Observation of the geo-?'s and reactor anti-?'s in Borexino

1
Observation of thegeo-?'s and reactor anti-?'s
in Borexino

• Yury Suvorov
• INFN (LNGS) / RRC Kurchatov Inst.

(on behalf of the Borexino collaboration)
APC 28th April 2010, Paris
2
Historical introduction
Gamow in the letter to Reines (1953) Dear
Fred, ...your background neutrinos may just be
coming from high energy ß-decaying members of U
and Th families in the crust of the Earth.
G. Marx, N. Menyard Mitteilungen der Sternwarte,
Budapest, 48 (1960) First estimation of the
anti-neutrino fluxes from U,Th and K.
?.?.?arkov ????????, ?., Nauka (1964) First
proposal to use the reaction of inverse ß decay
for the geo-nu registration.
Gernot Eder Nucl Phys 78 (1966) Terrestrial
neutrinos Idea to determine chemical
composition of some elements by the study of the
emited neutrinos.
APC 28th April 2010, Paris
3
Geo-?'s
Geo-neutrinos are the electron anti-neutrinos
produced in the beta decays of 40K, 238U and
232Th isotopes, which are naturally present in
the Earth interior.
Earth shines mainly in the anti-?'s.
APC 28th April 2010, Paris
4
Geo-? fluxes around the world
1 TNU 1 ev in 1032 protons / yr
Importance of the geological location of the
detector (Crust/Mantle)
Expected in Borexino rate is about 5 ev/yr
APC 28th April 2010, Paris
5
KamLAND
2005 First try of Experimental investigation
of geologically produced anti-neutrinos Nature
436, 499-503 (28 July 2005)
Assuming a Th/U mass concentration ratio of 3.9,
the 90 per cent confidence interval for the total
number of geoneutrinos detected is 4.5 to 54.2.
2008 - new results based on the greater
statistics yield 7327 ev ( 2.5s) The
indication of an excess of low-energy
anti-neutrinos consistent with an interpretation
as geo-neutrinos persists. Phys. Rev. Lett. 100
(2008) 221803
APC 28th April 2010, Paris
6
Borexino collaboration
About 160 collaborators from 6 countries
Heidelberg (Germany)
APC 28th April 2010, Paris
7
Borexino detector
2212 8 ETL 9351 PMTs mounted inside the SSS
Two Nylon balloons Inner Vessel
(8.5 m, V
340 m3) filled with 278 tons of scintillator
(PC 1.5 g/l of PPO) Inner Buffer (11.5 m)
filled with PC DMP
Water Tank (d18 m, V
2400 m3) Shielding from ? and n. Water Cerenkov
detector (Muon Veto) 208 PMTs
Stainless Steel Sphere (d 13.7 m,
Volume 1340 m3)
20 supporting legs
APC 28th April 2010, Paris
8
Anti-? detection
The unprecedentedly low intrinsic radioactivity
of Borexino, the high photon yield and large
number of free target protons ( 1.7x1031) offer
a unique tool for the anti-? study in the MeV
energy range.
Inverse beta decay reaction.
Correlated in space and time pair of signals
1) Prompt signal Positron 2 ? from
annihilation with e-, E?0.511 MeV After
t250µs 2) Delayed signal Neutron capture on
Hydrogen - ? of E?2.2 MeV
Eth1.806 MeV
APC 28th April 2010, Paris
9
Anti-? sources
Electron anti-? from the Nuclear Power plants
Electron anti-? from The Earth Interior
APC 28th April 2010, Paris
10
Reactor anti-?'s
Main contribution from 194 European nuclear
plants. 245 plants from the rest of the world
give 2.5 . 3 most powerful power plants in
France give 13 of the total signal. Mean-base
line 1000 km (60 of total flux)
Knowledge of the exact duty cycle and the fuel
composition of the nuclear plants is of the
crucial importance! Nominal thermal power and
monthly load factors for each European plant from
the IAEA and EDF.
APC 28th April 2010, Paris
11
Reactor anti-? spectrum
Expected in Borexino In the whole spectrum and
in case of 100 detection efficiency Without
oscill. 9.90.5 events/(100 ton yr) With
oscillations 5.70.3 events/(100 ton yr)
The diff. reactor anti-? spectrum in units of
(anti-?/MeVcm2)
?i - P.huber,T.Shwetz, Phys.Rev D70 (2004)
Only 35 in the geo-neutrino window 5 cpy!
APC 28th April 2010, Paris
12
Geo-? signal in Borexino
Ideal expected spectrum
Expected in Borexino signal from the geo-?'s
obtained from the BSE model (Mantovani et al.
2004) 2.5 events/100ton yr
APC 28th April 2010, Paris
13
Calibrations 2008-2009
Detailed study of the detector. Over 300 on-axis
and off-axis positions.
Detector response vs position 100 Bq 14C222Rn
in scintillator in gt100 positions. Improved
position reconstruction. We reduced the error on
the FV down to 3.8 (old value 6)
Energy scale and quenching ?? 14C, 222Rn in
scintillator ?? 222Rn in scintillator ???
139Ce, 57Co, 60Co, 203Hg, 65Zn, 40K, 85Sr,
54Mn Neutron 241Am-9Be (protons recoil study)
Improved understanding of energy scale (from
120 keV up to 9.3 MeV)
Monte Carlo code is tuned to take into account
non-linearities of the energy scale (ionization
quenching, electronics)
APC 28th April 2010, Paris
14
MC spectra GeoReactors
The expected prompt positron event spectrum in
Borexino after taking into account the position
and energy response of the detector, MC.
APC 28th April 2010, Paris
15
Candidates selection
The data set used for the analysis is 537.2 live
days (Dec. 2007 - Dec. 2009)
1) Qprompt gt 410 p.e. 2) 700 p.e. lt Qdelayed
lt1250 p.e. 3) ?Rlt 1m
4) 20µs lt ?t lt 1280µs 5) RIV Rprompt gt 0.25m

Total detection efficiency determined with MC
is 0.850.01
APC 28th April 2010, Paris
16
Background sources
Search for the background capable to mimic the
anti-? signature (fake anti-?'s)
Radiogenic fake anti-?'s 1) (a,n) reaction
Cosmogenic fake anti-?'s (muon induced
background) 1) Fast neutrons 2)
Spallation isotopes (9Li,8He) Accidental
coincidences
APC 28th April 2010, Paris
17
Fake anti-? (210 Po)
Two correlated in time and space events Prompt -
beta from decay of 16O, or recoil proton Delayed
gamma from the neutron capture on 1H
210Po is the main contaminant in Borexino
(12 cpd/t on avarage
in present data set) Low abundance of 13C (1.1
) Probability for the reaction to occur
(5.00.3) 10-8
0.0210.002 ev/(100 t yr)
APC 28th April 2010, Paris
18
Fast neutrons
Borexino shielding 2 m of water and 2.5 m of PC
(buffer)
Fast neutrons can also fake the anti-? signal
(two signals correlated in space and time)
1) recoil proton 2)
gamma from the neutron capture.
Muons can generate neutrons from the Rocks,
Water.
Flux of fast neutrons from the surrounding rock
in Hall C was studied with MC. Measurements
obtained by Macro and LVD (D.M. Mei and A. Hime,
Phys. Rev. D 73, 053004 (2006)) as an input for
the simulations.
APC 28th April 2010, Paris
19
Muon induced fast n's
3800 mwe Fµ 1.2 µ/m2/hr ltEµgt 320
GeV Rµ IV 4200 cpd Frockn
7.3x10-10cm-1s-1 ltErockngt 90 MeV
- Muon Veto inefficiency is 99.5 - 2ms veto
after each WT muon remove more then 99.5 of WT
neutrons
lt 0.01 ev / (100 ton yr) (90 C.L.)
APC 28th April 2010, Paris
20
Fast n's from the rock
Overall statistics of 5x106 neutrons generated on
the surface Water Tank. 10 MeV lt En lt 3.5 GeV.
160 events in the IV (reduction factor 104)
Only 1 pair of correlated in time events (fake
anti-neutrino)
lt 0.04 ev / (100 ton yr) (90 C.L.)
APC 28th April 2010, Paris
21
Spallation Isotopes
Cosmic ray muons crossing the scintillator can
create the radioactive isotopes by the spallation
of the 12C atoms. Some of these isotopes can
decay via ?-?n cascade perfectly mimicking
anti-nu signature.
Expected in Borexino rate (0.096
0.019) ev/d (T.Hagner et all Ast. Phys. 14
(2000) 33-47)
APC 28th April 2010, Paris
22
9Li-8He isotopes
APC 28th April 2010, Paris
23
9Li-8He isotopes
Dedicated study 2s window after each muon. 51
candidates selected in the whole period.
Measured rate 15.4 events (100 ton yr)
Application of 2s veto (8 tau) after each passing
muon sufficiently removes all such events,
however this brings to the 10 reduction of the
exposition.
Remaining background from Li-He lt 0.03 0.02
ev/100t/y
APC 28th April 2010, Paris
24
Accidental coincidences
The same energy cuts for the prompt and delayed
signals as in the anti-? selection. Off-time
coincidence window of (2-20) s.
Energy spectrum of the acc. coincidence
0.080 0.010 ev/(100t y)
Search for the time corellated background time
window of 2ms - 2s
lt0.003 ev/(100t y)
APC 28th April 2010, Paris
25
Background summary
Expected signal from georeactors 5 15
APC 28th April 2010, Paris
26
Selected anti-? candidates
- Data set from Dec. 2007 to Dec. 2009 - Total
live time 532.7 days - Cosmogenic veto 2s
after each detected muon is removed (10 of
live time) - The fiducial exposure after all
cuts 252.6 ton yr - Total number of selected
anti-? candidates 21
APC 28th April 2010, Paris
27
21 selected candidates
APC 28th April 2010, Paris
28
21 selected candidates
Expected background is 0.400.05! Signal/Backgrou
nd 501
APC 28th April 2010, Paris
29
Reactor anti-? signal
Expected reactor anti-? signal in the reactor
anti-? window (Egt1300 p.e.) 65 of total Without
oscillations 16.31.1 events With
oscillation 9.40.6
events Background 0.090.06
events Observed 6 events
Exclusion of the non-oscillation hypothesis at
99.6 C.L.
Long mean-base line - effective distance is
1000 km - flux of 105cm-2s-1
APC 28th April 2010, Paris
30
Observed Geo-? signal
In the geo-neutrino window Elt1300 p.e Expected
from reactors anti-? 5.00.3 events Expected
background 0.310.05
events Observed
15 events
The best estimates for the geo-? and reactor
anti-? are
APC 28th April 2010, Paris
31
Observed Geo-? signal
Scaling the obtained best estimation with
exposure of the 252.6 ton yr, the best Borexino
measurement of the geo-? signal becomes
Phys. Lett. B 687 (2010) 299-304
By studying the profile of the log-likelihood
with respect to the Ngeo we reject the null
hypothesis for geo-? signal at more then 3?
Rate in events/(100 ton yr)
Hypothesis of a geo-reactor present in
the Earths Core upper limit for a 3 TW at 95
C.L.
APC 28th April 2010, Paris
32
Summary and Perspectives
First observation of geo-?'s with more then 3s.
First indication of the electron anti-?
disappearance on the long (1000 km) base line
(2.9 s).
The possible presence of the natural geo-reactor
of gt 3TW is excluded at 95 C.L.
Spectroscopic study of the geo-? signal
- Fully radiogenic or Min. radiogenic model?
- U/Th mass ratio measurement?
APC 28th April 2010, Paris
33
The End