33RD INTERNATIONAL CONFERENCE,RIODE JANEIRO 2013 THE ASTROPARTICLE PHYSICS CONFERENCE

Measurement of -carbon interactions for better understanding of air showers with NA61/SHINE

H. P. DEMBINSKI1 FORTHE NA61/SHINE COLLABORATION2. 1 Institut fur¨ Kernphysik, KIT Karlsruhe, Postfach 3640, D - 76021 Karlsruhe 2 Full author list: https: // na61. web. . ch/ na61/ pages/ storage/ authors_ list. pdf [email protected]

Abstract: The interpretation of air shower measurements requires the detailed simulation of hadronic particle production over a wide range of energies and the forward phase space of secondary particles. In air showers, the bulk of particles is produced at later stages of the shower development and at equivalent beam energies in the sub-TeV range. NA61/SHINE is a fixed target experiment using secondary beams produced at the SPS at 1 CERN. Hadron-hadron interactions have been recorded at beam momenta between 13 and 350 GeVc− with a wide-acceptance spectrometer. In this article we present measurements of the inelastic cross-section and secondary particle yields of -carbon interactions, which are essential for modelling air showers.

Keywords: hadron interaction, QCD, air shower, pion

1 Hadronic interactions in air showers Cosmic rays initiate extensive air showers (EAS) when they collide with nuclei of the atmosphere. The EAS data recorded by experiments like the Pierre Auger Observa- tory [1], KASCADE [2] or IceTop [3], is used to infer their properties. The interpretation relies on simulations of the shower development during which electromagnetic and hadronic particle interactions are followed from primary 20 9 energies of & 10 eV down to energies of 10 eV. The penetrating muon component of an EAS is particu- larly important for the interpretation of data from ground detectors. Myons that reach ground typically originate from mesons produced in low energy interactions in the late stages of an air shower. Depending on the primary energy and detection distance, the relevant interaction energies are between 10 and 1000 GeV and the modeling of the corre- sponding low energy interactions contributes at least 10% Figure 1: Schematic layout of the NA61/SHINE experi- to the overall uncertainty of the predicted muon number at ment. ground (see e.g. Refs. [4–7]). Comprehensive and precise particle production measure- properties of the onset of de-confinement and a search for ments in this energy range are of great value. Unfortunately, the critical point of strongly interacting matter (see e.g. direct measurements are sparse, even for the most numer- Ref. [10]). ous projectile in air showers, the π-meson. Thus new data The layout of the NA61/SHINE detector is sketched in 1 with pion beams at 158 and 350 GeVc− on a thin car- Fig. 1. A set of scintillation and Cherenkov counters as bon target (as a proxy for nitrogen) were collected by the well as beam position detectors upstream of the spectrom- NA61/SHINE experiment at the CERN SPS and prelimi- eter provide timing reference, identification and position nary results from this data set are presented here. measurements of the incoming beam particles. Large time- projection-chambers (TPCs) inherited from the NA49 exper- iment [11] are used to measure the charge and momentum of 2 The NA61/SHINE experiment particles. The momentum resolution, σ(1/p) = σ(p)/p2, 4 1 NA61/SHINE (SHINE = SPS Heavy Ion and Ex- is about 10− GeV− c at full magnetic field and the track- periment) [8] is a multi-purpose fixed target experiment to ing efficiency is better than 95%. Particle identification is study hadron production in hadron-nucleus and nucleus- achieved by measuring the energy loss along the tracks in nucleus collisions at the CERN Super Synchrotron the TPCs and by determining their velocity from the time (SPS). Among its physics goals are precise hadron pro- of flight provided by large scintillator walls placed down- duction measurements for improving calculations of the stream of the TPCs. neutrino beam flux in the T2K experi- The centrality of nucleus-nucleus collisions can be esti- ment [9] as well as for more reliable simulations of hadronic mated using the measurement of the energy of projectile interactions in air showers. Moreover, p+p, p+Pb and nu- spectators with a calorimeter [12] located behind the time cleus+nucleus collisions are measured for a study of the of flight detectors. For nucleon-nucleus collisions, the cen- Hadron-carbon interactions measured with NA61/SHINE 33RD INTERNATIONAL COSMIC RAY CONFERENCE,RIODE JANEIRO 2013

1 0 1 0 10− 10 10− 10 103 103

102 102 2 c

2 10 10 −

1 1

1 1 / mb GeV 10− 10− T p d σ 2 2 2 lab d 10− 10− p d

3 1 + 1 3 10− h− at 158 GeV c− h at 158 GeV c− 10− NA61/SHINE preliminary NA61/SHINE preliminary 4 4 10− 10− 103 103

102 102 2 c

2 10 10 −

1 1

1 1 / mb GeV 10− 10− T p d σ 2 2 2 lab d 10− 10− p d

3 1 + 1 3 10− h− at 350 GeV c− h at 350 GeV c− 10− NA61/SHINE preliminary NA61/SHINE preliminary 4 4 10− 10− 1 0 1 0 10− 10 10− 10 1 1 pT/GeV c− pT/GeV c−

1 Figure 2: Inclusive pT-spectra of charged produced in π−+C interactions at 158 and 350 GeVc− . In each figure, 1 1 the particle momentum p ranges from 0.58 to 145 GeVc− in steps of log10(plab/GeVc− ) = 0.08 from top to bottom. The 1 markers alternate between circle and square for each step in log10(plab/GeVc− ) to guide the eye. Hadron-carbon interactions measured with NA61/SHINE 33RD INTERNATIONAL COSMIC RAY CONFERENCE,RIODE JANEIRO 2013

] 10

2 0.5 NA61/SHINE preliminary simulation

1 0.4 1 1 − [mb/(GeV/c)

T 0.3 σ 2

d -1

dp 10

/GeV c 10 1 0.2 data T − p

QGSJetII-3 lepton fraction EPOS1.99 0.1 10-2 Sibyll2.1 NA61/SHINE preliminary 2 10− 0.0 0.5 1 1.5 2 2.5 0 1 2 p [GeV/c] 10 10 10 T 1 plab/GeV c−

Figure 3: Transverse momentum spectrum of negatively Figure 4: Predicted fraction of negative leptons in charged 1 charged hadrons produced in π−+C interactions at particles produced in π−+C interactions at 158 GeVc− 1 1 158 GeVc beam momentum at plab =10.4 GeVc . beam momentum (average of VENUS and EPOS). − h i − trality is determined by counting low momentum particles are displayed in Fig. 3. The uncertainties shown are the from the target (so called ’gray ’) with a small TPC total uncertainties including the statistical uncertainty and around the target. systematics from model correction, normalization, trigger NA61/SHINE started data taking in 2007. During the last bias, calibration and track topology. four years, a wealth of data has been recorded by the exper- These preliminary measurements are already useful to 1 iment at beam momenta ranging from 13 to 350 GeVc− judge the quality of event generators used in air-shower with various beam particles and targets. simulations. An example of the pT distribution of negatively charged hadrons produced in π−+C interactions at a beam 1 momentum of 158 GeVc− is shown in Fig. 3 for particle 1 3 NA61/SHINE results for the momenta with plab =10.4 GeVc and compared to h i − interpretation of cosmic ray air showers predictions by QGSJETII-03 [19], SIBYLL2.1 [20] and EPOS1.99 [21]. As can be seen, none of these hadronic 6 The analysis presented here is based on 4.7 10 inelastic interaction models which are used to simulate air showers · 1 events collected with pion beams at 158 and 350 GeVc− on can reproduce that data and especially SIBYLL2.1 predicts a thin carbon target. We present the production cross-section a too steep spectrum at high transverse momenta. The and momentum spectra of produced charged hadrons [13]. underestimation of charged hadron production at large The production cross-section in π−+C interactions was transverse momenta is present at all momenta for EPOS1.99 determined in a similar manner as described in Ref. [14]. and SIBYLL2.1. Of all the models studied QGSJETII-03 The experimental interaction cross-section is corrected for describes our data best with only a small deficit of tracks residual contributions from elastic and quasi-elastic scat- with high pT at large particle momentum but slightly too tering as well as for the inelastic contribution to which the many particles at low traverse momenta. Comparisons with NA61/SHINE interaction trigger is not sensitive. The uncer- the latest models will be shown at the conference. tainties of the measurement are currently dominated by the A shortcoming of the present analysis is the model de- model-dependence of this correction. Preliminary values pendent correction of the charged hadron spectra for lep- . . are σprod = 172 2(stat ) 4(syst.) and 178 2(stat ) ton contamination (almost entirely e±). The contamination ± ± 1 ± ± 1 1 4(syst.) at 158 and 350 GeVc− respectively [15]. This exceeds 20 % at plab < 10GeVc− and pT < 0.1GeVc− measurement is compatible with previous results [16,17] as indicated in Fig. 4 and data in this range are not shown and already gives the most precise value of the production because of the associated large systematic uncertainty. 1 cross section at around 160 GeVc− . Currently work is ongoing to use the measured en- The momentum spectra of charged hadrons produced ergy loss along the tracks in the TPCs to identify the in these interactions are presented in Fig. 2. The spectra hadrons directly. The average restricted energy loss per were obtained within a fiducial phase space in the NA61 track dE/dx trunc is reconstructed from a truncated distri- detector, for which the detection and selection efficiency for butionh of thei collected charge clusters along the track with charged tracks is close to unity. Feed-down and track loss a procedure similar to Ref. [22]. Hadrons and electrons are are corrected based on the average predictions of the event well separated in dE/dx trunc in most of the phase space h i generators VENUS and EPOS fed into the detector simula- so that the e± contamination can be extracted by fitting a tion based on GEANT3 [18]. The trigger bias is corrected statistical model of the energy loss for hadrons and elec- for by studying the track loss in a sub-sample of unbiased trons to the observed distribution. An example is shown in beam-trigger data. Only phase-space regions for which the Fig. 5. This improvement will allow us to recover the lost overall model-dependent correction is below 20% and for phase space. The results will be presented at the conference which the total systematic uncertainty is smaller than 20% for the first time. Hadron-carbon interactions measured with NA61/SHINE 33RD INTERNATIONAL COSMIC RAY CONFERENCE,RIODE JANEIRO 2013

of the shower development below a TeV. 600 p = 1.2 GeV c 1 p = 0.063 GeV c 1 h labi − h Ti − Acknowledgment: This work was supported by the Hungarian 400 Scientific Research Fund (grants OTKA 68506 and 71989), the Polish Ministry of Science and Higher Education (grants 667/N- CERN/2010/0, NN 202 48 4339 and NN 202 23 1837), the 200 Federal Agency of Education of the Ministry of Education and Science of the Russian Federation (grant RNP 2.2.2.2.1547), /0.005 0 hadrons e± the Russian Academy of Science and the Russian Foundation ev for Basic Research (grants 08-02-00018 and 09-02-00664), the

N Ministry of Education, Culture, Sports, Science and Technology, Japan, Grant-in-Aid for Scientific Research (grants 18071005, × 200 19034011, 19740162, 20740160 and 20039012), the German

q − Research Foundation (grant GA 1480/2-1), Bulgarian National 400 Scientific Fondation (grant DDVU 02/19/ 2010), Ministry of − model Education and Science of the Republic of Serbia (grant OI171002), NA61/SHINE preliminary Swiss Nationalfonds Foundation (grant 200020-117913/1) and 600 data ETH Research Grant TH-01 07-3. − 0.1 0.0 0.1 0.2 0.3 − log10( dE/dx trunc/mip) References −h i [1] J. Abraham et al. [Pierre Auger Collaboration], Nucl. Instrum. Meth. A 523 (2004) 50. Figure 5: Histogram of the average restricted energy loss [2] T. Antoni et al. [KASCADE Collaboration], Nucl. Instrum. dE/dx trunc per particle in a small interval around plab = Meth. A 513 (2003) 490. h i 1 1 h i 1.2GeVc− and pT = 0.063GeVc− . Positively and neg- [3] R. Abbasi et al. [IceCube Collaboration], Nucl. Instrum. h i Meth. A 700 (2013) 188. atively charged particles are shown together by mapping [4] D. Heck et al., Proc. 28th ICRC, (2003) 279. the latter to the negative y-axis. Continuous lines indicate [5] H.-J. Drescher et al., Astropart. Phys. 21 (2004) 87. the fitted model and the shaded area the e±-contribution. [6] C. Meurer et al., Czech. J. Phys. 56 (2006) A211. [7] I. C. Maris¸ et al., Nucl. Phys. Proc. Suppl. 196 (2009) 86. [8] N. Antoniou et al. [NA61/SHINE Collaboration], CERN- SPSC-2007-004, (2007); CERN-SPSC-2007-019, (2007); 4 Conclusions and outlook N. Abgrall et al. [NA61/SHINE Collaboration], CERN-SPSC-2008-018, (2008). In this article, we summarized results from pion-carbon [9] K. Abe et al. [T2K Collaboration], Nucl. Instrum. Meth. A interactions measured with the multi-purpose experiment 659 (2011) 106. NA61/SHINE at the CERN SPS, which are of importance [10] M. Gazdzicki, M. Gorenstein, P. Seyboth, Acta Phys. Polon. B42 (2011) 307. for the modelling of cosmic ray air showers. It is worth [11] S. Afanasev et al. [NA49 Collaboration], Nucl. Instrum. noting that although the different measurements are moti- Meth. A 430 (1999) 210. vated by different physics goals, all the hadron-nucleus and [12] A. Ivashkin et al., arXiv:1205.4864. [13] M. Unger for the NA61/SHINE Collaboration, Proc. proton-proton data collected by NA61/SHINE are valuable ISVHECRI (2012), arXiv:1305.5281 [nucl-ex]. for the tuning of hadronic interaction models for the under- [14] N. Abgrall et al. [NA61/SHINE Collaboration], Phys. Rev. standing of air showers. The published NA61/SHINE data C 84 (2011) 034604. on p+C interactions at 31 GeVc 1 [14] have already been [15] M. Haug, Messung des Wirkungsquerschnitts von Pion- − Kohlenstoff-Wechselwirkungen mit Hilfe des NA61 used to fix [23] the UrQMD1.3.1 model [24] and to further Detektors, Diploma Thesis, KIT, 2012. strengthen [25] the case against the use of GHEISHA [26]. [16] A. Schiz et al., Phys. Rev. D 21 (1980) 3010. Both of these models are available in CORSIKA [27] to sim- [17] U. Dersch et al. [SELEX Collaboration], Nucl. Phys. B 579 ulate low energy interactions in air showers. Moreover, re- (2000) 277. [18] M. Ruprecht, Measurement of the Spectrum of Charged cent comparisons [28] of NA61/SHINE data to EPOS sug- Hadrons in π−+C Interactions with the NA61 Experiment, gest that high-energy interaction models need further tuning Diploma Thesis, KIT, 2012. even for ’well-known’ reactions like p+p. [19] S.S. Ostapchenko, Nucl. Phys. Proc. Suppl. 151 (2006), It is planned to further refine the particle identification 143;. [20] E.J. Ahn et al., Phys. Rev. D80 (2009) 094003. analysis based on dE/dx trunc and to include the time-of- [21] T. Pierog and K. Werner, Phys. Rev. Lett. 101 (2008), flight measurementsh in orderi to separate all hadrons. The 171101. spectra of identified hadrons will be used to study short- [22] G.I. Veres, Baryon Momentum Transfer in Hadronic and Nuclear Collisions at the CERN NA49 Experiment, Ph.D. comings of hadronic models in more detail. NA61/SHINE Thesis, Eotv¨ os¨ Lorand´ University, Budapest, 2001. will be able to validate the measurement of proton and anti- [23] V. Uzhinsky, arXiv:1107.0374 [hep-ph]. proton production in π+C interactions from Ref. [29], the [24] S.M. Bass et al., Prog. Part. Nucl. Phys. 41 225 (1998); main evidence that led to the proposal [21] of an increased M. Bleicher et al., J. Phys. G: Nucl. Part. Phys. 25, 1859 (1999). baryon production as a possibility to enlarge the number [25] M. Unger for the NA61/SHINE Collaboration, Proc. 35th of muons in air showers. The NA61/SHINE data set also ICHEP (2010), arXiv:1012.2604. offers the possibility to constrain the ρ0 production in π+C [26] H. Fesefeldt, Aachen Report No. PITHA-85-02 (1985). [27] D. Heck et al., FZK Report No. FZKA-6019, 1998. which may be equally important for muons observed in air [28] S. Pulawski for the NA61/SHINE Collaboration, talk at the showers as the baryon fraction (see e.g. Ref. [30]). 8th CPOD (2013). Together with the special cosmic ray runs for pion-carbon [29] D. S. Barton et al., Phys. Rev. D 27 (1983) 2580. interactions, the rich data set of NA61/SHINE will thus be [30] H. -J. Drescher, Phys. Rev. D 77 (2008) 056003. very useful to reduce uncertainties in air shower calculations and increase the knowledge on interactions in the late stages