- Email: [email protected]
Borexino: recent results, detector calibration and future perspectives
Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106 www.elsevier.com/locate/npbps
Borexino: recent results, detector calibration and future perspectives Marco Pallavicini (on behalf of the Borexino Collaboration)a∗ a
Dipartimento di Fisica - Universit`a di Genova via Dodecaneso, 33 - 16146 - Genova and INFN Sezione di Genova Borexino Collaboration: G. Bellini, J.Benziger, S.Bonetti, M.Buizza Avanzini, B.Caccianiga, L.Cadonati, F.Calaprice, C.Carraro, A.Chavarria, F.Dalnoki-Veress, D.D’Angelo, S.Davini, H.de Kerret, A.Derbin, A.Etenko, F.von Feilitzsch, K.Fomenko, D.Franco, C.Galbiati, S.Gazzana, C.Ghiano,M.Giammarchi, M.Goeger-Neff, A.Goretti, E.Guardincerri, S.Hardy, Aldo Ianni, Andrea Ianni, M.Joyce, V.Kobychev, Y.Koshio,G.Korga, D.Kryn, M.Laubenstein, M.Leung, T.Lewke, E.Litvinovich, B.Loer, F.Lombardi, P.Lombardi, L.Ludhova, I.Machulin, S.Manecki, W.Maneschg, G.Manuzio, Q.Meindl, E.Meroni, L.Miramonti, M.Misiaszek, D.Montanari, V.Muratova, L.Oberauer, M.Obolensky, F.Ortica, M.Pallavicini, L.Papp, L.Perasso, S.Perasso, A.Pocar, R.S.Raghavan, G.Ranucci, A.Razeto, A.Re, P.Risso, A.Romani, D.Rountree, A.Sabelnikov, R.Saldanha, C.Salvo, S.Sch¨ onert, H.Simgen, M.Skorokhvatov, O.Smirnov, A.Sotnikov, S.Sukhotin, Y.Suvorov,R.Tartaglia, G.Testera,D.Vignaud, R.B.Vogelaar, J.Winter, M.Wojcik, M.Wurm, A.Wright, J.Xu, O.Zaimidoroga, S.Zavatarelli, G.Zuzel
The Borexino experiment has been running since May 2007 at the Gran Sasso underground laboratory, in Italy. Solar neutrinos are detected with a large unsegmented liquid scintillator detector with unprecedented radioactive purity. The main results obtained include the measurement of the 7 Be solar neutrino flux, the measurement of the 8 B neutrino flux with electron recoil energy threshold of 3.0 MeV and the first clear detection of geo-neutrinos (see Aldo Ianni’s talk in these proceedings for further details). Borexino has recently completed a large calibration campaign, and better results on 7 Be solar neutrino measurement are expected soon. Short and medium term perspectives are summarized in the conclusions.
1. Introduction Borexino [1–4] detects low energy solar neutrinos via their elastic scattering on the electrons of a 300 t ultra-pure liquid scintillator target. The high light yield of the liquid scintillator (a mixture of Pseudocumene doped with 1.5 g/l of PPO) and the extreme radiopurity achieved during the purification processes allow the real time detection of solar neutrinos down to about 200 keV of electron recoil energy, being limited below this value by the presence of the unavoidable 14 C background. The main scientific goal is the measurement of the flux and of the energy spectrum of solar neutrinos with sub-MeV or few MeV energy but the program also includes the detection of geoneutri∗ e-mail:
[email protected]
0920-5632/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.nuclphysbps.2011.04.077
nos (see [5] and Aldo Ianni’s talk in these proceedings), of super-nova neutrinos and the search for very rare decays [6–10]. Because of the limited space, I will here summarize the main results on solar neutrinos. Solar neutrino oscillations enhanced by MSW effect [14] have been well established by radiochemical experiments [11] and by water Cerenkov detectors [12] and the parameters describing the oscillation phenomenon have been determined by SNO and KamLAND [13]. The parameters lie in the so called LMA (Large Mixing Angle) region of the plane θ12 Δm212 (tan2 (2θ12 ) = 0.45+0.06 −0.05 and −5 2 · 10 eV ) [22]. Δm212 = 7.69+0.21 −0.21 The LMA-MSW solution predicts two main features: the shape of the energy dependence of the neutrino survival probability Pee and the lack of day-night asymmetry. The Pee function
102
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
decreases with increasing energy: matter effects dominate at energies above 3 MeV while are absent below 1 MeV. The region in between is called the transition region. While the LMA-MSW predicts a well defined shape for the Pee function in the transition region, current experimental data do not constraint it at all, and some theoretical models, including non standard interactions, predict survival probability curves with different shapes [16]. Borexino is the only experiment that has measured the signal rate due to the 0.862 MeV 7 Be neutrinos [17][18]; besides, it has measured the one due to 8 B [19] with an energy threshold (3.0 MeV), lower than any previous experiment. A measurement of a null day night difference of the 7 Be flux provides a further independent confirmation of the LMA-MSW solution. Interest of this measurement is also related to the possibility to accommodate quite large effects that are predicted by in some alternative oscillation scenario, based for example on the mass varying model [23]. The neutrino count rate depends on both solar neutrino flux and oscillation parameters. The relevance of the measurements of the various components of the solar neutrino measurements in Borexino is then twofold: from one side they can increase the confidence in the oscillation scenario and from the other side, assuming the knowledge of the oscillation parameters, they provide a measurement of the absolute solar neutrino flux and may yield very valuable information about solar physics, particularly on the metallicity controversy and about the role of the CNO cycle in the Sun The precision of the actual data about solar neutrinos does not allow to distinguish between high and low metallicity models [21] but future high precision measurements of the 7 Be might give useful constraints. In the future, if the purification program will be fully successful, Borexino may detect CNO neutrinos, providing a direct determination of the Sun metallicity. 2. The 7 Be signal The 7 Be signal rate in Borexino is determined by fitting an energy spectrum which is obtained after the event selection described in [18]. The
energy spectrum corresponding to a live time of 192 days is shown in Fig. 1. The spectrum is cut below 250 keV. The energy calibration so far has been obtained by studying the β decay of 14 C with 156 keV end point (not shown) and through the spectral fit itself. The expected 7 Be spectral signature is a electron recoil spectrum with a Compton like shape and its features are visible in Fig. 1. The large peak in the same figure is due to the 5.3 MeV α decay of 210 P o, a daughter of 222 Rn out of equilibrium. The ionization quenching of the scintillator reduces the visible energy by a factor about 13 and brings the α peak in the energy region of the 7 Be signal. A positive side effect of this large background is its use to study the yield stability and the energy resolution of the detector. The 210 P o count rate decreases with time consistently with its mean life of 200 days. The study of the time correlated events belonging to the 238 U and 232 Th radioactive chains yields, under the hypothesis of secular equilibrium, an internal contamination for 238 U of (1.6± 0.1) · 10−17 g/g and for 232 T h of (6.8 ± 1.5) · 10−18 g/g. The concentration of these contaminants is more than an order of magnitude lower than the design value of 10−16 g/g and it is not therefore the main issue of the 7 Be analysis. On the contrary, the most important background is due to the β decay of 85 Kr with 687 keV end point having a rate of the same order of magnitude of the 7 Be signal and a spectral shape not too different. The analysis of the rare decays of 85 Kr into 85 Rb (branching ration 0.43 % but easy to tag thanks to the presence of time correlated events) yields 30 ± 5 counts/(days 100 t) after a live time of more than 1.5 year. Additional background is identified as 210 Bi and as 11 C. The last one is produced by the interaction of muons in the scintillator. The energy spectrum is fitted by two procedures [17][18]: one of them includes the 210 Po α peak and in the second one a statistical subtraction of this peak is applied. The two results are mutually consistent and they give the value of the 0.862 7 Be solar neutrino interaction rate of 49 ± 3stat ± 4syst cpd/100 t after 192 days of
103
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
counts/days*4 hits
Day (Red) and Night (Blue) spectrum 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 200
300
400
500
600 nhits
Figure 1. The Borexino energy spectrum (192 days of live time) and its fit. A similar fit procedure in which the 210 Po is statistically subtracted by exploiting the α-β separation capability of the detector gives consistent results.
live time. Assuming the flux of the Standard Solar Model with high metallicity the expected rate without oscillations is 74 ± 4 cpd/100 t which should reduce to 48±4 cpd/100 t using the LMAMSW oscillation parameters. The 7 Be measurement of Borexino confirms the prediction at low energy of the LMA oscillation model. A significant effort is in progress to reduce the errors associated to the measurement of the 7 Be signal rate: the main contributions are the imperfect knowledge of the energy calibration, of the detector response function and the fiducial volume. The three calibration campaigns already completed will reduce these errors significantly. 3. Day night asymmetry of the 7 Be signal A preliminary analysis of the day and night spectra provides a further confirmation of the prediction of the LMA model through the absence of a significant day-night asymmetry in the 7 Be flux.
Figure 2. Day (red) and night (blue) spectra of the selected events normalized to the night live time and in the energy region outside the 210 Po peak.
For each event, from the absolute time we compute the value of the Sun zenith angle at the Laboratori Nazionali del Gran Sasso latitude. Fig. 2 shows the day and night spectra corresponding to a total live time of 422.12 days with 212.87 days and 209.25 night. The horizontal axis represents the number of hits detected by the photomultipliers which is closely connected to the energy of the events. Events are selected as described in [17] the only difference concerns the choice of the fiducial volume being here only the spatial cut r<3 m applied. The day-night asymmetry Adn is defined n −Cd as Adn = C Cn +Cd where Cd and Cn are the counts during day and the night time. Adn has been evaluated for every bin (see Fig. 3) in the region from nhits > 250 to nhits < 700 as shown in figure. Here below nhits=350 the signal to background ratio is maximum while for nhits>350 the background (due to 11 C) is dominant. The binned day night asymmetry well fit with a constant function providing Afdnit = 0.007 ± 0.008 in the region (250,700) nhits and Afdnit = 0.014±0.013 in the region (250,350) nhits. This last value is consistent with Aaver dn = 0.011 ± 0.014 obtained using the integrated counts in the region with nhits (250,350).
104
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
Figure 3. Binned day nigh asymmetry as a function of nhits and the zoom in the region where the contribution of the 7 Be signal is maximum. The fit is performed with a constant function. The reduced χ2 is 66.5/72.
The absence of significant difference between the day and night signals in the high energy region is a check of the consistency of the data while the ref it sults about Aaver dn and Adn in the region (250,350) nhits show that the day night asymmetry of the 7 Be solar neutrino signal is zero within one standard deviation. This result is independent on the precision of the definition of the fiducial volume and on the knowledge of the detector response function. The day night asymmetry in the region nhits (250,350) here discussed includes the contribution both of the signal and of the background. Considering the statistical precision of the 7 Be flux determination in the day and night periods we get the contribution of the signal alone Aνdn = 0.02 ± 0.04stat . This is our preliminary result confirming the expectation of the LMA-MSW scenario. Further analysis and the evaluation of the contribution of possible systematic errors due to the selection of the data sample are in progress.
Figure 4. Comparison of the final spectrum after data selection (red dots) to Monte Carlo simulations (black line). The expected electron recoil spectrum (from 8 B neutrinos including oscillations) is also shown in filled blue histogram while the 208 Tl background is shown in green and the 11 Be background is cyan. Violet is the expected external background.
4. The low energy threshold 8 B signal The excellent radiopurity levels obtained by Borexino made possible a measurement of the 8 B solar neutrino flux with the unprecedented energy threshold of 3.0 MeV [19]. This value is mainly determined by the need to cut the residual γ background due to the Thallium decay in the PMT materials. The expected signal rate, including neutrino oscillations, is 0.26 ± 0.03 counts/day in 100 t. Data selection procedure (see [19] for details) includes the removal of short lived (τ < 2 s) cosmogenic isotopes by vetoing the detector for 5 s after each muon crossing the scintillator, the removal of 10 C by the triple coincidence with the parent muon and the neutron capture on proton and the statistical subtraction of the Thallium spectrum due to the internal radioactivity. The resulting 8 B neutrinos signal rate is 0.217 ± 0.038stat ± 0.008sys counts/(days 100 t) after 345.3 days of exposure, after the cuts described in [19].
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
Figure 5. Comparison of the final spectrum of 8 B neutrinos after data selection and background statistical subtraction (red dots) and Monte Carlo simulation of oscillated 8 B neutrinos interactions. The Monte Carlo is computed with BPS09(GS98) model [21] and oscillation parameters from [22] θ12 Δm212 (tan2 (2θ12 ) = 0.45 and Δm212 = 7.69 · 10−5 eV 2 ) .
The final spectrum is shown in Fig. 4 together with a comparison with the expected signal from Monte Carlo simulation. This result corresponds to a neutrino flux of (2.7 ± 0.4 ± 0.1) 106 cm−2 s−1 , in very good agreement with previous more precise measurements. The survival probabilities of both 7 Be and 8 B neutrinos are measured by the same experiment for the first time, yielding a further confirmation of the transition between the matter dominated regime above 3 MeV and the vacuum regime at lower energies. Eliminating the common sources of systematic errors the ratio between the two probabilities is 1.6 ± 0.33 confirming the expectation of the LMA-MSW oscillation scenario at 93% C.L. (see Fig. 6). 5. Conclusions and future programs Borexino has completed the first real time measurements of 7 Be and the low threshold 8 B signal
105
Figure 6. The electron neutrino survival probability of the LMA-MSW model and the experimental results including the new data of Borexino. The curve is computed with BPS09(GS98) model [21] and oscillation parameters from [22] θ12 Δm212 (tan2 (2θ12 ) = 0.45 and Δm212 = 7.69 · 10−5 eV 2 ). The dots are Borexino results from 7 Be and 8 B measurements. The error bars include also theoretical uncertainties on the expected flux from the solar models.
rate. A preliminary result about the day night asymmetry has also been obtained. All results confirm the current LMA-MSW scenario. Borexino has also performed the first clear detection of the geo-neutrinos, the anti-neutrinos emitted by the Earth because of the radio-activity content of the planet. A calibration campaign has been completed. The precise response function to α, β and neutrons have been determined by inserting suitable designed radioactive sources into the core of the detector. Particularly, the dependence on the source position has been studied very carefully. The results will contribute to a significant reduction of the errors on the fiducial volume and on the detector response function, yielding a 7 Be signal rate measurement with few percent accuracy and of the reduction of the systematic errors of the 8 B measurement. Neutron calibration data
106
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
has been already used on geo-neutrino analysis to calibrate the energy response at high energies. The possibility to purify the scintillator further in order to reduce the background due to 85 Kr and to 210 Bi is real. We have completed a first series of purification loops, and the results are encouraging. If fully successful, Borexino might attempt the direct detection of pep and pp, and, possibly, CNO solar neutrinos.
13. 14.
6. Acknowledgments This work was funded by: INFN (Italy), NSF (USA), BMBF, DFG and MPG (Germany), Rosnauka(Russia), MNiSW (Poland), and PRIN project JR4STW 2007 by MIUR (Italy). The support of the Laboratori Nazionali del Gran Sasso is kindly acknowledged.
15. 16.
17. 18.
REFERENCES 1. G. Alimonti et al. (Borexino Collaboration), Astropart. Phys. 16 (205-234), 2002 2. C. Arpesella et al. (Borexino Collaboration), Astropart. Phys. 18 (1-25), 2002 3. H. O. Back et al. (Borexino Collaboration), Nucl. Instrum. Meth. A 600 (98) 2008 4. H. O. Back et al. (Borexino Collaboration), Nucl. Instrum. Methods A 609 (58), 2009 5. G. Bellini et al. (Borexino Collaboration) Phys. Lett. B 687 (299-304), 2010 6. G. Bellini et al. (Borexino Collaboration) Phys. Rev. C 81 (034317), 2010 7. H. O. Back et al. (Borexino Collaboration), Phys. Lett. B 563 (23), 2003 8. H. O. Back et al. (Borexino Collaboration), Phys. Lett. B 525 (29), 2002 9. H. O. Back et al. (Borexino Collaboration), Eur. Phys. J. C 37 (421), 2004 10. H. O. Back et al. (Borexino Collaboration), Phys. Lett. B 563 (35), 2003 11. B. T. Cleveland et al., Astrophys. J. 496 (505), 1998 J. N. Abdurashitov et al. (SAGE Collaboration), Phys. Rev. Lett. 83, (4686), 1999 M. Altmann et al. (GNO Collaboration), Phys. Lett. B 616 (174), 2005 12. K. S. Hirata et al. (KamiokaNDE Collaboration), Phys. Rev. Lett. 63 (16), 1989 Y.
19. 20.
21. 22. 23. 24.
Fukuda et al. (Super-Kamiokande Collaboration), Phys. Rev. Lett. 81 (1562), 1998 J. P. Cravens et al. (SuperKamiokande Collaboration), arXiv:0803.4312v1 Q. R. Ahmad et al. (SNO Collaboration), Phys. Rev. Lett. 87 (071301), 2001 B. Aharmim et al. (SNO Collaboration), Phys. Rev. C 75 (045502), 2007 S. Abe et al. (KamLAND Collaboration), Phys. Rev. Lett. 100 ( 221803), 2008 S. P. Mikheyev and A. Yu. Smirnov, Sov. J. Nucl. Phys. 42 (913), 1985; L.Wolfenstein, Phys. Rev. D 17 ( 2369), 1978 A. Yu. Smirnov arXiv:hep-ph/0305106v1 V. Barger et al., Phys. Rev. Lett. 95 (211802), 2005; A.Friedland et al., Phys. Lett. B 594 (347), 2004 C. Arpesella et al. (Borexino Collaboration) Phys. Rev. Lett. 101 (9), 2008 C. Arpesella et al. (Borexino Collaboration) Phys. Lett. B 658, (101), 2008 G. Bellini et al. (Borexino Collaboration) Phys. Rev. D, 82 (033006), 2010 A. de Gouvea, A. Friedland, H. Murayama Phys. Rev. D 60 (093011), 1999; hepph/9910286v2 C. Pena Garay, A. Serenelli arXiv:0811.2424; A. Serenelli, arXiv:0910.3690 G. Fogli et al., Phys. Rev. Lett. 101 (141801), 2008 P. C. de Holanda, arXiv:0811:0567v1 J.N.Bahcall and P.N. Krastev Phys. Rev. C 56 (52839), 1997
Borexino: recent results, detector calibration and future perspectives Marco Pallavicini (on behalf of the Borexino Collaboration)a∗ a
Dipartimento di Fisica - Universit`a di Genova via Dodecaneso, 33 - 16146 - Genova and INFN Sezione di Genova Borexino Collaboration: G. Bellini, J.Benziger, S.Bonetti, M.Buizza Avanzini, B.Caccianiga, L.Cadonati, F.Calaprice, C.Carraro, A.Chavarria, F.Dalnoki-Veress, D.D’Angelo, S.Davini, H.de Kerret, A.Derbin, A.Etenko, F.von Feilitzsch, K.Fomenko, D.Franco, C.Galbiati, S.Gazzana, C.Ghiano,M.Giammarchi, M.Goeger-Neff, A.Goretti, E.Guardincerri, S.Hardy, Aldo Ianni, Andrea Ianni, M.Joyce, V.Kobychev, Y.Koshio,G.Korga, D.Kryn, M.Laubenstein, M.Leung, T.Lewke, E.Litvinovich, B.Loer, F.Lombardi, P.Lombardi, L.Ludhova, I.Machulin, S.Manecki, W.Maneschg, G.Manuzio, Q.Meindl, E.Meroni, L.Miramonti, M.Misiaszek, D.Montanari, V.Muratova, L.Oberauer, M.Obolensky, F.Ortica, M.Pallavicini, L.Papp, L.Perasso, S.Perasso, A.Pocar, R.S.Raghavan, G.Ranucci, A.Razeto, A.Re, P.Risso, A.Romani, D.Rountree, A.Sabelnikov, R.Saldanha, C.Salvo, S.Sch¨ onert, H.Simgen, M.Skorokhvatov, O.Smirnov, A.Sotnikov, S.Sukhotin, Y.Suvorov,R.Tartaglia, G.Testera,D.Vignaud, R.B.Vogelaar, J.Winter, M.Wojcik, M.Wurm, A.Wright, J.Xu, O.Zaimidoroga, S.Zavatarelli, G.Zuzel
The Borexino experiment has been running since May 2007 at the Gran Sasso underground laboratory, in Italy. Solar neutrinos are detected with a large unsegmented liquid scintillator detector with unprecedented radioactive purity. The main results obtained include the measurement of the 7 Be solar neutrino flux, the measurement of the 8 B neutrino flux with electron recoil energy threshold of 3.0 MeV and the first clear detection of geo-neutrinos (see Aldo Ianni’s talk in these proceedings for further details). Borexino has recently completed a large calibration campaign, and better results on 7 Be solar neutrino measurement are expected soon. Short and medium term perspectives are summarized in the conclusions.
1. Introduction Borexino [1–4] detects low energy solar neutrinos via their elastic scattering on the electrons of a 300 t ultra-pure liquid scintillator target. The high light yield of the liquid scintillator (a mixture of Pseudocumene doped with 1.5 g/l of PPO) and the extreme radiopurity achieved during the purification processes allow the real time detection of solar neutrinos down to about 200 keV of electron recoil energy, being limited below this value by the presence of the unavoidable 14 C background. The main scientific goal is the measurement of the flux and of the energy spectrum of solar neutrinos with sub-MeV or few MeV energy but the program also includes the detection of geoneutri∗ e-mail:
[email protected]
0920-5632/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.nuclphysbps.2011.04.077
nos (see [5] and Aldo Ianni’s talk in these proceedings), of super-nova neutrinos and the search for very rare decays [6–10]. Because of the limited space, I will here summarize the main results on solar neutrinos. Solar neutrino oscillations enhanced by MSW effect [14] have been well established by radiochemical experiments [11] and by water Cerenkov detectors [12] and the parameters describing the oscillation phenomenon have been determined by SNO and KamLAND [13]. The parameters lie in the so called LMA (Large Mixing Angle) region of the plane θ12 Δm212 (tan2 (2θ12 ) = 0.45+0.06 −0.05 and −5 2 · 10 eV ) [22]. Δm212 = 7.69+0.21 −0.21 The LMA-MSW solution predicts two main features: the shape of the energy dependence of the neutrino survival probability Pee and the lack of day-night asymmetry. The Pee function
102
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
decreases with increasing energy: matter effects dominate at energies above 3 MeV while are absent below 1 MeV. The region in between is called the transition region. While the LMA-MSW predicts a well defined shape for the Pee function in the transition region, current experimental data do not constraint it at all, and some theoretical models, including non standard interactions, predict survival probability curves with different shapes [16]. Borexino is the only experiment that has measured the signal rate due to the 0.862 MeV 7 Be neutrinos [17][18]; besides, it has measured the one due to 8 B [19] with an energy threshold (3.0 MeV), lower than any previous experiment. A measurement of a null day night difference of the 7 Be flux provides a further independent confirmation of the LMA-MSW solution. Interest of this measurement is also related to the possibility to accommodate quite large effects that are predicted by in some alternative oscillation scenario, based for example on the mass varying model [23]. The neutrino count rate depends on both solar neutrino flux and oscillation parameters. The relevance of the measurements of the various components of the solar neutrino measurements in Borexino is then twofold: from one side they can increase the confidence in the oscillation scenario and from the other side, assuming the knowledge of the oscillation parameters, they provide a measurement of the absolute solar neutrino flux and may yield very valuable information about solar physics, particularly on the metallicity controversy and about the role of the CNO cycle in the Sun The precision of the actual data about solar neutrinos does not allow to distinguish between high and low metallicity models [21] but future high precision measurements of the 7 Be might give useful constraints. In the future, if the purification program will be fully successful, Borexino may detect CNO neutrinos, providing a direct determination of the Sun metallicity. 2. The 7 Be signal The 7 Be signal rate in Borexino is determined by fitting an energy spectrum which is obtained after the event selection described in [18]. The
energy spectrum corresponding to a live time of 192 days is shown in Fig. 1. The spectrum is cut below 250 keV. The energy calibration so far has been obtained by studying the β decay of 14 C with 156 keV end point (not shown) and through the spectral fit itself. The expected 7 Be spectral signature is a electron recoil spectrum with a Compton like shape and its features are visible in Fig. 1. The large peak in the same figure is due to the 5.3 MeV α decay of 210 P o, a daughter of 222 Rn out of equilibrium. The ionization quenching of the scintillator reduces the visible energy by a factor about 13 and brings the α peak in the energy region of the 7 Be signal. A positive side effect of this large background is its use to study the yield stability and the energy resolution of the detector. The 210 P o count rate decreases with time consistently with its mean life of 200 days. The study of the time correlated events belonging to the 238 U and 232 Th radioactive chains yields, under the hypothesis of secular equilibrium, an internal contamination for 238 U of (1.6± 0.1) · 10−17 g/g and for 232 T h of (6.8 ± 1.5) · 10−18 g/g. The concentration of these contaminants is more than an order of magnitude lower than the design value of 10−16 g/g and it is not therefore the main issue of the 7 Be analysis. On the contrary, the most important background is due to the β decay of 85 Kr with 687 keV end point having a rate of the same order of magnitude of the 7 Be signal and a spectral shape not too different. The analysis of the rare decays of 85 Kr into 85 Rb (branching ration 0.43 % but easy to tag thanks to the presence of time correlated events) yields 30 ± 5 counts/(days 100 t) after a live time of more than 1.5 year. Additional background is identified as 210 Bi and as 11 C. The last one is produced by the interaction of muons in the scintillator. The energy spectrum is fitted by two procedures [17][18]: one of them includes the 210 Po α peak and in the second one a statistical subtraction of this peak is applied. The two results are mutually consistent and they give the value of the 0.862 7 Be solar neutrino interaction rate of 49 ± 3stat ± 4syst cpd/100 t after 192 days of
103
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
counts/days*4 hits
Day (Red) and Night (Blue) spectrum 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 200
300
400
500
600 nhits
Figure 1. The Borexino energy spectrum (192 days of live time) and its fit. A similar fit procedure in which the 210 Po is statistically subtracted by exploiting the α-β separation capability of the detector gives consistent results.
live time. Assuming the flux of the Standard Solar Model with high metallicity the expected rate without oscillations is 74 ± 4 cpd/100 t which should reduce to 48±4 cpd/100 t using the LMAMSW oscillation parameters. The 7 Be measurement of Borexino confirms the prediction at low energy of the LMA oscillation model. A significant effort is in progress to reduce the errors associated to the measurement of the 7 Be signal rate: the main contributions are the imperfect knowledge of the energy calibration, of the detector response function and the fiducial volume. The three calibration campaigns already completed will reduce these errors significantly. 3. Day night asymmetry of the 7 Be signal A preliminary analysis of the day and night spectra provides a further confirmation of the prediction of the LMA model through the absence of a significant day-night asymmetry in the 7 Be flux.
Figure 2. Day (red) and night (blue) spectra of the selected events normalized to the night live time and in the energy region outside the 210 Po peak.
For each event, from the absolute time we compute the value of the Sun zenith angle at the Laboratori Nazionali del Gran Sasso latitude. Fig. 2 shows the day and night spectra corresponding to a total live time of 422.12 days with 212.87 days and 209.25 night. The horizontal axis represents the number of hits detected by the photomultipliers which is closely connected to the energy of the events. Events are selected as described in [17] the only difference concerns the choice of the fiducial volume being here only the spatial cut r<3 m applied. The day-night asymmetry Adn is defined n −Cd as Adn = C Cn +Cd where Cd and Cn are the counts during day and the night time. Adn has been evaluated for every bin (see Fig. 3) in the region from nhits > 250 to nhits < 700 as shown in figure. Here below nhits=350 the signal to background ratio is maximum while for nhits>350 the background (due to 11 C) is dominant. The binned day night asymmetry well fit with a constant function providing Afdnit = 0.007 ± 0.008 in the region (250,700) nhits and Afdnit = 0.014±0.013 in the region (250,350) nhits. This last value is consistent with Aaver dn = 0.011 ± 0.014 obtained using the integrated counts in the region with nhits (250,350).
104
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
Figure 3. Binned day nigh asymmetry as a function of nhits and the zoom in the region where the contribution of the 7 Be signal is maximum. The fit is performed with a constant function. The reduced χ2 is 66.5/72.
The absence of significant difference between the day and night signals in the high energy region is a check of the consistency of the data while the ref it sults about Aaver dn and Adn in the region (250,350) nhits show that the day night asymmetry of the 7 Be solar neutrino signal is zero within one standard deviation. This result is independent on the precision of the definition of the fiducial volume and on the knowledge of the detector response function. The day night asymmetry in the region nhits (250,350) here discussed includes the contribution both of the signal and of the background. Considering the statistical precision of the 7 Be flux determination in the day and night periods we get the contribution of the signal alone Aνdn = 0.02 ± 0.04stat . This is our preliminary result confirming the expectation of the LMA-MSW scenario. Further analysis and the evaluation of the contribution of possible systematic errors due to the selection of the data sample are in progress.
Figure 4. Comparison of the final spectrum after data selection (red dots) to Monte Carlo simulations (black line). The expected electron recoil spectrum (from 8 B neutrinos including oscillations) is also shown in filled blue histogram while the 208 Tl background is shown in green and the 11 Be background is cyan. Violet is the expected external background.
4. The low energy threshold 8 B signal The excellent radiopurity levels obtained by Borexino made possible a measurement of the 8 B solar neutrino flux with the unprecedented energy threshold of 3.0 MeV [19]. This value is mainly determined by the need to cut the residual γ background due to the Thallium decay in the PMT materials. The expected signal rate, including neutrino oscillations, is 0.26 ± 0.03 counts/day in 100 t. Data selection procedure (see [19] for details) includes the removal of short lived (τ < 2 s) cosmogenic isotopes by vetoing the detector for 5 s after each muon crossing the scintillator, the removal of 10 C by the triple coincidence with the parent muon and the neutron capture on proton and the statistical subtraction of the Thallium spectrum due to the internal radioactivity. The resulting 8 B neutrinos signal rate is 0.217 ± 0.038stat ± 0.008sys counts/(days 100 t) after 345.3 days of exposure, after the cuts described in [19].
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
Figure 5. Comparison of the final spectrum of 8 B neutrinos after data selection and background statistical subtraction (red dots) and Monte Carlo simulation of oscillated 8 B neutrinos interactions. The Monte Carlo is computed with BPS09(GS98) model [21] and oscillation parameters from [22] θ12 Δm212 (tan2 (2θ12 ) = 0.45 and Δm212 = 7.69 · 10−5 eV 2 ) .
The final spectrum is shown in Fig. 4 together with a comparison with the expected signal from Monte Carlo simulation. This result corresponds to a neutrino flux of (2.7 ± 0.4 ± 0.1) 106 cm−2 s−1 , in very good agreement with previous more precise measurements. The survival probabilities of both 7 Be and 8 B neutrinos are measured by the same experiment for the first time, yielding a further confirmation of the transition between the matter dominated regime above 3 MeV and the vacuum regime at lower energies. Eliminating the common sources of systematic errors the ratio between the two probabilities is 1.6 ± 0.33 confirming the expectation of the LMA-MSW oscillation scenario at 93% C.L. (see Fig. 6). 5. Conclusions and future programs Borexino has completed the first real time measurements of 7 Be and the low threshold 8 B signal
105
Figure 6. The electron neutrino survival probability of the LMA-MSW model and the experimental results including the new data of Borexino. The curve is computed with BPS09(GS98) model [21] and oscillation parameters from [22] θ12 Δm212 (tan2 (2θ12 ) = 0.45 and Δm212 = 7.69 · 10−5 eV 2 ). The dots are Borexino results from 7 Be and 8 B measurements. The error bars include also theoretical uncertainties on the expected flux from the solar models.
rate. A preliminary result about the day night asymmetry has also been obtained. All results confirm the current LMA-MSW scenario. Borexino has also performed the first clear detection of the geo-neutrinos, the anti-neutrinos emitted by the Earth because of the radio-activity content of the planet. A calibration campaign has been completed. The precise response function to α, β and neutrons have been determined by inserting suitable designed radioactive sources into the core of the detector. Particularly, the dependence on the source position has been studied very carefully. The results will contribute to a significant reduction of the errors on the fiducial volume and on the detector response function, yielding a 7 Be signal rate measurement with few percent accuracy and of the reduction of the systematic errors of the 8 B measurement. Neutron calibration data
106
M. Pallavicini et al. / Nuclear Physics B (Proc. Suppl.) 217 (2011) 101–106
has been already used on geo-neutrino analysis to calibrate the energy response at high energies. The possibility to purify the scintillator further in order to reduce the background due to 85 Kr and to 210 Bi is real. We have completed a first series of purification loops, and the results are encouraging. If fully successful, Borexino might attempt the direct detection of pep and pp, and, possibly, CNO solar neutrinos.
13. 14.
6. Acknowledgments This work was funded by: INFN (Italy), NSF (USA), BMBF, DFG and MPG (Germany), Rosnauka(Russia), MNiSW (Poland), and PRIN project JR4STW 2007 by MIUR (Italy). The support of the Laboratori Nazionali del Gran Sasso is kindly acknowledged.
15. 16.
17. 18.
REFERENCES 1. G. Alimonti et al. (Borexino Collaboration), Astropart. Phys. 16 (205-234), 2002 2. C. Arpesella et al. (Borexino Collaboration), Astropart. Phys. 18 (1-25), 2002 3. H. O. Back et al. (Borexino Collaboration), Nucl. Instrum. Meth. A 600 (98) 2008 4. H. O. Back et al. (Borexino Collaboration), Nucl. Instrum. Methods A 609 (58), 2009 5. G. Bellini et al. (Borexino Collaboration) Phys. Lett. B 687 (299-304), 2010 6. G. Bellini et al. (Borexino Collaboration) Phys. Rev. C 81 (034317), 2010 7. H. O. Back et al. (Borexino Collaboration), Phys. Lett. B 563 (23), 2003 8. H. O. Back et al. (Borexino Collaboration), Phys. Lett. B 525 (29), 2002 9. H. O. Back et al. (Borexino Collaboration), Eur. Phys. J. C 37 (421), 2004 10. H. O. Back et al. (Borexino Collaboration), Phys. Lett. B 563 (35), 2003 11. B. T. Cleveland et al., Astrophys. J. 496 (505), 1998 J. N. Abdurashitov et al. (SAGE Collaboration), Phys. Rev. Lett. 83, (4686), 1999 M. Altmann et al. (GNO Collaboration), Phys. Lett. B 616 (174), 2005 12. K. S. Hirata et al. (KamiokaNDE Collaboration), Phys. Rev. Lett. 63 (16), 1989 Y.
19. 20.
21. 22. 23. 24.
Fukuda et al. (Super-Kamiokande Collaboration), Phys. Rev. Lett. 81 (1562), 1998 J. P. Cravens et al. (SuperKamiokande Collaboration), arXiv:0803.4312v1 Q. R. Ahmad et al. (SNO Collaboration), Phys. Rev. Lett. 87 (071301), 2001 B. Aharmim et al. (SNO Collaboration), Phys. Rev. C 75 (045502), 2007 S. Abe et al. (KamLAND Collaboration), Phys. Rev. Lett. 100 ( 221803), 2008 S. P. Mikheyev and A. Yu. Smirnov, Sov. J. Nucl. Phys. 42 (913), 1985; L.Wolfenstein, Phys. Rev. D 17 ( 2369), 1978 A. Yu. Smirnov arXiv:hep-ph/0305106v1 V. Barger et al., Phys. Rev. Lett. 95 (211802), 2005; A.Friedland et al., Phys. Lett. B 594 (347), 2004 C. Arpesella et al. (Borexino Collaboration) Phys. Rev. Lett. 101 (9), 2008 C. Arpesella et al. (Borexino Collaboration) Phys. Lett. B 658, (101), 2008 G. Bellini et al. (Borexino Collaboration) Phys. Rev. D, 82 (033006), 2010 A. de Gouvea, A. Friedland, H. Murayama Phys. Rev. D 60 (093011), 1999; hepph/9910286v2 C. Pena Garay, A. Serenelli arXiv:0811.2424; A. Serenelli, arXiv:0910.3690 G. Fogli et al., Phys. Rev. Lett. 101 (141801), 2008 P. C. de Holanda, arXiv:0811:0567v1 J.N.Bahcall and P.N. Krastev Phys. Rev. C 56 (52839), 1997