arXiv:0707.1115v1 [hep-ex] 7 Jul 2007 protons-on-target collected - PDF document

Flavor Physics and CP Violation Conference, Bled, 2007 1 The Search for ν µ → ν e Oscillations at MiniBooNE H. A. Tanaka, for the MiniBooNE collaboration Department of Physics, Joseph Henry Laboratories, Princeton University, Princeton, New Jersey, 08544 United States of America MiniBooNE (Mini Booster Neutrino Experiment) searches for the ν µ → ν e oscillations with ∆ m 2 ∼ 1 eV 2 /c 4 indicated by the LSND experiment. The LSND evidence, when taken with the solar and atmospheric neutrino oscillations, suggests new physics beyond the Standard Model. However, this evidence has not been confirmed by other experiments. MiniBooNE has completed its first ν µ → ν e oscillation search using a sample of ∼ 1 GeV neutrino events obtained with 5 . 58 × 10 20 protons delivered to the Booster Neutrino Beamline. The analysis finds no significant excess of ν e events in the analysis region of 475 − 3000 MeV. 1. Introduction a search for ν µ → ν e oscillations using 5 . 58 × 10 20 arXiv:0707.1115v1 [hep-ex] 7 Jul 2007 protons-on-target collected in neutrino mode[24]. The resulting neutrino beam is > 99% pure in ν µ and peaks The long-standing deficits in observed solar elec- at ∼ 700 MeV, with a small component of ν e coming tron neutrinos ( ν e ) [1, 2, 3, 4, 5, 6, 7] and atmo- from three-body decays of muons and kaons. spheric muon neutrinos ( ν µ ) [8, 9, 10, 11, 12] now have The MiniBooNE detector [25] is situated 541 me- a firm explanation in terms of the flavor oscillations ters downstream of the target, near the axis defined of the three neutrinos in the Standard Model. This by the beam. This distance, along with the energy interpretation has been confirmed with experiments of the ν µ beam, matches the distance/energy ratio of using terrestial neutrino sources[13, 14, 15, 16] and the LSND experiment ( ∼ 0 . 8m / MeV), with the result point to oscillations with mass-squared differences of 12 ∼ 8 × 10 − 5 eV 2 /c 4 and ∆ m 2 23 ∼ 3 × 10 − 3 eV 2 /c 4 that the oscillation probability is similar in the two ∆ m 2 experiments. The detector is a 12.2 meter diameter for the solar and atmospheric oscillations, respectively. sphere filled with 800 tons of undoped Marcol 7 min- However, the indications for ν µ → ν e oscillations reported by the LSND collaboration with ∆ m 2 ∼ eral oil with an index of refraction of ∼ 1 . 47. The sphere is divided into two concentric, optically decou- O (1 eV 2 )[17] cannot be accommodated within this pic- pled regions at 575 cm radius, resulting in “main” and ture. Taken with other constraints, such as the num- “veto” regions. The main region is instrumented on its ber of light neutrinos with standard weak couplings outer surface with 1280 inward-facing 8” photomulti- obtained from measurements of the Z width[18, 19], plier tubes (PMTs) covering 10% of the surface. The this result would require a dramatic departure from veto region is instrumented with 240 8” PMTs. Neu- the Standard Model in the form of “sterile” neutri- trino interactions within the main region are identified nos without standard weak couplings and/or exotic via the Cherenkov radiation produced by the charged forms of symmetry violation[20]. The LSND results particles emerging from the interaction and detected have not been confirmed, though other experiments on the PMT array. Cosmic muons entering the main searching for neutrino oscillations with similar values of ∆ m 2 have not had sufficient sensitivity to rule it region can be identified and rejected by the light pro- duced and detected in the veto region. The PMT out completely[21, 22]. activity in a 19.2 µ s window around each batch of The Mini Booster Neutrino Experiment (Mini- protons delivered from the Booster is recorded, where BooNE) searches for the neutrino oscillations indi- the expected arrival time of the neutrinos occurs 4.6 cated by the LSND result using an O (1 GeV) ν µ to 6.2 µ s after the start of the window. Other ac- beam produced by 1.6- µ s-long pulses of 8 GeV pro- tivity in the detector, in the form of random triggers tons from the Fermilab Booster synchrotron[23]. Typ- ically, ∼ 4 × 10 12 protons are delivered in such a pulse and calibration data, is recorded to study the detector response and non-beam backgrounds. at a rate of ∼ 4 Hz. The protons in each spill are measured by two toroids before they impinge on a 71- cm-long beryllium target, where the p -Be interactions 2. The Booster Neutrino Beam produce secondary mesons (dominated by pions, with a small contribution from kaons), which in turn decay in a 50-meter-long decay region following the target The predicted neutrino flux at the detector to produce the neutrino beam. The beryllium target is obtained from a GEANT4-based Monte Carlo is embedded within an electromagnet (“horn”) pulsed simulation[26]. The simulation includes a detailed de- synchronously with the beam with a 174 kA current scription of the beamline geometry, the spatial and to produce a toroidal magnetic field that focuses pos- kinematic properties of the primary proton beam inci- itive(negative) particles, resulting in a ν ( ν )-enhanced dent of the target, and pion and kaon production mod- beam. In its first results, MiniBooNE has performed els tuned to the available data. The primary protons fpcp07 122

2 Flavor Physics and CP Violation Conference, Bled, 2007 Figure 1: Absolute particle production rates in 8 GeV p -Be interactions for π + (left) and K + (right). The π + measure- ments show the double differential cross sections measured by the HARP experiment as a function of π + momentum in bins of the polar angle of the π + relative to the incoming proton. The K + measurements show invariant cross sections measured with proton energies of 9.5-24 GeV and extrapolated to 8 GeV using the Feynman scaling hypothesis. The solid red(black) lines for π + ( K + ) show the parametrization obtained by fitting the data, while the dashed lines show the variation in the parametrized cross sections when the parameters are varied according to their uncertainties. K 0 production. and secondary particles are tracked through the beam- The K + production in p -Be interactions has been line geometry, accounting for the magnetic field due to the horn, as well as electromagnetic and hadronic in- measured by a number of experiments. However, no teractions in the material (which can produce further measurements at 8 GeV proton energy exist. As a re- particles), until a particle decays to produce a neu- sult, measurements at energies between 9.5 and 24 trino. In the case where a decay chain has multiple op- GeV[29] are used, with the Lorentz-invariant cross portunities to produce neutrinos ( e.g. π + → µ + + ν µ , sections extrapolated to 8 GeV using an assumption µ + → e + + ν µ + ν e ), all decays producing neutrinos of scaling in the Feynman x variable[30]. The scaled are recorded. invariant cross sections are parametrized as a function of x , with the results shown on the right in Figure 1. The primary source of the neutrino flux is the decay As with the π + Sanford-Wang parametrization, this of π + from p -Be interactions. The HARP experiment Feynman scaling-based function is used directly in the at CERN has measured the momentum and angular simuation to determine the multiplicity and kinemat- distribution of π ± production for 8 GeV protons on ics of the outgoing K + in p -Be interactions. a thin beryllium target[27]. Similar measurements at Cross sections for three categories of hadron-nuclear 6 GeV /c and 12 . 3 GeV /c proton momentum are avail- interactions (elastic, quasi-elastic, and inelastic scat- able from the BNL E910 experiment. These measure- tering) have been adjusted to match existing measure- ments are summarized as absolute double differential ments for proton, neutron and charged pion on beryl- cross sections in pion momentum and angle (relative lium and aluminum[31, 32, 33, 34]. In some cases, to the incident proton) and parametrized using the theoretical guidance [35] is used to infer cross sec- Sanford-Wang function[28]. The parameters of the tions where direct measurements do not exist. In all function are obtained via a fit to the three datasets. other hadron-nucleus interactions, the GEANT4 de- The results are shown on the left in Figure 1, where fault cross sections are used. the parametrization (in red) is overlayed on the π + double differential cross sections from the HARP ex- The predicted neutrino flux at the MiniBooNE de- periment. The cyan curves show the variations in the tector is shown in Figure 2. The dominant part of function when the underlying parameters are varied the ν µ flux, particularly at < 2 GeV, is due to pion according to the uncertainties returned by the fit. The decays, while the kaon contribution becomes larger function is used directly in the simulation to deter- for energies > 2 . 5 GeV. For the ν e flux, muon decay mine the multiplicity and kinematics of π + mesons is the largest component, while kaon decays become emerging from p -Be interactions. A similar procedure dominant at > 1 GeV. A number of systematic un- is used for the π − production based on the HARP and certainties in the flux prediction are considered. The E910 data, while an E910 analysis of K 0 S production largest uncertainty arises from the modeling of the sec- is used to obtain a Sanford-Wang parametrization of ondary particle production, while uncertainties in the fpcp07 122