Cosmic ray velocity and electric charge measurements with the AMS/RICH detector: prototype results Lu´ısa Arruda, Fernando Bara˜o, Patr´ıcia Gonc¸alves, Rui Pereira LIP/IST Av. Elias Garcia, 14, 1o andar 8 1000-149 Lisboa, Portugal 0 e-mail: [email protected] 0 2 n a Abstract—The Alpha Magnetic Spectrometer (AMS) to be forchargedparticles,as wellasinformationontheir direction J installed on the International Space Station (ISS) will measure of incidence. The TOF operation at regions with very intense chargedcosmicrayspectraofelementsuptoiron,intherigidity 1 magneticfieldsforcestheuseofshieldedfinemeshphototubes range from 1GV to 1TV, for at least three years. AMS is a 3 and the optimization of the light guides geometry, with some large angular spectrometer composed of different subdetectors, includingaproximityfocusingRingImagingCHerenkov(RICH) of them twisted and bent. Moreover the system guarantees ] h detector. This will be equipped with a mixed radiator made of redundancy, with two photomultipliers on each end of the p aerogel and sodium fluoride (NaF), a lateral conical mirror and paddles and double redundant electronics. A time resolution - a detection plane made of 680 photomultipliers coupled to light of 140ps for protons is expected [4]. o guides. The RICH detector allows measurements of particle’s r electric charge up to iron, and particle’s velocity. Two possible The tracking system will be surrounded by veto counters t s methods for reconstructing the Cˇerenkov angle and the electric andembeddedina magneticfieldofabout0.9Tesla produced a charge with the RICH will be discussed. by a superconducting magnet. It will consist on a Silicon [ A RICH prototype consisting of a detection matrix with 96 Tracker [3], made of 8 layers of double sided silicon sensors photomultipliers, a segment of a conical mirror and samples 1 with a total area of ∼6.7m2. There will be a total of ∼2500 of the radiator materials was built and its performance was v silicon sensors arranged on 192 ladders. The position of evaluated using ion beam data. Results from the last test beam 2 performed with ion fragments resulting from the collision of a the charged particles crossing the tracker layers is measured 5 9 158GeV/c/nucleonprimarybeamofindiumions(CERNSPS)on with a precision of ∼10µm along the bending plane and 4 a lead target are reported. The large amount of collected data ∼30µm on the transverse direction. With a bending power allowed to test and characterize different aerogel samples and . (BL2) of around 0.9T.m2, particles rigidity is measured with 1 the NaF radiator. In addition, the reflectivity of the mirror was 0 evaluated. The data analysis confirms the design goals. an accuracy better than 2% up to 20GV and the maximal 8 detectable rigidity is around 1-2TV. Electric charge is also 0 I. THEAMS-02 EXPERIMENT measured from energy deposition up to Z∼26. v: The Alpha Magnetic Spectrometer [1] (AMS) is a particle The Ring Imaging Cˇerenkov Detector (RICH) [2] will i detector to be installed in the International Space Station be located right after the last TOF plane and before the X (ISS) for at least three years. The spectrometer will be able electromagnetic calorimeter. It will be described in detail in r a to measure the rigidity (R ≡ pc/|Z|e), the charge (Z), next section. the velocity (β) and the energy (E) of cosmic rays from some MeV up to ∼1TeV within a geometrical acceptance of ∼0.5m2.sr. Figure I shows a schematic view of the AMS spectrometer. At both ends of the AMS spectrometer exist the Transition Radiation Detector (TRD) (top) and the Elec- tromagnetic Calorimeter (ECAL) (bottom). Both will provide AMS with capability to discriminate between leptons and hadrons. Additionally the calorimeter will trigger and detect photons. The TRD will be followed by the first of the four Time-of-Flight (TOF) system scintillator planes. The TOF system [4] is composed of four roughly circular planes of 12cm wide scintillator paddles, one pair of planes above the magnet, the upper TOF, and one pair bellow, the lower TOF. There will be a total of 34 paddles. The TOF will provide a Fig.1. Awholeview oftheAMSSpectrometer. fast trigger within 200ns, charge and velocity measurements ThelongstayofAMSinspaceanditslargeacceptancewill RICH was designed to measure the velocity (β ≡ v/c) allowtheaccumulationofalargestatistic ofeventsincreasing of singly charged particles with a resolution ∆β/β of 0.1%, in several orders of magnitude the sensitivity of the proposed to extend the charge separation (Z) up to iron (Z=26), to physical measurements. With an average collection rate of contribute to e/p separation and to albedo rejection. 1000 events per second, a total of 109 protons per year and In order to validate the RICH design, a prototype with an around 104 antiprotons will be accumulated. arrayof9×11cellsfilledwith96photomultiplierreadoutunits The main goals of the AMS-02 experiment are: similartopartofthematrixofthefinalmodelwasconstructed. Theperformanceofthisprototypehasbeentestedwithcosmic • A precise measurement of the charged cosmic ray spec- muons and with a beam of secondary ions at the CERN SPS trum between ∼100MeV and ∼1TeV, and the detection produced by fragmentation of a primary beam in 2002 and of photons up to a few hundred GeV; 2003. The light guides used were prototypes with a slightly • A search for heavy antinuclei (Z ≥2), which if discov- ered would signal the existence of primordialantimatter; smaller collecting area (31×31mm2). Different samples of the radiator materials were tested and placed at an adjustable • Searchfornon-baryonicdarkmatterthroughthedetection supporting structure. Different expansion heights were set in of annihilation products appearing as anomalies of the cosmic-ray spectra (e+, p¯, γ and d¯); order to have fully contained photon rings on the detection matrixlike in the flightdesign.A segmentofa conicalmirror with 1/12 of the final azimuthal coverage, which is shown in II. THE AMSRICH DETECTOR left picture of Figure 3, was also tested. The RICH is a proximity focusing device with a dual TheRICHassemblyhasalreadystartedatCIEMATinSpain radiator configuration on the top made of 92 aerogel 25mm and is foreseen to be finished in July 2007. A rectangular thick tiles with a refractive index 1.050 and sodium fluoride grid has already been assembled and has been subject to a (NaF) tiles with a thickness of 5mm in the center covering mechanicalfittest,functionaltests,vibrationtestsandvacuum an area of 34×34cm2. The NaF placement prevents the loss tests. The other grids will follow. The refractive index of the of photons in the hole existing in the center of the readout aerogeltiles is being measuredand the radiator containerwas plane(64×64cm2),infrontoftheECALcalorimeterlocated subjected to a mechanical test. The final integration of RICH below. The radiator tiles are supported by a 1mm thick layer in AMS will take place at CERN in 2008. of methacrylate (n=1.5) free of UV absorbing additives. The detection matrix is composed of 680 multiplixelized photon readout cells each consisting of a photomultiplier coupled to a light guide, HV divider plus front-end (FE) electronics, all housed and potted in a plastic shell and then enclosed in a magnetic shielding with a thickness varying from 0.8 to 1.2mm. The photon detection is made with an array of multianode Hamamatsu tubes (R7600-00-M16) with a spectral response ranging from 300 to 650nm and a maximum quantum efficiency at λ∼420nm. To increase the photon collection efficiency, a light guide consisting of 16 solid acrylic pipes glued to a thin top layer (1mm) was Fig.2. Schematic viewoftheRICHdetector. produced.Itisopticallycoupledtotheactiveareaofphototube cathode through a 1 mm flexible optical pad. With a total height of 31 mm and a collecting surface of 34×34 mm2, it presentsareadoutpixelsize of8.5mmanda pitchof37mm. The light guide is mechanically attached through nylon wires to the photomultiplier polycarbonate housing. A high reflectivity conical mirror surrounds the whole set. The mirror was included to increase the device acceptance since around33%ofthe aerogelgeneratedphotonsimpacton the mirror. It consists of a carbon fiber reinforced composite substrate with a multilayer coating made of aluminium and SiO depositedontheinnersurface.Thisensuresareflectivity 2 higher than 85% for 420nm wavelength photons. The RICH hasa truncatedconicalshape with an expansion height of 46.3cm, a top radius of 60cm and a bottom radius of 67cm. The total height of the detector is 60.5cm and it Fig. 3. Protype with reflector (left). Top view of the test beam 2003 covers 80% of the AMS magnet acceptance. Figure 2 shows experimental setupusingCERNSPSfacility (right). a schematic view of the RICH detector. 1215. / 43 MCoenasntant 00..13118069EE+-0053 0.767 42E8-.2073 whereb=0.5122and0.105respectivelyforaerogelandNaF 104 Sigma 0.3837 0.6252E-03 in the prototype setup. The detector’s dimension is defined as D=100cm for flight setup and 20cm for prototype setup. For a more complete description of the method see Ref. [5]. 103 The Cˇerenkov photons produced in the radiator are uniformly emitted along the particle path inside the dielectric medium, L, and their number per unit of energy depends on the particle’s charge,Z, and velocity, β, and on the refractive 102 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 index, n, according to the expression: hres dN 1 Fig.4. BerylliumeventdisplaygeneratedinaNaFradiator.Thereconstructed γ ∝Z2L 1− (4) photon pattern (full line) includes bothreflected and non-reflected branches. dE β2n2 The outer circular line corresponds to the lower boundary of the conical (cid:18) (cid:19) mirror.Thesquareisthelimitofthenon-active region(left). Hit’s residuals Thereforetoreconstructthechargethefollowingprocedure distributionbelongingtophotonringsgeneratedinanaerogelradiator,n=1.05, is required: 2.5cmthick. • Cˇerenkov angle reconstruction. • Estimation of the particle path, L, which relies on the information of the particle direction provided by the III. BETA (β)AND CHARGE (Z)RECONSTRUCTION tracker. Achargedparticlecrossingadielectricmaterialofrefractive • Counting the number of photoelectrons. The number index n with a velocity β, greater than the speed of light in of photoelectrons related to the Cˇerenkov ring has to thatmedium,emitsphotons.Theapertureangleoftheemitted be counted within a fiducial area, in order to exclude photons with respect to the radiating particle is known as the the uncorrelated background. In particular it ensures the Cˇerenkov angle, θ , and it is given by (see Ref. [5]) exclusion of photons which are scattered in the radiator. c A distance of 13mm to the ring was defined as the limit 1 cosθc = (1) forphotoelectroncounting,correspondingtoaringwidth β n of ∼5 pixels. It follows that the velocity of the particle, β, is straightfor- • Evaluationofthephotondetectionefficiency.Thenumber ward derived from the Cˇerenkov angle reconstruction, which ofradiatedphotons(N )whichwillbedetected(n ) is γ p.e is based on a fit to the pattern of the detected photons. reduced due to the interactions with the radiator (ε ), rad Complex photon patterns can occur at the detector plane due the photon ring acceptance (ε ), light guide efficiency geo to mirror reflected photons, as can be seen on the left display (ε ) and photomultiplier efficiency (ε ). lg pmt of Figure 4. The event shown is generated by a simulated beryllium nucleus crossing the NaF radiator. The Cˇerenkov np.e. ∼Nγ εrad εgeo εlg εpmt (5) angle reconstruction procedure relies on the higly accurate The charge is then calculated according to expression 4, information (see Section I) of the particle direction (θ,φ) where the normalization constant can be evaluated from a provided by the tracker. The tagging of the hits signaling the calibrated beam of charged particles. For a more complete passage of the particle through the solid light guides in the description of the charge reconstruction method see Ref. [5]. detectionplaneprovidesanadditionaltrackelement,however, those hits are excluded from the reconstruction. The best IV. RESULTSWITHTHE RICH PROTOTYPE value of θ will result from the maximization of a Likelihood The RICH prototype was subject to cosmic muons and to c function, built as the product of the probabilities, p , that the in-beam tests using secondary nuclei from fragmentation of i detectedhitsbelongtoagiven(hypothetical)Cˇerenkovphoton 20GeV/c/nucleon lead (Pb) ions in a beryllium target and pattern ring, 158GeV/c/nucleon indium nuclei in a Pb target from the nhits L(θ )= psi[r (θ )]. (2) CERN SPS in 2002 and 2003, respectively [6]. c i i c In 2003 a monocromatic particle beam with a momentum i=1 This probability takes intYo account ri, the closest distance resolution 0.15%≤∆P/P≤1.5% was obtained. The optics of of the hit to the Cˇerenkov pattern and s the signal strength. the line was tuned to provide a beam as parallel as possible, i The probability of a hit belonging to the pattern is obtained with a divergence less than 1mrad. The beam section was by taking into account that it can either be part of the ∼1mm2 for the narrowbeamrunsand ∼1cm2 for the spread noise (essentially flat) or of the Cˇerenkov pattern (Gaussian beam runs. distributed). Expressing b as the photon background fraction, The beam nuclear compositioncould be selected according D as the detector’s dimensions and σ as the width of the to the desired A/Z value of the fragmentation products by residualsdistribution(ascanbeseeninrightplotofFigure 4), setting the beam line rigidity at the appropriate value. Three we can write: mainselectionvalueswereestablished:A/Z=2(4He,6Li,10B, b 12C,...); A/Z=2.25 to enhance the 9Be peak and A/Z=2.35 to Pi =(1−b)G(σ;ri)+ (3) enhance the indium peak. D Test Beam 2003 setFuipguinret3he(reigxhpte)rismhoewntsalaagreenaeHra8l-vSiPewS aotfCthEeR2N00.3Ttheestpbreoatom- mean npe190 MMCMIEEENCCCyyyy00002221.1...11100003353 3 223.0...223 c cccmmmm 1168 Test Beam 2002 CINy03.105 type was placed inside a light-tight container. The setup was 8 CINy02.104 3.0 cm 14 ciwnoimrtehpeplebrtoeepadomrw,tiiaothnTaOAl cFMhpSamrosbtioleitrcysop,netwtprolaaccokeregdradnlaoicywesnrcssitnrpetliaallmcaet,odtrwucoposmutrnuetaletmris- 4567 116802 MECy02.10M3ECy01.10M3ECy02.10C5INy02.103CINy02.104MECy03.10C3INy02.103 and during a certain period a plastic Cˇerenkov counter. The 3 4 two scintillators placed ∼1m apart in front of the prototype 2 2 cointainer,providedtheDAQtriggeraswellasanindependent 1 4 6 8 10 12 14 0 1 2 3 4 5 6 7 8 chargemeasurement.Thesilicon tracker prototypeprovideda p (GeV/c/nuc) ad Ladighadt Yieadld (adZ=1ad, 3 cadm)ad r r r r r r r r very precise measurementof the particle track parameters for Fig.5. Lightyieldasfunction ofprotonbeammomentumforthedifferent the event reconstruction as well as an external selection of aerogel samples tested in 2002 (left). Light yield comparison based on test charge. beam data 2002 and2003. All values wereextrapolated forfully cointained The purposes of the tests were testing flight front-end rings generated by a particle with β∼1 and in an aerogel radiator with a commonthickness of3cm(right). electronics,characterizetheperformanceoftheaerogelandthe NaFradiators,estimatingthemirrorreflectivityandevaluating sample producedin 2003 with 1.05 refractiveindexreflecting the global functionality of the prototype. A total number of theverygoodclarity(∼0.0055µm4/cm)oftheaerogelbatch. 11million eventswere recordedduringelevendays.Different Theresolutionoftheβ measurement,obtainedasexplained particle incidences were obtained by rotating the prototype in Section III, was estimated using a Gaussian fit to the setup with respect to the beam line (0o, 5o, 10o,15o,20o). reconstructed β spectrum, shown in left plot of Figure 6 for Theeventselectionwasmainlyintendedtoremovewrongly helium nuclei. Data were collected with the aerogel radiator reconstructed tracks, events with clusterized hits and events CINy03.105,2.5cmthicktogetherwithanexpansionheightof arising from later fragmentation. First, consistency between 35.31cm. The events shown correspond to particles inciding the external determination of the track transverse coordinates verticallyandgeneratingfullycontainedrings.Thebetarecon- and the estimation from the reconstructed ring is required. structed from a simulated helium data is also shown (shaded Theneventswithmorethanoneparticleclusterinthedetection histogram) superimposed with good agreement between data matrix are rejected. Futhermore, the Kolmogorov probability and Monte Carlo. of the event is calculated requiring an uniform azimuthal distribution of the ring hits. E3 1 The evaluation of the aerogel samples in order to make •DATA *1. 0.9 a final radiator choice was one of the key issues of these 102 MC Db/b 0.8 0.7 DATA tests. Different production batches from two manufacturers, 0.6 MC Matsushita Electric Co. (MEC) and Catalysis Institute of 0.5 10 Novosibirsk (CIN) with different refractive indexes, 1.03 and 0.4 0.3 1.05,wereanalyzed.Therequiredcriteriaforagoodcandidate 0.2 wereahighphotonyield,inordertoensureagoodringrecon- 1 0.1 struction efficiency and accurate β and charge measurements. -0.2-0.15-0.1-0.05 0 0.05 0.1 0.15 0.2 0 0 5 10 15 20 25 Theaerogellightyielddependsonthetile thicknessandon 1-BETA LIP x 10-2 Z(scint&std) its optical properties (refractive index and clarity). The light Fig.6. Comparisonofthe(β−1)∗103 distributionforheliumdata(black yield has been evaluated from the analysis of helium samples dots) and simulation (shaded) (left). Evolution of the β resolution with the collectedin2003andfromtheanalysisofprotondatasamples charge obtained for the same aerogel radiator. Simulated points for Z=2, 6, 16aremarkedwithfullsquares (right). gathered in 2002 with different beam momentum between 5 and 13GeV/c [7]. Thechargedependenceofthevelocityrelativeresolutionfor Figure 5 (left) shows the evolution of the light yield of the same radiator is shown in the right plot of Figure 6. The the different aerogel samples tested in 2002 with the proton differentchargeswereselectedusingexternalandindependent beam momentum. A fit to each set of data was applied measurements performed by the silicon tracker prototype and the light yield for a proton with β∼1, generating fully and by the two scintillators. The observed resolution varies contained rings in a radiator with a common thickness of according to a law ∝1/Z, as it is expected from the charge 3cm was extrapolated. Right plot of the same figure shows dependenceof the photon yield in the Cˇerenkov emission, up the normalized to 3cm thickness light yield for the different to a saturation limit set by the pixel size of the detection unit aerogel samples tested in 2002 and 2003. Two interesting cell. The function used to perform the fit is the following: features are enhanced. In one hand, the same sample of 2 A CINy02.103wasusedinbothyearsanditslightyieldanalysis σ(β)= +B2 (6) shows the same value which proves the setup stability and s(cid:18)Z(cid:19) the aerogel good performance after one year period; on the where, A meansthe β resolutionfor a singly chargedparticle otherhanditisnotoriousthehighestsignalcomesfromaCIN whileBmeanstheresolutionforaveryhighchargegenerating a large number of hits, here the resolution is dominated by TheRICHgoalofagoodchargeseparationinawiderange the pixel size as mentioned before. The fitted values are of nuclei charges implies a good mapping and monitoring of A = 0.872 ± 0.003 and B = 0.047 ± 0.001 for the run thepotentialnon-uniformitiespresentonthedetector.Inorder conditions stated above. Simulated data points for Z=2, 6, 16 tokeepthesystematicuncertaintiesbelow1%,theaerogeltile are marked upon the same plot with full squares. Once more thickness,therefractiveindexandtheclarityshouldnothavea the agreement between data and Monte Carlo measurements spreadgreaterthan0.25mm,10−4and5%,respectively;atthe for charges different of Z=2 is good. detectionlevelapreciseknowledge(< 5%level)ofthesingle The beta resolution for a β ∼1, helium nuclei impacting in unit cell photo-detection efficiency and gains is required. each of the aerogel samples tested in 2003 are summarized in Table I. The results are for a common expansion height HeLi zrec00.4.55 extrapolatedfromthevaluesmeasuredattheadjustedheights. BeB s C 0.4 All the tested radiators fulfill the RICH requirement for β H NOFNe 0.35 measurement. NaMg 0.3 Al I SiPSClArK 0.25 β resolution forZ=2,H=33.5cm CaSTciVCrMnCo 0.2 radiator CIN103 MEC103 CIN105 FeNi 0.15 σ(β)×103 0.421±0.003 0.435±0.002 0.459±0.004 0.1 TABLEI 0.05 BETARESOLUTIONFORAHELIUMPARTICLEWITHβ∼1OBTAINEDFOR 00 5 10 15 20 25 ALLTHEAEROGELSAMPLESTESTEDIN2003ANDEXTRAPOLATEDFORA Zrec RICH Z(scint&std) COMMONEXPANSIONHEIGHTOF33.5CM. Fig.7. Chargepeaksdistribution measuredwiththeRICHprototype using an=1.05aerogel radiator, 2.5cmthick. Individual peaks areidentified upto Z∼26(left).ChargeresolutionversusparticleZforthesameaerogelradiator. The distribution of the reconstructed charges in an aero- Thecurvegivestheexpectedvalueestimatedasexplainedinthetext(right). gel radiator of n=1.05, 2.5cm thick is shown in left plot of Figure 7. The reconstruction method used was the one describedinSection III.Thespectrumenhancesastructureof Runs with a mirror prototype were also performed and its well separated individual charge peaks over the whole range reflectivitywasderivedfromdataanalysis.Theobtainedvalue up to iron (Z=26). This spectrum was measured for a beam is in good agreement with the design value. selection of A/Z=9/4. V. CONCLUSIONS The chargeresolution for each nuclei, shown in rightpanel of Figure 7, was evaluated through individual Gaussian fits AMS-02 will be equippedwith a proximityfocusing RICH to the reconstructed charge peaks selected by the indepen- detector based on a mixed radiator of aerogel and sodium dent measurements performed by the scintillators and silicon fluoride, enabling velocity measurements with a resolution of trackerdetectors.Achargeresolutionforprotoneventsslightly about0.1% and extendingthe charge measurementsup to the better than 0.17 charge units is achieved and as expected the ironelement.Velocityreconstructionismadewithalikelihood best charge resolution is provided by this radiator due to its method. Charge reconstruction is made in an event-by-event higher photon yield. basis.Evaluationofbothalgorithmsonrealdatatakenwithin- beamtests atCERN, in October2003was done.The detector The charge resolution as function of the charge Z of design was validated and a refractive index 1.05 aerogel was the particle follows a curve that corresponds to the error chosen for the radiator, fulfilling both the demand for a large propagation on Z which can be expressed as: lightyieldandagoodvelocityresolution.TheRICHdetetector 1 1+σ2 ∆N 2 is being constructed and its assembling to the AMS complete σ(Z)= pe +Z2 . (7) setup is foreseen for 2008. 2s N0 (cid:18) N (cid:19)syst REFERENCES This expression describes the two distinct types of uncer- [1] C.Lechanoine-Leluc, Proc.29thICRC(Pune)3,381-384(2005). tainties that affect Z measurement: the statistical and the [2] F.Bara˜o, L.Arruda,etal.,Proc.29thICRC(Pune)9,299-302(2005). systematic. The statistical term is independent of the nuclei [3] C.Lechanoine-Leluc, Proc.29thICRC(Pune)9,299-302(2005). charge and depends essentially on the amount of Cˇerenkov [4] D.Casadeietal.,NuclearPhysicsB(Proc.Suppl.)113,133(2002). [5] F.Barao, L.Arrudaetal.,NIMA502,310(2003). signal detected for singly charged particles (N0 ∼ 14.7) and [6] P.Aguayo,L.Arrudaetal.,NIMA560,291-302(2006). on the resolution of the single photoelectron peak (σpe). The [7] R.Pereira, L.Arrudaetal.,In-beam aerogel light yield characterization. systematic uncertainty scales with Z, dominates for higher fortheAMS/RICHdetector ,AMSinternalnoteinpreparation charges and is around 1%. It appears due to non-uniformities at the radiator level coming from variations in the refractive index, tile thickness or clarity or due to non-uniformities at the photon detection efficiency like PMT temperature effects or light guide non-uniformities.