Recent CLEO Results on Tau Hadronic Decays J.E. Duboscq Cornell - - PowerPoint PPT Presentation

J.E. Duboscq Cornell University Tau04, Nara Japan

Recent CLEO Results on Tau Hadronic Decays

2

JED Tau04

The CLEO3 Detector Tau Decays to 3 Charged Hadrons + ν Structure of KKπ and Wess-Zumino

CLEO Hadronic Tau Results

PRL90:181802,2003 PRL92:232001,2004

3

JED Tau04

The CLEO3 Detector

Solenoid Coil Barrel Calorimeter RICH Drift Chamber Silicon / beampipe Endcap Calorimeter Iron Polepiece Barrel Muon Chambers Magnet Iron Rare Earth Quadrupole SC Quadrupoles SC Quadrupole Pylon

2230104-001

CL EO III

CLEOIII

4

JED Tau04

τ-→ h-h+h-ν decays predominantly to pions K-π+π- final state important to strange spectral function, ms, Vus K-K+π- state probes Wess-Zumino term K-K+K- state as yet unobserved

Tau to 3h± + ν

The Data Sample: 3x106 tau pairs at Υ(4s) produced at CESR

τ-→ h-h+h-ν

5

JED Tau04

Combine RICH and dE/dx Use DATA D*→Dπ, D→ Kπ to obtain PID ε and fake rates Cross check with wrong sign K in τ-→K+π-π+ν search (Use only loose dE/dx for π in KKπ - KKK not a background!)

Hadronic Particle ID

100 80 5 10 15 20 25 20 40 60 0.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 Efficiency (%) Fake Rate (%) Momentum (GeV/c) K K fakes fakes K

3350103-001

τ-→ h-h+h-ν

6

JED Tau04

Select 1 vs 3 tracks (using Thrust) Require e/µ/ρ/π tag Reject events w/ extra showers (3hπ0 rejection) Missing momentum, Evis cuts reject 2γ background Ks

0 rejection for Kππ mode

Use KORALB, JETSET, GEANT for efficiency (use data for PID)

Event Selection

τ-→ h-h+h-ν

7

JED Tau04

2000 1000 125 0.5 1.8 1.6 1.4 1.7 0.9 1.3

(3h) Mass (GeV/c2)

I

Events / 20 (MeV/c2) ( a ) ( c ) ( b ) ( d )

3350103-002 100

50 2 Data all MC Background Continuum Background

Largest τ bgd from other τ→3h(π0)ν modes Use MC to get feed-across For KKK use data to get feed- across KKπ Substructure tuned to fit data

3h Results

Mode Data τ bgd qq bgd ε(%) πππ 43543 3207±57 152±12 10.27±0.08 Kππ 3454 1475±38 57±8 11.63±0.12 KKπ 932 86±9 19±4 12.48±0.11 KKK 12 4±2 0.4±0.6 9.43±0.10

πππ Kππ KKK KKπ

τ-→ h-h+h-ν

8

JED Tau04

Very Good Data MC agreement Used 3π, Kππ tuning from TAU02 Tuned KKπ substructure: Less K*, more ρ’, no ρ’’

3h Substructure Plots

πππ KKπ Kππ

6000 2000 200 0.3 1.4 1.0 0.6 0.7 1.1 h+h Mass (GeV/c2)

I

Events / 20 (MeV/c2) Data all MC Background Continuum Background

3350103-003

250 125 4000 100 250 50 ( a ) ( b ) ( c ) ( d ) ( e )

KK ππ ππ Kπ Kπ

τ-→ h-h+h-ν

9

JED Tau04

3% PID systematic 2% each systematic for Lumi, σ(ττ), track finding 1% each syst for τ backgrounds, CC cuts

3h Systematics

PID Fake rate syst 0.1%/9%/2%/12% MC/Data studies, τ-→K+π-π+ν search qq background - MC vs data above tau mass syst = 0.2%/ 2%/1%/3% KKπ substructure 2%

τ-→ h-h+h-ν

10

JED Tau04

Final 3h Results

B(τ-→̟-̟+̟-ντ) = 9.13±0.05±0.46% B(τ-→K-̟+̟-ντ) = 0.384±0.014±0.038% B(τ-→K-K+̟-ντ) = 0.155±0.006±0.009% B(τ-→K-K+K-ντ) < 3.7x10-5 @90% CL First direct 3π result Best precision on KKπ Kππ consistent w/OPAL(0.360±0.082±0.048%) & CLEO2(0.346±0.023±0.056%), higher than ALEPH(0.214±0.037±0.029%) Most stringent limit on KKK

τ-→ h-h+h-ν

11

JED Tau04

Simplest τ decay picture: Vector (axial) current produces even (odd) numbers of pseudoscalars WZ Anomaly allows parity flip and allows a violation of this rule Golden mode τ→η̟̟0ν previously observed by CLEO (no axial component) τ→KKπν has both axial and vector (WZ) contribution WZ effects rate and substructure of KKπ

KKπ Structure - Wess Zumino Anomaly

KKπ - WZ Anomaly

12

JED Tau04

SM matrix element M∝LµJµ Define: Qµ=(q1+q2+q3)µ, si=(qj+qk) J is a sum over 4 form factors:

Structure of tau to 3hν Decays

J=∑ fi(q1,q2,q3,Q)Fi(s1,s2,Q) F1,F2 are axial terms f3=iεαβγq1αq2βq3γ F3 is the WZ vector term F4 is the scalar current (negligible) f i are kinematics - Fi are Form Factors (physics)

KKπ - WZ Anomaly

ν τ 3h

Kuhn, Mirkes, Z.PhysC56, 661(1992)

13

JED Tau04

Integrate over ν direction Two remaining Euler angles are kinematically determined dΓ(τ→KKπ)/dQ2ds1ds2 ∝ WA(F1,F2)+WB(F3) No interference between Axial and WZ term Measurement possible entirely by using Dalitz plot and Q2

Structure of tau to 3hν Decays

KKπ - WZ Anomaly

14

JED Tau04

The Physics We Fit

F1 ∝ BWa1(Q2) x ( BWρ(s2)+βρBWρ’(s2) ) a1→ρ(’)π , ρ(’)→KK F2 ∝ RF BWa1(Q2) x BWK*(s1) a1→K*K, K*→Kπ F3 ∝ RB

½ (BWρ(Q2)+λBWρ’(Q2)+δBWρ’’(Q2)) x

(BWω(s2)+αBWK*(s1))

ρ(’,’’)→K*K, K*→K̟ ρ(’,’’)→ω̟, ω →KK Five real fit parameters to KKπ, Kπ, KK masses

KKπ - WZ Anomaly Decker etal, ZPhysC.58,445(1993) Finkemeir & Mirkes, ZPhysC69, 243(1996)

Text

15

JED Tau04

Use 7 .09x106 τ pairs from CLEO3 Use same cuts as τ→3hν analysis 2255 signal events, 256±16±46 background Obtain consistent overall Branching Fraction Use unbinned extended Maximum Likelihood fit including background term PDF = PDF(KKπ) x PDF(KK) x PDF(Kπ) Use best known params for BW’ s

The Data and Fit Procedure

KKπ - WZ Anomaly

16

JED Tau04

Shown is total fit and contributions from Axial and WZ components ≈1/2 is from WZ

Fit Results

1.2 1.4 1.6 1.8 80 160 240 Events/20 MeV (a) 0.6 0.8 1.0 1.2 200 400 600 (b) 0.9 1.1 1.3 1.5 100 200 (c) K+ mass (GeV/c2)

I

K K+ mass (GeV/c2)

I I

K K+ mass (GeV/c2)

I

3351203-007

Data Fit Wess-Zumino Backgrounds Axial Vector

KKπ - WZ Anomaly α=0.471±0.060±0.034 δ=0.101±0.020±0.156 λ=-0.314±0.073±0.080 RB=3.23±0.26±1.90 RF=0.98±0.15±0.36

ΓWZ ΓTot =55.7±8.4±4.9%

17

JED Tau04

Relative rates in Kuhn & Mirkes model Axial current: τ→ a1 (→ρ(’)̟ , K*K) ν Vector current (WZ): τ→ ρ(’,’’) (→K*K , ω̟) ν

Substructure Result

RWZ =3.4±0.9±1.0%

ω̟

RAxial =2.50.8±0.4%

ρ(’)̟

RWZ =60.8±8.5±6.0%

K*K

RAxial =46.8±8.4±5.2%

K*K

Decay dominated by K*K, 50/50 WZ and Axial B(a1 to K*K)=2.2±0.5% consistent w/ previous CLEO ππ0π0 result Axial component much smaller than ALEPH CVC estimate from DM1, DM2 data

KKπ - WZ Anomaly

94

+6

• 8%

CERN EP99-026

18

JED Tau04

Angular Distributions

1 1 cos 100 200 Events (a) 1 1 100 200 (b) 1 1 100 200 (c)

I I

cos

I

cos Data Vector + Axial Vector Axial Vector Only All Backgrounds

3351203-008

β: ∠ P(KKpi) in lab frame, pKxpπ θ: ∠ P(τ) in lab, P(KKπ) in τ frame ψ: ∠ P(τ), P(lab) in KKπ frame Angles are all expressible in terms of observables Angles alone are not enough to extract WZ/Axial contributions

KKπ - WZ Anomaly

19

JED Tau04

Summary

3 First direct B(τ→3πν) result ✓ Best precision on B(τ→KKπν) ✓ B(τ→Kππν) consistent w/OPAL and CLEO, higher than ALEPH ✓ Most stringent limit on τ→KKKν ✓ First Study of WZ and Axial parts of τ→ KKπν ✓ Breakdown of KKπ in Kuhn+Mirkes model Using CLEO3, we have presented: