Mikihiko Nakao (KEK) December 13th, 2006, CKM workshop, Nagoya - PowerPoint PPT Presentation

Mikihiko Nakao (KEK) December 13th, 2006, CKM workshop, Nagoya mikihiko.nakao@kek.jp WG2 / 3 / 6-13-AM2 11:30 b → s γ measurements from B factories — M. Nakao 11:48 b → s γ branching fraction NNLO — P. Gambino 12:06 b → s γ moments, BF — T. Becher 12:24 b → s γ new results — M. Neubert 12:42 b → s ℓ + ℓ − , b → s + missing energy at B factories — M. Iwasaki

[Experimentalists’ picture of b → s γ ] γ γ t V * B tb b π V π ts X s W — Mikihiko Nakao — p.2 K b → s γ (quark level) B → X s γ (meson level) s Sensitivities to new physics Charged Higgs, SUSY, Left-right symmetric model. . . Photon as a probe for B decay properties universal parameters to improve V cb , V ub and b → s γ To measure | V ts | (without assuming CKM unitarity) ∼ 7% measurement error ⇔ NLO calculation ( ∼ 10% error) — review of measurements (this talk) — NNLO calculations are getting ready (next three talks. . . )

Integrated Luminosity (pb -1 ) (/day) Belle/KEKB 2006/12/05 07.24 on resonance, off resonance, energy scan has collected 700 fb − 1 1400 Integrated Luminosity (pb -1 ) (/day) 1200 1000 800 600 400 200 0 — Mikihiko Nakao — p.3 x 10 2 all data, on resonance, off resonance, energy scan 7000 Integrated Luminosity (pb -1 ) 6000 5000 4000 3000 2000 1000 0 5/3/1999 4/30/2001 4/28/2003 4/25/2005 4/23/2007 Belle log total : 690430 pb -1 Date BaBar/PEPII has collected 390 fb − 1

Integrated Luminosity (pb -1 ) (/day) Belle/KEKB 2006/12/05 07.24 on resonance, off resonance, energy scan has collected 700 fb − 1 1400 Integrated Luminosity (pb -1 ) (/day) 1200 b → s γ analysis up to 140 fb − 1 1000 800 600 400 200 0 — Mikihiko Nakao — p.3 x 10 2 all data, on resonance, off resonance, energy scan 7000 Integrated Luminosity (pb -1 ) 6000 5000 4000 3000 2000 1000 0 5/3/1999 4/30/2001 4/28/2003 4/25/2005 4/23/2007 Belle log total : 690430 pb -1 Date BaBar/PEPII has collected 390 fb − 1 b → s γ analysis up to 88 fb − 1

Full-inclusive BELLE Photon only, no B reconstruction Off-resonance subtraction (huge continuum background) B → X π 0 , π 0 → γγ subtraction — Mikihiko Nakao — p.4 Smeared by B momentum Semi-inclusive Standard M bc(ES) - ∆ E reconstruction Sum up as many modes (e.g., B → K πππγ ) 10 cm Cannot include all possible modes Trade-off on the minimum photon energy cut Large background ⇔ reduced model dependence as E γ cut is lowered

140 fb − 1 on- , 15 fb − 1 off-resonance Entries per 33 MeV 10 6 Scale factor calibration: ON − α OFF 10 5 E γ > 1 . 8 GeV Minimum E γ bin that has ∼ 1 σ signal 10 4 — Mikihiko Nakao — p.5 Continuum suppression 10 3 Not-so-fancy since no B reconstruction π 0 → γγ (similarly η → γγ ) 10 2 Reduction by π 0 -likelihood 10 B → π 0 X spectrum (continuum subtracted) B → π 0 X → γ Y subtraction 1 1.6 1.8 2 2.2 2.4 2.6 2.8 3 CM energy [ GeV ] Other backgrounds n , K 0 L , fake calorimeter cluster, beam b.g., . . . ¯ Relying on MC, may be problematic with more data In BaBar’s analysis, lepton-tag for other B to reduce background. Not obvious whether it improves figure-of-merit or not, depending on many other things.

1850801-007 25000 Data Events/100 MeV / 100 MeV) Spectator Model 0.8 B A B 20000 AR preliminary 40 Weights / 100 MeV -4 0.6 Partial Branching Fraction (10 15000 0.4 10000 — Mikihiko Nakao — p.6 5000 0.2 0 0 0 -5000 2 2.2 2.4 2.6 2.8 1.5 2 2.5 3 3.5 4 1.5 2.5 3.5 4.5 Reconstructed E* (GeV) E* γ [ GeV ] γ E (GeV) CLEO BaBar Belle 9 . 1 fb − 1 on Υ (4 S ) 81 . 5 fb − 1 on Υ (4 S ) 140 fb − 1 on Υ (4 S ) − 4 . 4 fb − 1 off-resonance − 9 . 6 fb − 1 off-resonance − 15 fb − 1 off-resonance E γ > 2 . 0 GeV E γ > 1 . 9 GeV E γ > 1 . 8 GeV [PRL87,251807(2001)] [hep-ex/0507001] [PRL93,061803(2004)] More data, lower photon energy cut Lower E γ cut by 0.1 GeV with roughly twice more data

Uncorrected results from photon counting in E γ bins give CLEO B ( B → X γ ) [2 . 0 , 2 . 7] GeV = (3 . 06 ± 0 . 41 ± 0 . 26) × 10 − 4 BaBar B ( B → X γ ) [1 . 9 , 2 . 7] GeV = (3 . 64 ± 0 . 29 ± 0 . 33 ± 0 . 22) × 10 − 4 — Mikihiko Nakao — p.7 Belle: B ( B → X γ ) [1 . 8 , 2 . 8] GeV = (3 . 51 ± 0 . 32 ± 0 . 29) × 10 − 4 In agreement each other, and with the SM predictions, after. . . Subtraction of b → d γ ( ∼ − 4% corr.) Corrections to the same E γ range (up to ∼ 10% corr.) Corrections to the E γ cut at B rest frame ( ∼ + 5% corr.)

82 fb − 1 on-resonance data 100 Missing Fractions (%) 80 X s system from one of 38 modes: 60 B → K πγ , K ππγ , K πππγ , 40 K ππππγ , K η ( π ( π )) γ , KKK ( π ) γ 20 — Mikihiko Nakao — p.8 ( K = K ± / K 0 S , π = π ± /π 0 ) 0 0.6 1.0 1.4 1.8 2.2 2.6 Divide events in 18 bins of M ( X s ) M (X s ) (GeV) 60 0.1/0.2 GeV step from 0.6 to 2.8 GeV M(X) = 1.4-1.5 GeV bin Events / 0.003 (GeV) 50 corresponds to E γ > 1 . 9 GeV 40 Select | ∆ E | < 0 . 10 — 0 . 07 GeV, 30 fit events in 5 . 22 < M ES < 5 . 29 GeV 20 10 0 5.22 5.24 5.26 5.28 5.3 M ES (GeV) Belle has also performed a simpler version of sum-of-exclusive analysis with less data ( ∼ 6 fb − 1 ) , lower M ( X s ) < 2 . 1 GeV and less modes

-3 × 10 0.2 Branching Fraction / 100 MeV Data B A B AR Kinetic scheme Shape Function scheme 0.15 M 2 B − M ( X s ) 2 0.1 E γ = 2 M B — Mikihiko Nakao — p.9 0.05 0 -0.05 1.9 2 2.1 2.2 2.3 2.4 2.5 2.6 E (GeV) γ Much better resolution (1–5 MeV) compared with E γ from calorimeter ( ∼ 40 MeV), no p ( B ) smearing BaBar B ( B → X s γ ) [1 . 9 , 2 . 6] GeV = (3 . 27 ± 0 . 18 + 0 . 55 + 0 . 04 − 0 . 09 ) × 10 − 4 − 0 . 40 Belle B ( B → X s γ ) [full range] = (3 . 36 ± 0 . 53 ± 0 . 42 + 0 . 50 − 0 . 54 ) × 10 − 4

1st moment (GeV) 1st moment: 2.4 � E γ � 2.3 — Mikihiko Nakao — p.10 Belle full inclusive ) 2 BaBar semi inclusive 2nd moment (GeV 2nd moment: BaBar full inclusive 0.04 CLEO full inclusive � ( � E γ � − E γ ) 2 � 0.02 (3rd moments are also measured by BaBar) 1.8 1.9 2 2.1 2.2 2.3 Minimum Photon Energy (GeV) Observables to be directly compared with predictions Universal parameters in operator product expansion (OPE) (several available schemes: kinetic scheme, shape function scheme. . . ) Kinetic scheme: m b ( b quark mass), µ 2 π (Fermi momentum) 2

Global fit (CLEO, Belle, BaBar data) to the moments from B → X s γ to the moments from B → X c ℓν Parameters are universal — Mikihiko Nakao — p.11 Fits to B → X c ℓν and B → X s γ are complementary Input to V ub from B → X u ℓν that recently reduced the | V ub | error significantly [Buchmüller-Flächer, hep-ph/0507023] Combined fit results — m b to less than 1% accuracy! π = 0 . 401 ± 0 . 019 (exp) ± 0 . 035 (OPE) GeV 2 m b = 4 . 590 ± 0 . 025 (exp) ± 0 . 030 (OPE) GeV, µ 2

OPE fit also improves the b → s γ measurement Shape function scheme Very small model Kinetic scheme dependence Kagan-Neubert scheme frac ( E γ > 1 . 8 GeV ) = 96 . 7 ± 0 . 6% — Mikihiko Nakao — p.12 frac ( E γ > 1 . 9 GeV ) = 93 . 6 ± 1 . 0% frac ( E γ > 2 . 0 GeV ) = 89 . 4 ± 1 . 6% Allow us to make an average at any E γ cut to compare with theory Belle, BaBar, CLEO cut ( 1 . 6 GeV is suitable now) However, this theoretical extrapolation is questionable? Lower E γ cuts are crucial to verify the predictions 1.6 GeV should be possible, can we go down to 1.0 GeV?

Average branching fraction for E γ > 1 . 6 GeV [Heavy Flavor Averaging Group (HFAG), hep-ex/0603003] B ( B → X s γ ; E γ > 1 . 6 GeV ) = (355 ± 24 (stat + sys) + 9 − 10(shape) ± 3 ( d γ ) ) × 10 − 6 NLO Buras Czarnecki Misiak Urban (NPB631:219,2002) — Mikihiko Nakao — p.13 CLEO -4 (3.29 0.53)x10 ± -1 [9.1 fb ] PRL87,251807(2001) BaBar +0.62 -4 (3.35 )x10 -1 [81.5 fb ] -0.51 PRD72,052004(2005) BaBar -4 (3.92 0.57)x10 ± -1 [81.5 fb ] hep-ex/0507001 Belle -4 (3.69 0.95)x10 ± -1 [5.8 fb ] PLB511,151(2001) Belle -4 (3.50 0.44)x10 ± -1 [140 fb ] PRL93,061803(2004) Average -4 (3.55 0.26)x10 ± HFAG hep-ex/0603003 2 3 4 5 6 -4 BF(B X )x10 → γ Very consistent with NLO SM, e.g., (357 ± 30) × 10 − 6

Average branching fraction for E γ > 1 . 6 GeV [Heavy Flavor Averaging Group (HFAG), hep-ex/0603003] B ( B → X s γ ; E γ > 1 . 6 GeV ) = (355 ± 24 (stat + sys) + 9 − 10(shape) ± 3 ( d γ ) ) × 10 − 6 NLO Hurth et al. (NPB704:56,2005) Asatrian et al. (PLB585:263,2004) Buras Czarnecki Misiak Urban (NPB631:219,2002) Kagan Neubert (EPJC7:5,1999) — Mikihiko Nakao — p.13 CLEO -4 (3.29 0.53)x10 ± -1 [9.1 fb ] PRL87,251807(2001) BaBar +0.62 -4 (3.35 )x10 -1 [81.5 fb ] -0.51 PRD72,052004(2005) BaBar -4 (3.92 0.57)x10 ± -1 [81.5 fb ] hep-ex/0507001 Belle -4 (3.69 0.95)x10 ± -1 [5.8 fb ] PLB511,151(2001) Belle -4 (3.50 0.44)x10 ± -1 [140 fb ] PRL93,061803(2004) Average -4 (3.55 0.26)x10 ± HFAG hep-ex/0603003 2 3 4 5 6 -4 BF(B X )x10 → γ Very consistent with NLO SM, e.g., (357 ± 30) × 10 − 6 Many NLO SM calculations — theory error?