Post on 08-Jul-2020
High-Precision Half-Life Measurements for the
Superallowed β+ Emitter 10C: Implications for Weak Scalar
Currents
Michelle R. DunlopUniversity of Guelph
Nuclear Structure, July 28 2016
Superallowed Fermi Beta Decays• In general, β decay ft values can be expressed as:
• For the special case of 0+ ⟶ 0+, we have a pure Fermi (Sβ = 0) allowed (Lβ = 0) transition. Transitions between isobaric analogue states are known as superallowed Fermi beta decays.
10C 0+ 10B 0+
T1/2, BR
ft =K
G2A|Mfi(GT )|2 +G2
V |Mfi(F )|2
Phase Space (Q-value) ConstantsMatrix
Element
Weak Coupling Strength
Testing Fundamental Properties of the Weak Interaction with Superallowed Decays• The Fermi beta decay transition is the isospin ladder operator.
For superallowed transitions between T=1 isobaric analogue states, the matrix element is simply:
• The superallowed ft values thus simplify:
{ {= 0 for Pure Fermi decays
= 2 for decays between T=1
IAS
{
constants (Conserved
Vector Current hypothesis)
ft =K
G2A|Mfi(GT )|2 +G2
V |Mfi(F )|2 ) K
2G2V
|Mfi(F )|2 = (T � TZ)(T + TZ + 1) = 2
A Selection of Beta Decay ft Values
Superallowed ft values
0 10 20 30 40Z of Daughter
3030
3040
3050
3060
3070
3080
3090
ft (s
)
10C14O
22Mg
26Alm
34Cl
34Ar
38Ca
38Km
42Sc
46V
50Mn
54Co
62Ga74Rb
ft ⇡ K
2G2V
ft(s)
{2%
= constant
ΔR = nucleus independent inner radiative correction: 2.361(38)%
δR = nucleus dependent radiative correction to order Z2α3: ≈1.4% - depends on electron’s energy and Z of nucleus
δNS = nuclear structure dependent radiative correction: -0.3% to 0.03%
δC = nucleus dependent isospin-symmetry-breaking (ISB) correction: 0.2% to 1.5% - strong nuclear structure dependence (radial overlap)
“Corrected” ft value
Experiment ◦ Half-life ◦ Q-value ◦ Branching Ratio
Calculated corrections (≈1%) (nucleus dependent)
Inner radiative correction (≈2.4%) (nucleus independent)
CVC Hypothesis
Ft = ft (1+ δR )(1+ δNS - δC) = 2Gv (1+ΔR)2
K
Superallowed Fermi Beta Decay: Theoretical Corrections
Corrected Superallowed Ft Values
• The superallowed data confirm the CVC hypothesis at the level of 1.2 x 10-4.
Search for Physics Beyond the Standard Model• The SM description of weak interaction is an equal mixture of vector and
axial-vector which maximizes parity violation.• We can seek evidence that the weak interaction is not pure “V-A” by
searching for contributions from scalar or tensor couplings in the weak interaction.
• In particular, the superallowed data are sensitive to contributions from scalar interactions.
� =p
1� (↵Z)2
Ft = constant ! Ft(1 + bF �hW�1i) = constantFt = constant ) Ft(1 + bF �hW�1i) = constant
bF = Fierz Interference Term
W = Total Positron Energy
SM description With Scalar Interaction
{The ⟨W-1⟩ dependence
means that it is the lowest-Z decays that are most
sensitive to contributions from a scalar interaction
Ft = constant ! Ft(1 + bF �hW�1i) = constant
• The Fierz interference term (bF) can be extracted by a linear fit of 1/Ft vs Ɣ⟨W-1⟩
• Current limit from Hardy and Towner, Phys Rev. C 91 025501 (2015): bF = -0.0028(26)
0 0.1 0.2 0.3 0.4 0.5 0.6γ <W-1>
32.35
32.40
32.45
32.50
32.55
32.60
32.65
32.70
32.75
1/Ft
(1/s)
x 1
0-5
bF = -0.0028 ± 0.0025
Ft0 = 3070.1 ± 2.0 s
10C
14O
22Mg
26Alm
74Rb
62Ga54Co
50Mn
46V
38Ca
42Sc
34Ar
34Cl
38Km
χ2/ν = 0.40bF = -0.0028 ± 0.0026
Search for Physics Beyond the Standard Model
Ft = constant ! Ft(1 + bF �hW�1i) = constant
• The Fierz interference term (bF) can be extracted by a linear fit of 1/Ft vs Ɣ⟨W-1⟩
• Current limit from Hardy and Towner, Phys Rev. C 91 025501 (2015): bF = -0.0028(26)
• Limits are dominated by the lightest superallowed decays 14O and, in particular, 10C0 0.1 0.2 0.3 0.4 0.5 0.6
γ <W-1>
32.35
32.40
32.45
32.50
32.55
32.60
32.65
32.70
32.75
1/Ft
(1/s)
x 1
0-5
bF = -0.0028 ± 0.0025
Ft0 = 3070.1 ± 2.0 s
10C
14O
22Mg
26Alm
74Rb
62Ga54Co
50Mn
46V
38Ca
42Sc
34Ar
34Cl
38Km
χ2/ν = 0.40bF = -0.0028 ± 0.0026
Search for Physics Beyond the Standard Model
The 10C half-life• The adopted 10C half-life is evaluated using the 4 most precise
measurements.• The inconsistencies in the dataset result in a highly inflated uncertainty on
the adopted world average half-life.
T1/2 = 19.3052 ± 0.0071 s
• The 10C half-life is important because it has a significant impact on the limits set on scalar currents.
• More than 200 individual superallowed measurements of comparable precision are currently used to set the limit on bF.
• If the half-life from either of the two peaks in the ideograph is adopted, the central value of bF is shifted by more than 0.5σ.
The 10C half-life
• An accurate and precise determination of the 10C half-life is thus critical to the limits set of bF set by the superallowed data.
bF = -0.0016 ± 0.0026 bF = -0.0031 ± 0.0026
TRIUMF-ISAC
• Up to 100 μA, 500 MeV protons from TRIUMF’s main cyclotron are accelerated onto targets which produce high-intensity secondary radioactive ion beams by the ISOL technique
• NiO target used to obtain CO+ and C+ ions• Radioactive ion beams of 10C16O (A=26) and 10C (A=10), with beam
intensities of ~1.5x105 pps and ~2.5x104 pps, respectively, were delivered to the detectors
The 10C decay scheme• 10C decays to the excited states in 10B giving us the opportunity to
measure the half-life by both direct measurements of the emitted β particle or by measuring the ɣ-ray emitted during the decay to the ground state.
0+
1+
3+
β+
10C
10Bstable
T1/2 = 19.3 s Q = 3648 keV
0+
1740
718
g.s.
1.4645%
98.530%
1022 keV
718 keV
ɣ Counting — The 8π Spectrometer
• Spherical array of 20 BGO Compton suppressed HPGe detectors• ~1% photopeak efficiency at 1.3 MeV• Beam implanted onto a tape at the centre of the array. The decay
activity was measured for 500 s (for ~25 half-lives). Tape moved into disposal box which is shielded by a lead wall (removes long lived contaminants out of view of the detector).
• The 10C half-life was measured by gating on the characteristic 718-keV ɣ-ray which follows 100% of 10C β decays.
ɣ Photopeak Counting
0 200 400 600 800 1000 1200 1400 1600 1800 2000Eγ (keV)
102
103
104
105
106
107Co
unts
per 0
.5 k
eV
705 710 715 720 725Eγ (keV)
103
104
105
106
107
Coun
ts
1022 keV
718 keV
• A total of 58 runs were taken, comprised of 562 cycles.• After a run, the shaping times and dead-times were varied. • The dead-time and pile-up corrected data are summed and a single fit to
the data is performed.
ɣ Photopeak Counting
103
104
105
106
Coun
ts pe
r 1.2
s
0 100 200 300 400 500Time (s)
-4
-2
0
2
4
Resid
uals
[(yi-y
fit)/σ
i]
T1/2 (10C) = 19.2969 ± 0.0052 s
χ2/ν = 0.87
(a)
0 10 20 30 40 50 60Run Number
19.1
19.2
19.3
19.4
19.5
19.6
Hal
f-life
(s)
Half-life Vs. Run
T1/2(10C) = 19.2970 ± 0.0052 s
χ2/ν = 1.19
(b)
Symbols = Different Shaping Times Colours = Different Applied Dead Times
• The data are grouped according to their experimental running conditions. The grouping with the largest 𝞆2/ν > 1 is used to inflate the uncertainty on the measured half-life, as recommended by the Particle Data Group (PDG).
ɣ Photopeak Counting — Systematics
0 1 2 3 4 5 6 7 8 9 1019.25
19.26
19.27
19.28
19.29
19.3
19.31
19.32
19.33
19.34
19.35
10C
Hal
f-life
(s)
All DataDead Time (µs): Variable, 27, 40Shaping Time (µs): 0.5, 1.0, 2.0Beam: Pre-Switching; Switching
Systematics
χ2/ν = 1.01 χ2/ν = 2.01 χ2/ν = 0.72
T1/2,ɣ(10C) = 19.2969 ± 0.0052sys ± 0.0052stat = 19.2969 ± 0.0074 sRelative precision of 0.03%
β Counting — 4π gas counter
• 4π continuous-flow proportional gas counter with tape transfer system• Directly detects β particles with ~100% efficiency• Cycles: Implant beam, move into gas counter and measured the
decay activity for ~500 s, move tape into tape disposal box to remove long-lived contaminants from the gas counter.
Implantation SiteBeam
Gas Counter
Tape Disposal
Box
Tape Spool
~ 28 cm
Atmosphere
High vacuum (~10-6 T)
• Radioactive beams of both 10C (A=10) and 10C16O (A=26) were delivered to the gas counter
β Counting — In-beam Contaminant
• In the A=26 beam, a contaminant with T1/2 = 9.97(8) min was identified
• T1/2 consistent with the that of 13N, which could be delivered as 13N2 (mass difference to 10C16O of 0.271 MeV) or HC13N (mass diff. of 1.673 MeV) in the A=26 beam.
• The literature T1/2 of 13N was included as a fixed parameter in the fit to the subset of data taken with the A=26 beam.
• A total of 82 runs were taken, comprised of 598 cycles.• After a run, parameters such as the dwell time, beam type, and gas
counter, (etc..) were changed before a new run was started.
β Counting
102
103
104
105
106
Coun
ts pe
r 1 s
10C decay13N contaminantBackgroundBest Fit Result
0 100 200 300 400 500Time (s)
-4
-2
0
2
4
Resid
uals
[(yi-y
fit)/σ
i] T1/2 (10C) = 19.2987 ± 0.0030 s χ
2/ν = 1.03
(a)
0 20 40 60 80Run Number
19.20
19.25
19.30
19.35
19.40
19.45
Hal
f-life
(s)
χ2/ν= 1.02
T1/2 (10C) = 19.3009 ± 0.0017 s
(b)
Open (Filled) Circles = 10C (10C16O) Radioactive Beam Colours = Different Gas Counters
0 1 2 3 4 5 619.298
19.299
19.3
19.301
19.302
19.303
10C
Hal
f-life
(s)
T1/2(13N): +/- 1σDeadtime: +/- 1σ
Systematics
19.28
19.29
19.30
19.31
19.32
19.33
19.34
19.35
19.36
Hal
f-life
(s)
All DataBias Voltage (V): 2400, 2450, 2500, 2550, 2600, 2700, 2750, 2800Threshold (mV): 70, 95, 120Dwell Time (s): 1.0, 1.2, 1.4Radioactive Beam: 10C16O, 10CGas Counter: 1, 2Initial 10C Activity (kHz): <6, 6-8, 8-10, 10-12, >12
χ2/ν = 0.92 0.13 0.54 0.06 0.620.62
T1/2(10C) = 19.3009 ± 0.0017 s
• Extensive investigation of systematics was performed.• Each grouping of the experimental parameters yield a 𝞆2/ν < 1 and hence
no inflation of the statistical uncertainty.• Vary the fixed parameters within their ±1σ limits. No statistically significant
change in the half-life is observed.
β Counting — Systematics
T1/2,β(10C) = 19.3009 ± 0.0017 sThe most precise (0.009%) superallowed T1/2 reported to date!
• The 6 most precise half-life measurements now yield T1/2 = 19.3015 ± 0.0015 s with a 𝞆2/ν = 1.90, which is an improvement in the uncertainty by a factor of nearly 3.
Results — 10C Half-life
T1/2(10C) = 19.3052 ± 0.0071 s
• With the updated 10C half-life and newly updated 14O Q-value (A. A. Valverde et al., Phys. Rev. Lett. 114, 232502 (2015)) and 14O branching ratio (P. A. Voytas et al. Phys. Rev. C 92, 065502 (2015)) measurements, a linear fit to the superallowed Ft data yields an updated Fierz interference term of:
Results — Scalar Current Limits
0 0.1 0.2 0.3 0.4 0.5 0.6
γ <W-1
>
32.35
32.40
32.45
32.50
32.55
32.60
32.65
32.70
32.75
1/F
t (1
/s)
x 1
0-5
bF = -0.0018 ± 0.0021
Ft0= 3070.8 ± 1.8 s
10C
14O
22Mg
26Al
m
74Rb
62Ga
54Co
50Mn
46V
38Ca
42Sc
34Ar
34Cl
38K
m
χ2/ν = 0.44
bF = -0.0018(21)and
Cs/CV = -bF/2 = 0.0009(11)
for the ratio of weak scalar to vector
couplings assuming left handed neutrinos
M. R. Dunlop et al., Phys. Rev. Lett. 116, 172501 (2016)
• Measurements of the 10C half-life were performed in order to resolve inconsistencies in the data set.
Conclusions
• Measurements of the 10C half-life were performed in order to resolve inconsistencies in the data set.
• Resolving the discrepancy in the 10C half-life is important because it is the lowest-Z superallowed decay and is most sensitive to contributions from scalar currents in the weak interaction.
Conclusions
• Measurements of the 10C half-life were performed in order to resolve inconsistencies in the data set.
• Resolving the discrepancy in the 10C half-life is important because it is the lowest-Z superallowed decay and is most sensitive to contributions from scalar currents in the weak interaction.
• We measured the half-life via both ɣ-ray photopeak counting and direct β counting, obtaining consistent results.
Conclusions
• Measurements of the 10C half-life were performed in order to resolve inconsistencies in the data set.
• Resolving the discrepancy in the 10C half-life is important because it is the lowest-Z superallowed decay and is most sensitive to contributions from scalar currents in the weak interaction.
• We measured the half-life via both ɣ-ray photopeak counting and direct β counting, obtaining consistent results.
• The β measurement (T1/2 = 19.3009(17) s) represents the most precise superallowed half-life measurement reported to date and the first to achieve a precision below 10-4.
Conclusions
• Measurements of the 10C half-life were performed in order to resolve inconsistencies in the data set.
• Resolving the discrepancy in the 10C half-life is important because it is the lowest-Z superallowed decay and is most sensitive to contributions from scalar currents in the weak interaction.
• We measured the half-life via both ɣ-ray photopeak counting and direct β counting, obtaining consistent results.
• The β measurement (T1/2 = 19.3009(17) s) represents the most precise superallowed half-life measurement reported to date and the first to achieve a precision below 10-4.
• The updated Ft data, which includes the 10C half-life measurements presented here and recent measurements of the 14O Q-value and BR, are used to calculate the limits on scalar currents yielding bF = -0.0018(21) which remains fully consistent with the absence of weak scalar currents.
Conclusions
Acknowledgements
G . C . B A L LT. B A L L A S TP. B E N D E RA . B . G A R N S W O R T H YD . M I L L E RB . M I L L SJ . PA R KM . M . R A J A B A L IC . U N S W O R T HZ . WA N G
G . F. G R I N Y E R
V. B I L D S T E I NA . D I A Z VA R E L AR . D U N L O PP. E . G A R R E T TB . H A D I N I AD . J A M I E S O NA . T. L A F F O L E YA . D . M A C L E A NA . J . R A D I C HE . R A N DC . E . S V E N S S O N
E . F. Z G A N G A R
C . A N D R E O I U
J . R . L E S L I E
R . A . E . A U S T I NA . VA L E N C I A
Backup Slides
• Superallowed data gives us bF, which we can relate to a scalar coupling as CS/CV = -bF/2 = +0.0009(11) given Cs = Cʹs
• For general CS ≠ CʹS we have bF = -(Cs/CV + Cʹs/CV) and need to use β-ν angular correlation coefficient to provide an additional constraint
Limits on Weak Scalar Currents
0 0.1 0.2 0.3 0.4 0.5 0.6
γ <W-1
>
32.35
32.40
32.45
32.50
32.55
32.60
32.65
32.70
32.75
1/F
t (1
/s)
x 1
0-5
bF = -0.0018 ± 0.0021
Ft0= 3070.8 ± 1.8 s
10C
14O
22Mg
26Al
m
74Rb
62Ga
54Co
50Mn
46V
38Ca
42Sc
34Ar
34Cl
38K
m
χ2/ν = 0.44
Testing Fundamental Properties of the Weak Interaction
0
@d0
s0
b0
1
A =
0
@Vud Vus Vub
Vcd Vcs Vcb
Vtd Vts Vtb
1
A
0
@dsb
1
A
Vud = GV /GF
• Using GV as determined from the superallowed ft data can be used to calculate Vud and test CKM unitarity.
• The most demanding test of the CKM unitarity comes from the sum of the squares of the top-row elements
• The superallowed data provide the most precise experimental measure for Vud.
V 2ud + V 2
us + V 2ub = 0.99978(55)
0 50 100 150 200 250 300 350 400Time (1.2 s/bin)
0
0.01
0.02
0.03
0.04
Prob
abili
ty o
f Pile
-Up
DataBest FitCosmic pile-upPile-up of 2 or more events
Pile-Up Correction0.5 µs Shaping Time
Pileup Correction
0 50 100 150 200 250 300 350 400Time (1.2 s/bin)
0
0.005
0.01
0.015
0.02
0.025
Prob
abili
ty o
f Pile
-Up
DataBest-fit CurvePile-up of 2 or more eventsCosmic Pile-up
Pile-Up Correction2.0 µs Shaping Time
0 50 100 150 200 250 300 350 400Time (1.2 s/bin)
0
0.01
0.02
0.03
Prob
abili
ty o
f Pile
-Up
DataBest FitCosmic pile-upPile-up of 2 or more events
Pile-Up Correction1.0 µs Shaping Time
0 20 40Number of Leading Channels Removed
19.27
19.28
19.29
19.3
19.31
19.32
19.33
Hal
f-life
(s)
No pile-up correction appliedPile-up correction applied
Chop PlotSummed Data
Gas Counter Rates
4 5 6 7 8 9 10 11 12 13 14 15 16Initial Count Rate Upper Limit (kHz)
19.29
19.295
19.3
19.305
19.31
Hal
f-life
(s)
High Rate Data Removal
0 2 4 6 8 10 12 14t=0 10C Activity (kHz)
19.28
19.29
19.30
19.31
19.32
19.33
19.34
19.35
Hal
f-life
(s)
Systematics10C Activity at t=0
χ2/ν = 0.692-4 kHz
4-6 kHz 6-8 kHz 8-10 kHz10-12 kHz
12-14 kHz
14-16 kHz
• Searched for possible contribution within the 718 keV photopeak from possible contaminants in the A = 26 beam.
ɣ Photopeak Counting - Contaminants
0 500 1000 1500 2000Eγ (keV)
0.1
1
10
100
1000
10000
Coun
ts
γ−β Coincidence Spectrum Gated on the first 7.2 seconds of the implation
1800 1805 1810 1815 1820Eγ (keV)
0.1
1
10
Coun
ts
(1) 26Na (T1/2 = 1.07128(25) s) • Absence of the
characteristic 1809 keV ɣ-ray from the decay of 26Na ɣ-β coincidence spectrum during the initial beam on period.
• Searched for possible contribution within the 718 keV photopeak from possible contaminants in the A = 26 beam.
ɣ Photopeak Counting - Contaminants
(2) 26mAl (T1/2 = 6.34602(54) s)
• No ɣ radiation. • Dedicated β counting
measurement with the mass separator was tuned to the mass of 26mAl.
• No significant contribution from 26mAl measured.
0 100 200 300 400 500Time (s)
100
1000
Coun
ts
26mAl Run10 Cycles
χ2/ν = 0.96
I(10C) = 733(11) c/sI(26mAl) = 46(21) c/s I(78Br) = 77.5(29) c/sBG = 6.4(17) c/s
R (10C, in beam) = 138.2(21) c/sR(26mAl, in beam) = 6.4(29) c/s
Values from fit for 10 cycles
Per cycle, 23.9 implant, 2.3 s cool time
R(78Br, in beam) = 186(70) c/s
• Searched for possible contribution within the 718 keV photopeak from possible contaminants in the A = 26 beam.
ɣ Photopeak Counting - Contaminants
(3) 13N (T1/2 = 9.9670(37) min)
• Observed in β counting expt • No ɣ radiation. • GEANT4 simulation of the
Bremsstrahlung production and annihilation radiation
• Less than 2 counts expected within the 718 keV peak in the entire experiment.
• The upper limits obtained for each contaminant was included in the fit to the data. No change in the half-life observed. Energy (keV)
710 712 714 716 718 720 722 724
Cou
nts
10
210
310
410
510 C10
N13C: 0.03%10N/13
718 keV photopeak
Contaminants
Mass difference between 26mAl
and 78Br is 12.274 MeV
The 10C beam intensity is suppressed by a factor of ~700 when the beam is tuned to 26mAl relative to the optimized tune to 10C16O. No such suppression factor is assumed when including the 26mAl component deduced from the
26mAl run. We expect the true amount of 26mAl to be even smaller than included in the above numbers.