields are extracted using an extended unbinned maximum likelihood fit which is simultaneous across the invariant mass distributions of the three possible final states (m(pK − K − ), m(pK − π − ) and m(pπ − π − )).The use of a simultaneous fit allows cross-feed background, where candidates from one final state are mis-identified as another, to be constrained to expected values.The control channel yield is extracted with a separate fit to the m(K + K − K − ) distribution.There are also partially reconstructed backgrounds present that have to be modelled in the fit.In the m(ph − h − ) distributions these arise from decays such as where the π 0 is not reconstructed.These backgrounds are modelled with an ARGUS function [14] convolved with a Gaussian function.
The projections of the simultaneous fit and the separate control channel fit are shown in Figure 9; the fitted yields and subsequent branching fraction times fragmentation fraction ratios are shown in Table 3.The significance of the Ξ − b → pK − K − signal yield is 8.7σ; this is the first observation of a charmless Ξ − b decay.There is also 3.4σ evidence for the decay Ξ − b → pK − π − , but all other signal yields have a significance of < 2.0σ meaning the charmless decay of an Ω − b baryon still remains unobserved.For these channels with signifiance < 2.0σ an upper limit is set on R ph − h − by integrating the likelihood after multiplying by a prior probability distribution which is uniform in the positive region.
Although it is not possible to measure absolute branching fractions, the observation of the Ξ − b → pK − K − decay means it is possible to measure the relative branching fractions of the Ξ − b decays.This cancels the dependance on fragmentation fractions, which is useful for comparisons with theoretical predictions.The measured relative branching fractions are, where the upper limit quoted is at the 90%(95%) CL.
It is of interest to study the intermediate resonances that occur in the Ξ − b → pK − K − channel; it is hoped that when larger data samples are available an amplitude analysis to dis-entangle the intermediate resonances and to search for CP violation can be performed.Whilst this is not currently possible, it is still of interest to inspect the intermediate resonances.    of reconstruction and selection efficiencies in the simulation has been validated with large control samples; the impact on the results of possible residual di↵erences between data and simulation is negligible.
For the , efficiency corrections for each candidate are applied using the method of Ref. [46] to take the variation over the phase space into account.Using this procedure, the efficiency-corrected and background-subtracted m(pK − ) min distribution shown in Fig. 3 Here m(pK − ) min indicates the smaller of the two m(pK − ) values for each signal candidate, evaluated with the ⌅ − b and the final-state particle masses fixed to their known values [40,44].The distribution contains a clear peak from the ⇤(1520) resonance, a structure that is consistent with being a combination of the ⇤(1670) and ⇤(1690) states, and possible additional contributions at higher mass.Compared to the pK − structures seen in the amplitude analysis of ⇤ 0 b !J/ pK − [47], the contributions from the broad ⇤(1600) and ⇤(1810) states appear to be smaller.A detailed amplitude analysis will be of interest when larger samples are available.
For channels without significant signal yields the efficiency averaged over phase space is used in Eq. (1).A corresponding systematic uncertainty is assigned from the variation of the efficiency over the phase space; this is the dominant source of systematic uncertainty 5

Channel
Yield Table 1: Fitted yields, efficiencies and relative branching fractions multiplied by fragmentation fractions (R ph − h 0− ).The two uncertainties quoted on R ph − h 0− are statistical and systematic.Upper limits are quoted at 90 (95) % confidence level for modes with signal significance less than 3 σ.Uncertainties on the efficiencies are not given as only the relative uncertainties a↵ect the branching fraction measurements.

Mode
Yield for those channels.The quantities entering Eq. ( 1), and the results for R ph − h 0− , are reported in Table 1.When the signal significance is less than 3 σ, upper limits are set by integrating the likelihood after multiplying by a prior probability distribution that is uniform in the region of positive branching fraction.The sources of systematic uncertainty arise from the fit model and the knowledge of the efficiency.The fit model is changed by varying the fixed parameters of the model, using alternative shapes for the components, and by including components that are omitted in the baseline fit.Intrinsic biases in the fitted yields are investigated with simulated pseudoexperiments, and are found to be negligible.Uncertainties in the efficiency arise due to the limited size of the simulation samples and possible residual di↵erences between data and simulation in the trigger and particle identification efficiencies [48].Possible biases in the results due to the vetoes of charm hadrons are also accounted for.The efficiency depends on the signal decay-time distribution, and therefore the precision of the ⌅ − b and ⌦ − b lifetime measurements [40][41][42] is a source of uncertainty.Similarly, the p T distribution assumed for signal decays in the simulation a↵ects the efficiency.Since the p T spectra for 6 This search is performed by studying asymmetries in the T operator, which is the unitary operator that reverses both momentum and spin three-vectors 1 .Scalar triple products are defined, where p h is the momentum of the final state hadron, h, in the Λ 0 b centre-of-mass frame.The π − with the largest momenta in the Λ 0 b rest frame is used (π − f ast ).Asymmetries can then be defined as, where N( N) is the number of Λ 0 b (Λ 0 b ) decays satisfying the given condition.Using these asymmetries, CP violation and P violation observables are defined: Any deviation from zero in these operators would indicate the presence of CP/P violation.The advantage of using triple product asymmetries is that they are largely insensitive to prodcution asymmetries and charge asymmetries caused by the LHCb detector.The event selection makes use of a BDT to reduce combinatorial background and optimised particle identification requirements.An extended unbinned maximum likelihood fit to the m(pπ − π + π − ) invariant mass distribution is used to extract the total signal yield of both Λ 0 b and Λ 0 b decays.This fit is shown in Figure 11 and has several components: signal decays are modelled by a Gaussian core with power law tails [13]; combinatorial background is modelled with an exponential function; partially reconstructed backgrounds from five body Λ 0 b decays where one particle is not reconstructed are modelled with an ARGUS [14] function convolved with a Gaussian and mis-ID backgrounds are modelled with shapes fixed from simulation.The signal yield from this fit is reconstructed Λ 0 b decays are described by an empirical function [41] conv Gaussian function to account for resolution effects.The shapes of back other b-hadron decays due to incorrectly identified particles, e.g.kaons iden or protons identified as kaons, are modelled using simulated events.These c of Λ 0 b → pK − π + π − and B 0 → K + π − π − π + decays for the Λ 0 b → pπ − π + π − s similar final states for the Λ 0 b → pπ − K + K − sample, as shown in Fig. 2. these contributions are obtained from fits to data reconstructed under th mass hypotheses for the final-state particles.The signal yields of Λ 0 b → p Λ 0 b → pπ − K + K − are 6646 ± 105 and 1030 ± 56, respectively.This is the firs of these decay modes.
Signal candidates are split into four categories according to Λ 0 b or Λ the sign of C T or C T in order to calculate the asymmetries defined in and (2).The reconstruction efficiency for signal candidates with C T > 0 that with C T < 0 within the statistical uncertainties of the control sample for C T , which indicates that the detector and the reconstruction program do measurement.This check is performed both on the Λ 0 b → Λ + c (pK − π + )π − sample and on large samples of simulated events, using yields about 30 time in data, which are generated with no CP asymmetry.The CP asymmetry the control sample is a T -odd CP (Λ + c π − ) = (0.15 ± 0.31)%, compatible with C The asymmetries A T and A T in the signal samples are measured with a unbinned maximum likelihood fit to the invariant mass distributions of signal categories, and are found to be uncorrelated.The values of a T -odd CP a then calculated from A T and A T .
In four-body particle decays, the CP asymmetries may vary over the ph to resonant contributions or their interference effects, possibly cancelling wh Many resonances are expected in the four body Λ 0 b → pπ − π + π − decay and it is expected different resonances will exhibit different levels of CP violation.It is entirely possible that the CP violation from different resonances would cancel if a phase space integrated asymmetry was calculated, which would considerably reduce the sensitivity of this search.Therefore, different regions of the phase space are investigated by calculating the a T −odd CP and a T −odd p observables for two separate binning schemes of the phase space.The first binning scheme, denoted "Scheme A", makes use of two body invariant masses in order to exploit the strong resonant structure from decays such as ∆(1232) ++ → pπ + ."Scheme B" consists of 10 bins in the angle Φ which is the angle between the p π − f ast and π − slow π + decay planes, shown in Figure 12.The objective of this binning scheme is to probe the interference between different resonant contributions.flavour.The product of p-values from both binning schemes is compared between these pseudoexperiments and the values obtained in data, this leads to a combined significance for CP violation in Λ 0 b → pπ − π + π − decays of 3.3σ.This is the first evidence for CP violation in a baryonic decay and it will be of great interest to study this asymmetry further when larger data samples are available.T and A b T .In four-body particle decays, the CP asymmetries may vary over the phase space due to resonant contributions or their interference e↵ects, possibly cancelling when integrated over the whole phase space.Therefore, the asymmetries are measured in di↵erent regions of phase space for the ⇤ 0 b !p⇡ − ⇡ + ⇡ − decay using two binning schemes, defined before examing the data.Scheme A, defined in Table 1, is designed to isolate regions of phase space according to their dominant resonant contributions.Scheme B exploits in more detail the interference of contributions which could be visible as a function of the angle Φ between the decay planes formed by the p⇡ − fast and the ⇡ − slow ⇡ + systems, as illustrated in Fig. 3 The asymmetries measured in ⇤ 0 b !p⇡ − ⇡ + ⇡ − decays with these two binning schemes are shown in Fig. 4 and reported in Table 2, together with the integrated measurements.For each scheme individually, the compatibility with the CP -symmetry hypothesis is evaluated by means of a χ 2 test, with χ 2 = R T V −1 R, where R is the array of a b T -odd CP measurements and V is the covariance matrix, which is the sum of the statistical and systematic covariance matrices.An average systematic uncertainty, whose evaluation is discussed below, is assigned for all bins.The systematic uncertainties are assumed to be fully correlated; their contribution is small compared to the statistical uncertainties.The p-values of the CP -symmetry hypothesis are 4.9 ⇥ 10 −2 and 7.1 ⇥ 10 −4 for schemes A and B, respectively, corresponding to statistical significances of 2.0 and 3.4 Gaussian standard deviations (σ).A similar χ 2 test is performed on a   (5)(6)(7)(8)(9)(10)(11)(12).Further splitting for |Φ| lower or greater than ⇡/2 is done to reduce potential dilution of asymmetries, as suggested in Ref. [19].Masses are in units of GeV/c

Summary
The study of charmless b hadron decays continues to yield many interesting results, and LHCb is leading the way in studying this sector.The branching fractions results for the B 0 (s) → K 0 S h + h − family of decays have been updated and the limits on the branching fraction of the B 0 s → K 0 S K + K − have been narrowed.These results set the foundations for further Dalitz plot studies of these decay channels.Several baryonic decays of b mesons have been observed for the first time; the decay B 0 s → pΛK − was the first baryonic decay of a B 0 s meson to be observed.The four body decays B 0 → ppπ + π − , B 0 s → ppK + π − and B 0 s → ppK + K − have been observed for the first time, which open up opportunities for CP violation studies in four body baryonic b meson decays.A stringent upper limit has been set on the branching fraction of the decay B 0 s → φη , B(B 0 s → φη ) = 0.82(1.01)× 10 −6 at the 90%(95%) CL, which is a very useful input to the theoretical models used to predict the properties of B 0 s → PV decays.In the baryonic sector, the observation of the decay Ξ − b → pK − K − is the first observation of a charmless Ξ − b decay and opens up opportunities to search for CP violation in charmless Ξ − b decays.Lastly, 3.3σ evidence for CP violation in Λ 0 b → pπ − π + π − decays has been found; this is the first evidence for CP violation in a baryonic beauty decay.With further studies and larger data samples it is hoped these differences could contribute to understanding the absence of anti-matter in the universe.

Figure 1 .
Figure 1.The different track types at LHCb.K 0 S particles that decay inside the VELO produce Long tracks and K 0S particles that decay outside the VELO produce downstream tracks[12].

Figure 1 :
Figure1: Invariant mass distributions of, from top to bottom, K 0 S K + K − , K 0 S K ± ⇡ ⌥ and K 0 S ⇡ + ⇡ − candidates, summing the three periods of data taking, with the selection optimisation chosen for the favoured decay modes for (left) downstream and (right) long K 0 S reconstruction categories.In each plot, data are the black points with error bars and the total fit model is overlaid (solid blue line).The B 0 (B 0 s ) signal components are the pink (orange) short-dashed (dotted) lines, while fully reconstructed misidentified decays are the green and dark blue dashed lines close to the B 0 and B 0 s peaks.The sum of the partially reconstructed contributions from B to open charm decays, charmless hadronic decays, B 0 !⌘ 0 K 0 S and charmless radiative decays are the red dash triple-dotted lines.The combinatorial background contribution is the gray long-dash dotted line.

) 8 Figure 2 .
Figure 2. Invariant mass distributions for each possible final state with the favoured mode selection.The Downstream(Long) reconstruction category is shown on the right(left).The B 0 (B 0s ) signal candidates are shown by the magenta (orange) short dashed line, partially reconstructed backgrounds are shown by the red double dot dash line, the combinatorial background is shown by the grey dot dash line and the blue and green long dashed lines are mis-ID backgrounds.The full fit is shown by the solid blue line.[1]

4 EPJFigure 2 :
Figure 2: Invariant mass distributions of, from top to bottom, K 0S K + K − , K 0 S K ± ⇡ ⌥ and K 0 S ⇡ + ⇡ − candidates,summing the three periods of data taking, with the selection optimisation chosen for the suppressed decay modes for (left) downstream and (right) long K 0 S reconstruction categories.Colours and line styles follow the same conventions as in Fig.1

S tracking and vertex reconstruction efficiency. A control 9 Figure 3 .
Figure 3. Invariant mass distributions for each possible final state with the suppressed mode selection.The Downstream(Long) reconstruction category is shown on the right(left).The individual fit components follow the same convention as Figure 2. [1]

Figure 1 :
Figure 1: Mass distributions for b-hadron candidates for (left) the p⇤⇡ − and (right) the p⇤K − sample for the combined long and downstream categories.The black points represent the data, the solid blue curve the result of the fit, the red dashed curve the B 0 s !p⇤K − contribution, the black (magenta) dotted curve the B 0 !p⇤⇡ − (B 0 s !p⌃ 0 K − ) and the green dash-dotted curve the contribution from B 0 !p⌃ 0 ⇡ − decays.The combinatorial background distribution is indicated by the shaded histogram.

Figure 4 .
Figure 4. Left (Right): The invariant mass distribution of B 0 → pΛπ − (B 0s → pΛK − ) channel candidate events.The grey shaded area is combinatorial background, partially reconstructed backgrounds are represented by the dot and dot dash green and pink lines and signal is represented by black and red dashed lines.The full fit is shown by the solid blue line.[3]

Figure 1 :
Figure 1: Invariant mass distributions for B 0 (s) candidates in the (top left) ppKK, (top right) ppK⇡, (bottom left) pp⇡⇡ final state and (bottom right) invariant mass distribution of B 0 !J/ K ⇤ (892) 0 in the ppK⇡ final state.The results of the fits are shown with blue solid lines.In the first three figures signals for B 0 and B 0 s decays are shown, respectively, with green dotted and red dot-dashed lines, combinatorial backgrounds are shown with black dashed lines and cross-feed backgrounds are shown with violet dot-dashed lines.In the bottom right figure the normalization signal is shown with a green dotted line, the K⇡ S-wave component is displayed with a red dot-dashed line and the combinatorial background with a black dashed line.

Figure 6 . 7 EPJ
Figure 6.Top Left, Top Right and Bottom Left: Signal channel simultaneous mass fit to all three final states.B 0 signal is shown by the dotted green line, B 0s signal by the dot dash red line and cross feed background is shown by the dot dash purple line.Bottom Right: Projection of the 3D control channel mass fit.[2] ] and background-subtracted m(hh 0 ) distributions from (top left) p right) B 0 !pp⇡⇡, and (bottom left) B 0 !ppK⇡ candidates and (bottom ributions from B 0 !ppK⇡ candidates.All distributions are normalized to

Figure 8 .
Figure 8. Left (Right): Distributions of m(π + π − γ) (m(η φ)) for the B 0s → η φ channel.The solid blue line represents the result of the two-dimensional simultaneous fit to both channels.The red dashed line shows the signal component, the green dashed line shows the combinatorial background with a real η , the blue dot-dash line shows combinatorial background without a real η and the black dot-dash line is the irreducible B 0 s → φφ background described in the text.[4]

baryon is 9 EPJ
[21].However a charmless decay of a Ξ − b or Ω − b Web of Conferences 158, 01005 (2017) DOI: 10.1051/epjconf/201715801005 QFTHEP 2017 yet to be observed, therefore a search for the decays Ξ − b , Ω − b → ph − h − where h=K − , π − is presented.It is not currently possible to measure the absolute branching fractions of Ξ − b and Ω − b decays because the fragmentation fractions, f Ξ −

10 EPJ
Figure 10 shows the background subtracted and efficiency corrected m(pK − ) min distribution for the Ξ − b → pK − K − channel; m(pK − ) min is the smaller of the two possible m(pK − ) values for each candidate.This distribution shows clear peaks consistent with the Λ(1520) and a combination of the Λ(1670) and Λ(1690) states.The less prominent peaks at higher mass suggest the possibility of further states being present.Web of Conferences 158, 01005 (2017) DOI: 10.1051/epjconf/201715801005 QFTHEP 2017

Figure 2 :
Figure 2: Mass distributions for b-hadron candidates in the (top left) pK − K − , (top right) pK − ⇡ − , (bottom left) p⇡ − ⇡ − and (bottom right) K + K − K − final states.Results of the fits are shown with dark blue solid lines.Signals for ⌅ − b and B − (⌦ − b ) decays are shown with pink (light green) dashed lines, combinatorial backgrounds are shown with grey long-dashed lines, cross-feed backgrounds are shown with red dot-dashed lines, and partially reconstructed backgrounds are shown with dark blue double-dot-dashed lines.

Figure 9 .
Figure 9. Top Left, Top Right and Bottom Left: Projections of the simultaneous fit to the m(ph − h − ) final states.Bottom Right: The control channel fit to m(K + K − K − ) mass distribution.The full fits are shown by the blue solid line, Ξ − b and B − signal components are shown by the pink dashed line, Ω − b signal components are shown by the green dashed line, cross-feed backgrounds are shown by the red dot-dash line, partially reconstructed backgrounds are shown by the blue double dot-dash line and combinatorial background is shown by the grey long dashed line.[5]

12 EPJFigure 2 :
Figure 2: Reconstructed invariant mass fits used to extract the signa invariant mass distributions for (a) Λ 0 b→ pπ − π + π − and (b) Λ 0 b → pπ − K + K − dec A fitis overlaid on top of the data points, with solid and dotted lines describing of the fit results for each of the components described in the text and listed Uncertainties are statistical only.

Figure 11 .
Figure 11.The invariant mass distribution of Λ 0 b → pπ − π + π − candidates with the fit results overlayed.The fit has several components: the Λ 0 b → pπ − π + π − signal component is shown by the solid red line, combinatorial background is shown by the blue dashed line, partially reconstructed backgrounds from 5 body decays are shown by the light blue double dot dashed line, mis-identified Λ 0b → pK − π + π − events are shown by the purple dot dash line and mis-identified B 0 → K + π − π − π + candidates are shown by the brown dashed line.The full fit is shown by the solid blue line.[6]

Figure 13
Figure 13 shows the results for a T −odd CP and a T −odd p in each binning scheme.A χ 2 test is used to assess the compatibility of each binning scheme with the null hypothesis of CP symmetry; the p-values are 4.9 × 10 −2 (2.0σ) and 7.1 × 10 −4 (3.4σ) for the binning schemes A and B respectively.The combined significance of the measured level of CP violation in Λ 0 b → pπ − π + π − decays is determined using a permutation test where 40,000 pseudoexperiments are generated with randomly assigned Λ 0 b /Λ 0 b

Figure 3 :
Figure 3: Definition of the Φ angle.The decay planes formed by the p⇡ − fast and the ⇡ − slow ⇡ + systems in the ⇤ 0 b rest frame.The momenta of the particles, represented by vectors, determine the two decay planes and the angle Φ 2 [−⇡, ⇡] [19] measures their relative orientation.
. Scheme B has 10 non-overlapping bins of width ⇡/10 in |Φ|.For every bin in each of the schemes, the ⇤ 0 b efficiencies for C b T > 0 and C b T < 0 are compared and found to be equal within uncertainties, and likewise the ⇤ 0 b efficiencies for C b T > 0 and C b T < 0. The analysis technique is validated on the ⇤ 0 b !⇤ + c (pK − ⇡ + )⇡ − control sample, for which the angle Φ is defined by the decay planes of the pK − and ⇡ + ⇡ − pairs, and on simulated signal events.

b 5 Figure 12 .Figure 4 :
Figure 12.A diagram showing the defnition of the angle Φ which is used for binning scheme B.[6]

Table 1 :
Definition of binning scheme A for the decay mode ⇤ 0 b !p⇡ − ⇡ + ⇡ − .Binning scheme A is defined to exploit interference patterns arising from the resonant structure of the decay.Bins 1-4 focus on the region dominated by the ∆(1232) ++ !p⇡ + resonance.The other eight bins are defined to study regions where p⇡ − resonances are present (5-8) on either side of the ⇢(770) 0 !⇡ + ⇡ − resonances