TWIST – the TRIUMF Weak Interaction Symmetry Test

Designed to achieve ~ 0.01% in the shape of the decay spectrum

Several data sets of 109 events eachA precision test of the weak

interaction in the Standard Model

A precision study of the + decay spectrum

NL Rodning – University of Alberta Nathan.Rodning@ualberta.ca

Outline

MotivationOverview of the

experimentStatus and plans

Louis Michel at TRIUMF

December 1999

Students Professional Staff

TRIUMF Willy Andersson Curtis Ballard Yuri Davydov Jaap Doornbos Wayne Faszer Dave Gill Peter Gumplinger Richard Helmer Robert Henderson John Macdonald Glen Marshall Art Olin David Ottewell Robert Openshaw Jean-Michel Poutissou Renee Poutissou Grant Sheffer Dennis Wright

Presently at SLAC

Alberta Andrei Gaponenko Peter Green Peter Kitching Rob MacDonald Maher Quraan Nathan Rodning John Schaapman Jan Soukup Glen Stinson British Columbia Blair Jamieson Doug Maas Mike HasinoffNorthern British Columbia Elie Korkmaz Tracy PorcelliMontreal Pierre Depommier

Regina Ted Mathie Roman TacikSaskatchewan Bill ShinTexas A&M Carl Gagliardi John Hardy Jim Musser Robert Tribble Maxim VasilievValparaiso Don Koetke Robert Manweiler Paul Nord Shirvel StanislausKIAE (Russia) Arkadi Khruchinsky Vladimir Selivanov Vladimir Torokhov

TWIST - Personnel

Allows for possible

- scalar

- vector

- tensor

interactions of right-handed and left-handed leptons

… The most general interaction, which does not presuppose the W

2

,,

,,||~ jei

LRji

TVSij egrate

5

5

ar Pseudoscal

Vector Axial

Tensor

Vector

Scalar

TWIST Motivation – testing the Standard Model

55.0||

424.0||

125.0||

066.0||

SLL

SRL

SLR

SRR

g

g

g

g

110.0||

060.0||

033.0||

VRL

VLR

VRR

g

g

g

0||

122.0||

036.0||

0||

TLL

TRL

TLR

TRR

g

g

g

g

96.0|| VLLg

e±

±

Actual knowledge of couplings-

Plenty of room for a surprise

e+ angle (radians)

e+ Reduced Energy

Decay distribution

The general interaction can be expanded in terms of the Michel

parameters

)34(

3

21)cos(

13)34(

3

233~ 2 xxP

xx

xxxxrate o

The above decay distribution is modified by radiative corrections, required to second order

These are being calculated by Andrzej Czarnecki and Andrej Arbuzov, at the University of Alberta

T

RLSRL

TRL

SRL

TLR

SLR

TLR

SLR

TRL

TLR

VRL

VLR

gggggggg

gggg

****

2222

ReReReRe43

||||2||||43

43

The Michel Parameters - The parameter largely determines the shape of the positron energy spectrum

The effect of large deviations in on the shape of the energy spectrum. The effect shown is roughly 500 times the TWIST sensitivity

****** 66Re21 T

RLSRL

VLR

TLR

SLR

VRL

SLL

VRR

SRR

VLL gggggggggg

The Michel Parameters - The parameter makes a subtle correction to the shape of the positron energy spectrum

The effect of large deviations in on the shape of the energy spectrum. The effect shown is roughly 500 times the TWIST sensitivity

As well, the Fermi coupling constant has a significant dependence on

)(41

119215

32

mmmG

eF 00025.0

50

F

F

G

G Which is roughly 30x larger than the total quoted uncertainty in GF

****

2222222

Re21

||2||8||2||4||2||21

||21

1

TLR

SLR

TLR

SLR

TRL

SRL

TRL

SRL

TLR

TRL

VLR

VRL

VRR

SLR

SRR

gggggggg

ggggggg

The Michel Parameters -

The effect of large deviations in on the energy-integrated angular distribution. The effect shown is roughly 500 times the TWIST sensitivity

The parameter determines the asymmetry in the energy-integrated spectrum.

The Michel Parameters – The parameter determines the energy dependence of the asymmetry in the spectrum.

The effect of large deviations in on the energy-integrated angular distribution. The effect shown is roughly 500 times the TWIST sensitivity

****22

22222

Re||4||2

||||||2||||43

43

TLR

SLR

TLR

SLR

TRL

SRL

TRL

SRL

TLR

TRL

VLR

VRL

VRR

SLR

SRR

gggggggggg

ggggg

The muon lifetime

The Michel parameter

The Michel parameter

The outgoing positron polarization

The rate of absorption of e from muon decay

A test of (V-A)Because the coupling constants often enter as sums of positive definite terms, it is possible to test (V-A) with far less than the 19 experiments one might expect for the determination of the 19 free parameters.

It has been shown, for example, that a rigorous test of the (V-A) postulate can be made by measuring:

TWIST will contribute two of these five required measurements

For example, a measurement of and provides a model independent test of five coupling constants set to zero in the standard model

The muon decay rate can be written as

Where Q describes the decay of a left- or right-handed muon into a left- or right-handed positron

RLRL

,,

Q

22

222

222

22

||||4

1Q

||3||||4

1Q

||3||||4

1Q

||||4

1Q

VRR

SRRRR

TRL

VRL

SRLRL

TLR

VLR

SLRLR

VLL

SLLLL

gg

ggg

ggg

gg

Dependence of the decay on Chirality

Coupling to right-handed muons The decay rate of right-handed muons into either right- or left-handed electrons is given by the sum

This combination of couplings happens to be equal to a combination of Michel parameters, so that

A determination of and provides a model-independent test for the existence of right-handed weak couplings to muons. Q 0 indicates a violation of the Standard Model, and the existence of right-handed couplings for muons Q = 0 indicates that right-handed couplings to the muon do not exist

22222 ||||41

||3||||41 V

RRSRR

TLR

VLR

SLRLRRR ggggg QQQR

916

31

121

RQ

Region consistent with no right-handed weak couplings to the muon

Existing limits on right-handed currents in muon decay

916

31

121

RQ

Existing limits- Discovery region

TWIST preliminary

Anticipated TWIST sensitivity to right-handed currents in

muon decay

916

31

121

RQ

TWIST final

Consider Left/Right Symmetric extensions to the Standard Model

1L

R

W

W

m

m

)sin()cos(

)sin()cos(

21

21

MMeM

MMM

iW

W

R

L

Consider a model with two weak bosons with the mass eigenstates M1 and M2

Parity violation at low energy is presumably due to

In general, the models may include a CP violating phase (), and a left/right mixing parameter

Left/Right Symmetric extensions to the Standard Model

Expect a non zero gRR and gLR

If tensor and scaler couplings are excluded (as unnecessary) from these extensions, then-

2

RLV

RR mmg1 V

RLVLR gg

22 ||2||4

3 VLR

VLL gg

0

222 ||||2||4

3 VRR

VLR

VLL ggg

222 2|| VLR

VRR

VLL ggg

For Left/Right Symmetric extensions

is sensitive to the Left/Right mixing

to the mixing and to the WRmass

and are unchanged by Left/Right extensions with manifest symmetry

A measurement of and determines the WR mass and its mixing

04

3

23

4

221

214

3

4

24

2

R

L

R

L

m

m

m

m

2

RLV

RR mmg1 V

RLVLR ggFor

(left/right Mixing)

WR Mass (GeV)

Left/Right Mixing constraints – Existing limits

Mixing consistent with accepted value of

WR mass consistent with D0 direct search – without assuming manifest left/right symmetry

D0 (and CDF) limit assuming that VR

ud = VLud without CP

violation

D0 Sensitivity to assumptions for the right-handed CKM matrix 200 GeV

Strovink exclusion assuming cos(R)/cos(R) = 1 and no CP violation

(left/right Mixing)

WR Mass (GeV)

Left/Right Mixing constraints – Existing limits

Strovink exclusion assuming cos(R)/cos(R) = 0

WR Mass (GeV)

(left/right Mixing)

Left/Right Mixing constraints – Anticipated

TWIST Sensitivity

Anticipated TWIST result

Anticipated TWIST sensitivity due only to the P measurement

Manifest l/r symmetry

Discovery potential

Testing SUSY

R-parity violating SUSY leads to the following deviations in the parameters at tree level

2

16

3 2

2322311

~24 mGF

where

4

02

2

4

3

2

So that

Deviations in bigger than 1% would show up with deviations in and bigger than 0.01% - with unchanged. Speculative, but a rather specific prediction.

The Experiment

Highly polarized muons enter the spectrometer one at a time

Unbiased trigger on muon entering system

Data sets of 109 muon decay events are obtained in roughly two weeks

The experiment is systematics limited. The high data rate is essential for systematics studies

The large acceptance of the device makes it possible essentially to make measurements of the Michel parameters under differing conditions – therefore improving the reliability of the result.

Full system commissioning November/December 2001

The TWIST spectrometer at M13

Chambers & half detectorPlanar drift chambers sample

positron track

Use 44 drift planes, and 12 PC planes

TWIST – Half Stack

Typical decay event

TWIST - BeamlineThe role of the beamline is to deliver highly polarized “surface muons” to the spectrometer

Pion decay at rest produces polarized muons

Polarized Muon Source

µ++

Back to back to conserve linear momentum

The has zero angular momentum

=> no angular momentum in the final system

TWIST channel resolution allows for the selection of muons produced within 25 microns of the surface of the production target

Depolarization: 0.0002 per 25 microns of target material

Protons incident on graphite produce numerous pions, some of which stop and decay into

muons

TWIST – RF Cuts

Backgrounds (extrapolated from higher momentum)

Cloud Muonspolarization ~ -0.3flux ~ 1%

Flight time through beamline

+

e+

Flight time through the channel show the time that pions are stopped in the pion production target prior to their decay

RF period ~ 1.5 pion lifetimes makes the TRIUMF beam perfect for TWIST

Anticipated Michel Spectrum in TWIST

Distributions in cos() and reduced energy (x). The fiducial volume will be cut at roughly

x > 0.30.5 < |cos()| <

0.95

Drift Plane GeometryThe chambers are

built to high precisionWires are placed to

within about three microns of their nominal position

Average deviation in wire position much less than three microns

Specification on the average wire spacing is exceeded by a factor of 30. Average deviation in wire positions is 2

Scatter in the individual wire placements is ~ 20 microns, a factor of 7 better than specification.

Plane-to-plane Alignments

Tracking ResidualsResults to date; further improvements in alignments and T-zero are underway.

Drift chamber resolution meets spec

Drift Distance (cm)

Distribution of tracking residualsRMS tracking residuals as a function of drift distance in wire cells.

Efficiency – artificial random hit losses accurately identified as inefficiencies by

tracking

TDC hits were rejected at random at the event unpacking stage to mimic an inefficiency. The inefficiency is accurately identified by the track reconstruction code. Efficiency differences of ~ 0.1% are accurately identified. Plane efficiencies are ~99.8%.

Planar Drift Chambers

Energy lost along positron track in the TWIST spectrometer vs. 1/cos(). The curves are for successive sets of detector planes. The slope is proportional to the amount of material in the path of the positron prior to the reconstruction.

The intercept is related to our ability to calibrate the energy scale.

All material seen by the outgoing positrons is in a planar geometry Effects of interactions with the detector are proportional to 1/cos()

Energy loss, Multiple scattering, Hard scattering (kinks) including wire scatters

Calibration of the energy scale through End-Point Fits

The energy calibration is obtained from the data itself. The endpoint of the spectrum has a “sharp edge” at 52.83 MeV.

The edge is rounded by finite resolution and by radiative corrections. As well, the edge is reduced at forward angles (opposite the muon spin).

The forward and backward data can be calibrated to approximately 3.8 and 1.3 keV, respectively. The resultant contribution to the uncertainty in the extracted Michel parameters is typically on the order of 1 part in 10,000.

Opposite muon spin

Toward muon spin

Upcoming schedule107 muon decay events are in hand

“practice” analysis during January-March 2002

Field mapping: April 2002First physics beam: Summer 2002

Preliminary measurement of andBeamline and depolarization studies:

2002/2003Preliminary data on P

Final publications: 2005/2006

ConclusionsThe TWIST experiment is underway

Anticipate preliminary measurements at ~0.1% of:and (Data this summer)P (Data during the summer of 2003)

Final precision on and and P at ~ 0.02%

TWIST will explore significant new space where evidence may be found for physics beyond the standard model

For left/right symmetric models, TWIST has a mass reach which is comparable to - and which complements - that of the Tevatron

Blank slide

Wire Position Mapping - Reproducibility

Wire scans are reliable and reproducible

Blue – Initial wire position scanRed – Replacement of broken wireGreen – Rescan of wire positions

Drift Chamber Noise

The chambers are extremely quiet. Spurious hits in coincidence with a decay positron are roughly one hit in 50 events, and occur at random in the tracking volume so that they are readily ignored by the tracking software.

Cross Talk – No effect on

decay positron tracking

Width

TDC Time

Cross talk is at the 0.2% level.

Cuts on- short width- Adjacency to a

wider, “real” hitReduce tracking

“confusion” to a few extra hits per 105 events – these are eliminated as track outliers

Drift plane efficiency and Operating Point

All DC planes behave the same to a high degree of uniformity. Three planes did not function due to misconnected preamp output cables. In a physics run, these cables would have been reconnected before data taking (at the cost of one week), but having 94% of the planes operational at first startup was excellent for the commissioning run.

Muon stopping distributionWe were fortunate to see a 3% pressure change in 12 hours of running during the commissioning run.

We look at the ratio of counts in the PC’s upstream and downstream of the target. The sensitivity of the ratio determines our ability to measure the range of the muons and consequently their stopping distribution. Note that the data and Geant show the same sensitivity to atmospheric pressure.

106 events are suitable for the determination of the centroid of the stopping distribution to an accuracy equivalent to 66 g/cm2 equivalent to roughly a ¼ keV variation in the average positron energy loss.

A 3 centroid shift would correspond to approximately ¾ keV shift in the positron energy.

Wire Recovery

When a large muon pulse is found on a wire, does that wire recover for use in positron tracking?

Yes.

At right is a histogram of the time at which a positron hits a wire which was hit by a muon a few microseconds earlier. The fit gives an accurate value for the muon lifetime.

Note that hits within 1 s after the incident muon are not used, due to the drift time required for collection of charge from the muon hits.

We are investigating the much rarer occurrence of a positron passing through the same segment of a wire cell that a muon traveled through within several microseconds.

Muon lifetime: 2.20 0.05 sWire efficiency is independent of time since a muon hits a given wire