M. Agostini et al. (Borexino Collaboration)
Phys. Rev. D 101, 012009 – Published 21 January 2020
This paper presents a comprehensive geoneutrino measurement using the Borexino detector, located at Laboratori Nazionali del Gran Sasso (LNGS) in Italy. The analysis is the result of 3262.74 days of data between December 2007 and April 2019. The paper describes improved analysis techniques and optimized data selection, which includes enlarged fiducial volume and sophisticated cosmogenic veto. The reported exposure of (1.29±0.05)×1032 protons ×year represents an increase by a factor of two over a previous Borexino analysis reported in 2015. By observing 52.6+9.4−8.6(stat)+2.7−2.1(sys) geoneutrinos (68% interval) from 238U and 232Th, a geoneutrino signal of 47.0+8.4−7.7(stat)+2.4−1.9(sys) TNU with +18.3−17.2% total precision was obtained. This result assumes the same Th/U mass ratio as found in chondritic CI meteorites but compatible results were found when contributions from 238U and 232Th were both fit as free parameters. Antineutrino background from reactors is fit unconstrained and found compatible with the expectations. The null-hypothesis of observing a geoneutrino signal from the mantle is excluded at a 99.0% C.L. when exploiting detailed knowledge of the local crust near the experimental site. Measured mantle signal of 21.2+9.5−9.0(stat)+1.1−0.9(sys) TNU corresponds to the production of a radiogenic heat of 24.6+11.1−10.4 TW (68% interval) from 238U and 232Th in the mantle. Assuming 18% contribution of 40K in the mantle and 8.1+1.9−1.4 TW of total radiogenic heat of the lithosphere, the Borexino estimate of the total radiogenic heat of the Earth is 38.2+13.6−12.7 TW, which corresponds to the convective Urey ratio of 0.78+0.41−0.28. These values are compatible with different geological predictions, however there is a ∼2.4σ tension with those Earth models which predict the lowest concentration of heat-producing elements in the mantle. In addition, by constraining the number of expected reactor antineutrino events, the existence of a hypothetical georeactor at the center of the Earth having power greater than 2.4 TW is excluded at 95% C.L. Particular attention is given to the description of all analysis details which should be of interest for the next generation of geoneutrino measurements using liquid scintillator detectors.
- Received 5 September 2019
Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Funded by SCOAP3.
Published by the American Physical Society
Neutrinos, the most abundant massive particles in the universe, are produced by a multitude of different processes. They interact only by the weak and gravitational interactions, and so are able to penetrate enormous distances through matter without absorption or deflection. Thus, they represent a unique tool to probe otherwise inaccessible objects, such as distant stars, the Sun, as well as the interior of the Earth.
The present availability of large neutrino detectors has opened a new window to study the deep Earth’s interior, complementary to more conventional direct methods used in seismology and geochemistry. For example, atmospheric neutrinos can be used as probe of the Earth’s structure . This absorption tomography is based on the fact that the Earth begins to become opaque to neutrinos with energies above ∼10 TeV. Thus, the attenuation of the neutrino flux, as measured by the signals in large Cherenkov detectors, provides information about the nucleon matter density of the Earth. Recently, IceCube determined the mass of the Earth and its core, its moment of inertia and verified that the core is denser than the mantle using data obtained from atmospheric neutrinos . A complementary information about the electron density could, in principle, be inferred by exploiting the flavor oscillations of atmospheric neutrinos in the energy range from MeV to GeV .
An independent method to study the matter composition deep within the Earth, can be provided by geoneutrinos, i.e., (anti)neutrinos emitted by the Earth’s radioactive elements. Their detection allows us to assess the Earth’s heat budget, specifically the heat emitted in the radioactive decays. The latter, the so-called radiogenic heat of the present Earth, arises mainly from the decays of isotopes with half-lives comparable to, or longer than Earth’s age ( 4.543×109 years): 232Th ( T1/2=1.40×1010 years), 238U ( T1/2=4.468×109 years), 235U ( T1/2=7.040×108 years), and 40K ( T1/2=1.248×109 years) . All these isotopes are labeled as heat-producing elements (HPEs). The natural Thorium is fully composed of 232Th, while the natural isotopic abundances of 238U, 235U, and 40K are 0.992742, 0.007204, and 1.17×10−4, respectively. In each decay, the emitted radiogenic heat is in a well-known ratio1 to the number of emitted geoneutrinos :
Obviously, the total amount of emitted geoneutrinos scales with the total mass of HPEs inside the Earth. Hence, geoneutrinos’ detection provides us a way of measuring this radiogenic heat.
This idea was first discussed by G. Marx and N. Menyhárd , G. Eder , and G. Marx  in the 1960s. It was further developed by M. L. Krauss, S. L. Glashow, and D. N. Schramm  in 1984. Finally, the potential to measure geoneutrinos with liquid scintillator detectors was suggested in the 1990s by C. G. Rothschild, M. C. Chen, and F. P. Calaprice  and independently by R. Raghavan et al. .
It took several decades to prove these ideas feasible. Currently, large-volume liquid-scintillator neutrino experiments KamLAND [12–15] and Borexino [16–18] have demonstrated the capability to efficiently detect a geoneutrino signal. These detectors are thus offering a unique insight into 200 years long discussion about the origin of the Earth’s internal heat sources.
The Borexino detector, located in hall-C of Laboratori Nazionali del Gran Sasso in Italy (LNGS), was originally designed to measure 7Be solar neutrinos. However thanks to the unprecedented levels of radiopurity, Borexino has surpassed its original goal and has now measured all2 the pp-chain neutrinos [20–22]. We report here a comprehensive geoneutrino measurement based on the Borexino data acquired during 3262.74 days (December 2007 to April 2019). Thanks to an improved analysis with optimized data selection cuts, an enlarged fiducial volume, and a sophisticated cosmogenic veto, the exposure of (1.29±0.05)×1032 protons ×year represents a factor 2 increase with respect to the previous Borexino analysis .
A detailed description of all the steps in the analysis is reported, and should be important to new experiments measuring geoneutrinos, e.g., SNO+ , JUNO , and Jinping . Hanohano  is an interesting, additional proposal to use a movable 5 kton detector resting on the ocean floor. As the oceanic crust is particularly thin and relatively depleted in HPEs, this experiment could provide the most direct information about the mantle. Finally, it is anticipated that using antineutrinos to study the Earth’s interior will increase in the future based on the availability of new detectors and the continuous development of analysis techniques.
This paper is structured as follows: Section II introduces the fundamental insights on what the geoneutrino studies can bring to the comprehension of the Earth’s inner structure and thermal budget. Section III details a description of the Borexino detector and the structure of its data. In Sec. IV, the ¯νe detection reaction—the inverse beta decay on free proton, that will be abbreviated as IBD through the text—is illustrated. It is shown that only geoneutrinos above 1.8 MeV kinematic threshold can be detected, leaving 40K and 235U geoneutrinos completely unreachable with present-day detection techniques. Section V deals with the estimation of the expected antineutrino signal from geoneutrinos, through background from reactor and atmospheric neutrinos, up to a hypothetical natural georeactor in the deep Earth. Section VI describes the nonantineutrino backgrounds, e.g., cosmogenic or natural radioactive nuclei whose decays could mimic IBD. The criteria to selectively identify the best candidates in the data, are discussed in Sec. VII, which involves the optimization of the signal-to-background ratio. Section VIII shows how the signal and background spectral shapes, expressed in the experimental energy estimator (normalized charge), were constructed and how the detection efficiency is calculated. Both procedures are based on Borexino Monte Carlo (MC) , that was tuned on independent calibration data. Section IX introduces the analyzed dataset and discusses the number of expected signal and background events passing the optimized cuts, based on Secs. V and VI. In Sec. X, the Borexino sensitivity to extract geoneutrino signals is illustrated. Finally, Sec. XI discusses our results. The golden IBD candidate sample is presented (Sec. XI A) together with the spectral analysis (Sec. XI B) and sources of systematic uncertainty (Sec. XI C). The measured geoneutrino signal at LNGS is compared to the expectations of different geological models in Sec. XI D. The extraction of the mantle signal using knowledge of the signal from the bulk lithosphere is discussed in Sec. XI E. The consequences of the new geoneutrino measurement with respect to the Earth’s radiogenic heat are discussed in Sec. XI F. Placing limits on the power of a hypothetical natural georeactor, located at different positions inside the Earth, is discussed in Sec. XI G. Final summary and conclusions are reported in Sec. XI. The acronyms used within the text are listed in alphabetical order in the Appendix.
II. WHY STUDY GEONEUTRINOS?
III. THE BOREXINO DETECTOR
IV. ANTINEUTRINO DETECTION
V. EXPECTED ANTINEUTRINO SIGNAL
VI. NONANTINEUTRINO BACKGROUNDS
VII. DATA SELECTION CUTS
VIII. MONTE CARLO OF SIGNAL AND BACKGROUNDS
IX. EVALUATION OF THE EXPECTED SIGNAL AND BACKGROUNDS WITH OPTIMIZED CUTS
X. SENSITIVITY TO GEONEUTRINOS
2 The upper limit was placed for hep solar neutrinos, the flux of which is expected to be about 3 orders of magnitude smaller than that of 8B solar neutrinos .
5 Recent speculations  about possible partitioning of some lithophile elements (including U and Th) into the metallic core are still debated [50,51]. This would explain the anomalous Sm/Nd ratio observed in the silicate Earth and would represent an additional radiogenic heat source for the geodynamo process.
7 The difference of ∼0.8 TNU with respect to the value reported in  is the result of the neutrino survival probability function calculated from each cell using the updated oscillation parameters. The oscillation amplifies the reduction of the signal due to the presence of surrounding carbonatic rocks poor in Th and U.
8 For year 2019 the data are not yet available, so we use the data of 2018.
The Borexino program is made possible by funding from Istituto Nazionale di Fisica Nucleare (INFN) (Italy), National Science Foundation (NSF) (USA), Deutsche Forschungsgemeinschaft (DFG) and Helmholtz-Gemeinschaft (HGF) (Germany), Russian Foundation for Basic Research (RFBR) (Grants No. 16-29-13014ofi-m, No. 17-02-00305A, and No. 19-02-00097A) and Russian Science Foundation (RSF) (Grant No. 17-12-01009) (Russia), and Narodowe Centrum Nauki (NCN) (Grant No. UMO 2017/26/M/ST2/00915) (Poland). We acknowledge the hospitality and support of the Laboratori Nazionali del Gran Sasso in Italy. The authors thank Matteo Albèri, Kassandra Raptis, and Andrea Serafini for their support.
APPENDIX: LIST OF ACRONYMS
|A||BSE model Anderson, 2007 |
|BDT||Boosted decision tree|
|BSE||Bulk silicate Earth|
|BTB||Borexino trigger board|
|BTB4||The same as MTB flag, see below|
|CC model||Cosmochemical bulk silicate Earth model|
|CLM||Continental lithospheric margin|
|DFV||Dynamical fiducial volume|
|DMP||Dimethylphthalate (DMP, C6H4(COOCH3)2)|
|e− or β−||Electron|
|e+ or β+||Positron|
|Ep||Energy of the prompt IBD candidate|
|Ed||Energy of the delayed IBD candidate|
|FADC||Flash analog-to-digital converter|
|FEB||Front end board|
|FFL||Far field lithosphere|
|FR model||Fully radiogenic bulk silicate Earth model|
|FWFD||Fast wave form digitizer|
|G4Bx2||geant4 based Borexino Monte Carlo code|
|GC model||Geochemical bulk silicate Earth model|
|GD model||Geodynamical bulk silicate Earth model|
|GR1, GR2, GR3||3 studied positions of georeactor inside the Earth|
|Hrad||Earth’s radiogenic heat|
|HCCrad||Earth’s continental crust radiogenic heat|
|Hmantlerad||Earth’s mantle radiogenic heat|
|HLSprad||Earth’s lithosphere radiogenic heat|
|HSC||Earth’s heat from the secular cooling|
|Htot||Integrated total surface heat flux of the Earth|
|HPEs||Heat producing elements|
|HSc||High scenario of the mantle signal prediction|
|IBD||Inverse beta decay|
|IDF||Inner detector flag|
|ISc||Intermediate scenario of the mantle signal prediction|
|J||BSE model Javoy et al., 2010 |
|LF||Load factor of nuclear power plants|
|L & K||BSE model Lyubetskaya & Korenaga, 2007 |
|LNGS||Laboratori Nazionali del Gran Sasso|
|LSc||Low scenario of the mantle signal prediction|
|M & S||BSE model McDonough & Sun, 1995 |
|MTB||Muon trigger board|
|MTF||Muon trigger flag|
|m w.e.||Meter water equivalent|
|¯νe||Electron flavor antineutrino|
|Nh||Number of detected hits|
|NP||Number of triggered PMTs|
|Npe||Number of detected photoelectrons|
|Pee||Survival probability of electron flavor neutrino|
|PC||Pseudocumene liquid scintillator, C6H3(CH3)3, 1,2,4-trimethylbenzene|
|Probability distribution function|
|PMNS||Pontecorvo–Maki–Nakagawa–Sakata mixing matrix|
|PPO||Fluorescent dye, C15H11NO, 2,5-diphenyloxazole|
|P & O||BSE model Palme and O’Neil, 2003 |
|Qp||Charge of the prompt IBD candidate|
|Qd||Charge of the delayed IBD candidate|
|RR||Rest of the region|
|SSS||Stainless steel sphere|
|SVM||Support vector machine|
|T||BSE model Taylor, 1980 |
|TMVA||Toolkit for multivariate data analysis|
|TNU||Terrestrial neutrino unit|
|T & S||BSE model Turcotte & Schubert, 2002 |
|URCV||Convective Urey ratio|
|W||BSE model Wang et al., 2018 |
|WE||Water extraction procedure of LS-purification|