UPDATED GEONEUTRINO MEASUREMENT WITH BOREXINO
LIVIA LUDHOVA FOR BOREXINO COLLABORATION
IKP-2, FORSCHUNGSZENTRUM JÜLICH AND RWTH AACHEN UNIVERSITY, GERMANY
SEPTEMBER 10TH, 2019 TAUP 2019, TOYAMA, JAPAN OUTLINE (OR WHERE IS THIS ENERGY COMING FROM?)
• What are geoneutrinos and why to study them • Expected geoneutrino signal at LNGS (Italy) • Borexino and antineutrino detection • Borexino geoneutrino measurement: fresh new results • Geological interpretation EARTH’S HEAT BUDGET Radiogenic heat & Integrated surface heat flux: Geoneutrinos can help!
Htot = 47 + 2 TW
Lithosphere Mantle Heat production in lithosphere “well” known Big uncertainty Mantle cooling 7 - 9 TW Heat production in mantle 1 – 27 TW
4 – 27 TW Core cooling 9 – 17 TW
Core cooling Mantle cooling Geoneutrinos: antineutrinos/neutrinos from the decays of long-lived radioactive isotopes naturally present in the Earth
238U (99.2739% of natural U) à 206Pb + 8 α + 8 e- + 6 anti-neutrinos + 51.7 MeV 232Th à 208Pb + 6 α + 4 e- + 4 anti-neutrinos + 42.8 MeV 235U (0.7205% of natural U) à 207Pb + 7 α + 4 e- + 4 anti-neutrinos + 46.4 MeV 40K (0.012% of natural K) à 40Ca + e- + 1 anti-neutrino + 1.32 MeV (BR=89.3 %) 40K + e- à 40Ar + 1 neutrino + 1.505 MeV (BR=10.7 %)
q the only direct probe of the deep Earth q released heat and geoneutrino flux in a well fixed ratio q to measure geoneutrino flux = (in principle) = to get radiogenic heat q in practice (as always) more complicated…..
Earth shines in geoneutrinos: flux ~106 cm-2 s-1 leaving freely and instantaneously the Earth interior (to compare: solar neutrinos (NOT antineutrinos!) flux ~1010 cm-2 s-1) GEONEUTRINOS AND WHY TO STUDY THEM
• Main goal: contribution of the Nuclear physics Abundance of Radiogenic radiogenic heat (mainly of the mantle) to the total Earth’s radioactive heat surface heat flux, which is an elements (main goal) important margin, test, and input at the same time for many geophysical and geochemical Distribution of radioactive elements models of the Earth; (geological models) • Further goals: U/Th bulk ratio, tests and discrimination among geological models, Earth composition models, study of the From geoneutrino mantle homogeneity or To predict: Geoneutrino flux stratification, insights to the measurement: processes of Earth’formation, additional sources of heat?, idea of U-based georeactor Neutrino geoscience: truly inter-disciplinary field! BOREXINO DETECTOR Laboratori Nazionali del Gran Sasso, Italy
3800 m.w.e 4300 muons/day crossing the inner detector 278 ton liquid scintillator (LS)
• the world’s radio-purest LS detector < 9 × 10-19 g(Th)/g LS , < 8 × 10-20 g(U)/g LS • ~500 hit PMTs / MeV • energy reconstruction: 5 keV (5%) @ 1 MeV NIM A600 (2009) 568 • position reconstruction: 10 cm @ 1 MeV + - • pulse shape identification (α/β, e /e ) Operating since 2007 ANTINEUTRINO DETECTION WITH LIQUID SCINTILLATORS
Electron antineutrino detection: delayed coincidence Energy threshold = 1.8 MeV • Inverse Beta Decay (IBD) σ @ few MeV: ~10-42 cm2 • Charge current, electron flavour only (~100 x more than scattering)
Eprompt = Evisible
= Te+ + 2 x 511 keV
~ Eantinu – 0.784 MeV
e+ νe
W
n p EXPECTED GEONEUTRINO SIGNAL AT GRAN SASSO
LOCAL AND GLOBAL GEOLOGICAL INFORMATION GEONEUTRINO ENERGY SPECTRA
• IBD) ~10-42 cm2 σ(
•
GEONEUTRINO SIGNAL AT LNGS
1 TNU (Terrestrial Neutrino Unit) = 1 event / 1032 target protons (~1kton LS) / year with 100% detection efficiency
S (U + Th) S(Th)/S(U) H (U + Th [TNU] +K) [TW] Local Crust (~500 km around LNGS) 9.2 ± 1.2 0.24 -
+4.9 +1.9 Bulk Lithosphere (observed) 25.9 -4.1 0.29 8.1 -1.4 Mantle = Bulk Silicate Earth model 2.5 – 19.6 0.26 3.2 – 25.4 – lithosphere (assuming for BSE chondritic value of 0.27) Total 28.5 – 45.5 0.27 (chondritic) 11.3 – 33.5 OPTIMIZED IBD SELECTION CUTS Efficiency: (86.98 ± 1.50)%
Charge of prompt Charge of delayed Time correlation Space correlation
Qp > 408 pe Qd > 700 (860) – 3000 pe dt = (2.5-12.5) µs + (20-1280) µs dR < 1.3 m
• Prompt spectrum • Neutron captures on proton Neutron capture τ = (254.5 ± 1.8) µs starts at 1 MeV (2.2 MeV) and in about 1% of 2 cluster event in 16 µs DAQ gate • 5% energy resolution cases on 12C (4.95 MeV) • Spill out effect at the nylon @ 1 MeV prompt delayed inner vessel border • Radon correlated 214Po(α + γ) decays from 214Bi and 214Po fast coincidences
Muon veto Dynamic Fiducial Volume Multiplicity α/β discrimination 2s || 1.6 s : 9Li(β + n) > 10 cm from IV (prompt) No event with Q >400 pe MLPdelayed > 0.8 2 ms: neutrons ±2 ms around promt/delayed • Exposure vs accidental bgr 214 • IV has a leak: shape reco from • Radon correlated Po(α+γ) • Several veto categories the data weekly • Strict and special muon tags • Suppressing undetected cosmogenic background, mostly Whole detector o multiple neutrons o Cylinder • Negligible exposure loss Only 2.2% exposure loss
GOLDEN CANDIDATES: 154
Prompt charge spectrum Delayed charge spectrum
• December 9, 2007 to April 28, 2019 • 3262.74 days of data taking
1 • Average FV = (245.8 ± 8.7) ton n+ H n+12C • Exposure = (1.29 ± 0.05) x 1032 proton x year • Including systematics on position reconstruction and muon veto loss, for 100% detection eff.
Distribution in time Radial distribution Distance to the Inner Vessel NEUTRINO BACKGROUNDS
Reactor antineutrinos Atmospheric neutrinos
Mueller et al 2011 With “5 MeV bump” Energy Geoneutrino Reactor > 1 MeV +1.5 +1.4 Signal [TNU] 84.5 -1.4 79.6 -1.3 window antineutrino +1.7 +1.6 # Events 97.6 -1.6 91.9 -1.5 Events 2.2 ± 1.1 3.3 ± 1.6 9.2 ± 4.6
• For all ~440 world reactors (1.2 TW total power) ü their nominal thermal powers (PRIS database of IAEA) ü monthly load factors (PRIS database) • Estimated 50% uncertainty on the prediction ü distance to LNGS (no reactors in Italy) • Indications of overestimation • 235U, 238U, 239Pu , and 241Pu fuel • Included in the systematic error ü power fractions for different reactor types • Atmospheric neutrino fluxes ü energy released per fission from HKKM2014 (>100 MeV) and FLUKA (<100 MeV) ü energy spectra (Mueller at al. 2011 and Daya Bay) • Matter effects included • Pee electron neutrino survival probability • IBD cross section • Detection efficiency = 0.8955 ± 0.0150 Charge spectrum after IBD selection cuts Accidentals
NON-ANTINEUTRINO BACKGROUNDS -1 Racc = (3029.0 ± 12.7) s including scaling factor 9 12 210 16 exp(-Rmuon x 2s) = 0.896 Li (β +n) events < 2s after muons C( Po(α), n) O due to the 2 s muon veto before delayed Y = (1.45 ± 0.22) x 10-7 IBD-like events in dt = 2 -20 s τmeasured = (0.260 ± 0.021) s n = 0.56 for 210Po in LS Charge of prompt εIBD-like
(0.260 $\pm$ 0.021)\, Distance from muon track
< 210Po rate> = (12.75 ± 0.08) cpd/ton SPECTRAL FIT with chondritic Th/U ratio
Reactor expectations with and without 5 MeV bump events
8σ
Geoneutrino 5σ # 3σ 1σ
# Reactor events Prompt charge [photoelectrons]: 1 MeV ~500 photoelectrons Resulting number of geoneutrinos (median value) • Unbinned likelihood fit of charge spectrum of 154 prompts • S(Th)/S(U) = 2.7 (corresponds to chondritic Th/U mass ratio of 3.9) 52.6+9.4 (stat)+2.7 (sys)events • Reactor signal unconstrained and result compatible with expectations −8.6 −2.1 9 • Li, accidentals, and (α, n) bgr constrained according to expectations +18.3 % total precision • Systematics includes atmospheric neutrinos, shape of reactor spectrum, −17.2 vessel shape and position reconstructions, detection efficiency +8,4 +2,4 GEONEUTRINO SIGNAL AT LNGS 47.0−7.7 (stat)−1.9 (sys)TNU
LOC = local crust = (9.2 ± 1.2) TNU
+1.4 FFL = far-field lithosphere = (4.0 _1.0) TNU
MANTLE (U + Th abundances) = BSE model – LITHOSPHERE
Intermediate scenario 2 layer distribution of U and Th in the mantle
J: Javoy at al., 2010 L&K: Lyubetskaya and Korenaga, 2007 T: Taylor, 1980 M&S: Mc Donough and Sun, 1995 A: Anderson, 2007 In agreement with expectations W: Wang, 2018 P&O: Palme and O’Neil, 2003 T&S: Turcotte and Schubert, 2002 SPECTRAL FIT with Th and U fit independently
3σ 2σ Th events Th 1
232 σ #
ratio Chondritic
232 238 Prompt charge [photoelectrons]: 1 MeV ~500 photoelectrons # 238U events Th / U ratio
Resulting number of geoneutrinos (median value) no sensitivity to measure Th/U ratio +46.8 50.4 events -44.05% total precision
• In agreement with the fit with Th/U fixed • Larger error MANTLE GEONEUTRINO SIGNAL
qobs= 5.4479 p value = 9.796 x 10-3 Likelihood
Prompt charge [photoelectrons]: 1 MeV ~500 photoelectrons # Mantle events
• Fit performed with signal from lithosphere constrained to Mantle signal (median value) (28.8 ± 5,6) events with S(Th)/S(U) = 0.29 +10.7 • Mantle PDF constructed with S(Th)/S(U) = 0.26, 23.7−10.1 events maintaining the global Th/U ratio as in CI chondrites +9.6 • Sensitivity study using log-likelihood ratio method: 21.2−9.1 TNU null hypothesis rejected with 99.0% C.L. RADIOGENIC HEAT Reminder: Htot = (47 ± 2) TW
CC = continental crust
Mantle radiogenic heat from U+Th: Earth radiogenic heat from U+Th+K: Convective Urey URCV ratio:
11.1 +13.6 +0.41 + 38.2 TW 0.78 24.6−10.4 TW −12.7 −0.28 At 90% C.L., mantle characteristics: Compatible with predictions, but least o Assuming 18% 40K mantle a(Th) >48 ppb & a(U) >13ppb (2.4σ) compatible with the contribution URCV >0.13 CosmoChemical model (CC) predicting o Lithospheric radiogenic heat U+Th+K lowest U+Th mantle abundances +1.9 8.1 -1.4TW GEOREACTOR
Upper limit (95% CL): • Hypothetical fission of Uranium deep in the Earth 18.7 TNU • Three locations considered 2.4 TW in the Earth’s center • 235U : 238U = 0.76 : 0.23 (Herndon) 0.5 TW near CMB at 2900 km • Fit with reactor spectrum constrained 5.7 TW far CMB at 9842 km • No sensitivity to oscillation pattern SUMMARY AND OUTLOOK More details you can find: arXiv: 1909.02257 • Borexino provided new geoneutrino analysis with all available data up to April 2019 Poster ü Optimized selection criteria Sindhujha ü Improved analysis Kumaran
ü Signal in agreement with geological predictions, with a preference for models predicting high U and Th abundances ü Null mantle signal excluded at 99.0% C.L. More related Borexino posters: ü Liudmila Lukianchenko: search for ü Estimates of mantle radiogenic heat, mantle minimal low energy (anti)neutrinos from U and Th abundances, and Urey convective ratio astrophysical sources ü No sensitivity to Th/U ratio ü Alina Vishneva: studies of non- standard neutrino properties • Geoneutrinos are proven a new tool to study the deep Earth and new generation of experiments are needed for firm geological conclusions!
Back up slides BOREXINO CALIBRATION
JINST 7 (2012) P10018 Internal calibration • ~300 points in the whole scintillator volume • LED-based source positioning system
External calibration 9 positions with 228Th source Optical (γ 2.615 MeV) fibers reaching Laser calibration each • PMT time equalisation PMT • PMT charge calibration (charge calib. also using 14C) Better than 1% (1.9%) precision BOREXINO MONTE CARLO for all relevant quantities in the solar analysis <2 (>3) MeV Astrop. Phys. 97 (2018) 136 C++ Borexino custom Echidna: C++ Borexino custom Geant-4 based Electronics simulation Reconstruction Tracking code Follows real DAQ conditions • Several energy estimators • Full detector geometry • PMT quality and calibration • Dark noise • Position reconstruction • Energy loss • Pulse-shape variables • Photon production & propagation • Trigger condition • Number of working channels on an • Output in the same format as event-by-event basis reconstructed data files
γ peaks from internal calibration • Tuning on calibration data. • Independently measured input parameters: emission spectra, attenuation length, PMT after-pulse, refractive index, effective quantum efficiencies. • Biasing technique for external background. • Simulation of pile-up events.
Page 23 EXPECTED GEONEUTRINO SIGNAL Expected “known and big” crustal signal
50 TNU The signal is small, we need big detectors!
1 TNU = 1 event / 1032 target protons / year cca 1 event /1 kton /1 year, 100% detection efficiency
Expected mantle signal: hypothesis of heterogeneous composition Motivated by the observed Large Shear Velocity Provinces at the mantle base
10.6 TNU To measure mantle signal is more challenging!
O. Šrámek et al. “Geophysical and geochemical constraints on geoneutrino fluxes from Earths mantle”, Earth Planet. Sci. Lett., 361 (2013) 356-366) HISTORY OF GEONU MEASUREMENTS KamLAND (Japan) Borexino (Italy) • The first investigation in 2005 • 99.997 CL observation in 2010 CL < 2σ Nature 436 (2005) 499 31 +4.1 7.09 x 10 target-proton year 9.9 – 3.4 geonu’s • Update in 2008 PRL 100 (2008) 221803 small exposure but low background level (December 2007 – December 2009) 73 + 27 geonu’s 1.5 x 1031 target-proton year 2.44 x 1032 target-proton year PLB 687 (2010) 299 • 99.997 CL observation in 2011 • Update in 2013 +29 106 – 28 geonu’s 14.3 + 4.4 geonu’s (March 2002 – April 2009) (December 2007 – August 2012) 3.49 x 1032 target-proton year 3.69 x 1031 target-proton year Nature Geoscience 4 (2011) 647 0-hypothesis @ 6 x 10-6
• Latest published result in 2013 PLB 722 (2013) 295–300 +28 • Latest in June 2015: 5.9 CL 116 – 27 geonu’s σ +6.5 +0.9 (March 2002 – November 2012) 23.7 (stat) (sys) geonu’s 4.9 x 1032 target-proton year (December 2007 – March 2015) PRD 88 (2013) 033001 5.5 x 1031 target-proton year • Preliminary update in 2016: 7.92σ CL 0-hypothesis @ 3.6 x 10-9 +28 164 – 25 geonu’s (LOW REACTOR) PRD 92 (2015) 031101 (R) (March 2002 – November 2016) • NEW UPDATE COMING SOON 32 6.39 x 10 target-proton year IMPROVED SELECTION, <20% PRECISION (H. Watanabe @ Neut. Res. And Thermal Evol. Earth) BULK SILICATE EARTH MODELS (BSE) Models predicting the composition of the Earth primitive mantle Various inputs: composition of the chondritic meteorites, correlations with the composition of the solar photosphere, composition of rock samples from upper mantle and crust, energy needed to run mantle convection….. Abundances of U/Th/K (and thus also radiogenic heat) in BSE = Lithosphere (crust + continental lithospheric mantle) + MANTLE
Lithosphere: 7-9 TW ( only ~0.2 TW in oceanic crust) “well” known MANTLE = BSE – CRUST 1-27 TW (different BSE models) Big uncertainty
Isotopic compositions of: 1) C1 carbonaceous chondrites 2) solar photosphere are highly correlated !
Was it the same in the primitive Earth? GOLDEN CANDIDATES: DISTANCE TO INNER VESSEL MUONS AND COSMOGENICS MUON EVENT STRUCTURE NEUTRON SOURCE CALIBRATION OPTIMIZATION OF DFV CUT PDFS USED IN SPECTRAL FIT ACCIDENTAL BACKGROUND RADON CORRELATED BACKGROUND INNER VESSEL SHAPE RECONSTRUCTION GEONEUTRINO ENERGY SPECTRA SYSTEMATICS