Latest results from MINOS 1. Overview. Making neutrinos 3. Detecting neutrinos 4. Results 5. The future David E. Jaffe Brookhaven National Laboratory for the MINOS Collaboration Argonne Athens Benedictine Brookhaven Caltech Cambridge Campinas Fermilab College de France Harvard IIT Indiana ITEP-Moscow Lebedev Livermore Minnesota-Twin Cities Minnesota-Duluth Oxford Pittsburgh Protvino Rutherford Sao Paulo South Carolina Stanford Sussex Texas A&M Texas-Austin Tufts UCL Western Washington William & Mary Wisconsin 1
Test the ν μ ν τ oscillation hypothesis Measure precisely Δm 3 and sin θ 3 Search for sub-dominant ν μ ν e oscillations Search for or constrain exotic phenomena Sterile ν, ν decay Compare ν, ν oscillations Test of CPT violation Atmospheric neutrino oscillations Phys. Rev. D73, 0700 (006) MINOS Physics Goals ν ν ν ν 3 ν ν 1 μ τ = U U U e1 μ 1 τ 1 Useful Approximations: e U U U ν μ Disappearance ( flavors): e μ τ U U U e 3 μ 3 τ 3 ν ν ν Δm 3 = m 3 m P(ν μ ν μ ) = 1 - sin θ 3 sin (1.7Δm 3 L/E) ν e Appearance: Units: Δm (ev ) L(km) E(GeV) P(ν μ ν e ) sin θ 3 sin θ 13 sin (1.7Δm 31 L/E) Where L, E are experimentally optimized and θ 3, θ 13, Δm 3 are to be determined 1 3
Disappearance measurement Generic long baseline ν μ disappearance experiment Predict unoscillated charged current (CC) spectrum at Far Detector (fixed L) Compare with measured Energy spectrum to extract oscillation parameters P( ν ν ) = 1 sin θ sin (1.67Δm μ μ 1 L / E) ν μ spectrum Unoscillated spectrum ratio Oscillated 1 Monte Carlo Monte Carlo ( Input parameters: sin θ = 1.0, Δm = 3.35x10-3 ev ) Neutral current (NC) background 3
Main Injector Neutrino Oscillation Search High power ν μ beam produced by 10 GeV protons from the Main Injector at FNAL Two functionally identical detectors: Near detector (ND) at Fermilab to measure the beam composition and energy spectrum Far Detector (FD), 735km away, in the Soudan Mine, Minnesota to search for evidence of oscillations L=735 km 4
Not to scale Neutrino production ~1 km Near Detector Moveable segmented graphite target variable beam energy Two parabolic magnetic focusing horns νor anti-ν beams 5
NuMI Neutrino Beam LE-10 configuration is most favorable for oscillation analysis and constitutes ~95% of total exposure Data taken in 5 other configurations for systematic studies LE-10 event composition: 9.9% ν μ, 5.8% ν μ, 1.3% ν e / ν e Expected number of Far Detector events without oscillations Beam LE-10 pme Target z position (cm) -10-100 FD Events* per 1e0 pot** 390 970 Position of oscillation maximum (Δm =0.00335eV,L=735km) phe -50 1340 *Events in fiducial volume **pot=protons-on-target 6
Neutrino detection FAR.54 cm thick magnetized (1.T) steel plates 4.1x1cm scintillator strips grouped into orthogonal U,V planes UV UV UV UV DETECTOR Steel Scintillator Veto Shield Orthogonal strips Coil Far Det Near Det Mass(kt) 5.4 1 Size(m3) 8x8x30 3.8x4.8x1 Steel/Scint. Planes 484/484 8/15 0 Oct 006 NEAR DETECTOR 7
Neutrino interaction identification ν μ CC Event NC Event ν e CC Event UZ VZ 3.5m 1.8m.3m Long muon track + hadronic activity at vertex Short showering event, often diffuse Short event with typical EM shower profile E ν = E shower + P μ Shower energy resolution: 55%/ E Muon momentum resolution: 6% range; 13% curvature 8
Pre-selecting ν μ CC Events Preselection for separating ν μ CC from NC events Data quality: Beam and detector monitoring cuts Preselection: At least one good reconstructed track Track vertex within detector fiducial volume Fiducial volume NEAR: 1m < z < 5m R<1m from beam center ν Calorimeter Spectrometer FAR: z>50cm from front face z>m from rear face R<3.7m from center of FD Fitted track must have negative charge (to reject ν μ ) 9
Selecting CC ν μ interactions Monte Carlo Related to Muon momentum Related to event inelasticity Related to de/dx P CC /P NC is the probability that a CC/NC event would be observed with these values where (P CC (P NC ), resp.) is the product of the three CC(NC) PDFs at those values PID= ( log( P CC ) log( PNC)) 10
Predicting the unoscillated FD energy spectrum p Target π + FD Decay Pipe E ν ~ 0.43E π / (1+γ π θ ν ) The unoscillated FD energy spectrum differs from the ND spectrum because the decay angles for neutrinos to reach the detectors differ Primary extrapolation method is matrix method that contains info of pion -body decay kinematics and beamline geometry (MC used to correct for energy resolution and acceptance) ND (Several methods were developed for the extrapolation) Near Energy spectrum Far Energy spectrum 11
MINOS Best-Fit Spectrum Best-fit spectrum for 1.7x10 0 POT MINOS MINOS NC Subtracted Δm 3 =.74 + 0.44 0.6 (stat + syst) 10 3 ev Measurement errors are 1σ, 1 DOF sin θ 3 = 1.00 0.13 (stat Normalization = 0.98 + syst) χ = nbins [ ( ei oi ) + oiln( oi ei )] + i= 1 nsys Δs σ j= 1 j s j 1
Fit includes penalty terms for three main systematic uncertainties Fit is constrained to physical region: sin (θ 3 ) 1 Allowed Region Δm sin 3 θ = 3.74 + 0.44 0.6 = 1.00 10 0.13 3 ev 13
Systematic Uncertainties Systematic shifts in the fitted parameters are computed using MC fake data samples for Δm =.7x10-3 ev and sin θ=1.0 The uncertainties considered and shifts obtained: Effect Shift in Δm (10-3 ev ) Shift in sin θ Near/Far normalization ±4% 0.050 0.005 Absolute hadronic energy scale ±11% 0.060 0.048 NC contamination ±50% 0.090 0.050 All other systematic uncertainties 0.044 0.011 Total systematic (summed in quadrature) 0.13 0.07 Statistical uncertainty (data) 0.36 0.1 Magnitude of systematic error is ~40% of statistical error for Δm Several systematic uncertainties are data driven improve with more data and study 14
Projected MINOS Sensitivity ν μ Disappearance MC MC MINOS MINOS sensitivity for different POT Current best values used as input: Δm 3=.74x10-3 ev sin θ 3 =1.00 Contours are 90% C.L. statistical errors only 15
ν μ ν e Oscillation Search P( ν μ ν e ) sin θ 3 sin θ 13 sin (1.7Δm 3 L/E) ν e CC Event (MC) Challenges to ν e CC signal selection Steel thickness.54cm = 1.44X 0, Strip width 4.1cm ~ Molière radius (3.7cm) typical few GeV ν e CC shower: 8planes x 4strips Backgrounds NC events (primary background) π 0 final states in hadronic system produce EM showers Intrinsic beam ν e are identical to signal High-y ν μ CC Hadronic shower dominates; muon track is very short or buried FD: Oscillated ν τ generally shower-like; τ decays to e - ~0% of the time ν e candidate identification based on compact shower with characteristic EM profile (several methods) ν μ CC NC ν e beam ν τ CC Total ν e osc 1.4 9.75. 1. 14.5 7.3 Neural Net selection results Oscillation parameters: sin (θ 13 ) = 0.1 Δm 3 =.7 10-3 ev sin (θ 3 ) = 1 POT = 4x10 0 16
Estimating ν e Backgrounds from Data Muon removal from CC events to estimate NC contribution Assumes similar hadron multiplicities/shower topologies Requires some corrections from MC Use horn-off data to resolve NC, ν μ CC background components NC component of background is enhanced after event selection Estimate μ + >e + ν e ν μ component from observed ν μ spectrum ν μ energy (GeV) 17
Projected MINOS Sensitivity ν e Appearance MINOS Preliminary Can improve on current best limit from CHOOZ Plot shows δ CP vs sin θ 13 for both mass hierarchies using MINOS νμ CC best fit values and 4x10 0 POT 10% systematic uncertainty on background included 18
Summary MINOS has completed a ν μ disappearance analysis of the first year of NuMI beam data Exposure used in analysis: 1.7x10 0 POT Results are consistent with the oscillation hypothesis with parameters: Δm 3 =.74 + 0.44 0.6 10 sin θ 3 = 1.00 0.13 3 Δm =.74 ± 0.8 10 Constraining the fit to sin (θ 3 ) = 1 yields: Systematic uncertainties under control and significant improvements expected with data driven studies & more statistics Accepted for publication in PRL (hep( hep-ex ex 0607088) Second year of running is underway. Stay tuned for new results on ν e appearance, sterile neutrinos, 3 ev 3 ev 19
Extras 0
Near Detector Located at FNAL 1040m from target 103m underground 980 ton mass 3.8m x 4.8m x 16m 8 steel + 153 scintillator planes Two distinct sections: Front: Calorimeter Every plane instrumented Back: Spectrometer One in five planes instrumented Fast QIE electronics Continuous (19ns) sampling in spill Plane installation fully completed on Aug 11, 004 1
Far Detector Located at Soudan mine, MN 735 km from target 705m underground 5.4 kton mass 8m x 8m x 30m 484 scintillator planes 8x optically multiplexed VA electronics Veto shield for cosmic ray rejection in atmospheric ν analysis GPS time stamping to synchronize FD to ND Main Injector spill times sent to FD for beam trigger 0 Oct 006 Veto Shield Coil Data taking since ~ September 001 Installation fully completed in July 003.
1 st Year of NuMI Running Spill Intensity (1e1 POT) First neutrinos in ND Start of LE running Integrated 10 0 POT First Analysis Data Set ( 0.93x10 0 POT) Final Analysis Data Set (1.7x10 0 POT) Main Injector Shut-down 3
Accumulated protons on NuMI target LE Now Scheduled shutdown Horn-1 Cooling problem Target problem 4
Near Detector Distributions Event rate is flat as a function of time Horn current scans on July 9 Aug 3 Different tunes in Feb Acceptance well reproduced Track angle w.r.t. vertical exhibits characteristic -3 o to Soudan MINOS Preliminary 1.7 10 0 POT X Vertex Z Vertex Track Angle (wrt vert.) ND ND ND ND Mean 9.76 RMS 15.75 Mean 9.80 RMS 15.55 Area normalised 5
Near Detector Energy Spectra LE-10 pme phe LE-10 pme phe Error envelopes shown on the plots reflect uncertainties due to cross-section modelling, beam modelling and calibration uncertainties 6
Hadron Production Tuning Parameterize Fluka005 prediction as a function of neutrino parent x F and p T Perform fit which reweights parent x F and p T to improve data/mc agreement Horn focusing, beam misalignments included as nuisance parameters in fits Small changes in x-section, neutrino energy scale, NC background also allowed LE-10/170kA LE-10/00kA LE-10/185kA pme/00ka Weights applied vs p z & p T LE-10/185kA MC phe/00ka Horn off Distribution of pions producing MINOS neutrinos 7
Beam Composition (MC) Composition of Charged-Current (CC) Events 9.9% ν μ 5.8% ν μ 1.% ν e 0.1% ν e
Selecting Far Detector Beam Events LE-10 configuration running from May 0 th 005 to March 3 rd 006 Total integrated POT: 1.7x10 0 Far Detector live time: 98.9% (POT weighted) Several software triggers in DAQ to read out FD activity: 4/5 plane trigger, minimum energy trigger, beam spill trigger Beam spill trigger reads out all activity in 100μs around spill signal (10μs duration) Possible due to GPS time stamping at ND & FD Event rate shows no time dependence MINOS FD Events/10 18 POT vs Time Events/10 18 POT 9
Selecting Far Detector Beam Events In addition to applying cut on event selection parameter apply cuts to reject cosmic ray (CR) background 53 o cut around beam axis Beam events have distinctive topology - tracks point to FNAL Demand that: -0μs < (event time spill signal) < 30μs Timing of neutrino candidates consistent with spill signal Two CR background estimates: Sideband analysis of region outside timing cut using full 1.7x10 0 POT sample upper limit of 0.5 events Using fake triggers in anticoincidence with spill.6m triggers no events selected upper limit of 0.5 events 30
Far Detector Distributions Predicted no oscillations (solid) Best fit (dashed) Track Vertex r (m ) MINOS 1.7 10 0 POT Event Classification Parameter MINOS 1.7 10 0 POT y = E shw /(E shw +P μ ) MINOS 1.7 10 0 POT 31