Study of b -hadron decays to Λ + c h − h ′− final states

,


Introduction
The large production rate of b hadrons in LHC collisions has enabled a vastly increased range of b-baryon decays to be studied.The majority of measurements in this sector have so far been made with Λ 0 b baryons, with only relatively few Ξ − b and Ω − b decays observed [1][2][3][4][5][6][7][8][9].New and improved measurements of the production and decays of b baryons will constrain QCD models and also reduce systematic uncertainties in future measurements of the properties of related decays.In particular, knowledge of production asymmetries is necessary to make precise searches for CP violation in b-baryon decays [4].Existing measurements of the production rates and asymmetries of Ξ − b baryons, obtained using the Ξ − b → J/ψΞ − decay [10], have large uncertainties so improvements are well motivated.Additionally, the phase-space distribution of multibody b-baryon decays can be probed to investigate the spectroscopy of charm baryons, including studies of potentially exotic hadrons.
At present there are no results on Ξ − b and Ω − b decays to Λ + c h − h ′− final states, where h and h ′ are K or π mesons.Figure 1 shows possible decay diagrams for three such processes, specifically those where the branching fractions are expected to be largest.The branching fractions is determined as The yield obtained from the fit to the X b -candidate mass distribution, N , is equal to n i=0 w i where w i are signal weights [13] obtained from the same fit for all the n candidates in the fit.Variation of the efficiency across the phase space of the decay is taken into account by use of candidate-by-candidate efficiencies, ϵ i , which depend on position in the decay phase-space and year of data taking.The efficiencies are determined from simulation with corrections, discussed below, applied in order to describe data more accurately.The efficiency-corrected yield is then N corr = n i=0 w i /ϵ i , and an effective average efficiency can be defined as ⟨ϵ⟩ = N/N corr .In case the yield of a particular channel is not significant, the candidate-by-candidate efficiency correction of Eq. (1) will not be reliable, and an average efficiency obtained directly from simulation with an assumption on the phase-space distribution is used instead.In this case, the effect of the unknown phase-space distribution will be a source of systematic uncertainty.
When decays of different b hadrons are considered, Eq. ( 1) is modified so that the left-hand side is multiplied by the appropriate ratio of fragmentation fractions, but the procedure is otherwise identical.The fragmentation fractions quantify the probability that a b quark produced in the pp collision hadronises into a particular b hadron.The fragmentation fractions can depend on pp collision centre-of-mass energy, and therefore all ratios are determined separately for Run 1 and Run 2. Combined results are obtained where appropriate, taking correlations of systematic uncertainties into account.
For a channel X b → Y with sufficient yield, the charge-conjugate processes can be separated to determine the X b production asymmetry, A prod , from the measured efficiency-corrected yield asymmetry, Here, the weights w i in all sums are obtained from a single fit to the sample including both charge conjugate final states, but sums over different candidates are performed for the two states.The sense of the asymmetry is defined as being between the X b hadron that contains a b quark and its Xb antiparticle that contains a b quark.The efficiency maps used to obtain the N corr values also differ for the two charge conjugate final states, and are denoted as ϵ and ε.The use of separate efficiency maps corrects for detection asymmetry, although potential mismodelling of ϵ and ε must be accounted for as a source of systematic uncertainty.For decays where no CP violation is expected, i.e.where A CP = 0 can be assumed, one has A meas = A prod .This is the case for the channels studied here, which are expected to be dominated by tree-level diagrams as shown in Fig. 1.Production asymmetries, and the relative fragmentation fractions of different hadrons, can depend on kinematics.The above approach can be extended to study such effects, by making measurements in bins of, for example, b-hadron transverse momentum.In this case the sum in Eq. ( 2) includes only events in the relevant bin, and the efficiency maps are those appropriate for that bin.

Detector and simulation
The LHCb detector [14,15] 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 siliconstrip vertex detector surrounding the pp interaction region [16], a large-area silicon-strip detector located upstream of a dipole magnet with a bending power of about 4 T m, and three stations of silicon-strip detectors and straw drift tubes [17,18] placed downstream of the magnet.The tracking system provides a measurement of the momentum, p, of charged particles with 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 pp collision vertex (PV), the impact parameter (IP), is measured with a resolution of (15 + 29/p T ) µm, where p T 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 detectors [19].Photons, electrons and hadrons are identified by a calorimeter system consisting of scintillating-pad and preshower detectors, an electromagnetic and a hadronic calorimeter.Muons are identified by a system composed of alternating layers of iron and multiwire proportional chambers [20].
The online event selection is performed by a trigger [21,22], 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.At the hardware trigger stage, events are required to have a muon with high p T or a hadron, photon or electron with high transverse energy in the calorimeters.For hadrons, the transverse energy threshold is 3.5 GeV.The software trigger requires a two-, three-or four-track secondary vertex with a significant displacement from any primary pp interaction vertex.At least one charged particle must have a transverse momentum p T > 1.6 GeV/c and be inconsistent with originating from a PV.A multivariate algorithm [23,24] is used for the identification of secondary vertices consistent with the decay of a b hadron.In the offline selection, trigger signals are associated with reconstructed particles.Selection requirements can therefore be made on the trigger selection itself and on whether the decision was due to the signal candidate, other particles produced in the pp collision, or a combination of both.
Simulation is used to model the effects of the detector acceptance and the imposed selection requirements, as well as to study potential backgrounds.In the simulation, pp collisions are generated using Pythia [25] with a specific LHCb configuration [26].The signal processes are generated with the FLATSQDALITZ model, which generates events uniformly in the square Dalitz plot as described in Ref. [27].Decays of unstable particles are described by EvtGen [28], in which final-state radiation is generated using Photos [29].The interaction of the generated particles with the detector, and its response, are implemented using the Geant4 toolkit [30] as described in Ref. [31].The underlying pp interaction is reused multiple times, with an independently generated signal decay for each [32].The response of the ring-imaging Cherenkov detectors in simulation is known to not match the response in data, so the corresponding particle identification (PID) variables in the simulation samples are corrected based on calibration data samples of D * + → D 0 π + , D 0 → K − π + decays for kaons and pions, and Λ + c → pK − π + decays for protons [33].The correction technique transforms the variables such that they match the data distributions, accounting for dependence on track kinematics and event multiplicity, The PID variables for Λ + c decay products are included in the first BDT algorithm, and are not used further in the selection.For the tracks originating directly from the X b decay, kaon and pion candidates are distinguished with requirements in the two-dimensional plane of the PID variables that characterise how well they fit each of those particle hypotheses.These requirements are optimised as described below, and are required to be mutually exclusive so that no candidate can appear in more than one final state.To select antiprotons in the B − → Λ + c ph − decays, a requirement on the consistency with the proton hypothesis is imposed.In order to ensure that reasonable PID information is available, fiducial requirements are imposed on the kinematics of all final-state tracks: 2.5 < p K,π < 100 GeV/c, 9.5 < p p < 100 GeV/c and 2 < η p,K,π < 5.In addition, all tracks are required to be inconsistent with being muons, based on information from the muon system.
The requirement on the second BDT output is optimised using the following figure of merit (FOM) [36], where B is the expected number of combinatorial background events in the signal region (5770 < m(X b ) < 5830 MeV/c 2 ), obtained by extrapolating yields in data in the upper mass sideband (5830 < m(X b ) < 6800 MeV/c 2 ), and ϵ sig is the signal efficiency determined from simulation.The requirements on the kaon and pion identification variables are optimised using where ϵ misID i is the misidentification rate for each potential cross-feed background i, taken from simulation with PID corrections applied, and r i = B(misID i)/B(sig) is an estimate of the branching fraction of the cross-feed background, B(misID i), relative to that of the signal mode, B(sig).The optimisation of the requirements on the second BDT output and the PID variables is iterated to obtain the best overall performance.The optimised requirements on the second BDT output reject around 99% of background while retaining 40-50% of signal depending on the final state and data-taking period.The Λ + c K − π − and Λ + c pK − samples contain misreconstructed candidates where the K − from the Λ + c decay is swapped with that directly from the X b decay.These are vetoed by removing candidates that satisfy 2260 < m(p ) < 2310 MeV/c 2 , where the subscript Λ + c indicates that the particle is one of the Λ + c decay products while the absence of a subscript indicates that the particle originates directly from the X b decay.Backgrounds also occur due to which are considered a background to this analysis, as well as cases where the extra track does not originate from the same b-hadron decay.Vetoes that remove candidates with 5500 < m(Λ + c π − ) < 5700 MeV/c 2 and 5500 < m(Λ + c K − ) < 5700 MeV/c 2 are applied in the relevant final states to remove these backgrounds.The loss in signal efficiency due to these veto requirements is considered as a source of systematic uncertainty.
Potential backgrounds from X b decays to the same final-state particles without intermediate Λ + c baryons are investigated by inspecting the X b -candidate mass distribution for candidates with Λ + c -candidate masses both below and above the region populated by signal.No significant contribution is visible and therefore no additional selection requirements are imposed to remove such backgrounds.
Backgrounds are expected from b-hadron decays to final states of Λ + c h − h ′− plus one or more low-momentum particles, subsequently referred to as partially reconstructed backgrounds.These include b-hadron decays to Σ c +(++) h − h ′− final states with Σ c +(++) → Λ + c π 0(+) .Similar arguments hold for partially reconstructed backgrounds to the B − → Λ + c ph − channels.Due to the missing particles, these backgrounds tend to peak at X b -candidate mass values below the signal peaks.In order to minimise dependence on precise modelling of these backgrounds, the mass ranges retained for the fits are 5550 < m(X b ) < 6800 MeV/c 2 for the signal channels and 5200 < m(X b ) < 6000 MeV/c 2 for the B − channels.
A small fraction (< 1%) of selected events contain more than one candidate.All are retained, and the presence of multiple candidates is considered as source of systematic uncertainty.

Determination of signal yields
Fits to the X b -candidate mass distributions are performed to determine the yields of the channels under study.Simultaneous unbinned extended maximum-likelihood fits are implemented to allow parameters to be shared between the samples.One simultaneous fit is performed for the three signal final states, and another for the two B − modes, with separate fits for the Run 1 and Run 2 samples.The fits include components to describe the signals, and contributions due to cross-feed, partially reconstructed decays and combinatorial background.
The signal components are described using the sum of two Crystal Ball (CB) functions [37], with a common central value and width, and independent tails to each side of the peak.The tail parameters and the relative fraction of the two CB functions are determined from fits to simulation samples and are fixed in the fit to data.
Cross-feed between the signal modes occurs due to misidentification of one of the particles originating directly from the X b decay.The X b -candidate mass distributions of these backgrounds depend on the momentum and misidentification probability of that particle.Therefore, these shapes depend (albeit fairly weakly) on the phase-space distribution of the misidentified decay.For the B − → Λ + c pπ − channel there are previous investigations of its resonant structure [11,12]; however these were made with samples much smaller than available in this study.Since the analysis of the B − modes is not blind, the phase-space distributions observed in data are used to weight the simulation and obtain a best estimate of the shape of the cross-feed background components.On the other hand, since the analysis is performed blind for the Ξ − b and Ω − b baryon decays, their phase-space distributions are taken from simulation without any phase-space weighting.This procedure is considered as a source of systematic uncertainty.The distribution for each sample under each K ↔ π misidentification hypothesis is fitted with a double CB function.Misidentification of protons as pions or kaons, and vice versa, does not lead to background contributions in the fitted mass ranges.
Residual partially reconstructed backgrounds due to the decays are described as a single component in the fit.The shape is investigated with a simplified simulation [38], and is modelled with a superposition of Gaussian kernels [39].Partially reconstructed backgrounds to the B − modes are found to be negligible in the chosen mass fit range.The combinatorial background is modelled with an exponential shape.The slope of the exponent in each final state is a free parameter in the fit.
The yields of all signal, partially reconstructed, and combinatorial components are free parameters in the simultaneous fit to the three signal final states.The yields of cross-feed backgrounds are constrained relative to the signal yield of the same decay when it is correctly reconstructed, based on knowledge of the relative misidentification rates and correct identification efficiencies.The shapes of all components are fixed, except that the slope parameters of the exponential function used to describe combinatorial background in each final state are free to vary independently of each other.In addition, a peak position offset and a width scale factor, which account for possible discrepancies between data and simulation and that are shared between all components described by a double CB function, are free parameters of the fit.In total there are seventeen free parameters in each simultaneous fit to the signal final states: six signal yields, three combinatorial background yields and three slope parameters, three partially reconstructed background yields, and the peak shift and width scale parameters.There are, in addition, ten Gaussian constraints that control the yields of the cross-feed backgrounds.
A similar configuration is used for the simultaneous fit to the two B − modes.In this case there are eight free parameters: two signal yields, two combinatorial background yields and two slope parameters, as well as the peak shift and width scale parameters.In addition, there are two Gaussian constraints that control the yields of the cross-feed backgrounds.
The performance of each fit is evaluated using ensembles of pseudoexperiments.For the fit to the signal final states, this is done prior to unblinding with guessed signal yields, and subsequently repeated with the yields observed in data.For the B − modes the yields observed in data are used in the generation of the pseudoexperiments.In all cases, the fits are found to return the signal yields with reasonable coverage and without large biases; small possible biases are accounted for as a source of systematic uncertainty.
The X b -candidate mass distributions for the three signal final states for both Run 1 and Run 2 are shown in Figs. 2 and 3 with linear and logarithmic y-axis scales, respectively, with results of the fits superimposed.The mass distributions and results of the fits to the two B − modes are similarly shown in Figs. 4 and 5. Pronounced peaks corresponding to Ξ − b → Λ + c K − π − and B − → Λ + c pK − decays, as well as the previously established B − → Λ + c pπ − decay, are observed.Smaller, but still clear, peaks corresponding to are also visible.The fitted signal yields are given in Table 1.The uncertainties on all fitted yields are found to be symmetric to within 20% and therefore symmetrised uncertainties are quoted in Table 1; the full likelihood curves are, however, used to set upper limits on channels with yields that are not significantly different from zero, as discussed below.The significance of each of the signals is discussed in Sec. 7.

Efficiency
The total efficiency includes effects from the geometrical acceptance of the LHCb detector, the reconstruction efficiency, and the probabilities to pass both trigger stages and offline selection requirements.All these effects are estimated using simulation, with data-driven corrections applied where appropriate.In particular, the efficiencies of the PID require-  ments can be obtained directly from the simulation samples, since the corresponding variables are corrected using calibration data.Separate efficiency maps for each channel are obtained for each year of data taking, using square Dalitz-plot coordinates [27] to represent the phase space of the three-body decay.The use of candidate-by-candidate efficiencies, depending on position in the phase space, ensures that the distribution of decays in data is accounted for when evaluating results using Eq. ( 1).The simulated samples are additionally weighted to correct for two sources of data-simulation disagreement.The first of these is due to the Ω − b lifetime, where improved measurements [1,40,41] are available compared to the value used when generating the simulation.A weighting factor is applied to the Ω − b signal samples based on the difference in the decay-time distribution associated to updating the lifetime from 5600 5800 6000 6200 6400 6600 6800 ]

Systematic uncertainties
As seen in Eqs. ( 1) and ( 2), each calculation of a ratio of fragmentation fractions times branching fractions or of a production asymmetry depends essentially on two inputs: the signal weights obtained from the X b -candidate mass fits and the efficiency maps as a function of phase-space position.Sources of systematic uncertainty are hence considered as being related to one of these inputs, although in some cases the choice of which to relate to is arbitrary.Since the B − → Λ + c pπ − control channel has the same topology as the signal channels, and the selection of different channels differs only by PID requirements, several potentially large systematic uncertainties are reduced in the ratios of fragmentation fractions times branching fractions.
Systematic uncertainties on the yields arise from the model chosen to fit the X bcandidate mass distributions, from fixed parameters in the fit model, and from the dependence of the shape of the misidentified Ξ − b → Λ + c K − π − component on its a priori unknown phase-space distribution.These uncertainties are evaluated by assessing the impact on the results of using alternative models for each of the fit components, varying fixed parameters within their uncertainties, and obtaining the misidentified Ξ − b → Λ + c K − π − shape from simulation weighted to match the phase-space distribution observed in data, respectively.The alternative models considered are a combination of Johnson functions and Gaussian shapes for the signal components, second-order polynomials for the combinatorial background, shapes obtained with kernel density estimates for the cross-feed components and modified kernel density estimates for the partially reconstructed backgrounds.Systematic uncertainties due to potential biases in the fit are determined from large ensembles of pseudoexperiments generated according to the yields obtained from, and the probability density functions used in, the fit to the X b -candidate mass distributions.Additionally, effects due to the presence of multiple candidates and from the impact of vetoing regions of the phase space are considered as being related to the yields.The former is evaluated from the impact on the results of allowing at most one candidate per pp bunch crossing, and the latter from changing the width of the veto window.
Systematic uncertainties on the efficiency maps occur due to the finite size of the simulation samples and potential inaccuracies in the simulation description of the detector and its response.The former is evaluated by varying the efficiency maps within their uncertainties, while for the latter several possible data-simulation inconsistencies are considered.These include the amount of detector material [42,43], the efficiency of the tracking algorithms [42] and the response of the hardware trigger [44].The impact of possible effects due to inaccuracies in the simulation description of the detector response are evaluated by weighting the simulated samples to correct the distributions of key variables.Uncertainties also arise due to the procedure to correct PID variables in simulation, and the granularity over phase space of the efficiency maps.The effect of correcting the Λ + c → pK − π + decay model in simulation to match the latest measurement [45] is also considered as a systematic uncertainty.Uncertainty in the b-baryon production kinematics is accounted for by considering the impact on the results of changing the weighting procedure to account for only the p T distribution, rather than both the p T and η distributions.Finally, for channels in which the signal yield is not significant, so that the candidate-by-candidate efficiency correction procedure is not reliable, an uncertainty is assigned due to the unknown phase-space distribution.
Systematic uncertainties on the production asymmetries are evaluated in the same way as for the ratios of fragmentation fractions and branching fractions, which is possible as the measurements are made in very similar ways, as seen in Eqs. ( 1) and ( 2).This includes systematic uncertainties associated to detection asymmetries, which arise because of the combined effect of different interaction probabilities of particles and antiparticles and the potential mismatch in the material description between data and simulation.
For the majority of measurements, the systematic uncertainties are significantly below the statistical uncertainties, as seen in Table 2. Some measurements involving unobserved modes have either or both fit model and veto-related systematic uncertainties that are comparable to the statistical uncertainty, which is understood due to the size of the backgrounds in the corresponding final states.Additionally, as the statistical precision of the Run 2 measurement of is around 5%, it is unsurprising that its systematic uncertainties are not negligible.In particular, effects due to uncertainty in the b-hadron production kinematics are significant due to the rapid variation in efficiency with transverse momentum.These effects, however, tend to cancel in ratios where decays of the same b hadron are considered in both numerator and denominator.
In order to combine results from Run 1 and Run 2, the effects of the systematic uncertainties are considered as being either 100% correlated or completely uncorrelated between the two data-taking periods.The dominant systematic uncertainties discussed above are all correlated between Run 1 and Run 2.

Results
Results for the ratios of fragmentation fractions and branching fractions are obtained for all channels studied relative to the B − → Λ + c pπ − normalisation channel.In addition, since significant signals are observed for , the relative branching fractions of the other Ξ − b and Ω − b decays, respectively, to those are also determined.These results are given separately for Run 1 and Run 2 in Table 2.
Results for ratios of branching fractions of the same b hadron do not depend on pp collision centre-of-mass energy, and therefore the Run 1 and Run 2 results can be combined.This is done taking correlations of systematic uncertainties into account, using the method described in Ref. [46], to obtain Table 2: Ratios of branching fractions, multiplied where appropriate by ratios of fragmentation fractions, separately for Run 1 and Run 2. In each result the first uncertainty is statistical and the second is systematic.

Decay Run 1 Run 2 Combined Statistical uncertainties only
where the first uncertainty is statistical and the second is systematic.
The significance of the signal in each mode is determined using Wilks' theorem, i.e. as 2∆(NLL) where 2∆(NLL) is double the difference in negative log-likelihood obtained in the best fit (with the signal yield free) and in another fit where the relevant signal yield is fixed to zero.Significances are determined separately for Run 1 and Run 2, using the likelihood functions obtained from the fits convolved with Gaussian functions to account for relevant sources of systematic uncertainty, and then combined taking correlations of systematic uncertainties into account.As the far in excess of 5 σ significance, they are not quantified further.Similarly, as the statistical significances of the are found to be below 3 σ, they are not quantified further.For the , the combined significances are found to be above 5 σ, as shown in Table 3.
For the modes where no significant signal is observed, upper limits on the relative ratios of fragmentation fractions times branching fractions are reported.The limits are obtained for the combined Run 1 and Run 2 ratios, so in cases where the decays in the numerator and denominator involve different b hadrons, the ratio of fragmentation fractions in the combination should be considered as an effective average of the values for Run 1 and Run 2 (dominated by the latter due to the larger sample size).A combined likelihood for Run 1 and Run 2 is obtained, taking correlations of systematic uncertainties into account, and upper limits at 90% (95%) confidence level (CL) are determined by integrating the likelihood in the physical region of non-negative relative fragmentation fractions times branching fractions.The combined limits are

B Ω
In channels with large enough yields, namely  11, with statistical uncertainties only (the majority of systematic effects are correlated bin-to-bin and therefore not relevant to the dependence).The limited statistical precision precludes any strong conclusion, except that the Run 1 result is higher than that of Run 2, as also seen in Table 2.As a cross-check, the relative branching fraction of B − → Λ + c pK − and B − → Λ + c pπ − decays is shown in the same kinematic bins in Fig. 12.No dependence is seen, as expected.
Similarly, the Ξ − b production asymmetry is studied in intervals of b-hadron kinematics using the Ξ − b → Λ + c K − π − channel, with the B − → Λ + c pπ − mode used as a control channel.The results are shown, with statistical uncertainties only, in Figs. 13 and 14 respectively.No strong dependence of the Ξ − b production asymmetry with either p T or η is observed, and all results are consistent with zero asymmetry within uncertainties.This is as expected since although a significant Λ 0 b production asymmetry, with a potential dependence on kinematics, has been observed [43], the effect is at the level of 2% or smaller and hence a comparably sized asymmetry would not be detectable in the current analysis.The results

Conclusion
The decays of the LHC.The Ξ − b → Λ + c K − π − mode is observed with large significance, and the are also observed with over 5 σ significance.Ratios of fragmentation fractions times branching fractions are determined for all channels relative to the control mode B − → Λ + c pπ − , and additionally ratios of branching fractions are determined relative to the most significant decay for each b baryon.The Ξ − b → Λ + c K − π − mode is also used to study the variation with kinematics of the Ξ − b /B − production ratio, and the Ξ − b production asymmetry.In both cases the variation is found to be consistent with zero; the integrated Ξ − b production asymmetry within the LHCb acceptance is also found to be consistent with zero.Furthermore, the B − → Λ + c pK − decay is observed with large significance, and its branching fraction measured relative to that of the B − → Λ + c pπ − control mode.These results significantly increase the currently available experimental knowledge of Ξ − b and Ω − b decays.The decay modes observed can in future be used to improve knowledge of the processes behind b-baryon production in high-energy pp collisions, including production asymmetries, and to study charm baryon spectroscopy.

Figure 2 :
Figure 2: Mass distributions of the signal channels with results of the fit superimposed: (top) Λ + c K − K − , (middle) Λ + c K − π − and (bottom) Λ + c π − π − final states for (left) Run 1 and (right) Run 2. The different components in the fit are shown as indicated in the legends, where the signal components include independent contributions from Ξ − b and Ω − b decays.

Figure 3 :
Figure 3: Mass distributions of the signal channels with results of the fit superimposed: (top) Λ + c K − K − , (middle) Λ + c K − π − and (bottom) Λ + c π − π − final states for (left) Run 1 and (right) Run 2 (log scale).The different components in the fit are shown as indicated in the legends, where the signal components include independent contributions from Ξ − b and Ω − b

Figure 4 :
Figure 4: Mass distributions of the B − modes with results of the fit superimposed: (top) Λ + c pK − and (bottom) Λ + c pπ − final states for (left) Run 1 and (right) Run 2. The different components in the fit are shown as indicated in the legends.

Table 1 :
Fitted yields for Ξ − b and Ω − b decays to the Λ + c h − h ′− final states, and for B − → Λ + c ph − decays.

Figure 5 :
Figure 5: Mass distributions of the B − modes with results of the fit superimposed: (top) Λ + c pK − and (bottom) Λ + c pπ − final states for (left) Run 1 and (right) Run 2 (log scale).The different components in the fit are shown as indicated in the legends.

Figure 6 :
Figure 6: Distributions of (top) p T and (bottom) η for B − → Λ + c pπ − decays in (left) Run 1 and (right) Run 2, obtained from background-subtracted data and compared to the uncorrected distributions from simulation.

Figure 7 :
Figure 7: Distributions of (top) p T and (bottom) η for Ξ − b → Λ + c K − π − decays in (left) Run 1 and (right) Run 2, obtained from background-subtracted data and compared to the uncorrected distributions from simulation.

Figure 11 :Figure 12 :
Figure 11: Variation of the ratio of fragmentation fractions times branching fractions of Ξ − b → Λ + c K − π − and B − → Λ + c pπ − decays with (left) b-hadron transverse momentum and (right) pseudorapidity, separately for Run 1 and Run 2, with statistical uncertainties only.