Flavor physics in the LHC era Zoltan Ligeti Lawrence Berkeley Lab
Flavor physics in the LHC era Zoltan Ligeti Lawrence Berkeley Lab • Introduction • Current status: sizable NP contributions allowed • Some key probes at LHCb and super-(KEK) B • High- p T flavor physics • Conclusions
Why is flavor physics interesting? • SM flavor problem: hierarchy of masses and mixing angles; why ν ’s are different • Empirical evidence that SM is incomplete: baryon asymmetry, dark matter, neutrino mass — at least two related to flavor • NP flavor problem: TeV scale (hierarchy problem) ≪ flavor & CPV scale ǫ K : ( s ¯ ∆ m B : ( b ¯ d ) 2 d ) 2 s ) 2 ∆ m Bs : ( b ¯ ∼ 10 4 TeV , ∼ 10 3 TeV , ∼ 10 2 TeV ⇒ Λ > ⇒ Λ > ⇒ Λ > Λ 2 Λ 2 Λ 2 – Many extensions of the SM have new sources of CP and flavor violation – The observed baryon asymmetry of the Universe requires CPV beyond the SM Not necessarily in flavor changing processes, nor necessarily in quark sector Flavor suppression destroys KM baryogenesis; flavor matters for leptogenesis • Flavor sector can be tested a lot better, many NP models have observable effects ZL — p.1
The name of the game in the LHC era • The question has been who sees NP first; once it’s seen, how to understand it? [Assume the LHC sees more than a Higgs ... ] • Concentrate on flavor physics topics where sensitivity can improve significantly (by an order of magnitude, or at least a factor of many) – Skip B → X s γ rate, near “hitting the theory wall” (best bound on many models) – ... some tension between sin 2 β and | V ub | [emphasized, e.g., by UTfit] – ... > 3 σ tension between LQCD f D s and D + s → ℓ + ν [Dobrescu & Kronfeld, arXiv:0803.0512] – Many measurements with complementary sensitivity will improve a lot – If all flavor effects < 1% in your favorite model (what is it?), I’ll have little to say • Lack of a “flavor theory” — there isn’t an obviously right / natural way for TeV-scale NP to duplicate GIM and CKM suppressions ZL — p.2
SUSY contributions to K 0 – K 0 mixing � 1 TeV � 2 � ∆ ˜ � 2 • (∆ m K ) SUSY m 2 � � (∆ m K ) exp ∼ 10 4 12 ( K d L ) 12 ( K d Re R ) 12 m 2 m ˜ ˜ K d L ( R ) : mixing in gluino couplings to left-(right-)handed down quarks and squarks Constraint from ǫ K : 10 4 Re � � ⇒ 10 6 Im � � ( K d L ) 12 ( K d ( K d L ) 12 ( K d R ) 12 R ) 12 • Classes of models to suppress each factors (i) Heavy squarks: ˜ m ≫ 1 TeV (e.g., split SUSY) m 2 (e.g., gauge mediation) (ii) Universality: ∆ m 2 D ≪ ˜ Q, ˜ ˜ (iii) Alignment: | ( K d L,R ) 12 | ≪ 1 (e.g., horizontal symmetries) • All SUSY models incorporate some of the above ZL — p.3
Where are we now?
The standard model CKM fit • Very impressive accomplishment • The level of agreement between the various measurements is often misinterpreted • Plausible TeV scale NP scenarios, consistent with all low energy data, w/o minimal flavor violation (MFV) • CKM is inevitable; the question is not if it’s correct, but is it sufficient? ZL — p.4
New Physics in FCNC processes • Mixing × ⇒ AND? OR Simple parameterization for each neutral meson: M 12 = M SM 12 (1 + he 2 iσ ) • Penguin decays γ γ b R s L b R s L t t × ⇒ AND? OR W H − Many operators for b → s transitions — no simple parameterization of NP • V td, ts only measurable in loops; likely also subleading couplings of new particles • Isolating modest NP contributions requires many measurements Compare NP-independent (tree) with NP-dependent (loop) processes ZL — p.5
Constraints on NP in B 0 d mixing • Overconstraining (“redundant”) measurements are crucial to bound new physics ρ, η determined from (effectively) tree level and loop-induced pro- cesses, separately M 12 = M SM 12 (1 + he 2 iσ ) aOnly the SM-like region is allowed, NP ∼ SM is still allowed; Think “MFV”: h ∼ (4 πv/ Λ flav . ) 2 ; is Λ flav . ≫ Λ EWSB ? even in the presence of NP in mixing • 10 – 20% non-SM contributions to most loop-mediated transitions are still possible ZL — p.6
B s mixing — ∆ m s • B 0 s – B 0 s oscillate 25 times on average before they decay — challenge to measure -1 CDF Run II Preliminary L = 1.0 fb 2 Amplitude -1 data ± 1 σ 95% CL limit 17.2 ps 1.5 -1 1.645 σ sensitivity 31.3 ps data 1.645 ± σ 1 data 1.645 (stat. only) ± σ 0.5 0 -0.5 -1 -1.5 -2 0 5 10 15 20 25 30 35 -1 ∆ m [ps ] s • ∆ m s = (17 . 77 ± 0 . 10 ± 0 . 07) ps − 1 √ B s Largest uncertainty: ξ = f Bs √ [CDF , hep-ex/0609040] f Bd B d Uncertainty σ (∆ m s ) = 0 . 7% is already Lattice QCD: ξ = 1 . 24 ± 0 . 04 ± 0 . 06 smaller than σ (∆ m d ) = 0 . 8% ZL — p.7
B s mixing phase — sin 2 β s • Next key measurement: time dep. CP asymmetry in B s → ψφ (as clean as sin 2 β ) In the SM: β s = arg( − V ts V ∗ tb /V cs V ∗ cb ) = 0 . 019 ± 0 . 001 • CDF & DØ disfavor large negative values: Testing a “squashed” UT: 0.10 CKM sol. w/ cos 2 β < 0 excluded area has CL > 0.95 CDF (no strong phase constraint & CL based on MC) (excl. at CL > 0.95) f i t t e r ε γ D0 (strong phase constraint & CL based on likelihood) K Pheno 2008 CKM fit sin 2 β 0.05 1.0 0.8 β s s 0.00 η 0.6 V α α 1 - CL ub m ∆ d 0.4 m & m ∆ ∆ -0.05 s d excluded at CL > 0.95 γ 0.2 CKM f i t t e r CDF, arXiv:0712.2397; DØ, arXiv:0802.2255 ε Pheno 2008 K -0.10 0.0 -0.10 -0.05 0.00 0.05 0.10 0.0 0.5 1.0 1.5 2.0 2.5 3.0 ρ 2 β (rad) s s Averaging complicated due to different assumptions, hopefully fixed by summer ZL — p.8
The D meson system • Complementary to K, B : CPV, FCNC both GIM & CKM suppressed ⇒ tiny in SM – 2007: signal for mixing > 5 σ [HFAG combination] – Only meson mixing generated by down-type quarks (SUSY: up-type squarks) a ∼ 10 − 2 Γ , since doubly- – SM suppression: ∆ m D , ∆Γ D < Cabibbo-suppressed and vanish in flavor SU (3) limit – CPV (mixing or direct) ≫ 10 − 3 would be sign of NP ( x = ∆ m/ Γ , y = ∆Γ / 2Γ ) – To do: Precise values of ∆ m and ∆Γ ? To do: Is CPV absent in mixing and decays? (not yet known if | q/p | ≃ 1 ) • Particularly interesting for SUSY: ∆ m D and ∆ m K ⇒ if first two squark doublets are within LHC reach, they must be quasi-degenerate (alignment alone not viable) ZL — p.9
The old/new B → Kπ puzzle ( T ) ( P ) • Q : new physics in CPV in B → Kπ ? A K + π − = − 0 . 097 ± 0 . 012 ( P + T ) ( C ) ( P ew ) A K + π 0 = 0 . 050 ± 0 . 025 ( P + T + C + A + P ew ) What is the reason for large difference? A K + π 0 − A K + π − = 0 . 147 ± 0 . 028 ( > 5 σ ) (Annihilation not shown) [Belle, Nature 452, 332 (2008)] SCET / factorization predicts: arg ( C/T ) = O (Λ QCD /m b ) and A + P ew small • A : huge fluctuation, breakdown of 1 /m exp., missing something subtle, new phys. • No similarly transparent problem with branching ratios, e.g., Lipkin sum rule looks Γ( B − → π 0 K − ) + ¯ Γ( B 0 → π 0 K 0 ) OK by now: ¯ 2 = 1 . 07 ± 0 . 05 (should be near 1 ) Γ( B − → π − K 0 ) + ¯ ¯ Γ( B 0 → π + K − ) ZL — p.10
Forthcoming progress
Questions we hope to gain insights on • The 3rd generation may differ from the 1st and 2nd by more than we know so far Large top Yukawa ⇒ maybe non-universal coupling to EWSB and NP sector Want to compare 3rd–1st and 3rd–2nd generation data with precision kaon data • Many processes have different sensitivities to various NP scenarios In SM: CPV only in flavor changing, charged current interactions of quarks With NP: possible in flavor diagonal processes, neutral currents, in lepton sector Does new physics give rise to operators forbidden (highly suppressed) in the SM? s σ µν F µν P R b vs. O ′ s σ µν F µν P L b E.g., O 7 = ¯ 7 = ¯ • Try to distinguish NP scenarios: One / many sources of CPV? Only in CC inter- actions? Couples to up / down sector? 3rd / all generations? ∆ F = 2 and / or 1 ? ZL — p.11
sin 2 β eff , α , γ — large improvements possible B →ρπ (WA) sin(2 β eff ) ≡ sin(2 φ e ff ) vs C CP ≡ -A CP H F AG H F A G 1.2 CK M 1 B →ρρ (WA) COMBINED f i t t e r LP 2007 LP 2007 C CP ≡ -A CP B →ππ (WA) PRELIMINARY 1 0.8 0.8 0.6 1 – CL 0.6 0.4 CKM fit no α meas. in fit 0.2 0.4 0 0.2 b → ccs φ K 0 -0.2 η′ K 0 0 0 20 40 60 80 100 120 140 160 180 K S K S K S π 0 K S -0.4 α (deg) ρ 0 K S ω K S CKM D(*) K(*) GLW + ADS WA -0.6 f 0 K 0 f i t t e r D(*) K(*) GGSZ Combined π 0 π 0 K S FPCP 07 K + K - K 0 Full Frequentist treatment on MC basis -0.8 1 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 sin(2 β eff ) ≡ sin(2 φ e ff ) 0.8 1 1 - CL 1 - CL Contours give -2 ∆ (ln L) = ∆χ 2 = 1, corresponding to 60.7 % CL for 2 dof 0.6 • E.g., S ψK − S φK = 0 . 29 ± 0 . 17 ; also for α & γ : 0.4 want ∼ 10 × smaller error ⇒ ∼ 100 × more data CKM fit 0.2 no meas. in fit γ • Need both LHCb and e + e − super B factory 0 0 0 20 20 40 40 60 60 80 80 100 100 120 120 140 140 160 160 180 180 γ γ (deg) (deg) ZL — p.12
Recommend
More recommend
