Measurement of K*(892)± production in inelastic pp collisions at the LHC

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Measurement of K*(892)± production in inelastic pp collisions at the LHC © 2022 European Organization for Nuclear Research Published version ALICE Collaboration ALICE Collaboration. (2022). Measurement of K*(892)± production in inelastic pp collisions at the LHC. Physics Letters B, 828, Article 137013. https://doi.org/10.1016/j.physletb.2022.137013 2022 Physics Letters B 828 (2022) 137013 Contents lists available at ScienceDirect Physics Letters B www.elsevier.com/locate/physletb Measurement of K∗(892)±production in inelastic pp collisions at the LHC .ALICE Collaboration a r t i c l e i n f o a b s t r a c t Article history: Received 27 May 2021 Received in revised form 27 February 2022 Accepted 9 March 2022 Available online 16 March 2022 Editor: M. Doser The first results on K∗(892)±resonance production in inelastic pp collisions at LHC energies of √s=5.02, 8, and 13 TeV are presented. The K∗(892)±has been reconstructed via its hadronic decay channel K∗(892)±→K0 S+π±with the ALICE detector. Measurements of transverse momentum distributions, pT-integrated yields, and mean transverse momenta for charged K∗(892) are found to be consistent with previous ALICE measurements for neutral K∗(892) within uncertainties. For pT>1 GeV/cthe K∗(892)±transverse momentum spectra become harder with increasing centre-of-mass energy from 5.02 to 13 TeV, similar to what previously observed for charged kaons and pions. For pT<1 GeV/cthe K∗(892)±yield does not evolve significantly and the abundance of K∗(892)±relative to Kis rather independent of the collision energy. The transverse momentum spectra, measured for K∗(892)±at midrapidity in the interval 0<pT<15 GeV/c, are not well described by predictions of different versions of PYTHIA 6, PYTHIA 8 and EPOS-LHC event generators. These generators reproduce the measured pT- integrated K∗±/K ratios and describe well the momentum dependence for pT<2 GeV/c. ©2022 European Organization for Nuclear Research. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Funded by SCOAP3. 1. Introduction Measurements of identified hadron production in high-energy proton-proton interactions provide key observables to characterize the global properties of the collisions. Particle production at high collider energies originates from the interplay of perturbative (hard) and non-perturbative (soft) Quantum Chromodynamic (QCD) processes. Soft scattering processes and parton shower hadronization dominate the bulk of particle production at low transverse momenta and can only be modeled phenomenologically. At the Large Hadron Collider (LHC) [1], the small Bjorken x regime is probed and contributions from hard-scattering processes are more relevant with increasing centre-of-mass energy. This produces a hardening of the transverse momentum spectra, as already observed in Refs. [2,3]. Measurements of strange hadrons such as the K∗(892) vector meson at different collision energies allow for testing and tuning perturbative QCD and low-transverse momentum phenomenological calculations [4–6], including strangeness production. In the following, K∗0denotes K∗(892)0and K∗(892)0, K∗± stands for K∗(892)+and K∗(892)−, while K∗indicates K∗0and K∗±. E-mail address: alice -publications @cern .ch. In heavy-ion collisions, due to their short lifetimes comparable with the lifetime of the hadronic phase of the system [7], resonances such as K∗(τ≈4fm/c) are sensitive probes of the dynamical evolution of the fireball. Re-scattering and regeneration in the hadron gas may change the number of resonances reconstructed via the hadronic decay channels compared to those predicted by thermal models at the chemical freeze-out, i.e. when the inelastic interactions stop. The K∗vector meson and its corresponding ground state, the K, have an identical quark content. They differ only in mass, lifetime and relative orientation of their quark spins. Therefore, the K∗/Kratio is an ideal observable to study the K∗properties and the freeze-out conditions in relativistic heavy-ion collisions. The integrated yield ratio K∗0/K exhibits a suppression with respect to pp collisions, which increases with the centrality of the collisions [8–11]. This could be explained as due to the dominance of re-scattering effects of K∗0decay products over regeneration processes in the hadronic phase of the collisions. Hints of the suppression of K∗0/K were observed also in highmultiplicity p–Pb and pp collisions [12–14]at LHC energies, suggesting the possible presence of re-scattering effects and thus of a hadronic phase with a short but finite lifetime in small collision systems. The observed multiplicity-dependent suppression should therefore be validated by measurements with an increased precision. This is particularly important for small systems such as pp and p–Pb because the K∗0/K ratios, measured in the highest and lowest multiplicity event classes differ by less than 2σ[12–14], with the largest uncertainty in the rahttps://doi.org/10.1016/j.physletb.2022.137013 0370-2693/©2022 European Organization for Nuclear Research. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Funded by SCOAP3. ALICE Collaboration Physics Letters B 828 (2022) 137013 tio being relative to the K∗0yield measurement. In this work, the K∗/K ratio is studied with increased precision by measuring the production yield of K∗± in pp collisions with the ALICE detector [15]. The production of charged and neutral K∗vector mesons is expected to be comparable. Indeed, they have a similar quark composition, K∗(892)+=(us), K∗(892)0=(ds), K∗(892)−=(us) and K∗(892)0=(ds), and their masses differ by about 0.004 GeV/c2, being M(K∗±) = 0.89166 ±0.0026 GeV/c2[16] and M(K∗0) = 0.89581 ±0.0019 GeV/c2[16]. At LHC energies, the measurement of the K∗± and K∗0strange vector mesons is quite challenging. These are reconstructed via their hadronic decay into a charged pion and a kaon: a neutral kaon for K∗± and a charged kaon for K∗0. Because of the different strategies used for their identification in ALICE, K0 Sare measured with a lower systematic uncertainty than charged kaons [3,13]. In this paper, transverse momentum (pT) distributions of K∗± resonances at midrapidity (|y|<0.5) are presented for the first time for inelastic pp collisions at the LHC. The evolution of the pTdistributions with the energy was investigated by studying pp collisions at the centre-of-mass energies of √s=5.02, 8, and 13 TeV. The similarity of the charged and neutral K∗production was checked by comparing K∗± results with existing K∗0measure- ments at the same collision energy [3,11,17]. These measurements are a useful probe of strangeness production and provide input to tune Monte Carlo event generators such as PYTHIA and EPOSLHC [4–6]as a function of collision energy. Furthermore, the measurements in inelastic pp collisions at √s=5.02, 8, and 13 TeV reported in this paper serve as reference data to study nuclear effects in p–Pb and Pb–Pb collisions. The paper is organized as follows. In Sec. 2the ALICE experimental setup is described, focusing on the detectors employed in the analysis presented here. Details on the event, track and particle identification as well as on the corrections applied to the measured raw yields and estimation of systematic uncertainties are discussed in Sec. 3. In Sec. 4, the results on the production of K∗± resonances are shown. These include the transverse momentum spectra, the mean transverse momenta, the per-event pT-integrated particle yields and the K∗±/K = (K∗+ +K ∗−)/(K++K −) ratio as a function of the collision energy. All these observables are compared with similar results for K∗0. The comparison of the pTspectra with different event generator (PYTHIA6, PYTHIA8 and EPOS-LHC) predictions is also presented. In Sec. 5results are summarized and conclusions are drawn. 2. Experimental setup A detailed description of the ALICE detector and its performance can be found in Refs. [15,18]. The sub-detectors used for the analysis presented in this paper are the Inner Tracking System (ITS) [15], the Time Projection Chamber (TPC) [19], and the V0 detectors [20]. All tracking detectors are positioned in a solenoidal magnetic field B=0.5T parallel to the LHC beam axis. Charged particle tracks are reconstructed by the ITS and the TPC. The ITS is the innermost barrel detector consisting of six cylindrical layers of high-resolution silicon tracking detectors. The innermost layers consist of two arrays of hybrid Silicon Pixel Detectors (SPD) located at an average radial distance rof 3.9 and 7.6 cm from the beam axis and covering |η|<2.0 and |η|<1.4, respectively. The SPD is used to reconstruct the primary vertex (PV) of the collisions, which is found as a space point to which the maximum number of tracklets (track segments defined by pairs of points, one point in each SPD layer) converges. The outer layers of the ITS are composed of two layers of silicon drift and two layers of silicon strip detectors, with the outermost layer positioned at r=43cm. The TPC is the main tracking device of ALICE. It is a large volume (90 m3) cylindrical drift chamber with Table 1 Number of minimum bias events after event selection (NMB), integrated luminosity (Lint), the trigger selection efficiency (εtrig), and the primary vertex reconstruction efficiency (εvertex) for the analyzed data sets. The uncertainty on εvertex is lower than 0.1%. √s(TeV) NMB (107)Lint (nb−1)εtrig εvertex 5.02 10.87 2.12 ±0.05 0.757 ±0.019 0.958 8.0 6.99 1.25 ±0.03 0.772 ±0.021 0.972 13.0 5.32 0.92 ±0.02 0.745 ±0.019 0.931 radial and longitudinal dimension of about 85 <r<250 cm and −250 <z<250 cm, respectively, covering for full-length tracks a pseudorapidity range of |η|<0.9 over the full azimuth. The endcaps of the TPC are equipped with multiwire proportional chambers segmented radially into pad rows. Together with the measurement of the drift time, the TPC provides three dimensional space point information, with up to 159 samples per track. The resolution on the position is 1100–800 μm on the transverse plane and 1250–1100 μm along z. Charged tracks originating from the primary vertex can be reconstructed down to pT≈0.1 GeV/c[18]. The TPC enables charged particle identification (PID) via the measurement of the specific ionization energy loss (dE/dx) with a resolution of about 5.2% [18]at low transverse momentum. A separation between π-K and K-p at the level of two standard deviations is possible for pT<0.8 GeV/cand 1.6 GeV/c, respectively. The V0 detectors are two forward scintillator hodoscopes employed for triggering and beam background suppression. They are placed along the beam axis on each side of the nominal interaction point (IP) at z=340 cm and z=−90 cm, covering the pseudorapidity regions 2.8<η<5.1 (V0A) and −3.7 <η<−1.7 (V0C), respectively. The pp data at √s=5.02 and 13 TeV used in this paper were collected in 2015 while data at √s=8TeV were collected in 2012. The data were collected with a minimum bias trigger requiring a hit in both V0 detectors, in coincidence with the arrival of proton bunches from both beam directions. The analyzed data are low pile-up samples in which the average number of interactions per bunch crossing are μ=0.019 ± 0.009, 0.02 ±0.01 and 0.068 ±0.003 for collisions at √s=5.02, 8, and 13 TeV, respectively. Contamination from beam-gas events is removed offline by using timing information from the V0 detector, which has a time resolution better than 1ns. The events in which pile-up or beam-gas interaction occurred are also rejected by exploiting the correlation between the number of SPD hits and the number of SPD tracklets, as discussed in detail in Ref. [18]. The events selected from the analysis are required to have a reconstructed primary vertex with its position along the beam axis being within 10 cm with respect to the nominal interaction point (the centre of the ALICE barrel). The events containing more than one reconstructed vertex are tagged as pile-up occurring within the same bunch crossing and discarded for the analysis. The size of the analyzed samples after selection and the corresponding pp integrated luminosities are given in Table 1. In the same table, the primary vertex reconstruction efficiency εvertex and the trigger selection efficiency εtrig are also reported. For each energy, the εtrig value, mainly defined by the charged particle multiplicity of the collision, is the ratio between the V0-triggered cross section [21–23] and the inelastic cross section [24] and the εvertex is the fraction of V0-triggered events for which a primary vertex is reconstructed. 3. Data analysis The K∗(892)±is a short-lived particle and its decay vertex cannot be distinguished from the primary collision vertex. It is reconstructed in ALICE via its main decay channel K∗±→K0 S+ π±, 2 ALICE Collaboration Physics Letters B 828 (2022) 137013 Table 2 The selection criteria parameters for K0 Scandidates. DCA stands for distance of closest approach, PV means primary vertex, θPA is the pointing angle, LmK0 S/pis the proper lifetime. The competing V0rejection window is 1.1157 ± 0.0043 GeV/c2while for the mass of the π+π−pairs the window is   mK0 S−mπ+π−  <4σmK0 S. K0 Sselection criteria Value Pion dE/dx(σ)<5 DCA of daughter to PV (cm/c)>0.06 DCA between daughters (σ)<1 Cosine of θPA >0.97 V0radius (cm) >0.5 Proper lifetime LmK0 S/p(cm) <20 Competing V0rejection window (GeV/c2)±0.0043 Mass K0 Swindow (σ)±4 Rapidity |y|<0.8 which has a branching ratio (B.R.) of (33.3 ±0.003)% [16], taking into account the B.R. of K∗± →K0+ π±decay and the probability of K0to be into a K0 Sstate. The K0 Sis reconstructed by exploiting its characteristic weak decay topology (K0 S→π++ π−) into two oppositely charged particles (V0topology) with branching ratio (69.2 ±0.05)% [16]. 3.1. Pion and K0 Sselection Particle identification for charged pions originating from the primary and secondary vertices (“primary and secondary pions”) is applied on a sample of high-quality tracks reconstructed with the TPC and the ITS. Informations from ITS are required only for primary tracks. The primary and secondary tracks reconstructed with the TPC are required to have crossed at least 70 readout rows out of a maximum 159. They are also requested to avoid large gaps in the number of expected tracking points in the radial direction. This is achieved by ensuring that the number of clusters expected, based on the reconstructed trajectory and the measurements in neighboring TPC pad rows, do not differ by more than 20%. Particles are required to have pT>0.15 GeV/cand to be located in the pseudorapidity range |η|<0.8 to avoid edge effects in the TPC acceptance. Furthermore, tracks of particles possibly originating from weak decays of pions and kaons are rejected when a kink in the track is observed [18]. Primary tracks are required to be associated with at least one cluster in the SPD and the goodness-of-fit values χ2per cluster of the track fit in the ITS and in TPC are restricted in order to select high-quality tracks. Primary tracks are required to have a distance of closest approach (DCA) to the primary vertex lower than 2cm along the beam axis and 7σin the transverse plane, where σ= (0.0015+0.0050 pT−1.1)cm with pTin units of GeV/c. Secondary tracks are required to have a DCA to the primary vertex larger than 0.06 cm. Selected pion candidates are identified by requiring that the specific ionization energy loss dE/dxmeasured in the TPC lies within nstandard deviations (σTPC) from the specific energy loss expected for pions, with nequal to 3 or 5 for primary and secondary pions, respectively. The selection criteria used for the K0 Sreconstruction are listed in Table 2. Candidates K0 Sare in the rapidity range |y|<0.8. The distance of closest approach between positively and negatively charged tracks is required to be smaller than one standard deviation with respect to the ideal value of zero and the cosine of the pointing angle (θPA), which corresponds to the angle between the V0momentum and the line connecting the secondary to the primary vertex, is required to be larger than 0.97. Only those V0candidates located at a radial distance larger than 0.5 cm (V0radius) are used in this analysis. Competing V0rejection is also applied: the V0mass is recalculated assuming that one of the pions is a (anti-)proton, and the V0candidates (about 2%) are rejected if their mass is compatible with the mass within ±0.0043 GeV/c2, which is about three times the typical mass resolution for the reconstructed in ALICE [25]. In addition, K0 Scandidates with a proper lifetime larger than 20 cm/care rejected to remove combinatorial background from interactions with the detector material. The proper lifetime is estimated as LmK0 S/p, where Lis the linear (3D) distance between the primary vertex and the V0decay vertex, pis the total momentum of K0 S, and mK0 S= 0.497611 GeV/c2is the nominal K0 Smass [16]. Finally, the invariant mass of π+π−pairs is required to be compatible with the nominal K0 Srest mass within ±4σmK0 S, with the K0 Smass resolution value increasing smoothly with the transverse momentum, from ≈3.5×10−3GeV/c2at pT≈0 to ≈6.2×10−3GeV/c2at pT=10GeV/ c. ALICE has measured K∗0exploiting its decay into K±+ π∓[3,9– 12,17,26], with pions and kaons reconstructed as primary particles and identified using energy loss and time-of-flight measurements. The crucial difference in the K∗± and K∗0reconstruction is the charged and neutral kaon identification. In particular, the neutral kaon reconstruction efficiency is larger for pT<0.2 GeV/cand for pT>2GeV/ c. At low pT, primary charged kaon detection depends on the tracking efficiency with a threshold of about 0.1 GeV/c, whereas at high pTthe larger efficiency in neutral kaon reconstruction is mainly connected to a loose charged particle selection based on the expected specific energy loss. 3.2. Signal extraction The raw yield of the K∗± is extracted from the same-event K0 Sπ±invariant mass distribution in different pTintervals between 0 and 15 GeV/c. The nominal mass value [16]is assigned to the K0 Swhen the K0 Sπ±invariant mass is estimated. The shape of the uncorrelated background is estimated using the invariant mass distribution of K0 Sπ±pairs selected from different events (event mixing method). To avoid any mismatch due to different acceptances and to ensure a similar event structure, particles from events with similar vertex positions along z(z<1cm) and track multiplicities n(n<5) are mixed. To reduce statistical uncertainties each event is mixed with 9 others. The mixed-event distribution is then normalized to the same-event distribution in the mass region 1.1 <MK0 Sπ±<1.2 GeV/c2and subtracted from the same-event distribution in each pTbin. The mixed-event background normalization range is varied for the study of systematic uncertainties. The K0 Sπ±invariant mass distributions in different pTranges obtained for the different collision energies are shown in the left panels of Fig. 1. Similar to previous K∗0analyses [3,9–12,17,26] the uncorrelated mixed-event background is subtracted from the same-event invariant mass distribution. The resulting distributions exhibit a characteristic peak on top of a residual background, as reported in the right panels of Fig. 1. The latter is due to the presence of correlated pairs from jets, multi-body decays of heavier particles and misreconstructed resonance decays. The resulting distribution is fitted with a combination of the non-relativistic Breit-Wigner function to describe the signal peak and a FBG function to describe the residual background. The fit, based on the minimization of the χ2, was performed according to the following expression: dN dMK0 Sπ±=C 2π 0 MK0 Sπ±−M02 +2 0 4 +FBG MK0 Sπ±(1) where M0and 0are the mass and the width of the K∗± [16]. The Cparameter is the integral of the peak function from 0 to ∞. The detector mass resolution for the reconstruction of K∗± is negligible 3 ALICE Collaboration Physics Letters B 828 (2022) 137013 Fig. 1. (Left panels) The K0 Sπ±invariant mass distributions at |y|<0.5 in pp collisions at √s=5.02, 8, and 13 TeV. The background shape estimated by the event-mixing technique is shown with empty red circles. Statistical uncertainties are shown with error bars. (Right panels) The K0 Sπ±invariant mass distributions in pp collisions at √s=5.02, 8, and 13 TeV after background subtraction. The solid red curve is the result of the fit with Eq. (1); the dashed red line describes the residual background distribution given by Eq. (2). Statistical uncertainties are shown with error bars. compared to its natural width, 0= (0.0508 ±0.0009) GeV/c2[16], and it is therefore not included in the peak model. The mass and width of K∗± were found to be compatible with the values reported in [16]. For the measurement of the yields, the width of K∗± was fixed to its natural value. Fits were performed with the width kept as a free parameter or fixed at 0.0517 or 0.0499 GeV/c2to estimate the systematic uncertainty. The shape of the correlated background in the invariant mass distribution of K0 Sπ±pairs is studied using the same samples of simulated events described in Sect. 3.3 that were used to estimate the Acceptance×Efficiency corrections. The produced particles and their decay products are propagated through the ALICE detector using GEANT3 [27]. Invariant mass distributions for K0 Sπ+and K0 Sπ−pairs are accumulated after applying the same event, track and particle identification selections as in data. The study shows that after subtracting the combinatorial background, the remaining background has a smooth dependence on mass. It is well described by the following function, already used in Refs. [28,29]: FBG MK0 Sπ±=MK0 Sπ±−mπ±+mK0 Sn ×expa+bMK0 Sπ±+cM2 K0 Sπ±(2) 4 ALICE Collaboration Physics Letters B 828 (2022) 137013 Fig. 2. Acceptance×Efficiency as a function of pTfor K∗± mesons, detected by their decay to K0 S+π±, with K0 Sreconstructed by their decay to π++π−. The K0 S→π++π−branching ratio is included in the efficiency estimation. Statistical uncertainties are shown with error bars. where n, a, b, and care fit parameters and mπ±and mK0 Sare the pion and K0 Smasses [16]. Examples of these fits for different pTintervals and different pp collision energies are shown in the right panels of Fig. 1. The typical fitting interval was 0.66 <MK0 Sπ±<1.1 GeV/c2. The K∗± raw yield (Nraw) is determined by integrating the combinatorial background-subtracted invariant mass distribution over the interval 0.79−0.99 GeV/c2, subtracting the integral of the residual background fit function over the same range, and correcting the result to account for the yield outside that range. The yield in the tails is estimated by integrating the non-relativistic Breit-Wigner function from mπ±+mK0 Sto 0.79 GeV/c2and from 0.99 GeV/c2to infinity. This correction to the total yield is about 13%. As an alternative used to estimate the systematic uncertainties, the K∗± yield is also obtained by integrating the peak fitting function in the allowed region (mπ±+mK0 S,∞). 3.3. Efficiency and acceptance To obtain the corrected resonance yields, the convolution between the geometrical acceptance (A) and the resonance reconstruction efficiency (εrec), which takes into account the criteria used to select primary charged pions and K0 S, is determined. The A ×εrec product takes into account also the branching ratio of K0 S→π++π−. For each collision energy, A ×εrec is determined using samples of about 50 million pp events simulated with different Monte Carlo event generators (PYTHIA6-Perugia 2011 tune [4,30], PYTHIA8-Monash 2013 tune [5,31], EPOS-LHC [6]) and a GEANT3-based simulation [27]of the ALICE detector response. The actual positions of the detectors (alignment), maps of dead or noisy elements, and time and amplitude calibrations are used in the reconstruction of real and simulated data. All the parameters taken into account for a careful calibration of the ALICE detector are listed in [18]. The residual differences between data and the sample of Monte Carlo simulation previously described are considered in the systematic uncertainty. For each pTinterval, the A ×εrec is calculated as the ratio Nrec/Ngen, where Nrec is the number of particles reconstructed in the K0 S+π±channel after all event and particle selections, while Ngen is the number of generated mesons decaying in the same channel. Both generated and reconstructed mesons have the rapidity in the range |y|<0.5. In general, the efficiency depends on the shape of the generated particle pTspectrum. Therefore, at the different collision energies, the efficiency for K∗± is estimated re-weighting iteratively the shape of the generated pTspectrum to measured shape. As an example the transverse momentum dependence of A ×εrec is reported in Fig. 2for the √s=5.02TeV sample. 3.4. Yield corrections The differential transverse momentum yield for inelastic pp collisions was calculated as 1 NINEL d2N dpTdy =Nraw NMB ×B.R.×pT×y fSL (A×εrec) ×εtrig ×εvertex.(3) The raw yields are corrected for the resonance branching ratio (B.R. = 33.3%) and A ×εrec in the K0 S+π±channel. Furthermore, these yields were normalized to the number of minimum bias events NMB and corrected for the vertex reconstruction efficiency εvertex as well as for the trigger selection efficiency εtrig. Values of NMB, εvertex, and εtrig for all collision energies are reported in Table 1. The signal-loss correction fSL takes into account the fraction of K∗± mesons in non-triggered inelastic events and it is estimated by Monte Carlo simulations. The latter is a pT-dependent correction factor which has its maximum at low pT(fSL ≈1.04 for pT<1GeV/ cand fSL ≈1.01 for pT>1GeV/ c). 3.5. Systematic uncertainties The measurement of K∗± production in pp collisions was tested for systematic effects due to uncertainties in signal extraction, track selection criteria and particle identification for primary pions, K0 Sreconstruction, global tracking efficiency for primary pions, primary vertex selection window, knowledge of the ALICE material budget and hadronic interaction cross section used in simulations and signal loss correction, as summarized in Table 3. The yieldweighted mean values are quoted for three separate transverse momentum intervals: low (0 <pT<1.2 GeV/c), intermediate (1.2 <pT<4GeV/c), and high-pT(4 <pT<15 GeV/c). The systematic uncertainties are dominated by the raw yield extraction, labeled as “Signal extraction” in Table 3and amount to about 3–6%. This includes the sensitivity in the choice of the normalization interval, the fitting range, the shape of the residual background function, the bin counting range and the constraints on the resonance width imposed in the fitting procedure. In addition to the default strategy described in Sec 3.2, the combinatorial background was normalized in different invariant mass regions. The sensitivity of the K∗± yield extraction to the fit range was studied by varying each interval boundary by ±0.005 GeV/c2. As an alternative to the function used to describe the shape of the residual background (Eq. (2)), a third- and a second-order polynomial function was used. In this last case, the fitting range was restricted to the region 0.74–1.1 GeV/c2, where the background is reasonably approximated by a second order polynomial shape. The integration limits were varied by ±0.01 GeV/c2. The sensitivity of the fit to the constraint on the K∗± signal width was estimated by using width values that take into account the current uncertainty on the PDG average value (0.0009 GeV/c2[16]) or by fitting the signal without any constraint. The contribution to the uncertainty related to the primary charged pion reconstruction, reported in Table 3, was estimated by varying simultaneously in the data and Monte Carlo events the track and the PID selections. This uncertainty ranges from 1 to 2%. In particular, the sensitivity of the track selection on the number of crossed rows, the number of reconstructed TPC space points and the distance of closest approach to the primary vertex was tested. To study the effect of PID on the signal extraction, the selection criteria based on the TPC energy loss were varied with respect to the default setting described in Sec. 3.1. PID criteria of 2.5σTPC and 4σTPC were used. 5 ALICE Collaboration Physics Letters B 828 (2022) 137013 Table 3 Sources and yield-weighted mean values of the relative systematic uncertainties (expressed in %) on the differential yields of the K∗± resonance at the three centre-of-mass energies under study for low, intermediate and high-pTranges. √s(TeV) 5.02 8.0 13.0 pT(GeV/c) 0–1.2 1.2–4 4–15 0–1.2 1.2–4 4–15 0–1.2 1.2–4 4–15 Signal extraction (%) 5.4 2.8 3.4 5.8 5.5 5.4 4.4 3.7 4.5 Primary pion reconstruction (%) 1.2 1.0 1.0 1.2 1.1 1.5 2.1 1.4 1.3 K0 Sreconstruction (%) 0.8 0.7 1.0 2.9 0.9 0.9 2.2 1.3 1.2 Global tracking efficiency (%) 1.0 1.0 1.4 3.0 3.0 3.0 1.0 1.0 1.0 Primary vertex (%) 2.3 0.7 1.4 1.5 0.6 1.5 1.0 0.6 0.7 Material budget (%) 3.1 1.7 0.7 3.1 1.7 0.7 3.0 1.6 0.7 Hadronic interaction (%) 1.1 1.1 0.5 1.1 1.1 0.5 1.1 1.1 0.5 Signal Loss (%) 1.4 0.6 0.4 0.9 0.4 0.1 1.6 0.7 0.5 Total (%) 7.1 3.9 4.3 8.1 6.8 6.6 6.6 4.8 5.1 Systematic uncertainties due to the V0topological and K0 Ssecondary track selections are reported in Table 3under label “K0 Sreconstruction”. These uncertainties were estimated by varying simultaneously in the data and Monte Carlo events the track and the PID selection criteria for the secondary tracks, and by varying all the topological selection criteria (DCA of decay products to PV and between decay products, cosine of pointing angle and V0radius). The sensitivity of the measurement to the competing V0rejection, the mass selection, the K0 Srapidity range and lifetime was also studied by varying the interval selections. Relative uncertainties in the range 0.7-2.9% were estimated for the three energies in all the pTintervals. The total systematic uncertainties associated with the K0 Smeasurement are lower than those for the charged ones [3,13]. In particular, by exploiting the topological identification of K0 S, the large uncertainties (amounting to about 6%) originating from track selection and the PID procedure for K±are avoided. In ALICE, the track reconstruction proceeds from the outermost to the innermost radius of the TPC. To have a high-quality track for a particle originating from the primary vertex, the segment of track reconstructed in the TPC should be matched to reconstructed points in the ITS. This is not necessary for secondary tracks that originate from weak decay vertices. The differences in matching probabilities of TPC tracks with reconstructed points in the ITS between data and Monte Carlo simulations define the global tracking efficiency uncertainty. This uncertainty is in the range 1–1.4% for the 5.02 TeV data set, while a constant value of 1% and 3% was estimated for the 13 and 8TeV data, respectively. These uncertainties are correlated across pTfor the inspected data sets. Variations in the selection window around the primary vertex position can modify the yield by about 0.6–2%. The uncertainty related to the knowledge of the ALICE material budget ranges from 3.1% to 1.7% for pT<4GeV/cand is about 0.7% at higher pT. The uncertainty connected to the knowledge of the hadronic interaction cross section in the detector material is about 1% for pT<4GeV/c. These effects are evaluated combining the uncertainties for a πand a K0 S, determined as in [3,32], according to the kinematics of the decay. For the signal loss correction an uncertainty of about 1.5% was estimated for pT<1.2GeV/cfor 5.02 and 13 TeV collisions, while a slightly lower value was estimated for the 8 TeV collisions. This, for each pTinterval, is the largest value between one half of (fSL −1) and the difference of signal-loss correction values estimated with different event generators. The total systematic uncertainty is 4–8% for all the considered pTintervals whereas the systematic uncertainties assigned to the K∗0measurements performed to date range from 9% to 18% depending on energy and pT[3,11,17]. This confirms that the systematic uncertainty on the K∗/Kratio can be reduced by studying the charged resonant state. Fig. 3. The pTspectra of K∗± in inelastic pp collisions at √s=5.02, 8, and 13 TeV (full symbols) are compared to the pTspectra of K∗0mesons (open symbols) at the same energies [3,11,17]. Statistical and systematic uncertainties are reported as error bars and boxes, respectively. The normalization uncertainties (2.51%, 2.72%, and 2.55% for 5.02, 8, and 13 TeV, respectively, see Table 1) are indicated as colored boxes and are not included in the point-to-point uncertainties. The ratio of each measured pTdistribution for K∗± mesons at √s=5.02 (red points), 8 (blue points) and 13 TeV (black points) to the K∗0spectrum at the same collision energy is reported in the bottom panels. The systematic uncertainty due to global tracking, material budget and hadronic interaction cross section of primary pions are equal for charged and neutral K∗, thus they cancel out in the propagation of the uncertainty to the final ratio. 4. Results and discussion 4.1. Energy dependence of pTspectra and model comparison The first measurement of K∗± meson production in inelastic pp collisions at √s=5.02, 8, and 13 TeV up to pT=15GeV/cis presented in Fig. 3. The pT-differential yields of K∗± are compared to those previously measured for K∗0in the same collision systems [3,11,17]. The spectra of the charged and neutral mesons are consistent within the uncertainties, as expected considering the similarity of their quark content and mass. A comparison between the measured pTspectra and predictions based on QCD-inspired event generators such as PYTHIA6 [4], 6 ALICE Collaboration Physics Letters B 828 (2022) 137013 Fig. 4. The K∗± pTspectra (black dots) measured in inelastic pp collisions at (a)√s=5.02TeV, (b) 8TeV, and (c) 13 TeV are compared to the distributions predicted by PYTHIA8-Monash 2013 [31](blue lines), PYTHIA6-Perugia 2011 [30](red lines), and EPOS-LHC [6](black lines). Statistical and systematic uncertainties are shown with error bars and empty boxes, respectively. The ratios of the rebinned predictions to the measured distributions are reported in the bottom panels. The shaded bands represent the fractional uncertainties of the data points. PYTHIA8 [5] and EPOS-LHC [6] provides useful information on the hadron production mechanisms. Event generators such as PYTHIA combine a perturbative formalism of hard processes with a non-perturbative description of hadronization that is simulated using the Lund string fragmentation model [38]. In the PYTHIA tunes considered here, multiple parton-parton interactions in the same event and the color reconnection mechanism are taken into account. These effects are important in hadron-hadron interactions at the high LHC energies. In particular, color string formation between final-state partons may mimic effects similar to those induced by collective flow in heavyion collisions [39]. The PYTHIA6-Perugia 2011 tune takes into account some of the lessons learnt from the early LHC data from inelastic pp collisions at 0.9 and 7 TeV. For instance, it takes into account the observed increase in baryon production in the strangeness sector by tuning the /Kratio on the ALICE [40,41] and CMS [42] data. On the other hand, the K∗0/K ratio is tuned on the LEP measurements [30]. Monash 2013 is an updated set of parameters for the PYTHIA8 event generator, with particular attention to heavyquark fragmentation and strangeness production. For all studied LHC collision energies the PYTHIA predictions overestimate by a factor of 1.5–2 the K∗0production at transverse momenta below 0.5 GeV/cand underestimate its production by about 10-20% at pT>1GeV/c[3,17,26]. The EPOS-LHC event generator differs significantly from PYTHIA in its modeling of both the hadronization and the underlying event. It is a microscopic model that relies on parton-based Gribov-Regge theory with an improved flow parameterization which takes into account the case of a very dense system in a small volume. This high density core is produced by the overlap of string segments due to multiple parton interactions in pp or multiple nucleon interactions dominating in nucleus–nucleus collisions. EPOS-LHC reproduces the increased baryon-to-meson ratios at intermediate pTas a consequence of radial flow in high-multiplicity pp events [13]. Both PYTHIA8 and EPOS-LHC are tuned to reproduce the charged particle multiplicity and the production of identified hadrons (such as π, K, p, , −) measured in pp collisions at √s=7TeV[6]. Fig. 4shows the comparison of the measured K∗± pTspectra at √s=5.02, 8, and 13 TeV with the PYTHIA6 (Perugia 2011 tune) [30] and the PYTHIA8 (Monash 2013 tune) generators [31], and EPOS-LHC [6]. The bottom panels show the ratios of the model predictions to the measured distributions for K∗± mesons. The agreement with data improves with the collision energy. The best agreement is reached with PYTHIA6-Perugia 2011 and PYTHIA8- Monash 2013 for 13 TeV collisions. None of the models considered for comparison is able to fully reproduce the data. For all three energies the models overestimate by a factor of 1.5–2 the yield for pT<0.5 GeV/cand underestimate it in the intermediate pTregion. EPOS-LHC predictions largely overestimate the data in the high-pTregion, whereas an agreement within the uncertainties is observed for PYTHIA6 and also for PYTHIA8 at √s=13 TeV. Agreement is also observed with PYTHIA6 for pT>4GeV/cat √s=8TeV. These results complement the observation reported in Ref. [3] confirming that a more accurate tuning of the models is needed to reproduce the phase-space distribution of strange hadrons. An evolution of the transverse momentum spectra with the collision energy is clearly observed in the left panel of Fig. 5, where 7 ALICE Collaboration Physics Letters B 828 (2022) 137013 Fig. 5. (Left panel) Ratios of transverse momentum spectra of K∗± in inelastic pp events at √s=8 and 13 TeV to corresponding spectra at 5.02 TeV. Statistical and systematic uncertainties are shown with error bars and empty boxes, respectively. The normalization uncertainties are shown as colored boxes around 1 and they are not included in the point-to-point uncertainties. Blue and red histograms represent the predictions for the same ratios from PYTHIA6 Perugia 2011, PYTHIA8 Monash 2013, and EPOS-LHC. (Right panel) Ratios of transverse momentum spectra of K∗±,K ++K−and π++π−in inelastic pp events at √s=13TeV[3]to corresponding spectra at 5.02 TeV [33]. Statistical and systematic uncertainties are shown with error bars and empty boxes, respectively. the ratios of the K∗± transverse-momentum spectra at √s=8 and 13 TeV to the one at √s=5.02 TeV are reported. The systematic uncertainties associated with the estimate of the material budget of the ALICE detector and the hadronic interaction cross section used in the simulations are the same for the different collision energies. Hence, they cancel out in the propagation of the uncertainties to the ratio. For pT>1GeV/ c, a hardening of the K∗± pTspectrum is observed from 5.02 to 13 TeV, which is indicative of an increasing contribution of hard scattering processes in particle production with the collision energy. In the right panel of Fig. 5the ratios of the K++K −and π++π−pTdistributions at √s=13 TeV [3]to the ones at √s=5.02 TeV [33]are compared to the same ratio for K∗±. Distributions of these ratios are similar for the different particle species as shown in ref. [3]for ratios of pTdistributions at √s=13 TeV to the one at √s=7TeV. These distributions, like the ones for K∗±, show a progressive and significant evolution of the spectral shape at high pTwith increasing collision energy and the shape independent of pTwithin uncertainties in the soft regime, pT<1GeV/c. In the left panel of Fig. 5the ratios of the K∗± transversemomentum spectra at √s=8 and 13 TeV to the one at √s=5.02 TeV predicted by PYTHIA6, PYTHIA8 and EPOS-LHC are also shown. PYTHIA6 and PYTHIA8 predict a larger hardening with the energy, while EPOS-LHC is consistent with data. 4.2. Energy dependence of dN/dy, pTand K∗±/Kratio The measurements of particle production and particle ratios in pp collisions are important, also as a baseline for comparison with heavy-ion reactions. The per-event pT-integrated K∗± yields (corresponding to 1/NINEL×dN/dy, hereby denoted as dN/dyfor brevity) for inelastic collisions and the mean transverse momenta pTare determined by integrating and averaging the transverse momentum spectra over the measured range and are listed in Table 4. For per-event pT-integrated yields and pTstatistical uncertainties are estimated varying the data randomly inside the estimated uncertainties of each bin. The systematic uncertainties are computed assuming a full correlation across pT. The uncertainty on dN/dyis estimated from the highest and lowest spectra allowed by the bin-by-bin systematic uncertainties whereas in the case of the pTthe allowed hardest and the softer pTdistribution are considered. Fig. 6. Particle ratios K∗±/K and K∗0/K, depicted as K∗/K, in pp [3,8–11,17,26,33–35], central d–Au [36], central p–Pb [12]and central A–A [8–10,35,37] collisions as a function of √sNN. For the d–Au data, the numerator yield is derived from a combination of K∗0and K∗± states. Bars represent the statistical uncertainties and boxes represent the systematic uncertainties. The points for K∗0for d–Au, Cu–Cu and p–Pb collisions and for K∗± for pp collisions have been shifted horizontally for visibility. Red, blue and black lines represent the K∗±/K ratio predicted with PYTHIA6-Perugia 2011 [30], PYTHIA8-Monash 2013 [31]and EPOS-LHC [6], respectively. The per-event pT-integrated yield of the K∗± in inelastic pp collisions increases from √s=5.02TeV to 13 TeV by 13.5 ±1.2%. The hardening of the K∗± transverse momentum spectra reported in Fig. 5manifests itself in the increasing mean transverse momentum. In pp collisions, the measured pTat √s=13 TeV is 11.1 ±0.3% larger than at √s=5.02 TeV. Similar increasing trend of per-event pT-integrated yields and mean pTare observed for K∗0across the same collisions energies [3,11,17]. Using the K∗± yields presented in this paper and the long-lived K±production measured by ALICE at the same pp collision energies [3,17,33], the values of the K∗±/K ratio were estimated and reported in Table 4. The value of dN/dyfor (K++K−)in pp collisions at √s=8TeV was estimated by fitting the data points at √s=0.9, 2.76 and 7TeV[17]with the polynomial function A(√s)n+B, where A, nand Bare the fit parameters and by extrapolating the value for √s=8TeV. Due to the fact that the same data samples were analyzed to extract both resonance and kaon yields, the uncertainties due to the absolute normalization cancel 8 ALICE Collaboration Physics Letters B 828 (2022) 137013 9Centro de Investigación y de Estudios Avanzados (CINVESTAV), Mexico City and Mérida, Mexico 10 Chicago State University, Chicago, IL, United States 11 China Institute of Atomic Energy, Beijing, China 12 Chungbuk National University, Cheongju, Republic of Korea 13 Comenius University Bratislava, Faculty of Mathematics, Physics and Informatics, Bratislava, Slovakia 14 COMSATS University Islamabad, Islamabad, Pakistan 15 Creighton University, Omaha, NE, United States 16 Department of Physics, Aligarh Muslim University, Aligarh, India 17 Department of Physics, Pusan National University, Pusan, Republic of Korea 18 Department of Physics, Sejong University, Seoul, Republic of Korea 19 Department of Physics, University of California, Berkeley, CA, United States 20 Department of Physics, University of Oslo, Oslo, Norway 21 Department of Physics and Technology, University of Bergen, Bergen, Norway 22 Dipartimento di Fisica dell’Università ‘La Sapienza’ and Sezione INFN, Rome, Italy 23 Dipartimento di Fisica dell’Università and Sezione INFN, Cagliari, Italy 24 Dipartimento di Fisica dell’Università and Sezione INFN, Trieste, Italy 25 Dipartimento di Fisica dell’Università and Sezione INFN, Turin, Italy 26 Dipartimento di Fisica e Astronomia dell’Università and Sezione INFN, Bologna, Italy 27 Dipartimento di Fisica e Astronomia dell’Università and Sezione INFN, Catania, Italy 28 Dipartimento di Fisica e Astronomia dell’Università and Sezione INFN, Padova, Italy 29 Dipartimento di Fisica e Nucleare e Teorica, Università di Pavia, Pavia, Italy 30 Dipartimento di Fisica ‘E.R. Caianiello’ dell’Università and Gruppo Collegato INFN, Salerno, Italy 31 Dipartimento DISAT del Politecnico and Sezione INFN, Turin, Italy 32 Dipartimento di Scienze e Innovazione Tecnologica dell’Università del Piemonte Orientale and INFN Sezione di Torino, Alessandria, Italy 33 Dipartimento di Scienze MIFT, Università di Messina, Messina, Italy 34 Dipartimento Interateneo di Fisica ‘M. Merlin’ and Sezione INFN, Bari, Italy 35 European Organization for Nuclear Research (CERN), Geneva, Switzerland 36 Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, University of Split, Split, Croatia 37 Faculty of Engineering and Science, Western Norway University of Applied Sciences, Bergen, Norway 38 Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Prague, Czech Republic 39 Faculty of Science, P.J. Šafárik University, Košice, Slovakia 40 Frankfurt Institute for Advanced Studies, Johann Wolfgang Goethe-Universität Frankfurt, Frankfurt, Germany 41 Fudan University, Shanghai, China 42 Gangneung-Wonju National University, Gangneung, Republic of Korea 43 Gauhati University, Department of Physics, Guwahati, India 44 Helmholtz-Institut für Strahlen- und Kernphysik, Rheinische Friedrich-Wilhelms-Universität Bonn, Bonn, Germany 45 Helsinki Institute of Physics (HIP), Helsinki, Finland 46 High Energy Physics Group, Universidad Autónoma de Puebla, Puebla, Mexico 47 Hiroshima University, Hiroshima, Japan 48 Hochschule Worms, Zentrum für Technologietransfer und Telekommunikation (ZTT), Worms, Germany 49 Horia Hulubei National Institute of Physics and Nuclear Engineering, Bucharest, Romania 50 Indian Institute of Technology Bombay (IIT), Mumbai, India 51 Indian Institute of Technology Indore, Indore, India 52 Indonesian Institute of Sciences, Jakarta, Indonesia 53 INFN, Laboratori Nazionali di Frascati, Frascati, Italy 54 INFN, Sezione di Bari, Bari, Italy 55 INFN, Sezione di Bologna, Bologna, Italy 56 INFN, Sezione di Cagliari, Cagliari, Italy 57 INFN, Sezione di Catania, Catania, Italy 58 INFN, Sezione di Padova, Padova, Italy 59 INFN, Sezione di Pavia, Pavia, Italy 60 INFN, Sezione di Roma, Rome, Italy 61 INFN, Sezione di Torino, Turin, Italy 62 INFN, Sezione di Trieste, Trieste, Italy 63 Inha University, Incheon, Republic of Korea 64 Institute for Gravitational and Subatomic Physics (GRASP), Utrecht University/Nikhef, Utrecht, Netherlands 65 Institute for Nuclear Research, Academy of Sciences, Moscow, Russia 66 Institute of Experimental Physics, Slovak Academy of Sciences, Košice, Slovakia 67 Institute of Physics, Homi Bhabha National Institute, Bhubaneswar, India 68 Institute of Physics of the Czech Academy of Sciences, Prague, Czech Republic 69 Institute of Space Science (ISS), Bucharest, Romania 70 Institut für Kernphysik, Johann Wolfgang Goethe-Universität Frankfurt, Frankfurt, Germany 71 Instituto de Ciencias Nucleares, Universidad Nacional Autónoma de México, Mexico City, Mexico 72 Instituto de Física, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, Brazil 73 Instituto de Física, Universidad Nacional Autónoma de México, Mexico City, Mexico 74 iThemba LABS, National Research Foundation, Somerset West, South Africa 75 Jeonbuk National University, Jeonju, Republic of Korea 76 Johann-Wolfgang-Goethe Universität Frankfurt Institut für Informatik, Fachbereich Informatik und Mathematik, Frankfurt, Germany 77 Joint Institute for Nuclear Research (JINR), Dubna, Russia 78 Korea Institute of Science and Technology Information, Daejeon, Republic of Korea 79 KTO Karatay University, Konya, Turkey 80 Laboratoire de Physique des 2 Infinis, Irène Joliot-Curie, Orsay, France 81 Laboratoire de Physique Subatomique et de Cosmologie, Université Grenoble-Alpes, CNRS-IN2P3, Grenoble, France 82 Lawrence Berkeley National Laboratory, Berkeley, CA, United States 83 Lund University Department of Physics, Division of Particle Physics, Lund, Sweden 84 Moscow Institute for Physics and Technology, Moscow, Russia 85 Nagasaki Institute of Applied Science, Nagasaki, Japan 86 Nara Women’s University (NWU), Nara, Japan 87 National and Kapodistrian University of Athens, School of Science, Department of Physics, Athens, Greece 88 National Centre for Nuclear Research, Warsaw, Poland 15 ALICE Collaboration Physics Letters B 828 (2022) 137013 89 National Institute of Science Education and Research, Homi Bhabha National Institute, Jatni, India 90 National Nuclear Research Center, Baku, Azerbaijan 91 National Research Centre Kurchatov Institute, Moscow, Russia 92 Niels Bohr Institute, University of Copenhagen, Copenhagen, Denmark 93 Nikhef, National institute for subatomic physics, Amsterdam, Netherlands 94 NRC Kurchatov Institute IHEP, Protvino, Russia 95 NRC Kurchatov Institute – ITEP, Moscow, Russia 96 NRNU Moscow Engineering Physics Institute, Moscow, Russia 97 Nuclear Physics Group, STFC Daresbury Laboratory, Daresbury, United Kingdom 98 Nuclear Physics Institute of the Czech Academy of Sciences, ˇ Rež u Prahy, Czech Republic 99 Oak Ridge National Laboratory, Oak Ridge, TN, United States 100 Ohio State University, Columbus, OH, United States 101 Petersburg Nuclear Physics Institute, Gatchina, Russia 102 Physics department, Faculty of science, University of Zagreb, Zagreb, Croatia 103 Physics Department, Panjab University, Chandigarh, India 104 Physics Department, University of Jammu, Jammu, India 105 Physics Department, University of Rajasthan, Jaipur, India 106 Physikalisches Institut, Eberhard-Karls-Universität Tübingen, Tübingen, Germany 107 Physikalisches Institut, Ruprecht-Karls-Universität Heidelberg, Heidelberg, Germany 108 Physik Department, Technische Universität München, Munich, Germany 109 Politecnico di Bari and Sezione INFN, Bari, Italy 110 Research Division and ExtreMe Matter Institute EMMI, GSI Helmholtzzentrum für Schwerionenforschung GmbH, Darmstadt, Germany 111 Russian Federal Nuclear Center (VNIIEF), Sarov, Russia 112 Saha Institute of Nuclear Physics, Homi Bhabha National Institute, Kolkata, India 113 School of Physics and Astronomy, University of Birmingham, Birmingham, United Kingdom 114 Sección Física, Departamento de Ciencias, Pontificia Universidad Católica del Perú, Lima, Peru 115 St. Petersburg State University, St. Petersburg, Russia 116 Stefan Meyer Institut für Subatomare Physik (SMI), Vienna, Austria 117 SUBATECH, IMT Atlantique, Université de Nantes, CNRS-IN2P3, Nantes, France 118 Suranaree University of Technology, Nakhon Ratchasima, Thailand 119 Technical University of Košice, Košice, Slovakia 120 The Henryk Niewodniczanski Institute of Nuclear Physics, Polish Academy of Sciences, Cracow, Poland 121 The University of Texas at Austin, Austin, TX, United States 122 Universidad Autónoma de Sinaloa, Culiacán, Mexico 123 Universidade de São Paulo (USP), São Paulo, Brazil 124 Universidade Estadual de Campinas (UNICAMP), Campinas, Brazil 125 Universidade Federal do ABC, Santo Andre, Brazil 126 University of Cape Town, Cape Town, South Africa 127 University of Houston, Houston, TX, United States 128 University of Jyväskylä, Jyväskylä, Finland 129 University of Kansas, Lawrence, KS, United States 130 University of Liverpool, Liverpool, United Kingdom 131 University of Science and Technology of China, Hefei, China 132 University of South-Eastern Norway, Tonsberg, Norway 133 University of Tennessee, Knoxville, TN, United States 134 University of the Witwatersrand, Johannesburg, South Africa 135 University of Tokyo, Tokyo, Japan 136 University of Tsukuba, Tsukuba, Japan 137 Université Clermont Auvergne, CNRS/IN2P3, LPC, Clermont-Ferrand, France 138 Université de Lyon, CNRS/IN2P3, Institut de Physique des 2 Infinis de Lyon, Lyon, France 139 Université de Strasbourg, CNRS, IPHC UMR 7178, F-67000 Strasbourg, France 140 Université Paris-Saclay Centre d’Etudes de Saclay (CEA), IRFU, Départment de Physique Nucléaire (DPhN), Saclay, France 141 Università degli Studi di Foggia, Foggia, Italy 142 Università di Brescia, Brescia, Italy 143 Variable Energy Cyclotron Centre, Homi Bhabha National Institute, Kolkata, India 144 Warsaw University of Technology, Warsaw, Poland 145 Wayne State University, Detroit, MI, United States 146 Westfälische Wilhelms-Universität Münster, Institut für Kernphysik, Münster, Germany 147 Wigner Research Centre for Physics, Budapest, Hungary 148 Yale University, New Haven, CT, United States 149 Yonsei University, Seoul, Republic of Korea IDeceased. II Also at: Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA), Bologna, Italy. III Also at: Dipartimento DET del Politecnico di Torino, Turin, Italy. IV Also at: M.V. Lomonosov Moscow State University, D.V. Skobeltsyn Institute of Nuclear, Physics, Moscow, Russia. VAlso at: Institute of Theoretical Physics, University of Wroclaw, Poland. 16