The first year of Borexino data
LIP
July 9, 2008 – Lisbon
Davide Franco
Milano University & INFN
Outline






The role of neutrinos in particle physics and
astro-particle
The physics of Borexino
The Borexino detector
The “radio-purity” challenge
The reached goals (7Be and mn)
Near and far future goals
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Neutrinos: cosmic messengers
Astrophysical
source
High energy protons
5·10Mpc
neutrinos
High energy gammas
10 Mpc
Low energy protons
deflected
Lisbon – July 9, 2008
Ideal messenger:
• neutral
• stable
• weakly interacting
Davide Franco – Università di Milano & INFN
Messengers from the Sun’s core
• Core (0-0.25 Rs)
• Nuclear reactions: T~1.5 107 °K
• energy chains pp e CNO (neutrino
production)
• Radiative region (0.25-0.75 Rs)
• Photons carry energy in ~ 105 y
• Convective region (0.75-1 Rs)
• Strong convection and turbulence
• Complex surface phenomena
• Corona (> 1 Rs)
• Complex magneto-hydrodynamic
phenomena
• Gas at T~ 106 °K
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
The Standard Solar Model


It is an evolution model f the Sun, starting from the origin up to now:
 Includes:

Hydrostatic equilibrium equations

Energetic and mass balance

Nuclear Reactions

Energy transport
 Depends on several parameters and/or assumptions:

Initial chemical composition

Nuclear cross sections

Magnetic field, rotation
The model predicts the actual temperature and hence neutrino production
 Main uncertainties due to chemical compound and nuclear cross
sections
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Solar nuclear reactions


pp chain
 Fusion of 4 protons in a 4He nucleus with emission of 27 MeV energy
 Main chain for “small-medium size” stars, like the Sun
CNO cycle
 Carbon-Nitrogen-Oxygen or CNO cycle converts hydrogen to helium
 It is dominant in population I “hot” stars
 In the Sun, the CNO cycle contributes to 1-2 %, but there is no any
experimental evidence
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Neutrino Production In The Sun
pp chain:
pp, pep, 7Be, and 8B n
Lisbon – July 9, 2008
CNO cycle:
13N, 15O, and 17F n
Davide Franco – Università di Milano & INFN
Solar Neutrinos
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Solar Neutrino Spectra
Gallex
GNO
Homestake
Sage
SNO
SuperK
(real time)
Borexino (real time)
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
The Standard Solar Model before 2004
One fundamental input of the Standard Solar Model is the metallicity of the Sun abundance of all elements above Helium:
The Standard Solar Model, based on the old metallicity derived by Grevesse and
Sauval (Space Sci. Rev. 85, 161 (1998)), was in agreement within 0.5 in % with
the solar sound speed measured by helioseismology.
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
The Sun Structure

The study of the Sun structure is supported by the “helium-seismology”
 Pressure wave propagation in the Sun
 Sun’s vibration modes depend on the internal structure
 The Standard Solar Model can be tested by looking at the p-mode (acoustic
waves), g-mode (density or gravity waves) and f-mode (surface density
waves)
Reconstructed speed of sound
Lisbon – July 9, 2008
Surface waves
Reconstruction
Davide Franco – Università di Milano & INFN
The Standard Solar Model after 2004
Latest work by Asplund, Grevesse and Sauval (Nucl. Phys. A 777, 1 (2006)) indicates
a metallicity lower by a factor ~2. This result destroys the agreement with
helioseismology
s-1]
pp
(1010)
pep
(1010)
hep
(103)
7Be
8B
13N
15O
17F
(109)
(106)
(108)
(108)
(106)
BS05
AGS 98
6.06
1.45
8.25
4.84
5.69
3.07
2.33
5.84
BS05
AGS 05
5.99
1.42
7.93
4.34
4.51
2.01
1.45
3.25
D
-1%
-2%
-4%
-12%
-23%
-42%
-47%
-57%
[cm-2
Solar neutrino measurements can solve the problem!
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Borexino goals: solar physics
• First ever observations of sub-MeV neutrinos in real time
• Check the balance between photon luminosity and neutrino
luminosity of the Sun
• CNO neutrinos (direct indication of metallicity in the Sun’s
core)
• pep neutrinos (indirect constraint on pp neutrino flux)
• Low energy (3-5 MeV) 8B neutrinos
• Tail end of pp neutrino spectrum?
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Borexino goals: neutrino physics
Test of the matter-vacuum
oscillation transition with 7Be,
pep, and low energy 8B
neutrinos
Check of the mass varying
neutrino model (Barger et al.,
PRL 95, 211802 (2005))
Limit
on
the
neutrino
magnetic
moment
by
analyzing the 7Be energy
spectrum and with Cr source
pep
8B
7Be
Moreover: geoneutrinos and supernovae
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Detecting neutrinos

Solar neutrino flux on Earth F ~ 1010 cm-2 s-1 with energy [0.1-10] MeV

Weakly interacting : s ~ 10-46 – 10-43 cm2

At 1 MeV (F ~ 108 cm-2 s-1 ) in water (s ~2 x10-46 cm2) in 30 days (T =
2.5 106 s) to see 1 event we need:
1
F s T
 2  10 31 atomi  600 tons
Huge active masses and extremely low
background are required
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
First requirement: underground
Second requirement: ultra high purity materials
and proper design to minimize contaminations
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Borexino Collaboration
Genova
Milano
Princeton University
APC Paris
Perugia
Dubna JINR
(Russia)
Lisbon – July 9, 2008
Kurchatov
Institute
(Russia)
Virginia Tech. University
Munich
(Germany)
Jagiellonian U.
Cracow
(Poland)
Heidelberg
(Germany)
Davide Franco – Università di Milano & INFN
Abruzzo
120 Km da Roma
Laboratori
Nazionali del
Gran Sasso
Laboratori esterni
Assergi (AQ)
Italy
~3500 m.w.e
Borexino – Rivelatore e impianti
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Detection principles and n signature


Borexino detects solar n via their elastic scattering off electrons in a volume of
highly purified liquid scintillator

Mono-energetic 0.862 MeV 7Be n are the main target, and the only considered so far

Mono-energetic pep n , CNO n and possibly pp n will be studied in the future
Detection via scintillation light:

Very low energy threshold

Good position reconstruction

Good energy resolution
Typical n rate (SSM+LMA+Borexino)
BUT…



No direction measurement
The n induced events can’t be
distinguished from other b events
due to natural radioactivity
Extreme radiopurity of the
scintillator is a must!
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
n-spectrum in Borexino
NW-1 = 0.25 – 0.8 MeV (7Be)
Events/3years/100 tons/0.05 MeV
NW-2 = 0.8 – 1.4 MeV (pep & CNO)
NW-1
NW-2
• LMA –BP2004 – LUNA
• 3 years statistics
• 100 tons
Source
Lisbon – July 9, 2008
Number of events in 3 y
NW-1
NW-2
7Be
28470
0
pep
1095
986
pp
1095
0
8B
44
33
13N
1260
66
15O
1533
593
All
33497
1634
Davide Franco – Università di Milano & INFN
Borexino Background
Expected solar neutrino rate in 100 tons of scintillator ~ 50 counts/day (~ 5 10-9 Bq/Kg)
Just for comparison:
Natural water
~ 10 Bq/kg in 238U, 232Th and 40K
Air
~ 10 Bq/m3 in 39Ar, 85Kr and 222Rn
Typical rock
~ 100-1000 Bq/kg in 238U, 232Th and 40K
BX scintillator must be 9/10 order of magnitude less
radioactive than anything on earth!
- Low background nylon vessel fabricated in hermetically sealed low radon clean
room (~1 yr)
- Rapid transport of scintillator solvent (PC) from production plant to underground
lab to avoid cosmogenic production of radioactivity (7Be)
- Underground purification plant to distill scintillator components.
- Gas stripping of scintlllator with special nitrogen free of radioactive 85Kr and 39Ar
from air
- All materials electropolished SS or teflon, precision cleaned with a dedicated
cleaning module
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Detector layout and main features
Scintillator:
270 t PC+PPO in a 150 mm
thick nylon vessel
Nylon vessels:
Inner: 4.25 m
Outer: 5.50 m
Carbon steel plates
Lisbon – July 9, 2008
Stainless Steel Sphere:
2212 PMTs
1350 m3
Water Tank:
g and n shield
m water Č detector
208 PMTs in water
2100 m3
20 legs
Davide Franco – Università di Milano & INFN
PMTs: PC & Water proof
Nylon vessel installation
Installation of PMTs on the sphere
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Water Plant
Storage area and Plants
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Counting Test Facility
CTF è un prototipo su piccola scala di
Borexino:
• ~ 4 tons of scintillator
• 100 PMTs
• Buffer of water
• Muon veto
• Vessel radius: 1 m
Lisbon – July 9, 2008
CTF ha dimostrato la
fattibilità di Borexino
Davide Franco – Università di Milano & INFN
May 15, 2007
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Borexino background
RadioIsotope
Name
m
Source
cosmic
Concentration or Flux
Typical
~200 s-1 m-2
Strategy for Reduction
Required
~
10-10
at sea level
Hardware
Software
Achieved
Underground
Cherenkov signal
<10-10
Cherenkov detector
PS analysis
(overall)
Ext. g
rock
Water Tank shielding
Fiducial Volume
negligible
Int. g
PMTs, SSS
Material Selection
Fiducial Volume
negligible
Water, Vessels
Clean constr. and handling
14C
Intrinsic PC/PPO
238U
Dust
Old Oil, check in CTF
~ 10-5-10-6 g/g
~ 10-18
< 10-16 g/g
~ 10-12
~ 10-18
~ 2 10-17
~ 7 10-18
Threshold cut
Distillation, Water Extraction
232Th
Organometallic (?)
(dust)
(in scintillator)
Filtration, cleanliness
7Be
Cosmogenic (12C)
~ 3 10-2 Bq/t
< 10-6 Bq/ton
Fast procurement, distillation
Not yet measurable
?
40K
Dust,
~ 2 10-6 g/g
< 10-14 g/g scin.
Water Extraction
Not yet measurable
?
PPO
(dust)
< 10-11 g/g PPO
Distillation
Not yet measurable
?
210Pb
Surface contam.
Cleanliness, distillation
from 222Rn decay
210Po
(NOT in eq. with
Surface contam.
Cleanliness, distillation
from 222Rn decay
222Rn
air, emanation from
210Po)
Spectral analysis
a/b stat. subtraction
~ 0.01 c/d/t
Delayed coincidence
< 0.02 c/d/t
?
~ 10 Bq/l (air)
< 1 c/d/100 t
Water and PC N2 stripping,
materials, vessels
~100 Bq/l (water)
(scintillator)
cleanliness, material selection
39Ar
Air (nitrogen)
~17 mBq/m3 (air)
< 1 c/d/100 t
Select vendor, leak tightness
Not yet measurable
85Kr
Air (nitrogen)
~ 1 Bq/m3 in air
< 1 c/d/100 t
Select vendor, leak tightness
Spectral fit
(learn how to measure it)
fast coincidence
Lisbon – July 9, 2008
~ 14
Davide Franco – Università di Milano & INFN
= 25±3
= 29±14
“Few” details

Detector & Plants

All materials carefully and painfully
selected for:

Low intrinsic radioactivity

Low Rn emanation

Good behaviour in contact with PC


Pipes, vessels, plants:

electropolished, cleaned with
detergent(s), pickled and
passivated with acids, rinsed with
ultra-pure water down to class 2050
The whole plant is vacuum tight

Leak requirements < 10-8 atm/cc/s

Critical regions (pumps, valves, big
flanges, small failures) were
protected with additional nitrogen
blanketing
Lisbon – July 9, 2008

PMTs (2212)

Sealing: PC and water tolerant

Low radioactivity glass

Light cones (Al) for uniform light collection in
fiducial volume

Time jitter: 1.1 ns (for good spatial resolution,
mu-metal shielding)

384 PMTs with no cones for m id

Nylon vessels

Material selection for chemical & mechanical
strength

Low radioactivity to get <1 c/d/100 t in FV

Construction in low 222Rn clean room

Never exposed to air
Davide Franco – Università di Milano & INFN
Expected Spectrum
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
The starting point: no cut spectrum
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Detecting (and rejecting)
cosmic muons

m pulses
m are identified by ID and OD

OD eff: ~ 99%

ID based on pulse shape analysis
m crossing the buffer only

Rejection factor

> 103 (conservative)
m crossing the scintillator
A muon
m track
in OD
ID efficiency
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Detecting (and rejecting) cosmogenic neutrons
t ~ 250 ms
n+p
t = 255 ± 5 ms
d+g
A dedicated trigger starts after each muon opening a gate for 1.6 ms.
An offline clustering algorithm identifies neutron in high multiplicity events
A muon
in OD
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Muon and neutron cuts
No cut
m cut
Residual background : << 1 c/d/100 t
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Position reconstruction

Position reconstruction algorythms (we have 4 codes right now)
 time of flight fit to hit time distribution
 developed with MC, tested and validated in CTF
 cross checked and tuned in Borexino with 214Bi-214Po events and 14C
events
z vs Rc scatter plot
Resolution
214Bi-214Po
(~800 KeV)
14±2 cm
14C
(~100 KeV):
41±4 cm
Rc  x 2  y 2
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Vessel Radius and Shape
Vessel radius and origin calibrated exploiting fast coincidences of
220Rn chain segments emanated from vessel nylon
212Bi-212Po
222Rn
and
214Bi-214Po
t = 432.8 ns
t = 236 ms
The two plots confirms the CCD camera calibration results: vessel origin is
shifted along positive z-axis by about 5 cm.
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Spatial resolution
22 cm @ ~400 keV
FV
220Rn-216Po
7Be
energy region (mainly 210Po)
m-induced neutrons
FV
214Bi-214Po
(~800 KeV)
14±2 cm
14C
15 cm @ ~2200 keV
Lisbon – July 9, 2008
(~100 KeV):
41±4 cm
Davide Franco – Università di Milano & INFN
Spectrum after FV cut (100 tons)
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
238U
content
t = 236 ± 8 ms
t = 236 ms
b
214Bi
3.2 MeV
214Po
a
210Pb
~700 keV eq.
Assuming secular equilibrium and
looking in the FV only:
0.02 cpd/tons corresponding to
238U = (1.9 ± 0.3)×10-17 g(U)/g
214Bi
214Po
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
232Th
212Bi
b
t = 432.8 ns
212Po
2.25 MeV
a
content
208Pb
t = 442 ± 57 ns
~1000 keV eq.
Events are mainly on the south vessel
surface (probably particulate)
Assuming secular equilibrium and looking
in the FV only :
0.00256 cpd/ton corresponding to
232Th = (6.8±1.5)×10-18 g(Th)/g
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
a/b discrimination
Full separation at high energy
Small deformation due to average
SSS light reflectivity
a particles
b particles
ns
250-260 pe; near the 210Po peak
2 gaussians fit
a/b Gatti parameter
Lisbon – July 9, 2008
200-210 pe; low energy side of the 210Po peak
2 gaussians fit
a/b Gatti parameter
Davide Franco – Università di Milano & INFN
210Po
No radial cut
contamination
R < 3.8 m
FV (R < 3 m)
Not from 210Pb
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
a/b statistical subtraction
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Energy calibration and stability


We have not calibrated with inserted sources (yet)
 Planned for the near future
So far, energy calibration determined from 14C end point spectrum
210Po a peak
 Energy stability and resolution monitored with
 Difficult to obtain a very precise calibration because:
14C intrinsic spectrum and electron quenching factor poorly known

Light yield determined from 14C fit
Light yield monitored with
210Po peak position
190
188
Po peak (nhits)
186
184
182
180
178
176
174
172
170
28-Apr
Lisbon – July 9, 2008
18-May
7-Jun
27-Jun
Davide Franco – Università di Milano & INFN
17-Jul
6-Aug
Energy scale
14C
MC vs data comparison of photoelectron time
distributions from 14C
MC-G4Bx
Data
11C
m-induced
neutrons
LY = 500 (1%) p.e./MeV
kB = 0.017 (15%) cm/MeV
Ph.Y. ~ 9000 photons/MeV
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
New results with 192 days of statistics
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
New results with 192 days of statistics
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Systematic and Final Result
Estimated 1σ Systematic Uncertainties* [%]
Total Scintillator Mass
0.2
Fiducial Mass Ratio
6.0
Live Time
0.1
Detector Resp. Function
6.0
*Prior
for LMA-MSW oscillations:
48±4 cpd/100 tons, which means:
to Calibration
Cuts Efficiency
7Be
Expected interaction rate in
absence of oscillations:
75±4 cpd/100 tons
0.3
Rate: 49±3stat±4syst cpd/100 tons , which means
Total
Lisbon – July 9, 2008
8.5
Davide Franco – Università di Milano & INFN
Before Borexino
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
After Borexino
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Constraints on pp and CNO fluxes
Combining Borexino 7Be results with other experiments,the expected
rate in Clorine and Gallium experiments is
where
measured over
predicted flux ratio
Survival Probability
• Ri,k and Pi,k are calculated in the hypothesis of high-Z SSM and MSW LMA
• Rk are the rates actually measured by Clorine and Gallium experiments
• f8B is measured by SNO and SuperK to be 0.87 ±0.07
• f7Be =1.02 ±0.10 is given by Borexino results
Plus luminosity constraint:
best determination of pp flux!
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Neutrino Magnetic Moment
Neutrino-electron scattering is the most sensitive test for mn search
EM current affects cross section:
spectral shape sensitive to μν
sensitivity enhanced at low
energies (c.s.≈ 1/T)
A fit is performed to the energy
spectrum including contributions
from 14C, leaving μn as free
parameter of the fit
Lisbon – July 9, 2008
Estimate
Method
10-11 μB
SuperK
8B
<11
Montanino et al.
7Be
<8.4
GEMMA
Reactor
<5.8
Borexino
7Be
<5.4
Davide Franco – Università di Milano & INFN
What next?
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
BOREXino
NA54 @ CERN: 100 and 190 GeV muon
beams on a 12C target: 11C represents
80% of all the muon-induced
contaminants and more than 99% in the
CNO pep-n energy window
Hagner et al., Astropart. Phys. 14, 33 (2000)
KamLAND
(from the Mitzui presentation at Neutrino06)
11C
Rate
(cts / day / 100 tons)
All energy
KamLAND
107
55
BOREXino
15
7.4
0.15
0.074
SNO+
Lisbon – July 9, 2008
0.8 – 1.4 MeV
Davide Franco – Università di Milano & INFN
11C
production and decay
m (+ secondaries) + 12C → m (+ secondaries) + 11C + n
n+p→d+g
Coincidence among:
• cosmic muon:
11C
→
11B
+
e+
+ ne
• rate at LNGS (3700 mwe): 1.16 hr-1 m-2
• average energy: 320 GeV
• gamma from neutron capture:
• energy: 2.2 MeV
• capture time: 250 ms
• positron from 11C decay:
• deposited energy between 1.022 and
1.982 MeV
• mean life: 30 min
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Large scintillator detector potential
m
PC+PPO
11C
n
g
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN

pep and CNO n fluxes
 software algorithm based on a three-fold coincidence analysis to
subtract efficiently cosmogenic 11C background
 Muon track reconstruction

8B

pp seasonal variations (?)

High precision measurements
 systematic reduction
 calibrations

geoneutrinos
at low energy region (3-5 MeV)
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN
Conclusion
o Borexino opened the study of the solar neutrinos in real time below
the barrier of natural radioactivity (4 MeV)
o Two measurements reported for 7Be neutrinos
o Best limits for pp and CNO neutrinos, combining information from
SNO and radiochemical experiments
o Opportunities to tackle pep and CNO neutrinos in direct
measurement
o Borexino will run comprehensive program for study of antineutrinos
(from Earth, Sun, and Reactors)
o Borexino is a powerful observatory for neutrinos from Supernovae
explosions within few tens of kpc
o Best limit on neutrino magnetic moment. Improve by dedicated
measurement with 51Cr neutrino source
Lisbon – July 9, 2008
Davide Franco – Università di Milano & INFN