Charm Mixing CPV and Rare D 0 decays

Charm Mixing, CPV and Rare D 0 decays at Ba. Bar William S. Lockman Representing the Collaboration Introduction Lifetime difference measurement Mixing in Wrong sign D 0→K and D 0→K 0 decays Time-integrated CPV measurements Radiative D 0 decays Summary PANIC 2008: International Conference on Particles and Nuclei November 9 -14, 2008 Eilat, Israel

Introduction • Mixing and CPV in the D 0 system were discussed over 30 years ago A. Pais and Treiman, Phys. Rev. D 12, 2744 (1975) • But evidence for D 0 mixing only recently observed: BABAR: PRL 98 211802 (2007) D 0 K decay time analysis 3. 9 BELLE: PRL 98 211803 (2007) D 0 K K vs K lifetime difference analysis 3. 2 BELLE: PRL 99 131803 (2007) D 0 Ks time dependent amplitude analysis 2. 2 CDF: PRL 100, 121802 (2008) D 0 K decay time analysis 3. 8 BABAR: PRD 78, 011105 R (2008) D 0 BABAR: ar. Xiv: 0807, 4544 (2008) D 0 K K vs lifetime difference analysis 3 dependent amplitude analysis 3. 1 all mixing results combined by HFAG: • K 0 time K ~10 Of all the neutral mesons, the D system exhibits the least mixing • short distance DC=2 suppression: • Long distance amplitudes predominant System: x: y: K 0 (1956) 0. 95 0. 99 Bd (1987) 0. 78 ≈0 Bs (2006) 26 0. 15 KK K . . . D 0 (2007) 0. 0098 0. 0075 • D mixing loop involves d-type quarks b quark loop suppressed: • s and d quark loops: GIM suppressed • Mass difference ampl. < O(10 -5) but hard to quantify • Recent estimates: |x| ≤ 1%, |y| ≤ 1% A. Petrov, Int. J. Mod. Phys. A 21: 5686 (2006). consistent with current observations • New Physics signature: CPV E. Golowich, J. Hewett, S. Pakvasa, A. Petrov, Phys. Rev. D 76 095009 (2007) W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 2

Flavor state Mixing and CPV • Flavor eigenstates can mix through weak interaction: Schroedinger eqn governs time evolution (off diagonal M and elements determine mixing) • Physical eigenstates D 1 and D 2 ≠ flavor eigenstates – In the limit of CP conservation: D 1 = CP D 2 = CP • If weak interaction splits the masses or widths of physical eigenstates, flavor state mixing will occur • Two parameters characterizing mixing: – where M 1, 2 and G 1, 2 are the masses and widths of physical states, resp. – The mixing rate is RM = (x 2+y 2)/2 – CPV in mixing is characterized by the asymmetry where + (-) indicates an initial D 0 (D 0) W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 3

Recent Measurements • Mixing measurements: – D 0 K + K - , + – D 0 K + - New Result × D 0 K(*)-l+n – D 0 K+ - × 0 × D 0 K s + - Ba. Bar × D 0 K s K+ K • Not today × Quantum Correlations Search for time integrated CPV: BELLE CLEO-c – D 0 K + K - , + – D 0 + - 0, K+K- 0 CDF × D+ K + K - + W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 4

Lifetime difference measurements • Using D*+→ +D 0 decays, we measure the lifetimes of – CP-mixed D 0 K Cabibbo-favored (CF) decays and – CP-even D 0 K K and + - singly Cabibbo-suppressed SCS decays • This allows a estimation of – where by and + (-) indicates an initial D 0 (D 0) • We also measure the CP asymmetry: • Relation to mixing parameters: – where AM is the mixing rate asymmetry and f characterizes CPV in the interference between mixing and decay • In the limit of CP conservation, y. CP = y and Y=0 W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 5

Decay time fits to determine (y. CP, Y) =409. 3± 0. 7 fs =401. 3± 2. 5 fs =404. 5± 2. 5 fs =407. 6± 3. 7 fs =407. 3± 3. 8 fs K and KK lifetimes differ! W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 6

Lifetime Difference Results 3. 2 evidence - no CPV PRL 98 211803 (2007) 540 fb-1 3. 0 evidence - no CPV PRD 78 011105(R) (2008) 384 fb-1 HFAG World Average: y. CP = (1. 072 ± 0. 257 )% Combining 384 /fb tagged and 91 /fb untagged (Ba. Bar): y. CP = (1. 03 ± 0. 33(stat. ) ± 0. 19(syst. ))% ar. Xiv 0808: 1297 (2008) W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 7

Mixing in “Wrong Sign” Decays (D 0→K ) Two types of WS Decays: – Doubly Cabbibo-supressed (DCS) – Mixing followed by Cabibbo-Favored (CF) decay D 0 0 D D 0 f f D 0 f Two ways to reach same final state interference! Discriminate between DCS and Mixing decays by their proper time evolution (assuming CP-conservation and |x| « 1, |y| « 1) : DCS decay K : W. Lockman Mixing Interference between DCS and mixing strong phase difference between CF and DCS decay amplitudes PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 8

Observations of Mixing in D 0→K Evidence for mixing from Ba. Bar (3. 9 s) and confirmation by CDF (3. 8 s) 384 fb-1 PRL 98, 211802 (2007) 3. 9 1. 5 fb-1 PRL 100, 121802 (2008) 3. 8 Two completely different experiments (Ba. Bar and CDF) yield nearly identical results: W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 9

Mixing in WS D 0 K+ - 0 Decays 384 fb-1 – new result: ar. Xiv: 0807, 4544 [hep-ex], submitted to PRL • Analysis formally similar to to wrong sign D 0 K+ - analysis but now mixing depends on position in Dalitz plot. D 0 0 D D 0 f • Final state can be reached in two ways, D 0 f yielding sensitivity to mixing by through analysis D 0 f of the time dependent WS decay rate (|x|, |y|<<1): DCS Interference Mixing • The measured mixing parameters are: where = phase difference between DCS D 0→r. K+ and CF D 0→r. K+ reference amplitudes (and cannot be determined in this analysis) W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 10

Mixing in WS D 0 K+ - 0 Decays · · · Find CF amplitude from timeintegrated fit to RS Dalitz plot · isobar model expansion Use this in time-dependent fit to WS plot to determine and mixing parameters. Results: signal box: 0. 1449< m<0. 1459 Ge. V/c 2 1. 8495

Time integrated CPV • Two diagrams (tree and penguin) in SCS decays can lead to CPV • Measured asymmetry includes direct and indirect terms • SM predictions for ACP are tiny: O(0. 001% - 0. 01%) F. Bucella et al. , Phys. Rev. D 51, 3478 (1995) S. Bianco et al. , Riv. Nuovo Cim. 26 N 7, 1(2003) S. Bianco, F. L. Fabbri, D. Benson, and I. Bigi, Riv. , Nuovo Cim. 26 N 7, 1 (2003). A. A. Petrov, Phys. Rev. D 69, 111901 (2004) Y. Grossman, A. L. Kagan, and Y. Nir, Phys. Rev. D 75, 036008 (2007) observation of ACP at ~0. 1% level would indicate NP • Whereas previous measurements of ACP had uncertainties of ~(1 -10)%, recent improvements in controlling experimental systematics have led to reduced errors ~(0. 2 -0. 4)% on ACP W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 12

Experimental Procedure • Measure the time integrated CP asymmetries • Relative s and s tracking efficiencies not equal – Use D 0→K- + tagged and untagged data to determine this • Due to Z/g interference and radiative corrections D 0 and D 0 are produced with a forward backward asymmetry in C. M. polar angle q* – compute the D 0 -D 0 flavor asymmetry vs cos in the center of mass W. Lockman – extract Acp and Afb by constructing even and odd functions of cos PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 13

Time integrated CPV in D 0 K K , Ba. Bar data sample 384 fb-1 PRL 100 061803 (2008) • No evidence for CP violation in either mode • 2 -3 x improvement on 2006 world average errors: W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 14

Search for CPV in D 0 K K 0, 0 phase space integrated asymmetries: Phys. Rev. D 78 051102 (2008) No evidence of CP violation in either decay mode. No significant difference between modes Used technique described earlier to correct for tracking asymmetries W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 15

New HFAG Average for ICHEP 08 http: //www. slac. stanford. edu/xorg/hfag/charm/index. html ar. Xiv: 0808. 1297 No-mixing point excluded at 9. 8σ W. Lockman PANIC 08 Conference, Eilat, Israel No-CPV point still allowed at 1σ Nov. 9 -14, 2008 16

Radiative D 0→fg and K*g Decays Phys. Rev. D 78, 071101 (2008) D 0→fg Cabibbo suppressed, D 0→K*0 g Cabibbo favored radiative D 0 decays dominated by long range processes Vector Meson Dominance Mode: Theoretical BF × 10 -5): D 0→fg 0. 1 -3. 4 D 0→K*0 g 7 -80 pole diagrams: o = weak transition P = pseudoscalar meson Results: Using world average B(D 0→K )=(3. 89± 0. 05)%: new W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 17

Summary • After 30 years of searching for it, the collective evidence for D 0 mixing is becoming compelling – The no-mixing point is excluded at ~10 s, including systematic uncertainties – However, no single measurement exceeds 5 s • Ba. Bar will be adding more measurements soon • Average values of the mixing parameters are x~1 %, y~0. 8 % – compatible with the upper range of standard model predictions • No evidence for CPV at the current experimental sensitivity (~0. 25 %) – systematic uncertainties are likely to diminish as more B-factory data is analyzed • Theoretical predictions in accord with measured branching fractions W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 18

Extra W. Lockman PANIC 08 Conference, Eilat, Israel Nov. 9 -14, 2008 19

Ba. Bar Generic Mixing Analysis Identify the D 0 flavor at production using the decays – select events around the expected D 0 decay vertex – The charge of the soft pion determines the flavor of the D 0 Identify the D 0 flavor at decay using the charge of the Kaon Beam spot: x ~ 100 m, y ~ 6 m right-sign (RS) wrong-sign (WS) Vertexing with beam spot constraint determines decay time, and decay time error, W. Lockman PANIC 08 Conference, Eilat, Israel D 0 production vertex Nov. 9 -14, 2008 20