Precision Measurement of Boron-to-Carbon
ratio in Cosmic Rays from 2 GV to 2 TV
with the Alpha Magnetic Spectrometer on
the International Space Station.
V. Formato on behalf of the AMS collaboration
CERN and INFN - Sezione di Perugia
05/09/2016 - ECRS - Torino
B/C in cosmic rays
The B/C flux ratio is a powerful tool to study the properties of cosmic ray
propagation as Boron is assumed to be produced purely from collision of primary
cosmic rays, such as Carbon and Oxygen, with the interstellar medium (ISM).
Galactic
Halo
C, N, O, …, Fe + ISM ! B
+X
ISS
Galactic
Disk
CRs are commonly modelled as a relativistic gas diffusing into a magnetised plasma.
Diffusion models based on different assumptions predict different behaviour for B/C ~ Rδ.
With Kolmogorov interstellar turbulence model a δ = -1/3 is expected, while Kraichnan
theory leads to δ = -1/2.
2
AMS in a nutshell
TRD
Identify e+, e-
Particles and nuclei are defined by
their charge (Z)
and energy (E ~ P)
TOF
Z, 𝜷
1
TRD
±Z
TOF
2
3-4
5-6
7-8
Tracker
Silicon Tracker
Z, P
Magnet
TOF
RICH
RICH
9
ECAL
E of e+, e-, γ
ECAL
Z, 𝜷
Z, P are measured
independently by the Tracker,
RICH, TOF and ECAL
3
3
Charge measurements in AMS
AMS Event Display
Tracker L1 - 6.2
Run/Event 1420006000 / 10014 GMT Time 20
Side
Front
Tracker, R = p/Z
Full Span MDR (Z=6) ≈ 2.6
TV
TRD - 5.6
Upper ToF - 6.0
Inner Tracker - 6.0
TOF, β
Δβ (β=1, Z=5,6) ≈
0.01
Lower ToF - 5.7
RICH, β
Δβ (β=1, Z=6) ≈ 5×10-4
Tracker L9 - 5.8
DAQ
Header
ECAL - 6.0
z
z
Level1
x y
y x
Level3
4
Particle TrTofTrdTrdHEcal No 0 Id=14 p= 31.4± 0.78 M= 8.03± 7.8 θ=2.99 φ=6.12 Q= 6 β= 0.9688± 0.0370 βh= 0.947± 0.040 θ_M -40.
Nuclei identification in AMS
Fe
Ni
Zn
H
He
Li
C
Be B
N O
Ne Mg
F
Na
Al Si P S
Cl Ar K Ca
Sc Ti V Cr
5
Fe
Ni
B/C Event selection
Selection
• Tracker and TOF Charges compatible with Z=5, 6.
• Track passing through L1 with good charge.
• Tracks with at least 5 points and a good fit (𝝌2Y L2-L8 < 10).
• Rigidity above geomagnetic cutoff (R>1.2 RC).
! Statistics for 60 months: 8.3M Carbon and 2.3M Boron.
AMS Event Display
Run/Event 1420006000 / 10014 GM
Side
Front
Long Lever Arm Analysis
• Tracker Layer 9 Charge compatible with Z=5, 6.
• Full Span Track with a good fit (𝝌2Y L1-L9 < 10).
! Highest possible MDR (2.6 TV for C, 3.0 TV for B).
DAQ
Header
z
z
Level1
x y
y x
Level3
Particle TrTofTrdTrdHEcal No 0 Id=14 p= 31.4± 0.78 M= 8.03± 7.8 θ=2.99 φ=6.12 Q= 6 β= 0.9688± 0.0370 βh= 0.947±
AMS Event Display
Run/Event 1420006000 / 10014 GM
Side
Front
Large Statistics Analysis
• No requirement on L9.
• Track with a good fit (𝝌2Y L1-L8 <10).
! Factor 5 more events, and less interacting events.
DAQ
Header
z
6
x y
z
Level1
y x
Level3
Particle TrTofTrdTrdHEcal No 0 Id=14 p= 31.4± 0.78 M= 8.03± 7.8 θ=2.99 φ=6.12 Q= 6 β= 0.9688± 0.0370 βh= 0.947±
Boron sample purity
The high redundancy in charge measurements allows to spot and
identify nuclei fragmenting through hadronic inelastic interactions.
Tracker L1 - 5.9
TRD - 5.7
Upper ToF 1 - 5.8
Upper ToF 2 - 10.9
Inner Tracker - 4.9
Lower ToF - 5.0
RICH - 5.5
7
Boron sample purity
Contamination < 3%
Selection efficiency > 96%
Systematics on the knowledge of the
charge spectra are included in final error.
Events / ( 0.0266667 )
16.600 < R
ρ (GV) < 18.000
nBoron = 46374 ± 220
• Data
Fit
103
nCarbon = 4560 ± 82
nNitrogen = 547 ± 31
nOxygen = 1141 ± 36
B
102
C
10
O
N
1
4
4.5
5
5.5
6
6.5
8
7
7.5
8
8.5
9
L1 Charge
Interactions above L1
Background generated above L1 is calculated using MC and light nuclei fluxes measured by AMS.
MC interaction channels (ex. C + C, Al ! B + X) have been verified with data (see below).
Background is up to 9% for B, associated systematic error is below 1%.
C
Data
MC
He
Li
B
Be
9
[27-36] GV
Systematics on Rigidity Resolution
The MC rigidity resolution functions were verified with the ISS data in multiple ways.
One of the comparison is the validation of the spatial resolution of the inner tracker.
Systematic errors arising from the understanding of the resolution matrix and the bin-to-bin
migration unfolding procedures account for 1% below 200GV and 3% at 2.5 TV.
3
1.5
3
× 10
a)
5
Boron Data
Boron MC
×10
Carbon Data
Carbon MC
b)
4
Events
Events
1
σ = 8 µm
0.5
3
2
σ = 10 µ m
1
0
−40
−20
0
∆y [µm]
20
0
−40
40
10
−20
0
∆y [µm]
20
40
Systematics on Acceptance
AMS Event Display
Side
L2
L8
DAQ
x y
Boron
Run/Event 1420006000 / 10014 GMT Time 2014-365.06:07:00
Front
z
C Data/MC
Surv.
Prob Ratio Probability
B Data/MC Surv.
Prob. Ratio
L8-L9
Survival
Ratio
Probability of detecting an hit on L9 depends on the inelastic interactions in the materials
between L8 and L9 (“1/3” of the AMS material), allowing the control of flux normalisation.
Total systematic error including all AMS materials is about 2%.
L9
Header
1
0.95
a)a
0.9
1.02
10
102
Carbon
1
DAQ
Header
z
Level1
Level3
Level1
y x
Level3
HEcal No 0 Id=14 p= 31.4± 0.78 M= 8.03± 7.8 θ=2.99 φ=6.12 Q= 6 β= 0.9688± 0.0370 βh= 0.947± 0.040 θ_M -40.4 o Coo=(-30.15,23.35,53.05) LT 0.90 θ_G 0.72 φ_G 1.88
0.98
b)
11
10 R [GV]
102
Measurement verification
B/C measured using events passing through L1-L9
divided by the one measured using events passing through L1-L8.
The observed agreement verifies:
(i) acceptance: the amount of material traversed is different
(ii) unfolding: bin-to-bin migration is different due to different resolution
L1
L1
(B/C)
/ (B/C)
(B/C)
/
(B/C)
L1-L9
L1-L8L1-L8
L1-L9
1.3
L8
1.2
1.1
L9
1
5%
0.9
Data
0.8
0.7
Total Error
20%
10
10
12
R [GV]
2
10
3
B/C current result
Boron-to-Carbon Ratio
0.4
(A. Oliva, ICRC 2015)
0.3
0.2
AMS-02
PAMELA (2014)
TRACER (2006)
CREAM-I (2004)
ATIC-02 (2003)
AMS-01 (1998)
Buckley et al. (1991)
CRN-Spacelab2 (1985)
Webber et al. (1981)
HEAO3-C2 (1980)
Simon et al. (1974-1976)
Dwyer & Meyer (1973-1975)
Orth et al. (1972)
0.1
0.05
0.04
0.03
0.02
1
10
10
13
2
10
3
Kinetic Energy (GeV/n)
Conclusions
• The B/C is an important measurement for the understanding of the
propagation of the CRs.
• AMS measures B/C from 1.9 GV to 2.6 TV, with 2.3 million boron and
8.3 million boron nuclei with a typical accuracy of 3% at 100 GV.
• AMS B/C is able to reject a class of models explaining the positron
fraction as an effect of secondary production.
• The accuracy of the B/C will be significantly improved, in particular
at the highest rigidities, during the lifetime of the ISS.
14