NO A Liquid Scintillator Production Stuart Mufson, Indiana - PowerPoint PPT Presentation

NO ν A Liquid Scintillator Production Stuart Mufson, Indiana University FroST 2016 March 19, 2016

The NO ν A Experiment N uMI O ff-axis 𝝃 e A ppearance E xperiment Ash River NuMI = N eutrinos at the M ain I njector � Long-baseline (anti-)neutrino oscillation experiment � 14 mrad off-axis with L/E ~ 400 km/GeV � Two functionally identical detectors, optimized for ν e identification Fermilab Bloomington 2

Neutrino beamline � Slip-stacking Booster batches since March 2015 • Beam power record: 521 kW • 85% uptime � Beam back Oct-Nov 2015 after upgrades to achieve 700 kW, NOvA design power 3

The NO ν A Detectors � Designed for electron ID � Massive, Low-Z, 65% active � Near Detector: 300 ton,1 km from source • 100m underground, 20,000 channels � Far Detector: 14 kton, 810 km from source • On the surface, 3m concrete+barite overburden; 344,000 channels 4

The NO ν A Far Detector in perspective 5

Far Detector site: Ash River, MN 6

Near Detector at Fermilab � Near Detector fully instrumented and cooled 7

The NO ν A Collaboration 231 collaborators from 41 ins:tu:ons and 8 countries 8

NO ν A Design Philosophy • NO ν A was optimized to search for the rare ν e 's at Ash River that have oscillated from the ν µ 's in the NuMI beam. • The primary design requirement for the NO ν A far detector was the efficient detection of ν e interactions at 2 GeV. • Furthermore, to minimize infrastructure costs, the plan was to operate the massive 14.4 kt far detector on the surface. • After considering several technologies, the NO ν A collaboration built a segmented liquid scintillator detector. • liquid scintillator was chosen over plastic because of its significant cost advantage for massive detectors.

Segmented NO ν A Detector Design • A large water Cherenkov detector on the surface does not yield enough light per unit path length for fine segmentation • Cherenkov detectors perform best on the relatively simple event topologies resulting from E < 1 GeV neutrino interactions - above this energy, the Chrenekov threshold does not provide good hadronic energy reconstruction. - particle ID becomes complicated by multiple overlaping Chrenkov rings • Further: to provide a veto and enough distance for ring formation on the nearest wall, Super-K only analyzes events recorded in the inner 22.5 kt of its 50 kt detector mass. - A segmented detector, is sensitive over a significant fraction of its active volume, enabling a smaller detector to achieve the same physics.

Alternative Technologies Considere d • Resistive Plate Chamber (RPC) sampling calorimeters - inferior particle ID efficiency for detecting ν e ’s - more expensive to construct - uninstrumented absorber regions provide paths for comic-rays to penerate into the fiducial volume - degraded performance as the gas-handling system ages • Low-Z sampling calorimeters with particle board as the absorber uses building materials with sufficient structural strength to support a massive detector - only half the ν e detection efficiency of a segmented liquid scintillator detector - require a significant increase in detector mass for the same physics reach

Alternative Technologies Considere d • Liquid Argon TPC’s - enormous potential for neutrino physics was recognized - however, the largest detector operated at the time of the NOvA technology decision in 2007 was ICARUS, with 500 tons of imaging mass that needed to be scaled up by a factor of thirty - deemed insufficiently mature

NOvA Liquid Scintillator NO ν A liquid scintillator is an organic fluorescent compound, which have a long history in particle physics It was formulated to meet the requirements of the NO ν A experiment, but based initially on commercially available scintillators component purpose mass mass mass mass fraction (kg) fraction (kg) blends: #1, #2 blends: #3 – #25 mineral oil solvent 94.91% 691,179 94.63% 7,658,656 pseudocumene scintillant 4.98% 36,2677 5.23% 423,278 PPO waveshifter 0.11% 801 0.14% 11,331 bis-MSB waveshifter 0.0016% 11.7 0.0016% 129 Stadis-425 antistatic 0.001% 7.3 0.001% 81 Vitamin E antioxidant 0.001% 7.1 0.001% 78 Total 728,247 8,093,264

NO ν A liquid scintillator cell (extruded PVC) with WLS fiber to 1 APD in the range 390 – 440 nm that excite the WLS fiber wavelength shifter bis-MSB which subsequently decays to photons • these down-converted scintillation photons excite the second which in turn decays and emits in the range 340 – 380 nm • these UV photons excite the wavelength shifter PPO, the range 270 – 320 nm • pseudocumene molecules decay by emitting photons in scintillant – pseudocumene • A charged particle incident on the cell excites the primary fibers pixel To 1 APD pixel To 1 APD pixel L L L typical charged particle path pical pical charged charged particle particle path path W W D D W D W = 3.8 cm D = 5.9 cm L = 15.5 m

Normalized emission profiles for blended NO ν A scintillator 1.0 Emission Profiles NOvA Scintillator & Waveshifters normalized emission profile 0.8 0.6 NOvA scintillator PPO 0.4 bis-MSB 0.2 0.0 300 350 400 450 500 550 600 650 wavelength (nm) The scintillation light is dominated by the emission from bis-MSB.

Far detector performance Light yield at far end of cells Efficiency for far end for 6-7.5 cm path length TDR requirement ← far side near side ➞ TDR requirement attenuation capture gain transparency length efficiency noise reflectivity #photons/cm threshold Scintillator × PVC × Fiber × APD 90% efficiency for seeing hits on muon tracks at the far end of the cells – meeting TDR specifications

550 𝜈 s exposure of the Far Detector 17

Close-up of neutrino interaction in the Far Detector Top view Beam direc?on Color denotes deposited charge Side view 18

Toll Blender Operations Toll Blending Operations Mineral Oil (III) scintillator blend tank (I) waveshifters (II) pseudocumene anti-static tanker trailers to Vit. E Ash River/Fermilab fluor mix tank (IV) (I) Mineral oil brought to blender from storage/rail cars and pumped into the scintillator blend tank (II) Fluors and additives mixed in the fluor mix tank at the NOvA mixing station, then pumped into the blend tank. (III) Scintillator blended with bubbles of N 2 gas (IV) Scintillator was shipped to the NOvA detectors by tanker trailer

Scintillator Blending Operations mineral oil storage 120,000 gal blend tanks Wolf Lake Terminals, Hammond, IN 600,000 gal epoxy-lined tank, Westway Terminals, Riverdale, IL blending with Pulsair system using N 2 bubbles 20

Fluor Blend Loop Fluor mix tank powder blending tank “Powder” blending shed fluor blending pumps

QC: Attenuation Length of Mineral Oil and Blended Scintillator (mineral oil transmission)/(glass transmission) 1.00 mineral oil requirement 0.95 scintillator requirement Renkert 70-T NOvA scintillator blended with Renkert 70-T experimental calibration curve 2 4 6 8 10 12 14 16 transmission of large attenuation length (m) quantities were rapidly tintometer calibrated measured relative to glass at IU with a with a Lovibond spectrophotometer tintometer

Attenuation Length of Mineral Oil and Blended Scintillator 20 1 (scintillator transmission)/(glass transmission) Scintillator Transmission by Blend 18 Mineral Oil Transmission (delivered by rail & barge) 0.99 16 requirement 14 0.98 12 10 0.97 8 0.96 6 4 0.95 requirement 2 0 0.94 0.96 0.98 1.00 1.02 1.04 0 5 10 15 20 25 transmission relative to glass blend # • 2 m scintillator attenuation length requirement established by simulation • mineral oil attenuation length requirement by experimental program to find mineral oil that could meet the 2 m scintillator requirement

QC: Chemical composition of fluor blend (PPO/pseudocumene) ratio by mass (bis-MSB/pseudocumene) ratio by mass 3.0 3.5 [bis-MSB]/[PS] x 10^4 [PPO]/[PS] x 10^4 2.5 3.0 : blends #1, #2 : blends #1, #2 : blends #3 - #25 : blends #3 - #25 2.0 2.5 5 10 15 20 25 5 10 15 20 25 blend # blend # • chemical composition determined with GC-MS for PPO and pseudocumene, and HPLC for bis-MSB • if fluor ratios correct, adding the correct quantity of mineral oil will result in scintillator with the proper ratio of fluors

QC: Scintillator Light Yield 25% of anthracene Gamma Test Light Yield for Blends #3 - #25 0.42 0.4 (Compton edge)/(alpha peak) 0.38 0.36 0.34 0.32 0.3 t fit function: exp( ) 0.28 t 0 0.26 0.24 test finds the ratio of the 0 100 200 300 400 500 days since blend #3 Compton edge (gammas) relative increasing ratio due to radiological to a fiducial Alpha peak: damage in plastic scintillator -- (Compton edge/Alpha peak) scintillator was certified so long as test followed radiation damage law -- a measure of the light output of scintillator as a function ( ) light yield ∝ 1/exp − t / t 0 of its composition

Summary • successfully blended 8.8 kt of liquid scintillator for the NO ν A experiment detector volume mass (gal) (kg) near detector 40,141 130,672 far detector 2,674,041 8,690,929 Total 2,714,182 8,821,511 • scintillator was produced commercially at Wolf Lake Terminals, Hammond, IN • scintillator met all performance requirements • details can be found NIM A 799 (2015) 1-9.

backups