Neutron captures update Pierre Lasorak, Aran Borkum & Tyler Alion 1
Intro/Content • Neutrons from the surrounding rock are expected to create 100 Hz of capture in a 10 kT module. • Been demonstrated that neutron captures is one of the main background for • SN triggering, • solar neutrino triggering • and solar analysis. • Important to properly characterise this background. • In this talk: • Geometry developments • Water shielding • Other sources of neutrons • Neutron spectrum • Impact on the SN trigger. Pierre Lasorak 2 20/06/2019
Geometry development • The 1x2x6 isn’t the best for neutron simulation: • Neutrons have low cross section on Argon and populate the whole detector volume, some of them escape the volume. • Hard to scale the rate from the 1x2x6 to a full 10kT detector. • Need to have a more informed analysis if we want to motivate the addition of water panels. Pierre Lasorak 3 20/06/2019
Proposed solution • Overall revamping of the geometry for these studies: • Utilisation of GGD (for General Geometry Description): • Provides a simple python interface to the nasty GDML that is used in ROOT and GEANT4. • https://github.com/DUNE/duneggd/tree/master/python/duneggd/ larfd • Integration of the 1x2x6 and 10kT in GGD has been done • Simple change to the configuration to generate whichever geometry (number of APA) one wants. • Creating smaller representative geometry that is easily scalable to the 10kT. • Addition of representative cryostat geometry: • Foam, steel of the cryostat. • Water panels to test neutron shielding Pierre Lasorak 4 20/06/2019
Testing Water APAs on Full Cryo, • Quite a lot of tests have been done: 1x2x6 Full 10kt Shielded Outside Few APA • Up to event display for the 1x2x6: G ✔ ✔ ✔ ✔ … GGD Tests G ✔ ✔ ✔ ✔ Ma • 3 wire views are looking OK ROOT Tests D versions Geometry ✔ ✔ ✔ ✔ … easily • Crossing TPC muons are looking L Sorting configured OK Channel ✔ ✔ ✔ ✔ A Mapping Te • 10 kT hasn’t been tested up to R ✔ th G4 Overlaps event display (but DetSim is slow…) S pa ✔ ✔ — — G4 / IDEs to O ons shot down the possible F ✔ DetSim / EVD config. T Passes ✔ … … … existing FCL e placement seems to Muon in 3 views Muon crossing the boundary Pierre Lasorak 5 20/06/2019
Water shielding results • Truth study up to GEANT4 stage. • Test against analysis from John Beacom’s group. • Similar reduction is observed. es 10kT with water shielding panels Pierre Lasorak 6 20/06/2019
Still on the TODO list • Finish the tests of the geometries. • Add representative external volumes to get an idea of neutron shielding. • Better representation of the cavern. • Add the photon detectors in the geometry. Pierre Lasorak 7 20/06/2019
Other sources of neutrons? LAr 80cm • From Paola Sala’s Internal SS (1.2mm) presentation at CERN. Plywood (1.2cm) • Completely different results! Polyurethane foam (37.6cm) • Main contributor is the � Plywood (2x1.2cm) + SS (1mm) foam (0.2 Hz) � � • 0.06 Hz of capture from the surrounding rock � Polyur ethane foam (37.6cm) • Needs to be investigated. Plywood (1.2cm) � External SS membrane (1cm) Mass Disint/year Fission Alpha,n [n/year] [t] 10mBq/kg [n/year] 10mBq/kg External SS SS460ML 250 7.9e10 71 000 53 490 Foam polyurethane 34 1.1e10 9 600 22 516 7.5E12 6.7e6 Internal SS SS304 4 0.13e10 3 400 1 168 Wood plywood 12 0.4e10 1 200 3 257 Pierre Lasorak 8 20/06/2019
Neutron spectrum • Along with the overall neutron rate, the neutron spectrum is also a subject to discussion: • Found that there is little difference in the neutron capture rate in the 10kT with changing spectrum. • These analyses should probably be repeated with water shielding. Neutron capture Rates LArSoft 104 Hz NeuCBOT 108 Hz SOURCES4 101 Hz Pierre Lasorak 9 20/06/2019
Impact on the SN triggers • Assuming simple 10 seconds counting 0.07 0.06 trigger: 0.05 Rate [Hz] 0.04 • No shape discrimination as shown at 0.03 0.02 the last CM. 0.01 0 3 K 4 A C N P R T 9 r 2 P y o o P A e a A A u p l d t A r o r a g t t o g r o n l o o n i n o • LMC is much better if you reduce the n u n n m neutron background. 1 1 SN trig eff Galaxy far side 0.9 0.95 0.8 SN trig eff LMC 0.7 0.9 0.6 0.85 0.5 0.4 0.8 0.3 0.2 0.75 0.1 0.7 0 3 3 2 1 2 − 10 10 10 10 1 10 10 10 Neutron capture rate [Hz] Neutron capture rate [Hz] Pierre Lasorak 10 20/06/2019
Impact on SN trigger PDF PDF • Using the shape information to inform the trigger decision. 1 − 10 Neutron capture rates: 1 kHz • Principle, simple extension of the counting algorithm: 400 Hz − 2 10 100 Hz • Integrate for 10 seconds the sumADC of the clusters 10 Hz − 3 10 1 Hz • Compare (likelihood ratio) with the shape of the background. • If the likelihood exceeds a certain value → trigger. − 4 10 • Divided the Likelihood into 2 parts: 0 20 40 60 80 100 120 140 160 180 200 SADC PDF PDF • Shape → helps when the background is high. Doesn’t Signal+Background 1.4 Individual SN event efficiency = 41% Background depend on the normalisation of the background. Background rate = 0.181469 Hz 1.2 1 • At the LMC, the efficiency doesn’t change if the neutron 400 Hz 0.8 capture are 100 Hz or more. 0.6 • Normalisation (i.e. counting) → is only efficient at low 0.4 0.2 background. 0 0 20 40 60 80 100 120 140 160 180 200 SADC • Effects turns on when the neutron capture gets around 10 Hz PDF PDF Signal+Background 0.8 Individual SN event efficiency = 41% Background • Consistent with previous study. 0.7 Background rate = 0.009440 Hz 0.6 1 Hz 0.5 • Unfortunately I didn’t have time to run enough toys to get a 0.4 quantifiable answer. 0.3 0.2 0.1 0 0 20 40 60 80 100 120 140 160 180 200 SADC Pierre Lasorak 11 20/06/2019
Conclusion • On-going development with the geometry will be able to inform low-energy triggering studies • Quantifiable impact of the external volumes (foam, water, shape of the cavern etc) on the neutron rates. • Important to cross check and make sure all the potential neutron sources are identified. • Neutron spectrum shape shouldn’t really matter, all the neutrons bounce elastically and thermalise before capturing. • Impact on the trigger: • Counting only trigger: big impact at the LMC (efficiency goes from 10 % → 50 % if neutron rate goes from 100 Hz → 10 Hz). • Shape trigger: effect on the LMC is mitigated at neutron capture rate high rates, more studies needed to understand what happens at lower rates. • Future: • Other backgrounds have been implemented in the radiological generator Bismuth-Polonium backgrounds. • Probably worth a request for a production when we are happy with all these new things? Pierre Lasorak 12 20/06/2019
Recommend
More recommend