e oscillation study in MINOS Tingjun Yang Stanford

  • Slides: 10
Download presentation
 e oscillation study in MINOS Tingjun Yang, Stanford University APS April Meeting 2006,

e oscillation study in MINOS Tingjun Yang, Stanford University APS April Meeting 2006, Dallas, Texas Outline: • Introduction • e identification in the MINOS detectors • Background studies using the MINOS near detector • Conclusion 1 2022/1/11

Goal of e oscillation study: measuring 13 i = ( 1, 2, 3, …):

Goal of e oscillation study: measuring 13 i = ( 1, 2, 3, …): mass eigenstates with mass mi = (m 1, m 2, m 3, …), mij=mi 2 -mj 2 = ( e, , , …): flavor eigenstates MNS matrix |Ue 3|2 = sin 2( 13) Normal Inverted 3 2 1 Solar sin 2(q 13) 2 1 Atmospheric 3 e m t Results from CHOOZ m 2 = 0. 003 e. V 2 • No evidence of oscillations in mode disappearance • sin 2(2 13) <0. 12 at 90% CL for | m 32|2 = 3 10 -3 e. V 2 2 2022/1/11

Goal of e oscillation study: measuring 13 (continued) P( e) = sin 2( 23)

Goal of e oscillation study: measuring 13 (continued) P( e) = sin 2( 23) sin 2(2 13) sin 2(1. 27 m 132 L/E) - ignoring matter effect, solar terms and CP violating phase - E: neutrino energy(Ge. V) L: distance neutrino travels(km) – 735 km • Neutrino beam provided by 120 Ge. V protons from the Fermilab Main Injector. LE p. ME • A Near detector at Fermilab to measure energy spectrum and understand the background • A Far detector deep underground in the Soudan Mine, Minnesota, to search for e signals from oscillation p. HE Position of osc. maximum for 3 Dm 2=0. 003 e. V 2 2022/1/11

signal/background separation in the MINOS detectors MINOS far detector: 5. 4 kton mass, 8

signal/background separation in the MINOS detectors MINOS far detector: 5. 4 kton mass, 8 8 30 m, 484 steel/scintillator planes MINOS near detector: 1 kton mass 3. 8 4. 8 15 m, 282 steel and 153 scintillator planes steel thickness: 2. 54 cm ~ 1. 44 X 0 strip width: 4. 12 cm (Molière radius ~3. 7 cm) Primary Background: NC interaction l + N l + X Transverse Direction Signal: e CC interaction: e + N e + X Beam direction • compact, with typical EM shower profile Beam direction • often diffuse and scattered Other background components: beam e, high-y CC interactions, 4 oscillated in the far detector 2022/1/11

signal/background separation in the MINOS detectors (continued) A lot of effort has been devoted

signal/background separation in the MINOS detectors (continued) A lot of effort has been devoted to the shower reconstruction in order to distinguish between electromagnetic shower and hadronic shower. A few different discriminating techniques have been tried to enhance signal/background separation: cuts, Multivariate Discriminant Analysis, ANN based on shower sampling, ANN based on shower reconstruction. ANN PID at the FD CC One example analysis – Neural Net sin 2(2 13) = 0. 04, | m 31|2 = 2. 5 10 -3 e. V 2, sin 2(2 23) = 1, POT=15 e 20 eosc was scaled up by a factor of 10 for clarity. Figure of Merit = signal/sqrt(background) = 1. 26 CC NC ebeam CC Total background eosc 15. 6 54. 1 10. 6 4. 3 84. 6 11. 6 5 2022/1/11

Estimating NC background using the muon-removal technique Transverse Direction Remove the muon in a

Estimating NC background using the muon-removal technique Transverse Direction Remove the muon in a selected CC event and use the rest of the event as a fake NC event. muon track removed Beam direction This technique is a direct estimate of the NC background after some corrections, provided that the difference in hadron multiplicity does not change the event topology too much: • CC selection efficiency and purity • CC oscillation probability in the far detector • CC/NC cross section ratio 6 2022/1/11

_ Constraining the e flux from measurements Primary source of low energy _beam e

_ Constraining the e flux from measurements Primary source of low energy _beam e is a measurement of low energy can be used to constrain the e flux No. of events True energy of beam e at the ND True energy of at the ND Ecut E (Ge. V) The majority of beam e background in the energy region we are interested in is from + decay No E (Ge. V) from m+ above this energy (Ecut) This is what we are trying to measure 7 2022/1/11

Estimating background uncertainties using horn off data True energy of true at the ND

Estimating background uncertainties using horn off data True energy of true at the ND If we turn off the horns, the pions will not get focused and the peak in the neutrino energy spectrum will disappear. After we apply the same e selection cuts, we will get a NCenriched sample. Non = NNC + NCC + Ne Noff = r. NC*NNC + r. CC*NCC+re*Ne (1) (2) Can be solved to get NC and CC background Non, Noff: selected e candidates with horn on and horn off – will be measured Ne: beam e background with horn on – from MC r. NC(CC, e)=NNC(CC, e)off/NNC(CC, e) – from MC NNC, NCC: NC, CC background with horn on – will be calculated based on eqn. (1) and (2) 8 2022/1/11

Estimating background uncertainties using horn off data (continued) The advantage of this technique: can

Estimating background uncertainties using horn off data (continued) The advantage of this technique: can separate different backgrounds and estimate the uncertainty of each component. MC simulation: 1. 5 e 18 POTs horn off and ~1 e 19 POTs horn on data: Non = 608. 4 Non = 0 Noff = 189. 1 Noff = 13. 8 r. NC = 0. 425 r. CC = 0. 107 re = 0. 165, r/r =10% Ne = 96. 8 Ne = 19. 4 – assign a 20% systematic error Expected background at Near Detector for 1. 5 e 18 POTs Tot. bg. NC CC background 608. 4 94. 2 361. 5 64. 2 150. 1 66. 1 96. 8 19. 4 error 15. 5% 17. 8% 44. 0% 20% ebeam 9 2022/1/11

Other contributions and ongoing work: • hand scanning – an independent cross check, valuable

Other contributions and ongoing work: • hand scanning – an independent cross check, valuable inputs to automated analysis • analysis tools development • cosmic ray background study • e -related hadron production study Sensitivity (90% CL Exclusion) • With our current data set, we will be able to approach CHOOZ’s limit. Dm 2=0. 003 e. V 2 • With five times more data, we will improve CHOOZ’s limit by a factor of 2. If 13 is not too small, we may see a signal and make the first measure of 13. 10 2022/1/11