HARP Hadron production experiments for neutrino physics Jaap Panman CERN Hadron production for neutrino experiments: Neutrino fluxes in conventional beams Prediction of atmospheric neutrino flux Neutrino factory design Hadronic interaction generators HARP measurements p-al (KK) Outlook p-be (MiniBooNE) MIPP p-ta (neutrino factory) NA49 p-c (atmospheric neutrinos) 1
Hadron yields for neutrino beams Different generators applied to MiniBooNE beam New measurements essential! (Dave Schmitz -NBI5) Typical errors Absolute rates: ~15% Ratios: ~5% Past: single arm spectrometers at CERN: Eichten et al., Atherton et al., SPY Present: Open geometry, heavy ion-like experiments BNL-E91, HARP, NA49, MIPP SPY: 1996, ±15mrad, ± 3mrad
The HARP experiment TPC Total Acceptance Forward Spectrometer 3
Pion Id. (forward spectrometer) e 1 p (GeV/c) 3 4 5 6 7 CAL CHERENKOV CALORIMETER p k TOF TOF CHERENKOV CHERENKOV TOF CHERENKOV TOF pi/p response for beam particles Cherenkov pi/e response for negative particles Below pion threshold Above pion threshold e p p e Nphe Nphe 4
HARP measurement for KK beam KK: Disappearance experiment to confirm atmospheric oscillation Oscillation probability at 5 km from the source for atmospheric parameters: maximum effect at ~1GeV Non-oscillated spectrum must be measured near the neutrino source oscillations: energy-dependent suppression of the spectrum 5
Far-Near flux ratio beam 5km far near For a point-like source the flux, in the absence of oscillations scales like 1/R (E )far = R(E ) (E )near If the near detector does not see a point-like source it is necessary to multiply by a factor R(E ) to obtain the predicted spectrum in the far detector. flux: 99% from + decay The determination of R(E ) is essential since the signal is a distortion of the energy spectrum (a wrongly determined R(E ) could distort oscillations) Oscillation peak R(E ).5 1. 1.5..5 6
HARP measurement for KK Al target Protons 1.9 GeV HARP Data taken with the parameters of the KK beam (p-al at 1.9 GeV/c) ( thin 5% target) 7
Cross-Section calculation d dpd N p correction factors p, Npot CORRECTIONS: Track reconstruction efficiency ~ G 1 % (Data) Geometrical correction ~ G 5-8% (Analytical) Absorption/decay in upstream detector ~ G 15 % (MC) Spectrometer acceptance ~ G 1-5 % (MC) Tertiary production in upstream materials ~ H 5 % (MC) Electron veto efficiency correction ~ G 5 % (Data) Kaon subtraction ~ H 1-3 % (Data) PID efficiency ~ G < 5% (Data) PID p- migration ~ H < 5% (Data) 8
Event Selection RPC TPC MWPC ITC BS HALO A BC B TDS HALO B 3 4 1 TOF B BC A TOF A target FTP Event selection for protons on target ( normalization trigger ): impact point and direction of primaries (BS, TDS, HALO A, HALO B) protons: identified bytof and Cherenkovs (TOF A, TOF B, BC A, BC B) Event selection for proton inelastic interactions ( physics trigger ): normalization trigger && forward trigger scintillator plane (FTP) 9
Track Reconstruction NOMAD drift chambers: 5 modules x 4 (chambers/module) x 3 (planes/module) Top view x z NDC4 Vertex4: use for fit Vertex4: Do not use for fit NDC1 dipole magnet NDC NDC5 target beam Vertex: use for fit Vertex: Do not use for fit 1 B NDC3 Two independent ways to reconstruct tracks: Vertex tracks: 3D track segment downstream, plus vertex match used to measure pion yield Vertex4 tracks: 3D track segment downstream, plus 3D segment in NDC1 used to measure track reconstruction efficiency in data Number of reconstructed tracks in analysis: 1, 1
dipole magnet NDC NDC5 Data MC 1. 1.1 1.999 down stream efficiency 1.1 1.999.998.997.996.996 -.15 -.1 -.5 -.5.1.15 Data MC 1..995. -.8 qx ( rad) 1.3 1.1 1.999.996 5 6 7 8..4.995.75-1 -8 1 p qx qy 6.5 GeV/c mrad 8 mrad 3 4 5 reconstruction efficiency reconstruction efficiency 1.8.6.4 -. -.15 -.1 -.5 -.5.1 1..15 Data MC 1.8.6.4 Data MC 1.8.6.4. -.8 qx ( rad). 1.. reconstruction efficiency 6 p ( GeV/c). reconstruction efficiency overall efficiency nearly constant and ~9% (in used phase space) conditions: Match downstream with vertex TOF measurement Data MC.8 qy ( rad) Data MC p ( GeV/c) 1..6 1.996 4.999.997 3 -. 1.1.998 -.4 1..997 1 -.6 1.3.998.995 Data MC 1..997 -. Essential: downstream efficiency nearly 1% 1.3.998.995 B down stream efficiency 1.3 down stream efficiency down stream efficiency Reconstruction efficiency 3 4 5 6 7 8 p ( GeV/c) -.4 -...4 1..6.8 qy ( rad) Data MC 1.8.6.4.75-1 -8. 1 -.6 1 p qx qy 6.5 GeV/c mrad 8 mrad 3 4 5 6 11 p ( GeV/c)
Detector response measured from data 99% pions in a sample of negative particles with e-veto + p + and - have the same behaviour k+ Above pion Cherenkov threshold pions are suppressed to less than 1% by Nphe<3 /p/k are clearly separated by the TOFW below 3 GeV Fit the inclusive beta distribution to a triple Gaussian with fixed shapes and free normalization p k+ 1
d s / (dp dw) (mb / (GeV/c sr)) HARP Results (p-al at 1.9 GeV/c) 6 6 6 3-6 mrad 6-9 mrad 9-1 mrad 4 4 4 6 4 6 6 1-15 mrad 4 6 6 15-18 mrad 4 4 Red histo: HARP S-W parametrization 4 6 4 6 4 6 18-1 mrad 4 4 6 p (GeV/c) HARP results in black, parametrization of HARP results in red 13
Error Evaluation systematic error evaluation performed, to quantify errors on: d dpd p, Typical error: 8.%.75 p 6.5GeV c, 3 1mrad Error on total cross-section: 5.8% Dominant error contributions: Overall normalization Tertiary subtraction Momentum scale Statistics 14
Parametrization of HARP Data HARP data on inclusive pion production fitted to Sanford-Wang parametrization: d p Al dpd where: X p, c c1 p 1 p p beam exp pc c3 c c6 p beam 4 5 c8 p c7 pbeam cos X : any other final state particle p beam 1.9 : proton beam momentum GeV c p, : d dpd c1, momentum GeV c, angle rad units: mb GeV c sr, where d d cos, c8 : emprical fit parameters Sanford-Wang parametrization used to: Use HARP data in KK beam MC Translate HARP pion production uncertainties into flux uncertainties Compare HARP results with previous results (similar beam momentum, phase space) 15
Comparing HARP With Previous Results Forward pion production (< mrad). + Proton beam momentum between 1 and 15 GeV/c, ds Restrict comparison to: / (dp dw) (mb / (GeV/c sr)) Reasonable agreement between HARP and previous results on 8 6 pbeam =1.9 GeV/c q=89 mrad sn=16 Y 4 6 pbeam =1.1 GeV/c q=61 mrad s N=5 Y 4 5 sn=15 Y 4 p beam =1.1 GeV/c q=61 mrad s N= Y 4 5 8 Abbott 9 6 pbeam =14.6 GeV/c q=164 mrad s N=15 Y 4 5 6 5 8 q=134 mrad Vorontsov 83 Abbott 9 pbeam =14.6 GeV/c X 8 Vorontsov 88 8 6 Red curve: HARP S-W parametrization 8 Sugaya 98 p Al Abbott 9 6 p beam =14.6 GeV/c q= mrad s N=15 Y 4 5 5 p (GeV/c) 16
KK Neutrino Flux Predictions Similar neutrino energy shapes arising from KK default and HARP pion production assumptions in KK beam MC Far/Near ratio Neglecting neutrino cross-sections, efficiencies, etc.: F N F N KK default HARP.864 1.846 1 6 6 HARP pion production uncertainty: <1% error on overall F/N flux ratio (but there are more error sources) 17
MiniBooNE Neutrino Beam Log scale Drawing not to scale Decay region 1.8 m 5 m 5 m 45 m e + + E (GeV) e+ e Relative neutrino fluxes K + e+ e,k- K ±. e. e,k+ p Target +horn K + + + + K ( ) ~3% of all 's ~9% of all flux } } e ~.6% of all 's ~3% of all 's 18
HARP Beryllium Thin Target Results Preliminary double differential + production cross sections from the Be 5% target are available.75 < p < 5 GeV/c 3 < q < 1 mrad q (mrad) preliminary p (GeV/c) Momentum and Angular distribution of pions decaying to a neutrino that passes through the MB detector. Data with target replica and thick targets of different lengths shape of target visible 19
Comparison with older data: Be
Neutrino factory maximize ± production rate Optimize: target material and geometry primary beam energy collection scheme Experimental data are poor (small acceptance, few materials) and old (Allaby et al.197, Eichten et al. 197) Measure pt distribution with high precision (<5%) Existing simulation packages show large discrepancies on pion yields and distributions 1
Why do we need measurements? Total Yield of + and : +GEANT4.5 Stephen Brooks, Kenny Walaron NuFact 5. GEANT4 Pi+ LHEP-BIC GEANT4 Pi- LHEP-BIC Pion/(Proton*Energy(GeV)) GEANT4 Pi+ QGSP GEANT4 Pi- QGSP GEANT4 Pi+ QGSP_BIC GEANT4 Pi- QGSP_BIC.15 GEANT4 Pi+ QGSP_BERT GEANT4 Pi- QGSP_BERT GEANT4 Pi+ LHEP GEANT4 Pi- LHEP GEANT4 Pi+ LHEP-BERT.1 GEANT4 Pi- LHEP-BERT GEANT4 Pi+ QGSC GEANT4 Pi- QGSC MARS15 Pi+ MARS15 Pi-.5. 1 3 4 5 6 7 8 9 1 Proton Energy (GeV)
HARP large angle analysis Large angle detectors TPC Pattern recognition Momentum de/dx RPCs TOF 3 3
Elastic scattering: TPC response d P c Pa P 1Gev/c 3 Gev/c a X b c a d b d M p p p ( p p p ) missing mass peak prec (GeV/c) q (degree) 4
PID with de/dx in TPC positives negatives 5
Preliminary measurement of pion yields 3 GeV/c p-ta data Backward production + - Forward production 6
Hadron production for atmospheric neutrinos Input for precise calculation of the atmospheric neutrino flux (from yields of secondary, K) Uncertainty now dominated by hadron interaction model About 3% uncertainty in extrapolations Target Target length material (l%) Cryogenic targets or carbon Beam Momentum (GeV) #events (millions) Be C Al Solid targets Cu (1) 5 Sn Ta 1 Pb KK Al MiniBooNE Be Cu button Cu Cu skew Cu Water Negative only % and 5% 15.7 +8.9.56 +1.9, +15 1.71 +1 1.69 6 cm ±3 ±5 ±8 ± 1 ± 15 58.43 H 18 cm ±3, ±8, ±14.5 13.83 H 1, 1 +1.5, +8(1%) 9.6 8 D1 H1 33.16 +1.9 5, 5, 1, replica N7 Cryogenic targets ±3 ±5 ±8 ± 1 ± 15 7
Daughter energy Atmospheric neutrinos: range covered 1 MIPP NA49 1 TeV Boxes show importance of phase space region for contained atmospheric neutrino events. New measurements. 1 HARP 1 1 GeV 1 GeV 1 G. Barr 1 1 TeV 1 Parent energy 8
Error in neutrino flux ratio: Up/Horiz. Most important sources of uncertainty HARP energy regime pions HARP energy regime kaons Next analysis with forward spectrometer: carbon G. Barr 9
Outlook HARP after analysis for KK and MiniBooNE beams: tantalum at large angles for the neutrino factory carbon (forward spectrometer) for atmospheric neutrinos thick targets to understand better reinteractions many more targets, momenta (e.g. Cryogenic targets) MIPP at FNAL large energy range: will go to higher energies possibility to use NA49 at CERN SPS TK beam atmospheric neutrinos and muons 3
MIPP: FNAL-E97 Approved November 1 Technical run 4 Physics data taking 5 Uses 1GeV Main Injector Primary protons to produce secondary beams of K p from 5 GeV/c to 1 GeV/c to measure particle production cross sections of various nuclei including hydrogen. Using a TPC they measure momenta of ~all charged particles produced in the interaction and identify the charged particles in the final state using a combination of de/dx, ToF, differential Cherenkov and RICH technologies. Open Geometry- Lower systematics. TPC gives high statistics. 31
MIPP Physics Interest Particle Physics-To acquire unbiased high statistics data with complete particle id coverage for hadron interactions. Study non-perturbative QCD hadron dynamics, scaling laws of particle production Investigate light meson spectroscopy, pentaquarks?, glueballs Investigate strangeness production in nuclei- RHIC connection Nuclear scaling Propagation of flavor through nuclei Atmospheric neutrinos Cross sections of protons and pions on Nitrogen from 5 GeV- 1 GeV Improve shower models in MARS, Geant4 Make measurements of production of pions for neutrino factory/muon collider targets Proton Radiography Stockpile Stewardship- National Security MINOS target measurements pion production measurements to control the near/far systematics Nuclear Physics Service Measurements Upgrade programme: Speed up TPC DAQ by using ALICE ALTRO/PASA chips. green light to acquire these chips from CERN ($8K). Speed up rest of DAQ. 3
MIPP TPC and Cherenkov DATA EOS TPC de/dx spectrum in MIPP run RICH rings 33
RICH ring 5 GeV p-c event TOF bars 34
MIPP Timeline Run till next shutdown in current mode Acquire Altro/PASA chips Design New TPC Sticks Get approval for proposal: Run will be aproved when results from existing data shown Get new collaborators Run in 6 (end of 6) in upgraded mode with current beam. Design lower momentum beam. Beam Cherenkovs may need redesign (too much multiple scattering) Lots of graduate student theses Possible to affect shower simulators on 7 time frame. 35
TK far-near ratio.5 p OA deg. MARS FLUKA ND 8m SK 95km Flux @SK = Flux @ND [Far/near ratio] From MC simulation The Difference of Far/near ratio is ~1.9±.4% 1.8 1 Eν [GeV] Double ratio (FLUKA/MARS) 1. 1. (Energy range:.4>1.[gev]) needed p-c data at 4 GeV/c Eν[GeV] 36
An existing facility: NA49 particle ID in the TPC is augmented by TOFs rate somehow limited (optimized for VERY high multiplicity events). order 16 event per week is achievable (electronic upgrade needed!) NA49 is located on the H fixed-target station on the CERN SPS. secondary beams of identified, K, p; 4 to 35 GeV/c momentum Measurements relevant for atmospheric neutrinos and NuMI have been performed in with two beam settings (1 and 158 GeV/c) with a 1% Carbon target New collaboration forming around hadron/heavy ion physics, and yields for atmospheric muons/neutrinos and neutrino beams 37
Summary (HARP) The HARP thin target analysis for Al (KK) and preliminary data for Be (MiniBooNE) have been shown. These will significantly improve the knowledge of these beams The result is in general compatible with and more precise than, older data available in Be and Al. Further results (Thick target, +/ -, K/, A and E dependence, Carbon) will appear over the next 6-1 months. Our goal is to make a major contribution to the understanding of neutrino fluxes for accelerator neutrinos as well as atmospheric fluxes. Preliminary +/ - yields in the large angle region have been obtained for Ta. The Ta analysis will be completed over the next few months. Other targets and energies will be analyzed at large angle. Our goal is to make a major contribution to the design of the Neutrino Factory. The elastic analysis provides a clean way to calibrate the momentum scale of the detector and to assess the impact of distortions. 38
Conclusions Neutrino oscillation experiments move from discovery to precision measurements Knowledge of neutrino cross-section and neutrino production is essential Hadron production measurements should be seen as integral part of the Neutrino Experiments Present trends (see HARP & MIPP) are Full-acceptance detectors (single arm spectrometers in the past) High statistics Characterization of the actual neutrino beam targets to reduce MC extrapolation to the minimum Direct interest of neutrino experiments in hadron production atmospheric neutrino predictions need similar measurements 39
4