Dinko Po cani c, for the Nab Collaborationgalileo.phys.virginia.edu/.../nab_fnpb-prac_2010-12-15.pdf2010/12/15 · Nab experiment: progress update Dinko Po cani c, for the Nab Collaboration

Nab experiment: progress update

Dinko Počanić, for the Nab Collaboration

University of Virginia

FnPB PRAC Review15 December 2010

D. Počanić (UVa) Nab progress update 15 Dec ’10 1 / 27

n-decay program at FnPB

The FnPB neutron decay program at SNS

I Nab: a precise measurement of

◦ a, the electron-neutrino correlation in neutron decay, and◦ b, the Fierz interference term, never before measured in n decay.

I Polarized program (abBA/PANDA): precise measurements of

◦ A, the electron asymmetry in neutron decay,◦ B, the neutrino asymmetry in neutron decay,◦ C , the proton asymmetry in neutron decay; also◦ independent measurements of a and b.

Typical goal uncertainties: δv/v ≤ 10−3, and δb ≤ 3× 10−3.


Motivation, goals

Neutron Decay Parameters (SM)

dw

dEedΩedΩν' keEe(E0 − Ee)2

×[

1 + a~ke ·~kνEeEν

+ bm

Ee+ 〈~σn〉 ·

(A~ke

Ee+ B

~kν

Eν

)+ . . .

]where:

a =1− |λ|2

1 + 3|λ|2A = −2

|λ|2 + Re(λ)1 + 3|λ|2

B = 2|λ|2 − Re(λ)

1 + 3|λ|2λ =

GA

GV(with τn ⇒ CKM Vud)

also:C = κ(A + B) where κ ' 0.275 .


Motivation, goals

Goals of the Nab experiment

I Measure the electron-neutrino parameter a in neutron decay

with accuracy of∆a

a' 10−3

current results:−0.1054± 0.0055 Byrne et al ’02−0.1017± 0.0051 Stratowa et al ’78−0.091± 0.039 Grigorev et al ’68

I Measure the Fierz interference term b in neutron decay

with accuracy of ∆b ' 3× 10−3

current results: none


CKM matrix: Vud

Status of A and λ in n decay

-0.13 -0.12

Beta Asymmetry A

-0.11 -0.10

Abele-09

Pattie-09

Abele-02

Liaud-97

Yeroz-97

Bopp-86

Average:= -0.1186(9)A

Uncertainty of the averagescaled up by factor 2.3×

Global fit χ2/dof = 28/5 !

Statistical probability for thisχ2 is 5× 10−5.H. Abele, private communication (2009).R.W. Pattie, et al., PRL 102, 012301 (2009).H. Abele et al., PRL 88, 211801 (2002).P. Liaud et al., NP A 612, 53 (1997).B. Yerozolimsky et al., PL B 412, 240 (1997).P. Bopp et al., PRL 56, 919 (1986).


CKM matrix: Vud

Status of A and λ in n decay (cont’d)

Nab goal value of ∆a:

⇒ ∆λ ' 3.5× 10−4

i.e., an order of magn.improvement.

∆λ

λ' 0.27

∆a

a' 0.24

∆A

A


Beyond Vud

n-decay Correlation Parameters Beyond Vud

I Beta decay parameters constrain L-R symmetric, SUSY extensions tothe SM. [Reviews: Herczeg, Prog. Part. Nucl. Phys. 46, 413 (2001),N. Severijns, M. Beck, O. Naviliat-Čunčić, Rev. Mod. Phys. 78, 991 (2006),

Ramsey-Musolf, Su, Phys. Rep. 456, 1 (2008)]

I Fierz int. term, never measured for the n, along with B, offers asensitive test of non-(V − A) terms in the weak Lagrangian (S ,T ).[S. Profumo, M. J. Ramsey-Musolf, S. Tulin, PRD 75, 075017 (2007)]

I Measurement of the electron-energy dependence of a and A canseparately confirm CVC and absence of SCC.

[Gardner, Zhang, PRL 86, 5666 (2001), Gardner, hep-ph/0312124]

I A connection exists between non-SM (e.g., S ,T ) terms in d → ueν̄and limits on ν masses. [Ito + Prézaeu, PRL 94 (2005)]


Beyond Vud non-V − A interaction terms

Updated limits for RH S and T currents n decay

-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

RS/L

V

RT/L

A

neutron and nuclear decays(survey, 95% C.L.)

“present limits”(68% C.L.)

Δχ2

C.L.

2.30 68.3%

90%

95.4%

4.61

6.17neutrino

mass(68% C.L.)

neutrino mass(68% C.L.)

muon decay“90% C.L.”

Present limits (n decay data)(SM values at origin of plot.)

S V

-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15-0.15

-0.10

-0.,05

0.00

0.,05

0.10

0.15

R /L

RT/L

A

neutron and nuclear decays(survey, 95% C.L.)

Δχ2

C.L.

2.30 68.3%

90%

95.4%

4.61

6.17



“future limits”(68% C.L.)


Projected limits; Grey contours:

β compilation [Sev-06]

Improvement from more precise a = −0.1030(1); using b ≡ 0.[G. Konrad, W. Heil, S. Baeßler, D. Počanić, F. Glück, arXiv 1007.3027.]



Limits for LH S and T currents n decay

LS/L

V

-0.3 -0.2 -0.1 0.0 0.1 0.2 0.3-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

LT/L

A Δχ2

C.L.

2.30 68.3%

90%

95.4%

4.61

6.17

neutron andnuclear decays(survey, 68% C.L.)

superallowed

0 →0 decays(68% C.L.)

+ +

“present limits”(68% C.L.)


nuclear decays

( ( In), 90% C.L.)P107

Present limits (n decay data)(SM values at origin of plot.)

-0.04 -0.02 0.00 0.02 0.04

-0.04

-0.02

0.00

0.02

0.04

LS/L

VL

T/L

A

Δχ2

C.L.

2.30 68.3%

90%

95.4%

4.61

6.17

“future limits”(68% C.L.)

superallowed

0 →0 decays(68% C.L.)

+ +

neutron andnuclear decays(survey, 68% C.L.)

nuclear decays

( ( In), 90% C.L.)P107

0.04

Projected limits assuming

a = −0.1030(1) ; b = 0 ± 0.003

[G. Konrad, W. Heil, S. Baeßler, D. Počanić, F. Glück, arXiv 1007.3027.]



Right-handed W bosonsAdding RH gives non-zero δ = m21/m

22, ζ:

W1 = WL cos ζ + WR sin ζ , and W2 = −WL sin ζ + WR cos ζ .

Mixing angle ζ

-0.2 -0.1 0.0 0.10.00

0.02

0.04

0.06

0.08

0.10

Mass

rati

oδ 250Δχ

2C.L.

2.30 68.3%

90%

95.4%

4.61

6.17300

350

400

450500550600

Mas

s m

[G

eV]

2

leptonscattering(90% C.L.)

μ decays(68% C.L.)

μ decays(90% C.L.)

DØ (95% C.L.)

0 →0decays

(68% C.L.)

+ +

“presentlimits”

(68% C.L.)

Present limits

Δχ2

C.L.

2.30 68.3%

90%

95.4%

4.61

6.17

Mixing angle ζM

ass

rati

oδ

400

500

600

900

700800

1000

Mass

m [G

eV

]2

-0.10 -0.05 0.00 0.050.00

0.01

0.02

0.03

0.04

0.05Δχ

2C.L.

2.30 68.3%

90%

95.4%

4.61

6.17

0 →0decays

(68% C.L.)

+ +

μ decays(90% C.L.)

DØ (95% C.L.)

μ decays(68% C.L.)

leptonscattering(90% C.L.)

“future limit”(68% C.L.)

Projected limits

[G. Konrad, W. Heil, S. Baeßler, D. Počanić, F. Glück, arXiv 1007.3027.]


Beyond Vud Fierz interference term

The Fierz interference term bb can be estimated from nuclear beta decays:

bF =CSCV

|CS |2 + |CV |2bGT =

CTCA|CT |2 + |CA|2

These terms vanish for pure ν(R) coupling.b 6= 0 only for S , T coupling to ν(L). (leptoquarks?)From 0+ → 0+ decays [Towner + Hardy ’98]:

|bF | '|CS ||CV |

≤ 0.0077 (90 % c.l.)

From analysis of GT decays [Deutsch + Quin, ’95]:

bGT = −0.0056(51) 'CT|CA|

(small FT from πe2γ!?)

⇒ a ∼ 10−3 measurement of bn is very interesting!D. Počanić (UVa) Nab progress update 15 Dec ’10 11 / 27

Beyond Vud Second class currents

Correlation Parameters with Recoil Correction[Gardner, Zhang, PRL 86, 5666 (2001), Gardner, hep-ph/0312124]

Most general form of hardonic weak current consistent with (V-A):

〈p(pp)|Jµ|n(pn,P)〉 =

ūp(pp)

(f1(q

2)γµ − if2(q2)

Mnqµ +

f3(q2)

Mnqµ + g1(q

2)γµγ5

− ig2(q2)

Mnσµνγ5qν +

g3(q2)

Mnγ5q

µ

)un(pn,P)

a,A,B⇒ λ =g1

f1while τn ∝ (f1)2 + 3(g1)2

However, f2 (weak magnetism) and SCC’s (g2, g3), remain unresolved inbeta decays (best tested in A=12 system). With recoil corrections,Gardner and Zhang find:

a(Ee) = func(f2) while A(Ee) = func(f2, g2)


Beyond Vud Second class currents

0.4 0.5 0.6 0.7 0.8 0.9

x

-0.008

-0.006

-0.004

-0.002

0

0.002

0.004a(

1) a

nd A

(1)


Comparison w/other experiments

Current experiments aiming to measure a

1. Nab: goal is to measure ∆a/a ∼ 10−3I Best statistical sensitivity,I Challenging but manageable systematics, esp. in asymm. design.

2. abBA: goal is to measure ∆a/a ∼ 10−3I Similar to Nab, but with spectrometer configured for A,B/C,I Detection function is very broad, syst. uncert. for a very demanding.

3. aCORN: goal is to measure ∆a/a ∼ 0.5− 2 %I Funded, under construction,I Uses only part of neutron decays.

4. aSPECT: aims to measure ∆a/a ∼ 10−3I Funded and running; recently overcame trapping problems,I Stat. sensitivity not as good as Nab due to integration; presently∼ 2 %/day—will likely improve on publ. results, not < 1 % this yr,

I Easier determination of detection function than in Nab at the presentlevel of accuracy. Singles measurement!


Nab measurement principles

Nab Measurement principles: Proton phase space

Ee (MeV)

p p2

(M

eV2 /

c2)

cos θeν = -1

cos θeν = 1

cos θeν = 0

proton phase spaceprobability (arb. units)

Ee =

75 keV

236 keV

450 keV

700 keV

0

0.5

1

1.5

0 0.2 0.4 0.6 0.8

NB: For a given Ee , cos θeν is a function of p2p only. Slope = a


Nab measurement principles

Nab principle of operation

I Collect and detectboth electron andproton from neutronbeta decay (magneticfield, detectors at bothends)

I Measure electronenergy and protonTOF and reconstructdecay kinematics(Magnetic field shape,silicon detectors atboth ends).

SegmentedSi detector

SegmentedSi detector

TOF region(field ∙ )r BB 0

Uup (upper HV)

Udown (lower HV)

magnetic filterregion (field )B0

decay volume(field ∙ )r BB,DV 0

Neutronbeam


Nab apparatus and installation

Nab spectrometer: magnetic field profile

466.25

0.03

5.00

25.28 43.81

37.50

0.50

481.25

14.7525.90

67.09

0.47

41.66

14.7725.90

4.34

4.34

10.5220.5838.16

3.13

3.13

29.94

16.4130.24

3.19

3.28

4.9312.92

8.00

16.77

c1i

z

r

c6i

c4i

c5i

c3ic2i

c1o

c6o

c4o

c5o

c3oc2o

Mag

net

ic f

ield

[T]

B

z [m]

z [cm]

00

0

1

20

-1 2

-20

3

10

4

-10

5

-30

1

2

3

4

5

Bz(on axis)

Bz(on axis)

Bz(off axis)

Mag

net

ic f

ield

[T]

B

0

1

2

3

4

5

Decayvolume

Decayvolume

Si detector

Filter

4 m flight path is omitted here



Si detector prototypes (15 cm diameter)

LANL group has full-size prototypes from Micron Corp.Full thickness t = 2 mm; dead layer thickness td ≤ 100 nm.Detailed testing currently under way at LANL.



Spectrometerinstallation:



Nab anti-magnetic shield (AMS)

X

Y 3020

10010

5

30

50

20100

Antimagnetic solenoids

Inner solenoids

Top view: Side view:

Spectrometer magnet

Nab: Passive AntiMagnetic Screen (WBS 4)

Staging area for Nab would save FnPB beam time during AMS testing.



Nab:DAQ

Det

ecto

r &

pre

amps

PIX

IE-

16

R5400n

contr

oll

er

PNChp

30 kV

rear

I/O

module

FS725

Rb freqP

IXIE

-

16

PXI

crate

PX

I-P

CI

8336

…

x8

…

x16

…

x16

DAQ

workstation

Analysis

workstation

RAID

storageD

etec

tor

&

pre

amps

PIX

IE-

16

R5400n

contr

oll

er

PNChp

30 kV

rear

I/O

module

PIX

IE-

16

PXI

crate

PX

I-P

CI

8336

…

x8

…

x16

…

x16

optical

signal

Isolation

transformer

power

120 V

Isolation

transformer


Schedule, milestones

Milestone Completion0.a Start of project Jul 20110.b Detector prototype detects protons Sep. 20110. Magnet design ready for bidding Sep. 2011

1.a Order for magnet placed (design & option to build) Dec. 20111.b Acceptance of engineering drawings Dec. 20121.c Delivery of magnet Sep. 20131. Spectrometer magnet accepted Dec. 2013

2.a Passive Anti-Magnetic screen: magnetic design finished Sep. 20122. Passive Anti-Magnetic screen built Dec. 2013

3.a Detector test chamber available Mar. 2012[ . . . ]

3.g Electrode system ready Mar. 20143. Main detectors work in spectrometer Jun. 2014

4.a Shielding calculation for Nab accepted Jun. 2013[ . . . ]

4.d Shielding and utilities ready Jun. 20144. Spectrometer ready for data taking Sep. 2014

5.a Magnetometer calibrated Sep. 20125.b Magnetic field mapping system constructed Dec. 20135. Magnetic field of spectrometer mapped Mar. 2014

6. Data acquisition Sep. 2015

7. Data analysis Sep. 2016


Nab collaborators

R. Alarcon1, L.P. Alonzi2§, S. Baeßler2∗, S. Balascuta1§, J.D. Bowman3†,M.A. Bychkov2, J. Byrne4, J.R. Calarco5, V. Cianciolo3, C. Crawford6,

E. Frlež2, M.T. Gericke7, F. Glück8, G.L. Greene9, R.K. Grzywacz9,V. Gudkov10, F.W. Hersman5, A. Klein11, M. Lehman2§, J. Martin12,

S. McGovern2§, S.A. Page6, A. Palladino2§, S.I. Penttilä3‡, D. Počanić2†,R. Rodgers2§, K.P. Rykaczewski3, W.S. Wilburn11, A.R. Young13.

1Arizona State University 2University of Virginia3Oak Ridge National Lab 4University of Sussex5Univ. of New Hampshire 6University of Kentucky7University of Manitoba 8Uni. Karlsruhe/RMKI Budapest9University of Tennessee 10University of South Carolina11Los Alamos National Lab 12University of Winnipeg13North Carlolina State Univ.†Co-spokesmen ∗Experiment Manager‡On-site Manager §Graduate Students

Home page: http://nab.phys.virginia.edu/


http://nab.phys.virginia.edu

Additional slides Detector response

Additional slides



Electron energy response

histoE_NEntries 85243

Mean 1.263

RMS 0.658

Number of bounces

1 2 3 4 5 6 7 8

Yie

ld

10

210

310

410

510

histoE_NEntries 85243

Mean 1.263

RMS 0.658

Number of bounces

above threshold(10 keV upper det.)

(40 keV lower det.)

all bounces

histoEe300idealsum

Entries 85243Mean 298.9RMS 9.131

detected Ee [keV]0 100 200 300 400 500 600 700 800

Yie

ld

1

10

210

310

410

510

histoEe300idealsum

Entries 85243Mean 298.9RMS 9.131

Ee in both detectors

histoEe300escEntries 85243Mean 1.203RMS 12.1

non-detected Ee [keV]0 50 100 150 200 250 300

Yie

ld

1

10

210

310

410

510

histoEe300escEntries 85243Mean 1.203RMS 12.1

Ee loss in escaped particles

in dead layerescaped particle



Detector response:

Electron energy

Note that detectors are made so big that radial walk wouldn’t be visible. We have to study the

edge effect in more detail.

I am quite embarrassed to admit that nothing of what I found is new. E.g, in [Abe93] the power

of the backscattering suppression for electrons on silicon detectors, as well as the remaining tails,

are shown well.

Energy Green off Green ok Red off Red ok

150 keV 1807 78082 5762 74127

300 keV 2103 83140 3999 81244

450 keV 2025 59592 2809 58808

600 keV 1112 24865 1301 24676

750 keV 71 989 83 977

Tab. 1: Errors in electron energy

reconstruction for different

algorithms. “Off” is defined as

being farther away than 5 keV

from the nominal energy.

I summarize the results for other electron energies in the Table 1. The electron response spectra

look similar (see Fig. 4) and can be described analytically with a fit which assumes that some

electrons (a relative amount of about 1 %) are detected with an energy too high by 30 keV, and

some electrons (a relative amount of 7·10-5

·Ee·keV-1

, so 2.1% at 300 keV) are in an exponential

tail with a “growth constant” of 0.13·Ee). The analytical function describes the detector response

sufficiently good for a further study of the sensitivity of Nab to the non-perfect detector response

for electrons.

Fig. 4: Detector response functions for electrons with energies which are multiples of 150 keV. The “data points”

are from the simulation, and the lines show our analytical model.

A consequence for Nab is that the determination of a (and b) is affected by the knowledge of the

low energy tail. To study the effect on the a determination, I used the analytic model for Nab, in

which the magnetic field is simplified so that the proton TOF can be given as an analytical

function. The model includes the neutron beam width, but computes protons only on axis and

omits the electrostatic acceleration. Fig. 5 shows the 1/TOF2 response for protons with different

energies.

Proton TOF

10

100

1000

10000

100000

0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02

1/TOF^2 [us^-2]

Yie

ld

Ee = 75 keVEe = 75 keV, Ee responseEe = 225 keVEe = 225 keV, Ee responseEe = 375 keVEe = 375 keV, Ee responseEe = 525 keVEe = 525 keV, Ee responseEe = 675 keVEe = 675 keV, Ee response

Fig. 5: 1/TOF

2 spectra for protons. The solid lines show the 1/TOF

2 spectra for perfect electron detection. The

dashed lines show the modification due to the low energy tail in the electron energy response. The ideal shape (sharp edges, slope of top line proportional to a) is modified by the TOF

response function, and the neutron beam width. In particular, the TOF response leads to a tail at

the low energy (low 1/TOF2) side. The electron energy response leads to tails at the other side,

too. The high energy tail could serve to verify the detection function, this is not studied.

There is some structure in the tails which are probably due to the rather coarse binning for the

electron energies. The main result is that the fitted value for a is sensitive to the size of the tail,

and an uncertainty the in the relative amount of electrons in the tail of 2% (out of the couple of

percent of electrons which are in the tail) would lead to ∆a/a ~ 10-3

.

Fig. 6: Camel hump curve, showing

the time difference (defined as the

impact time on the upper minus the

impact time on the lower detector)

between hits in different detectors.

Furthermore, the event reconstruction goes wrong if our timing resolution is not good enough to

determine the earlier detector. This is only important for observables for which the detector

being hit first is significant. In Fig. 6, we show the time difference between different hits in the

detectors. The black line shows cases in which the lower detector is hit first and the upper



histoETOF300du

Entries 5134Mean 33

RMS 6.017

TOF [ns]-200 -150 -100 -50 0 50 100 150 200

Yie

ld

1

10

210

310

histoETOF300du

Entries 5134Mean 33

RMS 6.017

Camel hump curve

lower det hit first upper det hit first

TOF = time of upper det. hit − time of lower det. hit


n-decay program at FnPBOverall motivation and goalsCKM matrix: VudBeyond Vudnon-V-A interaction termsFierz interference termSecond class currents

Comparison w/other experimentsNab measurement principlesNab apparatus and installationSchedule, milestonesNab collaboratorsAdditional slidesDetector response

Documents

Dinko Po cani c, for the Nab Collaborationgalileo.phys.virginia.edu/.../nab_fnpb-prac_2010-12-15.pdf2010/12/15 · Nab experiment: progress update Dinko Po cani c, for the Nab Collaboration