1 / 53

Constraining c harm penguins

Constraining c harm penguins. Brian Meadows Cincinnati. . TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: A A A A A A A A A A A A A A A A A A A A A A A A A A A. AD 1999:. Prologue.

tucker
Download Presentation

Constraining c harm penguins

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Constraining charm penguins Brian Meadows Cincinnati . TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: AAAAAAAAAAAAAAAAAAAAAAAAAAA

  2. AD 1999: Prologue • ”A comprehensive program of CP studies in heavy flavour decays has to go beyond observing large CP asymmetries in nonleptonic B decays and finding that the sum of the three angles of the unitarity triangle is consistent with 180◦. There are many more correlations between observables encoded in the KM matrix; those can be expressed through five unitarity triangles in addition to the one usually considered.” • -- IkarosBigi and A. Sanda • http://arxiv.org/abs/hep-ph/9909479

  3. Weak phases in Bd and D decays bdtriangle – Bd decays. All three phases are large. Tree phases ¯c are tiny BUT b-penguin phase c= is large. cutriangle – D decays Bevan, Inguglia, BM: Phys.Rev. D83 (2011) 051101

  4. It is probably beyond experimental ability to measure bc • But it is important to check that it is very small. • Also interesting to check other phases in triangle. • B-factory methods are possible approach. • Make t-dependent measurement of CP asymmetry for decays to CPeigenstatesTDCPV. (Talk by G. Inguglia). • Comparison of TDCPV for 2 modes can also provide measurement of D0mixing phase. • Effect of penguins will need to be estimated – an interesting measurement in any case.

  5. D0h+h- (K+K-, p+p-or r+r-) K+K- :zero Tree (T): CKM phase p+p-:bc q h+ u q c Exchange (E): CKM phase same as T SoCombine T and E as “T” h+ q c u D0 h- D0 u h- u u u q u h+ c q Penguin (P): CKM phase gc b,s,d q D0 h- u u

  6. Standard Model Penguins “cu” triangle condition • Small - could be larger with U-spin or QCD effects • Weak phase large (~ g) • Change iso-spin D I = ½ (c u) U-spin breaking a tunable parameter controls level of Ps, Pd effect on P. c u b, s, d SM: Brod, Grossman, Kagan, Zupan, JHEP 1210 (2012) 161 SM Penguins:

  7. TDCPV in D0 decays Bevan, Inguglia, BM, Phys.Rev. D84 (2011) 114009 D0 D0 f (CP eigestate) D0 Mixing phase Arg{q/p} = M See GianlucaInguglia’s talk Mixing allows D0and D0 to interfere, exposing weak phases. Assume only T amplitude: Define , then Measure using time-dependent CP asymmetry.

  8. Time-Dependent CP Asymmetry (TDCP) D0and D0oscillations leads to time-dependent CP asymmetry. Measure this for D0  h+h- and obtain • Asymmetry grows with | t/t |: • ACP for D0 is much smaller than for B0 and is almost linear in t. • Slope of line / arg {f} • |ACP |is largest for large t Direct CPV shifts asymmetry at t=0 From Penguin

  9. Time-Dependent CP Asymmetry (TDCP) D0and D0oscillations leads to time-dependent CP asymmetry. For D0  h+h- we expect • Asymmetry grows with | t/t |: • ACP for D0 is much smaller than for B0 and is almost linear in t. • Slope of line / arg {f} • |ACP |is largest for large t Direct CPV shifts asymmetry at t=0 Effect of Penguin (h = p or r)

  10. Penguins in Dppor rr decays With similar definitions for CP Conjugate modes I =3/2and1/2 : so includes P I =3/2 : No P 1A. Bevan, BM, in progress Penguin contributions, , to can be estimated from I -spinrelations between amplitudes A for different charge modes: Bose symmetry allows only I = ½ and I =3/2amplitudes. The former is possible for T and P but the latter only for T. 10

  11. In only T contributes so phase of is + and for it is – • Can rotate these to coincide Then re-label • The “+-” and “00” amplitudes include penguins. • Any phase difference between and , therefore, is (due to penguins).

  12. Apply to current charm data • Take current information on pp decay rates (from PDG): We create ensembles of MC simulated experiments, based on the magnitudes of these amplitudes. We assume no CP asymmetry yet We compute for each sample, noting the ambiguity in the relative orientation between the and triangles.

  13. D0  pp D0 rr • When, eventually, an asymmetry becomes evident in any of these modes, the positions of the peaks may be more interesting. Both solutions are clearly visible with a width that suggests an uncertainty in of about 2.70.for pp and 4.60for rr.

  14. Drr decays • Analysis is similar to that for D  ppbut with complications. Again, there are mostly I = 1 or I = 2 final states. BUT • The r0 interferes with w0, introducing an I = 1 component • r resonances are broad and interfere with other resonances and each another. • Transversity amplitudes for rrhave different CPand could have different penguin contributions. Therefore they require separate treatment. • So proper amplitude analyses of D  4p modes are required Our study assumes that just one transversity state dominates and ignores all the above complications.

  15. Current data on Drr Small r0r0 rate a hint that P is small ? Large fL justifies use of a single spin state.

  16. 16 I -spin analysis ofDrpdecays Neither side of the equation has a Pcomponent. I =3/2only I =3/2only amplitudes have been rotated by weak decay phase Ar+p- 2Ar0p0 ~ ~ Ar-p+ Ar+p- ~ 2Ar0p0 Ar-p+ amplitudes have been rotated by weak decay phase ~ \/2Ar+p0 \/2Ar0p+ \/2Ar+p0 ~ \/2Ar0p+ • For rp, Bose statistics does not apply, so there are five I -spin amplitudes contributing to a pentagonal relationship:

  17. 17 bcfrom D0rpdecays • This determines 26 invariant quantities related to D0  rpamplitudes , their magnitudes and relative phases, and their time-dependences. • These determine and the magnitudes and phases of the P’s and T’s. Ar+p- 2Ar0p0 ~ ~ Ar-p+ Ar+p- ~ 2Ar0p0 Ar-p+ A time-dependent Dalitz plot fitto D0 (and D0) p+p-p0decays provides all the required information. [Quinn, Snyder, Phys.Rev.D48(1993) 2139-2144]

  18. Toy “bc scan” on BaBar data Precision for Using the complex A’s from the BaBar fit to time-integrated D0 3p decays, we make fits for P and A at various values and plot the p-value for the best fit at each point. Two peaks are observed, respectively at and , each with a half-width of ~ 10.

  19. 19 I -spin constraints inDrpdecays I -spin conservation requires these 4 lines to have same lengths. Ar+p- 2Ar0p0 ~ ~ Ar-p+ Ar+p- ~ 2Ar0p0 Ar-p+ ~ \/2Ar+p0 \/2Ar0p+ \/2Ar+p0 ~ \/2Ar0p+

  20. Systematic Limitations • These studies rely on experimental determination of the relative magnitudes of D0 and D+ decay amplitudes. • Systematic limitations should be from p0 efficiency and will probably be reached by Belle2 or by 5 ab-1t-charm. Modes#p0 requiredComment p+p- :p0p0 :p+p0 0 : 1 : 1 Normalize p0p0 to Ksp0 r+r-: r0r0: r+r0 2 : 0 : 1 3 distinct Dalitz plots r+p-: r0p0: r+p0 0 : 0 : 0 Use just p+p-p0Dalitz plot • For BaBar, systematic is ~3% perp0. • If e+e-g used, this is ~0.6%, but lose factor 80 in sample size.

  21. 21 The effect of penguin amplitudes on weak decay phase was examined. rr complications ignored and no CP asymmetry assumed. pp rr 1A. Bevan, BM, in progress • For D0rr the analysis is similar, but complicated by need for • Separation of events intoCP=§1(transversity) eigenstates. • A full amplitude analysis to separaterr from I =1and any other amplitudes inD04pdecays. • Simplified analysis, using PDG values for branching fractions, and assuming no CPV asymmetries allows an estimate for precision for in the pp case, and for from rr decays.

  22. 22 D++0 BF and asymmetry CLEOc 0.818 fb-1 at (3770) BaBar 124 fb-1 at Y(4S) 1,227 Events (30% purity) 2,649 Events (55% purity) B+0 / BK-++= (1.29±0.04±0.05)×10−2 ACP = (2.9  2.9  0.3) x 10-2 B+0 / BK-++= (1.33±0.11±0.09)×10−2 ACP ~ (xxx  6.2) x 10-2 Phys.Rev. D74 (2006) 011107 Phys.Rev. D81 (2010) 052013 For ( ) then (I = 3/2) thus excluding any SM penguin contribution. CP asymmetry in these decays would require NP !! BaBar and CLEO measured this mode relative to D+K-++

  23. 23 0 0 BF and asymmetry BaBar 471 fb-1 at Y(4S) CLEOc 0.818 fb-1 at (3770) Preliminary 26,010 events (55% purity) 1,567 events (63% purity) B00/BK = (2.06 ± 0.07 ± 0.10)×10−2 ACP - NOT possible B00/BK00= (6.88 ± 0.08 ± 0.33)×10−2 ACP ~ (xxx  1.2) x10-2 Submitted to Phys.Rev. D Phys.Rev. D81 (2010) 052013

  24. 24 Projections for ACP Measurements % % • For D000BaBar measures BF, not ACP which we estimate. • For ACP measurements, we observe that most systematic uncertainties cancel except for uncertainties in signal and background shapes. • We assume these should shrink with the sqrt of data size (?)

  25. 25 Summary • Time-integrated CPV asymmetries have yet to be seen but, when they are, a look at the effect of penguins should be possible with a precision within the 1-2 degree range. • LHCbis working extremely well, and is clearly ready to lead the way in measurements of D decays with charged tracks, • but it will leave much for e+e- machines to do with the modes with ¼0’s and other neutrals. • Experiments at charm threshold have a particular role to play in studies with D decays with one or more p0 or g, and should be optimized for this role.

  26. Backup Slides

  27. c The triangles See Bigi and Sanda, hep-phy/9909479 (1999) Bd decays • Bigi and Sanda: • In addition to ,  and , the angles, c and s should be measured also, if possible. • LHCb is working on s using Bs(f0) decays. • SuperB and Belle2 should also be able to study Bs(‘) at Y(5S) BaBar/Belle ~1 (280)  s Bs decays LHCb/CDF/D0 ~2 (10) cu triangle D decays ~4 (.050)

  28. What is Interesting about Charm • Charm was “invented” to account for small FCNC interactions in nature (GIM mechanism). • In this scenario, for the charm sector, • Mixing is also greatly suppressed; • Many charm particle decays are also extremely small. • CP violation (CPV) is also expected to be small, mostly because weak phases are small (Arg{Vcd} ~ ¸4); • With “SM backgrounds” so small, charm is a good place to look for new physics (NP). • Charm also allows study of the role of the up-type quarks.

  29. A ¼0 Trigger ? p+ Need to trigger back up g (found offline) e+ 3 charged track Displaced vertex. e- At least 1 (or 2) e’sID’d. Three tracks do not point back to PV Invariant mass < D+ Consider D+ ¼+¼0 1/80¼0’s decay thus. |____ e+e-°

  30. Epilogue • “ An important goal in charm physics is not just • to observe CP Violation in D decays • but also to understand its origin” • -- IkarosBigi • “ Thanks, Ikaros – we are still looking.”

  31. Experimentalstudies of penguins? Superscripts on amplitudes Arefer to charges of final state p (or r) mesons. Anti-particle amplitudes are rotated by the weak decay phase expected for Tree amplitudes. 1A. Bevan, BM, in progress Methods developed to estimate penguin contributions in Bdecays were recently examined for use in Ddecays1. They look promising. I-spin relationships among different final states provide information on SM penguin amplitudes (Since they only contribute to I = ½ transitions). For D0,+p0+p0or r0+r0 final states, Bose symmetry allows only I = ½ and I =3/2 reduced I -spin amplitudes resulting in a triangle relationship: 32

  32. Experimentalstudies of penguins? Superscripts on amplitudes Arefer to charges of final state p (or r) mesons. Anti-particle amplitudes are rotated by the weak decay phase expected for Tree amplitudes. Indicates effective phase change introduced by penguin amplitudes. 1A. Bevan, BM, in progress I -spin relationships among different final states provide information on SM penguin amplitudes (Since they only contribute to I = ½ transitions). Methods developed to estimate penguin contributions in Bdecays were recently examined for use in Ddecays1. They look promising. For D0,+p0+p0or r0+r0 final states, Bose symmetry allows only I = ½ and I =3/2 reduced I -spin amplitudes resulting in a triangle relationship: 33

  33. 34 The effect of penguin amplitudes on weak decay phase was examined. rr complications ignored and no CP asymmetry assumed. pp rr 1A. Bevan, BM, in progress • For D0rr the analysis is similar, but complicated by need for • Separation of events intoCP=§1(transversity) eigenstates. • A full amplitude analysis to separaterr from I =1and any other amplitudes inD04pdecays. • Simplified analysis, using PDG values for branching fractions, and assuming no CPV asymmetries allows an estimate for precision for in the pp case, and for from rr decays.

  34. 35 I-spin analysis ofDrpdecays Neither side of the equation has a penguin component. Anti-particle amplitudes are rotated by the weak decay phase to align the Tree amplitudes. Precision for • For rp, Bose statistics does not apply, so there are five I -spin amplitudes contributing to a pentagonal relationship between modes: • Time-dependentDalitz plot fitfinds the sixD0(rp)0amplitudes: where tilde’s refer to CP conjugation, reversing sign of weak phase. • All T’s, P’s and can be determined from the TDDP fit.

  35. Direct and Indirect CPV in D0 decays Gersabek, 2011 and Time-integrated CP asymmetry Mean decay time asymm. LHCbhas excellentresolution in decay time t, but rejects short times. Babar have relatively poor t resolution but include eventscloser tot=0. • Two physical observables we measure are • In presence of direct CPV, the first depends on decay mode f. • Since D0 decays are not exponential, both observables depend on the (experiment-dependent) time-span for the observations.

  36. Direct and Indirect CPV HFAG (Gersabek) A c2 fit leads to values: Central values are ~ 4¾ from “no CPV” point (where CL= 2x10-5). No CPV The difference in time-integrated asymmetry includes both direct and indirect components but the difference is mostly direct (with small time dependence due to finite integration time):

  37. BaBar Peak luminosity 1.21034 cm–2s–1 Integrated luminosity 531 fb–1 • Main purpose: Study CP violation in asymmetric e+e - (4S)  BB • Experiment far exceeded the design goals • Luminosity order of magnitude larger • Many more measurements and discoveries.

  38. CPV in multi-body decay modes PhysRevD.78.051102 384 fb-1 • Extended search within h+h-¼0 modes: • CPV is unlikely to be seen in all channels – but perhaps in one Search each channel - e.g. D0 0 + 0 • Each channel can be normalized to whole Dalitz plot. Systematic uncertainties from s+ tagging or from production asymmetries become 2ndo`rder effects • CPV is signalled by differences in phase behaviour between D0 and D0. Dalitz plot for these 3-body final states yields information on phase behaviour between channels. • BaBar, Belle and LHCb are using several search strategies • Model-independent searches for CPV in exclusive parts of phase space. • Model-dependent searches based on fits to the Dalitz plot distributions

  39. Two Model-Independent Searches for CPV in D0-+0 and K-K+0by BaBar “Miranda method ?” Phys.Rev.D (TBP, 2008) Dalitz plots for D0 and for D0 are normalized and compared, bin-for-bin Unbiassed frequentist test yields 16.6% conf. level there is no difference. Legendre polynomial moments of D0-D0 differences (to order 8) are normalized and compared, in each channel. Unbiassedfrequentist test indicates 23-66% conf. levels there are no differences in the various channels. [+-]+ 0 channel [+0]+ - channel

  40. Model-dependent Search for CPV in D0-+0 and K-K+0 Phys.Rev.D (TBP, 2008) Dalitz plots for D0 and for D0 were fitted to isobar model expansions of interfering amplitudes in each channel. Differences in magnitudes and phases For each amplitude were insignificant.

  41. Measure TD CPV asymmetry SM: NP: NP?  ~?? P P T T  ~c ~ 0.040  ~gc ~ 670 The time-dependence of CPV asymmetry of weak decays of D0 to a CP eigenstate measures the phase M – 2 where M is the mixing phase and  is the weak decay phase. Differences between D0+- and D0K+K- can, therefore, be used to measure . This can be useful in understanding the difference between SM and NP for the differential asymmetry observed by LHCb between these two modes.

  42. D0 K+K- and +- Arxiv:0807.0148v1 (2008) NEW Phys.Rev.Lett.100:061803 (2008) • No evidence for CPV • Systematic uncertainties ~ 0.1% (Likely scale with luminosity-1/2) !! • No significant difference between KK and 

  43. u u K+ K+ c c s d,s,b s D0 s s D0 K- K- I = ½ or u 3 / 2 u + + c c d d,s,b d D0 D0 d d - - u u u u u u u u I-spin and U-spin I = ½ I = ½ U=½ (s d ) I = ½ There are differences in I–and U–spin in each amplitude The relation between K+K- and +- modes is a change U=½ (s d) that, if SU(3)flav.is not broken, results in a change in sign of the CP asymmetries.

  44. I , U and V-spin Conservation Y d u I-spin I3 V-spin U-spin s I-spin symmetry breaking sources: • EWpenguins - suppressed by factors/. • Differentuanddquark masses. • E/Minteractions. BUT Effects are O(1%) - comparable to some CPVasymmetries observed. Three SU(2) sub-groups of flavour SU(3): • Lipkin: • “I-spin, U-spin, V-spin •  V-all spin” U-spinsymmetry is probably broken. • Ratios of D 0 decay rates to K-+, K-K+and K+- differ from Cabibbo suppression values. • U-spin symmetry predicts that has yet to be experimentally tested. Feldman, Nandi, Soni, arXiv: 1202.3795

  45. I -spin breaking,due to electromagnetic interactions and to u and d quark mass differences areCPconserving. That due to EW penguin amplitudes are suppressed by ~(s/). • GZK keep this breaking a 2nd order effect in comparison with predicted asymmetries, by writing their sum rules mostly in terms ofCP differences of rates or amplitudes

  46. I-spin Tests for NP It is hard for the SM to account for ¢ACP of ~1%, but maybe not impossible. But how can we tell if NP is required? In the SM, the CPV asymmetries come only from I = 1/2 penguin amplitudes. So CPV symmetries from a I = 3/2 decay amplitude would be a clear signal for NP. Recognizing that I–spin breaking has similar magnitude to CPV asymmetries, Grossman, Kagan and Zupan (GKZ) recently proposed a number of sum rules that could, when sufficient data are available, expose any CPV effects in I = 3/2 amplitudes. Phys.Rev. D85 (2012) 114036

  47. t (ps) t (ps) Large and pure samples from D*+D0+decays fit to combined Ks and KsKK samples give most precise measurement to date Ks +- Signal : 541K purity 98.5% S-wave +- S-wave K0- P- and D-waves K-matrix model LASS model Breit-Wigner model KsK+K- Signal : 80K purity 99.2% S-wave K+K-Coupled-channel Breit-Wigner a0(980) All other waves Breit-Wigners

  48. Work in progress – Mark Mattson, ICHEP 2010 • Techniques pioneered by Babar, extended and used by Belle, virtually eliminate major systematic effects: • F-B production asymmetry • Use odd moments • Charge efficiency asymmetry • Use data to calibrate, NOT Monte Carlo • Now used by CDF. Time-Integrated CPV from TeVatron Interesting  interestinger …

  49. New Time-Integrated CPV Results from Belle

  50. Mixing Measurements at BaBar and Belle Good vertex resolution allows measurement of time-dependence of D0 decays. Can eliminate distortion from B decays by cutting low momentumD0 ’s Excellent particle ID (Dirc and dE/dx) allows clean K/ separation • D0’s from D*+ D0+ decays: • Tag flavor of D0 by the sign of the “slow pion” in D* decays • Allow clean rejection of backgrounds • BUT untagged events can be used too !

More Related