Measuring γ. γ is the phase of V ub. Can be determined using B ± decays. These diagrams result in the same final state for D 0 K + K -, K S π + π.

Transcription

1 Measuring γ γ is the phase of V ub. Can be determined using B ± decays. These diagrams result in the same final state for D 0 K + K -, K S π + π. A V cb V us A T A V ub V cs A CT Phase differs by γ, Amp by A CT /A T different A s for different final states Can also use doubly Cabbibo suppressed decays Fermilab Academic Lectures, May,

2 Results Analysis is very complicated & sums over many final states (including D 0 π ) Results for γ BaBar Belle LHCb (67±12) o All K S ππ+dcs + CLEO K S ππ Belle Fermilab Academic Lectures, May,

3 Measuring α The B 0 π + π & ρ + ρ decays can occur via If (a) is dominant, then by measuring a fcp, we measure Can tell by seeing the size of π 0 π 0 & ρ 0 ρ 0. (a) not dominant for π + π, but OK for ρ + ρ. However its not a CP eigenstate, but this can be dealt with BaBar: α = ( ), Belle: (84.9 ±12.9) 3

4 Charm CPV CP Violation in charm is not expected at at a level >~10-3, so is an excellent place to look for New Physics SM Fermilab Academic Lectures, May,

5 Are these measurements consistent? CKM fitter group Does a frequentist analysis Also UT fit group does a Bayesian analyis Fermilab Academic Lectures, May,

6 Seeking New Physics HFP as a tool for NP discovery While measurements of fundamental constants are fun, the main purpose of HFP is to find and/or define the properties of physics beyond the SM HFP probes large mass scales via virtual quantum loops. An example, of the importance of such loops is the Lamb shift in atomic hydrogen A small difference in energy between 2S 1/2 & 2P 1 /2 that should be of equal energy at lowest order Fermilab Academic Lectures, May,

7 Flavor Physics as a NP discovery tool Another example of the importance of such loops are changes in the W mass M w changes due to m t M w changes due to m H Fermilab Academic Lectures, May,

8 Limits on New Physics It is oft said that we have not seen New Physics, yet what we observe is the sum of Standard Model + New Physics. How to set limits on NP? One hypothesis: assume that tree level diagrams are dominated by SM and loop diagrams could contain NP Tree diagram example Fermilab Academic Lectures, May, 2014 Loop diagram example 8

9 What are limits on NP from quark decays? Tree diagrams are unlikely to be affected by physics beyond the Standard Model Note γ is a CP violating angle but is measured via Tree diagrams here Fermilab Academic Lectures, May,

10 CP Violation in B o & K o Only Absorptive (Imaginary) part of mixing diagram should be sensitive to New Physics. Lets compare Fermilab Academic Lectures, May,

11 They are Consistent But consistency is only at the 5% level Limits on NP are not so strong Fermilab Academic Lectures, May,

12 Compare measurements look for discrepancies Generic Analyses B o mixing and (s) CP. Parameterize NP as h & σ Fermilab Academic Lectures, May,

13 Limits on New Physics 99.7% cl 99.7% cl B 0 B 0 s New Physics amplitudes could be ~20% of Standard Model J. Charles et al arxiv: Fermilab Academic Lectures, May,

14 Ex. of Strong Constraints on NP Inclusive b sγ, (Eγ > 1.6 GeV) Measured (3.55±0.26)x10-4 (HFAG) Theory (3.15±0.23)x10-4 (NNLL) Misiak arxiv: Ratio = 1.13±0.11, Limits most NP models Example 2HDM m(h + ) < 316 GeV B(b sγ) 2HDM tanβ=2 } } SM Theory Misiak et. al hep-ph/ , See also A. Buras et. al, arxiv: Measurement Fermilab Academic Lectures, May,

15 Theorists task A given theoretical model must explain all the data Model must thread through all experimental constraints (12 axe handles). One measurement can, in principle, defeat the theorist, but we seek a consistent pattern. Fermilab Academic Lectures, May,

16 Top Down Analyses Here we pick models and work out their consequences in many modes. Ex. (circa 2010): Straub: axxiv: Fermilab Academic Lectures, May,

17 B s µ + µ SM branching ratio is (3.5±0.2)x10-9 [Buras arxiv: ], NP can make large contributions. Standard Model MSSM ~tan 6 β Many NP models possible, not just Super-Sym Fermilab Academic Lectures, May,

18 Discrimination LHCb uses B h + h - to tune cuts for a multivariate analysis Other variables to discriminate against bkgrd : B impact parameter, B lifetime, B p t, B isolation, muon isolation, minimum impact parameter of muons, B s production is measured by using the LHCb measured ratio f s /f d. New value of 0.259±0.015 Fermilab Academic Lectures, May,

19 Production fractions: B DXµν use equality of Γ sl & known τ s D + LHCb Dfb LHCb D + K - π + π + Dfb: 9406±110 Prompt D + Fake D + D s K - K+ - π + D s Dfb: 2446±60 Fermilab Academic Lectures, May, 2014 Also D o, Λ b 19

20 P T & η dependence Ncorr(B s Dµ) is D s µ+dkµ Also using hadronic B s & B 0 decays find Fermilab Academic Lectures, May,

21 Evidence for B s µ + µ Avg: B(B s µ + µ )=(2.9±0.7)x10-9 Avg: B(B 0 µ + µ )=( )x10-10 (not significant) Fermilab Academic Lectures, May,

22 Implications Only this range allowed Fermilab Academic Lectures, May,

23 Flavor as a High Mass Probe i Already excluded ranges from box diagrams, take c i ~1 Ways out 1. New particles have large masses >>1 TeV 2. New particles have degenerate masses 3. Mixing angles in new sector are small, same as in SM (MFV) 4. The above already implies strong constrains on NP See: Isidori, Nir Fermilab Academic Lectures, & Perez May, arxiv: ; Neubert EPS 2011 talk

24 Some hints of discrepancies with SM Fermilab Academic Lectures, May,

25 B K ( * ) l + l Similar to K*γ, but more decay paths + new particles in loops Several variables can be examined, e.g. muon forward-backward asymmetry, A FB is well predicted in SM Fermilab Academic Lectures, May,

26 Theory K ( * ) l + l Decay described by 3 angles & dimuon invariant mass (q 2 ) For each bin in q 2 F L is fraction of longitudinally polarized K* 0 A FB, forward-backward asymmetry SM prediction of q 2 for A FB crossing 0 is (Beneke) Fermilab Academic Lectures, May,

27 B o K* o l + l J/ψ } Ψ2 } Other data from CDF, BaBar & Belle Conforms to SM predictions by Bobeth et al. & Matias et al Fermilab Academic Lectures, May,

28 Forward-Backward asymmetry No evidence of deviation from SM so far Fermilab Academic Lectures, May,

29 B - K l + l Resonances found in high q 2 region One would think they would be in K* o l + l also Should affect theory predictions Ψ(3770) Ψ(4160) Fermilab Academic Lectures, May,

30 More variables Back to K ( * ) l + l, new observables in formalism designed to less sensitive to hadronic form-factors Descotes-Genon et al arxiv: Fermilab Academic Lectures, May,

31 Possible deviation Could be something, but significance depends on theoretical model, & deviation is only in one place 1 fb -1 Fermilab Academic Lectures, May,

32 Rare Decays - Generic C i O i for SM, C i O i are for NP. Operators are for P R,L = (1±γ 5 )/2 O =O with P R,L P L,R Each process depends on a unique combination Fermilab Academic Lectures, May,

33 Other Processes Other processes probe different operators Let δci=c i (NP)-C i (SM) Examples: Fermilab Academic Lectures, May,

34 Maximizing deviations Filled bands: B K*µ + µ, K*γ & B s µ + µ Dashed: all q 2 for K*µ + µ Orange: only 1<q 2 <6 GeV 2 for K*µ + µ Some suggest a 7 TeV Z Gauld et al arxiv: ! Buras, Girrbach arxiv: Descotes-Genon et al arxiv: Fermilab Academic Lectures, May,

35 B - τ - ν problem? B - τ - ν, tree process: sin2β, CPV in e.g. B o J/ψ K s : Box diagram Measurement not in good agreement with SM prediction based on CKM fit New Belle measurement in using 1 method. Discrepancy may be resolved, but 3 other determinations need to be checked New Belle Result Can be new particles instead of W - but why not also in D + (s) l + ν? World average of 4 measurements Fermilab Academic Lectures, May,

36 B D ( * ) τν Also, tree level BaBar result Similar to B - τ - ν analysis Fully reconstruct one B, keep events with an additional D ( * ) plus an e - or µ -. Signal is wide, background, especially D**l ν, needs careful estimation Fermilab Academic Lectures, May,

37 B D ( * ) τν ΙΙ Results given in terms of ratio to B D ( * ) lν SM Theory BaBar value Diff. R(D) 0.297± ±0.058± σ R(D*) 0.252± ±0.024± σ Sum is 3.4σ above SM Also inconsistent with type II 2HDM 2HDM Fermilab Academic Lectures, May,

38 Other searches Fermilab Academic Lectures, May,

39 Several ways of looking for presence of heavy ν s (N) in heavy quark decays if they are Majorana (their own anti-particles) and couple to ordinary ν s Majorana ν s Modes analogous to ν less nuclear β decay Simplest Channels: B - D + l - l - & B - D* + l - l l - & l - can be e -, µ - or τ -. Fermilab Academic Lectures, May,

40 Limits on D ( * )+ l - l Upper limits in e - e - mode not competitive with nuclear β decay Others unique since measure coupling of Majorana ν to µ - Mode Exp. u. l. x 10-6 B - D + e - e - Belle < 2.6 B - D + e - µ - Belle < 1.8 B - D + µ - µ - Belle < 1.0 B - D + µ - µ - LHCb < 0.69 B - D* + µ - µ - LHCb < 3.6 Belle [arxiv: ] Fermilab Academic Lectures, May,

41 Can also look for Majorana ν (N), where N W + µ On-Shell ν A. Atre, T. Han, S. Pascoli, & B. Zhang [arxiv: ] Many other ways of searching: K π N µ e γ τ µ + π π.. Fermilab Academic Lectures, May,

42 B π + µ µ LHCb search as a function of Majorana neutrino mass and lifetime Fermilab Academic Lectures, May,

43 The Dark Sector Could it be that there are 3 classes of matter? SM particles with charges [SU(3)xSU(2)xU(1)] Dark matter particles with dark charges Some matter having both ( mediators ) Searches for dark photons A mediator, couples to b-quarks (see arxiv: hep/ph) BaBar B(Y(1S) 90% cl Other experiments Fermilab Academic Lectures, May,

44 Search Summary Parameterize by mixing ε ε Needed to explain g-2 Dark photon mass ma From B. Echenard arxiv: Fermilab Academic Lectures, May,

45 Tetraquarks, both heavy & light Fermilab Academic Lectures, May,

46 Belle 2008: B 0 J/ψπ K +. Claimed resonant signal decaying into J/ψπ at 4430 MeV a charged charmonium state, not possible with only cc Tetraquark candidate Z (4430) - PRL 100, (2008) Veto K*(890) Residual K* background Fermilab Academic Lectures, May,

47 BaBar shows that moments of K + π resonances can reflect in mass peak Data are compatible with Belle Difference is in interpretation But not BaBar Fermilab Academic Lectures, May,

48 Belle does 4D amplitude fit New fit confirms observation, but questions remain Fermilab Academic Lectures, May,

49 LHCb full fit for 1 + Z p value of 12% arxiv: Fermilab Academic Lectures, May,

50 Argand diagram arxiv: Fermilab Academic Lectures, May,

51 Scalar octet problem 0 + vs 1 - meson masses (charge = 0) For 1 -, adding an s quark increases meson mass Suggestions that 0 + mesons are tetraquarks For qq, σ f 0 (500) & f 0 (980) are mixed with f 0 (980) mostly ss As tetraquarks Fermilab Academic Lectures, May,

52 Suggested B s test Here f 0 f 0 (980), σ=f 0 (500) qq model qq model Stone & Zhang, Phys.Rev.Lett. 111 (2013) 6, tetraquark model tetraquark model Large f 0 expected in qq, no σ rate for tetraquark Fermilab Academic Lectures, May,

53 Suggested B 0 test Here f 0 f 0 (980), σ=f 0 (500) qq model qq model Stone & Zhang, Phys.Rev.Lett. 111 (2013) 6, tetraquark model tetraquark model Small f 0 expected in qq, half of σ rate in tetraquark Fermilab Academic Lectures, May,

54 B s & B 0 signals B 0 B s arxiv:

55 B s results Huge f 0, no σ arxiv: Fermilab Academic Lectures, May,

56 B 0 results Nice σ, no f 0 σ arxiv: Fermilab Academic Lectures, May,

57 Not tetraquarks In qq model mixing Tetraquark prediction of 0.5 ruled out at 8σ Fermilab Academic Lectures, May, 2014 arxiv:

58 Future Acts LHCb Upgrade: run at cm -2 /s (x5), & double trigger efficiency on purely hadronic final states. Much improved sensitivities to New Physics at higher mass Implemented by having a purely software trigger Requires entire detector to be read-out at 40 MHz e + e - Super Belle Time scales are on the order of 5 years Fermilab Academic Lectures, May,

59 Conclusions Heavy Flavor physics is very sensitive to potential New Physics effects at high mass scales LHCb has started to make world class measurements of flavor physics. We hope to find physics beyond the Standard Model or derive limits that strongly constrains theories of New Physics. The LHCb upgrade is necessary to improve sensitivities. Many other interesting results have not been mentioned Fermilab Academic Lectures, May,

60 Theory conquers Fermilab Academic Lectures, May,

61 The End 61 Fermilab Academic Lectures, May, 2014

62 Common Analysis APS W. Altmannshofer, P. Paradisi & D. M. Straub arxiv: v2 1σ & 2σ allowed B(B X s l + l - ) S(K*γ) K* o l + l - low q 2 B(b sγ) K* o l + l - high q 2 Many more such generic constraints Fermilab Academic Lectures, May,

63 Also B Dh - D - π + Take ratios, use theory P t dependence now evident, implications for ATLAS, CMS analyses D - K + D s - π + Fermilab Academic Lectures, May,

64 B s lifetime results here use only fully reconstructed decays K + K - is taken as CP even (A Γ =-1) Ovals show 39% cl, while bands 68% cl τ s =1.509±0.010 ps, Γ s = 0.092±0.011 ps -1, y s = Γ s /2Γ s = 0.07±0.01 (from Anna Phan) Γ s & Γ s Lenz arxiv: Contours [log(l)]=0.5 only full reconstructed B s decays used Fermilab Academic Lectures, May,

65 a sl By definition at t=0 M f is zero as is M f _ Here f is by construction _ flavor specific, f f Can measure eg. B s D+ s µ ν, versus B s D- s µ + ν, Or can consider that muons from two B decays can be like-sign when one mixes and the other decays, so look at µ + µ + vs µ - µ - a sl is expected to be very small in the SM, a sl =( Γ/ M) tanφ 12, where tanφ 12 =Arg(-Γ 12 /M 12 ) d In SM (B o ) a sl =-4.1x10-4, (B s ) a sl =+1.9x10-5 Fermilab Academic Lectures, May, 2014 arxiv: [hep-ph] 65 s

66 D o a sl Using dimuons (3.9σ) Indication from D0 that its B s Separate dimuons into B d and B s samples using muon impact parameter Find Fermilab Academic Lectures, May,

67 New D0 Analysis s Measure a sl using D s µ ν events, D s φπ ± Detect a µ associated with a D s decay D + D + s zero suppressed s Find a sl =(-1.08±0.72±0.17)% d Also measure a sl using D + µ ν, D + Kπ + π + d a sl =(0.93±0.45±0.14)% Fermilab Academic Lectures, May,

68 a sl according to D0 s a sl =(-1.81±0.56)% d a sl =(-0.22±0.30)% 3σ from SM arxiv: SM Fermilab Academic Lectures, May,

69 LHCb measurement Use D s µ ν, D s φπ ±, magnet is periodicaly reversed. For magnet down: D + D s + D - D s - Effect of B s production asymmetry is reduced to a negligible level by rapid mixing oscillations Calibration samples (J/ψ, D* + ) used to measure detector trigger, track & muon ID biases Fermilab Academic Lectures, May,

70 a sl not D0 LHCb finds B-factory Results consistent with SM Expect φ s to grow as s sin[2 β s +arg(m 12 )] for finite a sl. s a sl Fermilab Academic Lectures, May,

71 Λ b Fraction Significant p t dependence s = 7 TeV LHCb Preliminary ~3 pb -1 s = 7 TeV LHCb Preliminary ~3 pb -1 In general agreement with CDF measured at <p t >~10 GeV/c Fermilab Academic Lectures, May,