aa r X i v : . [ h e p - e x ] N ov Proceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 Charm Physics Opportunities at a Super Flavor Factory
D. Asner
Carleton University, Ottawa, Canada
The primary physics goals of a high luminosity e + e − flavor factory are discussed, including the possibilitiesto perform detailed studies of the CKM mechanism of quark mixing, and constrain virtual Higgs and non-standard model particle contributions to the dynamics of rare B u,d,s decays. The large samples of D mesonsand tau leptons produced at a flavor factory will result in improved sensitivities to rare D processes - mixing, CPviolation and rare decays - and lepton flavor violation searches, respectively. Recent developments in acceleratorphysics have demonstrated the feasibility to build an accelerator that can achieve luminosities of O (10
6) cm − s − at √ s = 10 GeV. The capablity to run at √ s = 3 .
770 GeV with luminosity of 10 cm − s − is included inthe initial design. This report emphasizes the charm physics that can be probed at a Super Flavor Factory. These proceedings aim to present a brief overviewof the Super B effort with a special emphasis on thecharm physics program of such a facility. In the in-terest of completeness (and time) some passages fromthe Super B Conceptual Design Report[1] are repro-duced here.
1. Introduction
Elementary particle physics in the next decade willbe focused on the investigation of the origin of elec-troweak symmetry breaking and the search for exten-sions of the Standard Model (SM) at the TeV scale.The discovery of New Physics will likely produce aperiod of excitement and progress recalling the yearsfollowing the discovery of the
J/ψ . In this new world,attention will be riveted on the detailed elucidation ofnew phenomena uncovered at the LHC; these discov-eries will also provide strong motivation for the con-struction of the ILC. High statistics studies of heavyquarks and leptons will have a crucial role to play inthis new world.The two asymmetric B Factories, PEP-II [2] andKEKB[3], and their companion detectors, B A B AR [4]and Belle[5], have over the last seven years produced awealth of flavour physics results, subjecting the quarkand lepton sectors of the Standard Model to a seriesof stringent tests, all of which have been passed. Withthe much larger data sample made possible by a Su-per B Factory, qualitatively new studies will be possi-ble. These studies will provide a uniquely importantsource of information about the details of the NewPhysics uncovered at hadron colliders in the comingdecade.The continued detailed studies of heavy quark andheavy lepton (henceforth heavy flavour ) physics willnot only be pertinent in the next decade; they willbe central to understanding the flavour sector of NewPhysics phenomena. A Super Flavour Factory suchas Super B will be a partner with LHC, and eventu-ally, ILC, experiments, in ascertaining exactly whatkind of New Physics has been found. The capabilities of Super B in measuring CP -violating asymmetries invery rare b and c quark decays, accessing branchingfractions of heavy quark and heavy lepton decays inprocesses that are either extremely rare or forbiddenin the Standard Model, and making detailed investiga-tions of complex kinematic distributions will provideunique and important constraints in, for example, as-certaining the type of supersymmetry breaking or thekind of extra dimension model behind the new phe-nomena that many expect to be manifest at the LHC.The Super B Conceptual Design Report[1] is thefounding document of a nascent international enter-prise aimed at the construction of a very high lumi-nosity asymmetric e + e − Flavour Factory. A possiblelocation for Super B is the campus of the University ofRome “Tor Vergata”, near the INFN National Labo-ratory of Frascati.The exciting physics program that can be accom-plished with a very large sample of heavy quark andheavy lepton decays produced in the very clean envi-ronment of an e + e − collider; with a peak luminosityin excess of 10 cm − s − at the Υ (4 S ) resonanceis described in Ref.[1] and summarized below. Thisis program complementary to that of an experimentsuch as LHC b at a hadronic machine. The physicsreach of LHC b and Super B in the b -sector are com-pared in Figure 1. Luminosities of 10 cm − s − atthe ψ (3770) are expected. This report focuses on thecharm physics that can be probed both near the Υ (4 S )resonance and near charm production threshold.The conceptual design of a new type of e + e − col-lider that produces a nearly two-order-of-magnitudeincrease in luminosity over the current generation ofasymmetric B Factories is described in Ref.[1]. Thekey idea is the use of low emittance beams producedin an accelerator lattice derived from the ILC Damp-ing Ring Design, together with a new collision region,again with roots in the ILC final focus design, butwith important new concepts developed in this designeffort. Remarkably, Super B produces this very largeimprovement in luminosity with circulating currentsand wallplug power similar to those of the current B Factories. The design of an appropriate detector,
Proceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007
Figure 1: Comparison of Super B with 50 ab − and andupgraded LHCb 100 fb − . Design luminosity for Super B is15 ab − /year. Design luminosity for LHCb is 2 fb − /year.This comparison assumes that Super B does not integrateluminosity at the Υ (5 S ). An upgraded LHCb could inte-grate luminosity at a 10 times greater rate than LHCb. based on an upgrade of B A B AR as an example, is alsodiscussed in some detail in Ref. [1]. B By measuring mixing-dependent CP -violatingasymmetries in the B meson system for the first time,PEP-II/ B A B AR and KEKB/Belle have shown that theCKM phase accounts for all observed CP -violatingphenomena in b decays. The Unitarity Triangleconstruction provides a set of unique overconstrainedprecision tests of the self-consistency of the threegeneration Standard Model. Figure 2 shows thecurrent status of the Unitarity Triangle construc-tion, incorporating measurements from B A B AR andBelle, as well as the B s mixing measurement ofCDF; the addition of CP asymmetry measurements,together with the improvement in the precision of CP -conserving measurements, has made this uniquelyprecise set of Standard Model tests possible.The fact that the CKM phase has now beenshown to be consistent with all observed CP -violatingphenomena is both a triumph and an opportunity.In completing the experimentally-verified StandardModel ansatz (except, of course, for the Higgs), it ! -1 -0.5 0 0.5 1 1.5 2 " -1.5-1-0.500.511.5 ! -1 -0.5 0 0.5 1 1.5 2 " -1.5-1-0.500.511.5 % % d m & K ’ K ’ d m & & s m & ub V $ sin2 < 0 $ sol. w/ cos2(excl. at CL > 0.95) excluded area has CL > 0.95 e xc l uded a t C L > . Summer 2007
CKM f i t t e r
Figure 2: Global fit of the Unitarity Triangle constructionas of LeptonPhoton 2007 conference. intensifies the mystery of the creation of the baryon-antibaryon asymmetry of the universe: the observed CP -violation is too small for the Standard Model toaccount for electroweak baryogenesis. This intriguingresult opens the door to two possibilities: the matterantimatter asymmetry is produced by another mech-anism, such as leptogenesis, or baryogenesis proceedsthrough the additional CP -violating phases that natu-rally arise in many extensions of the Standard Model.These extra phases produce measurable effects in theweak decays of heavy flavour particles. The detailedpattern of these effects, as well as of rare decay branch-ing fractions and kinematic distributions, is, in fact,diagnostic of the characteristics of New Physics at orbelow the TeV scale,By the end of this decade, the two B Factories willhave accumulated a total of ∼ − . Even at thislevel, most important measurements pertinent to theUnitarity Triangle construction will still be statisticslimited: an even larger data sample would provideincreasingly stringent tests of three-generation CKMunitarity. There are two main thrusts here. Thefirst is the substantial remaining improvement thatcan still be made in the Unitarity Triangle construc-tion. Here measurements in B , D and τ decay playan important role, as do improvements in lattice QCDcalculations of hadronic matrix elements. This impor-tant physics goal is NOT , however, the sole, or eventhe primary, motivation for a Super B Factory. Theprecision of our knowledge of the Unitarity Trianglewill perforce improve to the limit allowed by theo-retical uncertainties as we pursue the primary goal:improving the precision of the measurement of CP roceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 b → s transitions, to a level where there is substan-tial sensitivity to New Physics effects. This requiresdata samples substantially larger than the current B Factories will provide. Some of these measurementsare accessible at the LHC [6], but the most promisingapproach to this physics is Super B , a very high lumi-nosity asymmetric B Factory, which is also, of course,a Super Flavour Factory, providing large samples of b and c quark and τ lepton decays.Super B , having an initial luminosity of 10 cm − s − , will collect 15 ab − in a New SnowmassYear [7], or 75 ab − in five years. A data samplethis large will make the Unitarity Triangle tests, intheir manifold versions, the ultimate precision test ofthe flavour sector of the Standard Model, and openup the world of New Physics effects in very rare B , D , and τ decaysA primary tool for isolating new physics is the time-dependent CP asymmetry in decay channels that pro-ceed through penguin diagrams, such as the b → s ¯ ss processes B d → φK and B d → ( K ¯ K ) CP K or sim-ilar transitions such as B d → η ′ K , B d → f K , B d → π K , B d → ρ K , B d → ωK , and B d → π π K . The dominant contribution to these decaysis the combination of CKM elements V tb V ∗ ts ; these am-plitudes have the same phase as the charmonium chan-nels b → c ¯ cs , up to a small phase shift of V ts withrespect to V cb . New heavy particles contribute newloop amplitudes, with new phases that can contributeto the CP asymmetry and the S coefficient of the time-dependent analysis, so that the measured CP violationparameter could be substantially different from sin 2 β .Physics beyond the Standard Model can affect rare B decay modes, through observables such as branch-ing fractions, CP -violating asymmetries and kine-matic distributions. These decays do not typicallyoccur at tree level, and thus their rates are stronglysuppressed in the Standard Model. Substantial en-hancements in the rates and/or variations in angu-lar distributions of final state particles could resultfrom the presence of new heavy particles in loop di-agrams, resulting in clear evidence of New Physics.Moreover, because the pattern of observable effectsis highly model-dependent, measurements of severalrare decay modes can provide information regardingthe source of the New Physics. An extended run at the Υ (5 S ) is also contemplated; such a run would yield awealth of interesting new B s decay results.The Super B data sample will also contain unprece-dented numbers of charm quark and τ lepton decays.This data is also of great interest, both for its capac-ity to improve the precision of existing measurementsand for its sensitivity to New Physics. This interestextends beyond weak decays; the detailed explorationof new charmonium states is also an important ob-jective. Limits on rare τ decays, particularly lepton- flavour-violating decays, already provide importantconstraints on New Physics models. Super B may havethe sensitivity to actually observe such decays. Theaccelerator design will allow for longitudinal polariza-tion of the e − beam, making possible uniquely sensi-tive searches for a τ electric dipole moment, as well asfor CP -violating τ decays.Some measurements in charm and τ physics are bestdone near threshold. Super B also has the capabilityof running in the 4 GeV region. Short runs at spe-cific center-of-mass energies in this region, represent-ing perhaps 10% of data taking time, would producedata samples substantially larger than those currentlyenvisioned to exist in the next decade. The many ad-vantages of analysis at threshold are enumerated inSection 2.1 B Design
Given the strong physics motivation, there has beena great deal of activity over the past few years aimedat designing an e + e − B Factory that can produce sam-ples of B mesons 50 to 100 times larger than will existwhen the current B Factory programs end. Severalapproaches were tried before the design[1] describedbriefly here was developed.Upgrades of PEP-II [8] and KEKB [9] to Super B Factories that accomplish this goal have been pro-posed at SLAC and at KEK. These machines are ex-trapolations of the existing B Factories, with highercurrents, more bunches, and smaller β functions (1.5to 3 mm). They also use a great deal of power ( ≥ B Factory designs stimulated a new ap-proach, using low emittance beams, to constructing aSuper B Factory with a luminosity of 10 , but withreduced power consumption [10].The current machine concept, which has roots inILC R&D: a very low emittance storage ring, basedon the ILC damping ring minimum emittance growthlattice and final focus, that incorporates several novelaccelerator concepts and appears capable of meetingall design criteria, while reducing the power consump-tion, which dominates the operating costs of the facil-ity, to a level similar to that of the current B Factories.
Proceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007
Due to similarities in the design of the low emittancerings and the final focus, operation of Super B canserve as a system test for these linear collider compo-nentsBy utilizing concepts developed for the ILC damp-ing rings and final focus in the design of the Super B collider, it is possible to produce a two-order-of-magnitude increase in luminosity with beam currentsthat are comparable to those in the existing asym-metric B Factories. Background rates and radiationlevels associated with the circulating currents are com-parable to current values; luminosity-related back-grounds such as those due to radiative Bhabhas, in-crease substantially. With careful design of the in-teraction region, including appropriate local shield-ing, and straightforward revisions of detector compo-nents, upgraded detectors based on B A B AR or Belle area good match to the machine environment: in this dis-cussion, we use B A B AR as a specific example. Requireddetector upgrades include: reduction of the radius ofthe beam pipe, allowing a first measurement of trackposition closer to the vertex and improving the ver-tex resolution (this allows the energy asymmetry ofthe collider to be reduced to 7 on 4 GeV); replace-ment of the drift chamber, as the current chamberwill have exceeded its design lifetime; replacement ofthe endcap calorimeter, with faster crystals having asmaller Moli`ere radius, since there is a large increasein Bhabha electrons in this region.The Super B design has been undertaken subjectto two important constraints: 1) the lattice is closelyrelated to the ILC Damping Ring lattice, and 2) asmany PEP-II components as possible have been in-corporated into the design. A large number of PEP-IIcomponents can, in fact, be reused: The majority ofthe HER and LER magnets, the magnet power sup-plies, the RF system, the digital feedback system, andmany vacuum components. This will reduce the costand engineering effort needed to bring the project tofruition.The Super B concept is a breakthrough in colliderdesign. The invention of the “crabbed waist” finalfocus can, in fact, have impact even on the currentgeneration of colliders. A test of the crabbed waistconcept is planned to take place at Frascati in 2007; apositive result of this test would be an important mile-stone as the Super B design progresses. The low emit-tance lattice, fundamental as well to the ILC dampingring design, allow high luminosity with modest powerconsumption and demands on the detector.Super B appears to be the most promising approachto producing the very high luminosity asymmetric B Factory that is required to observe and explore thecontributions of physics beyond the Standard Modelto heavy quark and τ decays.
2. Charm Physics at Super B It is a truth universally accepted that charm studiesplayed a seminal role in the evolution and acceptanceof the Standard Model. Yet the continuing impor-tance of this sector is not widely appreciated, sincethe Standard Model electroweak phenomenology forcharm decays is on the dull side: the CKM parame-ters are known, D D oscillations are slow, CP asym-metries are small or absent and loop-driven decays areextremely rare.Yet on closer examination, a strong case emerges intwo respects, both of which derive from this apparentdullness: • Detailed and comprehensive analyses of charmtransitions will continue to provide us with newinsights into QCD’s nonperturbative dynamics,and advance us significantly towards establish-ing theoretical control over them. Beyond theintrinsic value of such lessons, they will also cal-ibrate our theoretical tools for B studies; thiswill be essential to saturate the discovery po-tential for New Physics in B transitions. • Charm decays constitute a novel window intoNew Physics.Lessons from the first item will have an obvious im-pact on the tasks listed under the second. They mightactually be of great value even beyond QCD, if theNew Physics anticipated for the TeV scale is of thestrongly interacting variety.The capabilities of a Super Flavour Factory are wellmatched to these goals. It allows uniquely clean de-terminations of CKM parameters, with six of the ninematrix elements impacted by charm measurements.New Physics signals can easily exceed Standard Modelpredictions by considerable factors such that there willbe no ambiguity in interpreting them, yet they are un-likely to be large; this again requires the clean environ-ment and huge statistics that a Super Flavour Factorycan provide.A number of other facilities either currently runningor soon to commence operation provide competitionin the area of charm physics. The current B Factoryprogram is expected to produce a sample of about 10 charm hadrons from operation at or near the Υ (4 S )resonance. The CLEO- c experiment at CESR is op-erating in the charm threshold region, and anticipatescollecting a total of 5 × D D pairs and about7 × D ∗ + s D − s + D + s D ∗− s through coherent produc-tion. The BESIII experiment at BEPCII expects first e + e − collisions in 2008, and will collect large charmo-nium samples, in addition to exceeding the CLEO- c data set in open charm production. Although therewill be no successors to the Fermilab fixed targetcharm production experiments, the LHC will producecopious quantities of charm (notably, charm physics roceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 b physics program); these areexpected to result in very large samples of charmedhadrons in final states with reconstructible topologies.Most of the benchmark charm measurements willstill be statistics-limited after the CLEO- c , BESIIIand B Factory projects, and many will not be achiev-able in a hadronic environment. Super B is the idealmachine with which to pursue these measurements totheir ultimate precision. Operation near the Υ (4 S )will provide enormous samples of charm hadrons,in a clean environment and with a detector well-suited for charm studies. The charm physics programwould benefit further from the ability to operate inthe threshold region, in order to exploit the quan-tum correlations associated with coherent production.The expected lower luminosity at threshold would bepartly compensated by the higher production cross-section, resulting in a comparable charm productionrate. To estimate the reach of Super B from operationat the charm threshold, we have assumed a simpledependence of the luminosity on the center-of-massenergy: L peak ∝ s . Thus, we expect that Super B (which will integrate ∼
15 ab − per year operatingat the Υ (4S)) can accumulate ∼
150 fb − per monthwhen operated at the ψ (3770). The production rate of charm during threshold run-ning at a Super B and Υ (4S) running is comparable.Although the luminosity for charm threshold runningis expected to be an order of magnitude lower, theproduction cross section is 3 times higher than at √ s = 10 .
58 GeV. Charm threshold data has distinctand powerful advantages over continuum and b → c charm production data accumulated above B produc-tion threshold. Charm Events at Threshold are ExtremelyClean:
The charged and neutral multiplicites in ψ (3770) events are only 5.0 and 2.4 - approximately1/2 the multiplicity of continuum charm productionat √ s = 10 .
58 GeV.
Charm Events at Threshold are pure DD : Noadditional fragmentation particles are produced. Thesame is true for √ s = 4170 MeV production of DD ∗ , D + s D − s , D + s D ∗− s , and for threshold productionof Λ c ¯ Λ c . This allows use of kinematic constraints,such as total candidate energy and beam constrainedmass, and permits effective use of missing mass meth-ods and neutrino reconstruction. The crisp definitionof the initial state is a uniquely powerful advantageof threshold production that is absent in continuumcharm prodution. Double Tag Studies are Pristine:
The pure pro-duction of DD states, together with low multiplic-ity and large branching ratios characteristic of many D decays permits effective use of double-tag studies in which one D meson is fully reconstructed and therest of the event is examined without bias but withsubstantial kinematic knowledge. The techniques pi-oneered by Mark III and extended by CLEO-c[13, 14]allow precise absolute branching fraction determina-tion. Backgrounds under these conditions are heavilysuppressed which minimizes both statistical errors andsystematic uncertainties. Signal/Background is Optimum at Threshold:
The cross section for the signal ψ (3770) → DD isabout 1/2 the cross section for the underlying contin-uum e + e − → hadrons background. By contrast, for c ¯ c production at √ s = 10 .
58 GeV the signal is only1/4 of the total hadronic cross section.
Neutrino Reconstruction:
The undetected energyand momentum is interpreted as the neutrino four-vector. For leptonic and semileptonic charm decaysthe signal is observed in missing mass squared dis-tributions and for double-tagged events these mea-surements have low backgrounds. The missing massresolution is about one pion mass. For semileptonicdecays the q resolution is excellent, about 3 timesbetter than in continuum charm reconstruction at √ s = 10 .
58 GeV. Neutrino reconstruction at thresh-old is clean.
Quantum Coherence:
The production of D and D in a coherent C = − ψ (3770) de-cay is of central importance for the subsequent evo-lution and decay of these particles. The same is truefor DD ( n ) π ( m ) γ produced at √ s ∼ C = − m and C = +1 for odd m . The coher-ence of the two initial state D mesons allows both sim-ple and sophisticated methods to measure DD mixingparameters, strong phases, CP eigenstate branchingfractions, and CP violation[15, 16, 17, 18, 19]. Detailed analyses of (semi)leptonic decays of charmhadrons provide a challenging test bed for validatinglattice QCD (LQCD), which is the only known frame-work with realistic promise for a truly quantitativetreatment of charm hadrons that can be systemati-cally improved . Such studies form the core of theongoing CLEO- c and the nascent BESIII programs;they are also pursued very profitably at the B Fac-tories. Central goals are measuring the decay con-stants f D + and f D s and going beyond total rates forsemileptonic D + , D and D + s decays. on the Cabibboallowed and forbidden level by extracting the form fac-tors etc. It is essential to analyze lepton spectra andperform “meaningful” Dalitz plot studies. To quan-tify “meaningful” we can compare to analyses on K e decays. With a sample size of 30,000 events as it be-came available in 1977 one was able to begin extract-ing dynamical information. Precise measurements arepossible now with NA48/2 and E685 each having ac- Proceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 cumulated 400,000 events. For charm we are nowherenear that level yet: CLEO- c will have about 10,000semileptonic charm decays – comparable to kaon stud-ies in the late 1970s. Since for charm the phase space islarger (actually a good thing, since it opens up moredomains of interest) it seems reasonable to aim forsample sizes of 10 events. Again, this is well beyondthe reach of CLEO- c and most probably of BESIII aswell. Such high quality studies will greatly improveour understanding of hadronization and provide aneven richer test bed for LQCD with the lessons tobe learned of crucial importance for extracting V ub from semileptonic B decays. Our knowledge of charmbaryon decays is also rather limited; e.g. , no precisiondata on absolute branching ratios or semileptonic de-cay distributions exist. CLEO- c will not run abovethe charm baryon threshold, and BESIII cannot. In the Standard Model the leptonic decay width isgiven by [20]:Γ( D + → ℓ + ν ) = G F π f D + m ℓ M D + (cid:18) − m ℓ M D + (cid:19) | V cd | Γ( D + s → ℓ + ν ) = G F π f D + s m ℓ M D + s − m ℓ M D + s ! | V cs | . (1)Taking | V cd | and | V cs | from elsewhere, one uses Eq.(1)to extract f D + and f D + s . The ratio R ℓ of the lep-tonic decay rates of the D + s and the D + is propor-tional to ( f D + s /f D + ) , for which the lattice calculationis substantially more precise. A significant deviationfrom its predicted value would be a clear sign of NewPhysics, probably in the form of a charged Higgs ex-change [21]. On the other hand, the ratio of the ratesof tauonic and muonic decays for either D + or D + s isindependent of both form factors and CKM elements,and serves as a useful cross-check in this context.CLEO- c has published a measurement of f D + [22,23, 24], and several measurements of f D + s [25, 26, 27].These measurements have benefitted from a “double-tag” method uniquely possible at threshold, where a D +( s ) D − ( s ) pair is produced with no extra particles. Thelatest results are f D + = (222 . ± . +2 . − . ) MeV . (2) f D s = (275 ± ±
5) MeV (3) f D + s /f D + = 1 . ± . ± . . (4) B A B AR has also measured f D s = (283 ± ± ±
14) MeV[28]. The central values for f D + s and f D + s /f D + are slightly above, but consistent with, thepresent LQCD calculations. It is important to note that the desired 1–3% accuracy level has not yet beenreached on either the experimental or theoretical side.While LQCD practitioners expect to reach this levelover the next decade, the experimental precision islikely to fall significantly short of this goal, even afterBESIII. Since larger statistics will certainly allow re-duction of the systematic errors in the current results,it is clear that data accumulated by Super B from arelatively short run ( ∼ O (1%) level will have impor-tant consequences for B d and B s oscillations, since itwould give us demonstrated confidence in the theoret-ical extrapolation to f B d and f B s /f B d . In the area of semileptonic decays, CLEO- c hasmade the most accurate measurements for the inclu-sive D and D + semileptonic branching fractions – B ( D → Xℓν ℓ ) = (6 . ± . ± . B ( D + → Xℓν ℓ ) = (16 . ± . ± . D + s . Such data provide important“engineering input” for other D and B decay studies.However, a central goal must be to go beyond the totalrates for these decays and to extract the form factors etc. In order to do so, it is essential to analyze leptonspectra and perform “meaningful” Dalitz plot studies.To quantify “meaningful”, it is instructive to com-pare to analyses on K e decays. With a sample size of30,000 events which became available in 1977, one wasable to begin extracting dynamical information. Pre-cise measurements are now possible, with NA48/2 andE685 each having accumulated 400,000 events [30, 31].For charm we are nowhere near that level: CLEO- c will have about 10,000 semileptonic charm decays –comparable to kaon studies in the late 1970s. Sincefor charm the phase space is larger, thereby openingmore domains of interest, a reasonable target samplesize is 10 events, which is far beyond the reach ofCLEO- c , and most probably, of BESIII.Three-family unitarity constraints on the CKM ma-trix yield rather precise values for | V cs | and | V cd | . Us-ing these, one can extract the form factors from anal-yses of exclusive semileptonic charm decays. Both thenormalization and q dependence must be measured.Existing LQCD studies do not allow us to determinethe latter from first principles; instead a parametriza-tion originally proposed by Becirevic and Kaidalov( BK ) is used [32]. Recent and forthcoming resultsfrom CLEO- c , B A B AR and Belle [33, 34] are expectedto be statistics limited, and will not reach the desired1–3% level.The current status can be characterized by com-paring the measured value of the ratio R sl , which isindependent of | V cd | , to that inferred from a recent roceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 R sl = s Γ( D + → µ + ν µ )Γ( D → πeν e ) = (cid:26) . ± .
019 (exp)0 . ± .
028 (theo) . (5)The values are nicely consistent, yet both are still farfrom the required level of precision.While operation in the Υ region will produce largequantities of charm hadrons, there are significantbackgrounds and one pays a price in statistics whenusing kinematic constraints to infer neutrino mo-menta, etc. . On the other hand, even a limited run atcharm threshold will generate the statistics requiredto study (semi)leptonic decays with the desired accu-racy. Assuming that systematic uncertainties in track-ing and muon identification will provide a limit to theprecision at the 0 .
5% level, we estimate the integratedluminosity from threshold running required to achievea similar statistical uncertainty. As shown in Table Iwe expect to be able to measure f D + , f D s and their ra-tio with better than 0 .
5% statistical uncertainty withintegrated luminosities of at least 100 fb − . Table I Statistics required to obtain 0 .
5% statisticaluncertainties on corresponding branching fractions usingdouble-tagged events, when running at threshold.Channel Integrated luminosity(fb − ) D + → µ + ν µ D + s → µ + ν µ For semileptonic decays, a case-by-case study is nec-essary. One also has to distinguish between merelydetermining the branching ratio and performing a“meaningful” Dalitz plot analysis, as discussed above.The required integrated luminosities are given in Ta-ble II. It is clear that the ∼
150 fb − anticipated fromone month of running in the threshold region wouldprovide the desired statistics for most measurements.Note that while D s mesons are not produced at the ψ (3770), short runs at other energies are possible. Studies of leptonic decay constants and semilep-tonic form factors will yield a set of measurements,including | V cd | and | V cs | , at the few percent level.These measurements will constrain theoretical calcu-lations, and those that survive will be validated foruse in a variety of areas in which interesting physicscannot be extracted without theoretical input. This broader impact of charm measurements extends be-yond those measurements that can be performed di-rectly at charm threshold, and has a large impact onthe precision determination of CKM matrix elements.The determination of | V td | and | V ts | is limited byignorance of f B p B B d and f B s p B B s ; improved de-terminations of f B and f B s are required. Precisionmeasurements of f D and f D s can validate the the-oretical treatment of the analogous quantities for B mesons. Similarly, improved form factor calculationsin the decays D → πℓν and D → ρℓν and inclusivesemileptonic charm decays will enable improved pre-cision in | V ub | and | V cb | .The precision measurement of the UT angle γ de-pends on decays of B mesons to final states con-taining neutral D mesons. A variety of charm mea-surements impact these analyses, including: improvedconstraints on charm mixing amplitudes, – impor-tant for the GLW method [36, 37], measurements ofrelative rates and strong phases between Cabibbo-favoured and -suppressed decays measurement of therelative rate and relative strong phase δ between D and D decay to K + π − – important for ADSmethod[38, 39], and studies of charm Dalitz plotstagged by hadronic flavor or CP content [40, 41, 42].Note that the latter two measurements can only beperformed with data from charm threshold. At present three-family unitarity constraints yieldmore precise values for | V cs | and | V cd | than direct mea- Table II Statistics required to obtain 0 .
5% statistical un-certainties on corresponding branching fractions (column2) or one million signal events (column 3) using doubletagged events, when running at threshold.Channel Integrated luminosity Integrated luminosity(fb − ) (fb − ) D → K − e + ν e D → K ∗− e + ν e
17 425 D → π − e + ν e
20 500 D → ρ − e + ν e
45 1125 D + → K S e + ν e D + → ¯ K ∗ e + ν e D + → π e + ν e
75 1900 D + → ρ e + ν e
110 2750 D + s → φe + ν e
85 2200 D + s → K S e + ν e D + s → K ∗ e + ν e Proceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 surements. Since it is conceivable that a fourth familyexists (with neutrinos so heavy that the Z could notdecay into them), one would like to obtain more accu-rate direct determinations. This should be possible ifLQCD is indeed validated at the O (1%) level throughits predictions on form factors and their ratios.From four-family unitarity, and using current ex-perimental constraints [43] we can infer for a fourthquark doublet ( t ′ , b ′ ): | V cb ′ | = p − | V cd | − | V cs | − | V cb | < ∼ . , (6) | V t ′ s | = p − | V us | − | V cs | − | V ts | < ∼ . . (7)These loose bounds are largely due to the 10% erroron | V cs | . While significant progress can be guaranteed for theStandard Model studies outlined above, the situationis much less certain concerning the search for NewPhysics. No sign of it has yet been seen, but we haveonly begun to approach the regime of experimentalsensitivity in which a signal for New Physics could re-alistically emerge in the data. The interesting regionof sensitivity extends several orders of magnitude be-yond the current status.New Physics scenarios in general induce flavor-changing neutral currents that a priori have no reasonto be as strongly suppressed as in the Standard Model.More specifically, they could be substantially strongerfor up-type than for down-type quarks; this can oc-cur in particular in models that reduce strangeness-changing neutral currents below phenomenologicallyacceptable levels through an alignment mechanism.In such scenarios, charm plays a unique role amongthe up-type quarks u , c and t ; for only charm allowsthe full range of probes for New Physics. Since topquarks do not hadronize [44], there can be no T ¯ T oscillations (recall that hadronization, while hard tobring under theoretical control, enhances the observ-ability of CP violation). As far as u quarks are con-cerned, π , η and η ′ do not oscillate, and decay electro-magnetically, not weakly. CP asymmetries are mostlyruled out by CP T invariance. Our basic contentioncan then be formulated as follows: charm transitionsprovide a unique portal for a novel access to flavor dy-namics with the experimental situation being a priori quite favourable. The aim is to go beyond “merely”establishing the existence of New Physics around theTeV scale – we want to identify the salient features ofthis New Physics as well. This requires a comprehen-sive study, i.e. , that we also search in unconventionalareas such as charm decays.
In a scenario in which the LHC discovers directevidence of SUSY via observation of sleptons or squarks, the Super Flavour Factory program becomeseven more important. The sfermion mass matricesare a new potential source of flavor mixing and CP violation and contain information about the SUSY-breaking mechanism. Direct measurements of themasses can only constrain the diagonal elements ofthis matrix. However, off-diagonal elements can bemeasured through the study of loop-mediated heavyflavor processes. As a specific example, a minimal fla-vor violation scenario such as mSUGRA with mod-erate tan β , could result in a SUSY partner massspectrum that is essentially indistinguishable from anSU(5) GUT model with right-handed neutrinos. How-ever the mSUGRA scenario would be expected toyield no observable effects in the heavy flavor sec-tor, whereas the SU(5) model is expected to producemeasurable effects in time-dependent CP violation inpenguin-mediated hadronic and radiative decays.While there is no compelling scenario that wouldgenerate observable effects in charm, but not in beautyand strange decays, it is nevertheless reassuring thatsuch scenarios do exist. One should keep in mindthat New Physics signals in charm CP asymmetriesare particularly clean, since the Standard Model back-ground (which often exists in B decays) is largely ab-sent. The consequence is that New Physics could pro-duce signals that exceed Standard Model predictionsby an order of magnitude or more – something thatis of great help in interpreting the signals. We willfocus on the most promising areas; more details canbe found in several recent reviews [17, 45, 46]. D D oscillations Oscillations of neutral D mesons driven by the twoquantities x D = ∆ M D / Γ D and y D = ∆Γ D / D lead to an effective violation of the Standard Model∆ C = ∆ Q and ∆ C = ∆ S rules in semileptonic andnonleptonic channels. The status of the StandardModel prediction can be summarized as [17]: whileone predicts x D ∼ O (10 − ) ∼ y D , at present one can-not rule out x D , y D ∼ . D D oscillations (or another processwith origin beyond the Standard Model). The wrong-sign hadronic decay D → K + π − is sensitive to linearcombinations of the mass and lifetime differences, de-noted x ′ D and y ′ D . The relation of these parametersto x D and y D is controlled by a strong phase differ-ence. Direct measurements of x D and y D indepen-dent of unknown strong interaction phases, can alsobe made using time-independent studies of amplitudespresent in multi-body decays of the D , for example, D → K S π + π − . Direct evidence for y D = 0 can alsoappear through lifetime differences between decays to CP eigenstates. The measured quantity in this case, roceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 Figure 3: Likelihood contours in the ( x D , y D ) plane fromHFAG [54]. These preliminary world averages use all avail-able charm mixing results. y CP , is equivalent to y D in the absence of CP violation.Another approach is to study quantum correlationsnear threshold [17, 18, 19] in e + e − → D D ( π ) andin e + e − → D D γ , which yield C -odd and C -even D D pairs, respectively.Very recently, several new results have suggestedthat charm mixing may be at the upper end of therange of Standard Model predictions. B A B AR findsevidence for oscillations in D → K + π − with 3 . σ significance [47], while Belle sees a 3 . σ effect in D → K + K − , with results using D → K S π + π − supporting the claim [48]. These results are consis-tent with previous measurements, some of which hadhinted at a mixing effect [49, 50, 51, 52, 53]. Theresults are not systematics limited, and further im-provements are anticipated.The charm decays subgroup of the Heavy FlavorAveraging Group [54] is preparing world averages ofall the charm mixing measurements, taking into ac-count correlations between the measured quantities.A preliminary average is available, giving: x D = (cid:0) . +3 . − . (cid:1) × − and y D = (cid:0) . +2 . − . (cid:1) × − .Contours in the ( x D , y D ) plane are shown in Fig. 3.The significance of the oscillation effect in the prelim-inary world averages exceeds 5 σ .At present no clear signal has emerged. Since nosingle measurement exceeds 5 σ significance, it is tooearly to consider charm oscillations as definitively es-tablished. Nonetheless, even if one accepts the cen-tral The interpretation of these new results in termsof New Physics is inconclusive. For one thing, it is not yet clear whether the effect is caused by x D = 0or y D = 0 or both, though the latter is favored andthis point may be clarified soon. As shown in Ta-ble III, Super B will be able to observe both lifetimeand mass differences in the D system, if they lie inthe range of Standard Model predictions. It shouldbe noted that the full benefit of measurements in the D → K + π − system (and indeed for other hadronicdecays) can only be obtained if the strong phases aremeasured. This can be achieved with a short ( ∼ CP asymmetries that do provideunequivocal New Physics signals. Table III Summary of the expected precision on charmmixing parameters. For comparison we put the reach ofthe B Factories at 2 ab − . The estimates for Super B assume that systematic uncertainties can be kept undercontrol.Mode B Factories Super B (2 ab − (75 ab − D → K + K − y CP × − × − D → K + π − y ′ D × − × − x ′ D × − × − D → K S π + π − y D × − × − x D × − × − Average y D × − × − x D × − × − CP Violation With and Without Oscillations
Several factors favor dedicated searches for CP vio-lation in charm transitions: • Within the Standard Model, the effective weakphase is highly diluted, namely ∼ O ( λ ), and can ariseonly in singly-Cabibbo-suppressed transitions, whereone expects asymmetries to reach the O (0 . D ± → K S π ± [17] where the CP impurity in K S induces an asymmetry of 3 . × − .CLEO-c measures A CP = ( − . ± . ± . Proceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 • Strong phase shifts required for direct CP viola-tion to emerge are, in general, large, as are the branch-ing ratios into relevant modes. Although large finalstate interactions complicate the interpretation of anobserved signal in terms of the microscopic parametersof the underlying dynamics, they enhance its observ-ability. • With the Standard Model providing one ampli-tude, observable CP asymmetries can be linear in NewPhysics amplitudes – unlike the case for rare decays –thus increasing the sensitivity. • Decays to multibody final states contain more dy-namical information than given by their widths; theirdecay distributions as described by Dalitz plots or T -odd moments can exhibit CP asymmetries that mightbe considerably larger than those for the width. Finalstate interactions, while not necessary for the emer-gence of such effects, can produce a signal that can bedisentangled from New Physics effects by comparing T -odd moments for CP conjugate modes [55]. • The distinctive channel D ∗± → Dπ ± provides apowerful tag on the flavor identity of the neutral D meson.The notable “fly in the ointment” in searching for CP violation in the charm sector is that D D os-cillations are slow. Nevertheless one should acceptthis challenge: CP violation involving D D oscilla-tions is a reliable probe of New Physics: the asym-metry is controlled by sin(∆ m D t ) × Im ( q/p )¯ ρ ( D → f ). In the Standard Model both factors are small,namely ∼ O (10 − ), making such an asymmetry un-observably tiny – unless there is New Physics (see, e.g. , [56, 57]). D D oscillations, CP violation andNew Physics might thus be discovered simultaneouslyin a transition. Such effects can be searched for infinal states common to D and D such as CP eigen-states ( e.g. , D → K + K − ) doubly Cabibbo sup-pressed modes ( e.g. , D → K + π − ) or three-body fi-nal states ( e.g. D → K S π + π − ). Undertaking time-dependent Dalitz plot studies[48, 53] requires a highinitial overhead, yet in the long run this should payhandsome dividends, since Dalitz plot analyses can in-voke many internal correlations that, in turn, serve tocontrol systematic uncertainties. Such analyses mayallow the best sensitivity to New Physics. Direct CP violation CP violation in ∆ C = 1 dynamics can be searchedfor by comparing partial widths for CP conjugatechannels. For an observable effect two conditions haveto be satisfied simultaneously: a transition must re-ceive contributions from two coherent amplitudes with(a) different weak and (b) different strong phases.While condition (a) is just the requirement of CP vi-olation in the underlying dynamics, condition (b) isneeded to make the relative weak phase observable.Since the decays of charm hadrons proceed in the nearby presence of many hadronic resonances induc-ing virulent final state interactions (FSI), requirement(b) is in general easily met; thus it provides no draw-back for the observability of a CP asymmetry – albeitit does for its microscopic interpretation.As already mentioned CKM dynamics does not sup-port any CP violation in Cabibbo allowed and doublysuppressed channels due to the absence of a secondweak amplitude. In singly Cabibbo suppressed tran-sitions one expects CP asymmetries, albeit highly di-luted ones of order λ ∼ − or less [56]. CP asymmetries involving oscillations For final states that are common to D and ¯ D decays one can search for CP violation manifestingitself with the help of D – ¯ D oscillations in quali-tative – though certainly not quantitative – analogyto B d → J/ψK S . Such common states can be CP eigenstates – like D → K + K − /π + π − /K S η ( ′ ) –, butdo not have to be: two very promising candidatesare D → K S π + π − , where one can bring the fullDalitz plot machinery to bear, and D → K + π − vs. ¯ D → K − π + , since its Standard Model am-plitude is doubly Cabibbo suppressed. Undertakingtime-dependent Dalitz plot studies requires a higherinitial overhead, yet in the long run this should payhandsome dividends exactly since Dalitz analyses caninvoke many internal correlations that in turn serveto control systematic uncertainties. Time-integrated CP asymmetries have beensearched for and sensitivities of order 1% [several%] have been achieved for Cabibbo-allowed and-singly suppressed modes with two [three] body finalstates [58]. A Dalitz-plot analysis of time-integrated CP asymmetries provides constraints O (10 − )[59].Time-dependent CP asymmetries ( i.e. , those in-volving D D oscillations) are still largely terraincognita .Since the primary goal is to establish the interven-tion of New Physics, one “merely” needs a sensitivitylevel above the reach of the Standard Model; “merely”does not mean this can easily be achieved. As far asdirect CP violation is concerned, this means asym-metries down to the 10 − or 10 − level in Cabibbo-allowed channels and down to the 1% level or betterin doubly Cabibbo-suppressed modes. In Cabibbo-singly-suppressed decays one wants to reach the 10 − range (although CKM dynamics can produce effects ofthat order, future advances might sharpen the Stan-dard Model predictions). For time-dependent asym-metries in D → K S π + π − , K + K − , π + π − etc. , and in D → K + π − , one should strive for the O (10 − ) and O (10 − ) levels, respectively.When striving to measure asymmetries below the1% level, one has to minimize systematic uncertain- roceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 x D and y D , which requiresexcellent vertex detectors; ii) Dalitz plot consistencychecks; iii) quantum statistics constraints on distribu-tions, T -odd moments, etc. [18]. In this section we briefly summarize the experimen-tal reach of Super B for New Physics sensitive channelsin the charm sector. Table IV shows the expected90% confidence level upper limits that may be ob-tained on various important rare D decays, includ-ing suppressed flavor-changing neutral currents, lep-ton flavor-violating and lepton number-violating chan-nels, from one month of running at the ψ (3770). It isexpected that the results from running at the Υ (4 S )will be systematics limited before reaching this preci-sion.For studies of D D mixing, running in the Υ re-gion appears preferable, and, if the true values of themixing parameters are unobservably small, the upperlimits on both x D and y D can be driven to below 0 . D → K + π − , K + K − , K S π + π − , etc. ) Therefore, Super B can study charm mixing if x D and y D lie within the ranges predicted by the Stan-dard Model, and recently observed. The sensitivityto mixing-induced CP violation effects obviously de-pends strongly on the size of the mixing parameters. Ifone or both of x D and y D are O (1%), as indicated bythe most recent results, Super B will be able to makestringent tests of New Physics effects in this sector.The situation for searches of direct CP violationis clearer: the Super B statistics will be sufficient toobserve the Standard Model effect of ∼ × − in D + → K S π + [17], and other channels can be pursuedto a similar level. Within three body modes, uncer-tainties in the Dalitz model are likely to become thelimiting factor. However, model-independent T -oddmoments can be constructed in multibody channels,and limits in the 10 − region appear obtainable. B One does not have to be an incorrigible optimist toargue that the best might still be ahead of us in theexploration of the weak decays of charm hadrons. De-tailed studies of leptonic and semileptonic charm de-cays will allow experimental verification of improve-ments in lattice QCD calculations, down to the re-quired O (1%) level of precision. This will result insignificant improvements in the precision of CKM ma-trix elements. The possibility to operate with e + e − collision energies in the charm threshold region furtherextends the physics reach and the charm program ofthe Super Flavour Factory. Table IV Expected 90% confidence level upper limits thatmay be obtained on various important rare D decays, from1 month of Super B running at the ψ (3770).Channel Sensitivity D → e + e − , D → µ + µ − × − D → π e + e − , D → π µ + µ − × − D → ηe + e − , D → ηµ + µ − × − D → K S e + e − , D → K S µ + µ − × − D + → π + e + e − , D + → π + µ + µ − × − D → e ± µ ∓ × − D + → π + e ± µ ∓ × − D → π e ± µ ∓ × − D → ηe ± µ ∓ × − D → K S e ± µ ∓ × − D + → π − e + e + , D + → K − e + e + × − D + → π − µ + µ + , D + → K − µ + µ + × − D + → π − e ± µ ∓ , D + → K − e ± µ ∓ × − While no evidence for New Physics has yet beenfound in charm decays, the searches have only recentlyentered a domain where one could realistically hopefor an effect. New Physics typically induces flavor-changing neutral currents. Those could be consider-ably less suppressed for up-type than for down-typequarks. Charm quarks are unique among up-typequarks in the sense that only they allow to probe thefull range of phenomena induced by flavor changingneutral currents, including CP asymmetries involvingoscillations.There is little Standard Model background to NewPhysics signals in charm CP asymmetries, and whatthere is will probably be under good control by thetime Super B starts operating. Baryogenesis – neces-sary to explain the observed matter-antimatter asym-metry in our Universe – requires a new source of CP violation beyond that of the Standard Model. Suchnew sources can be probed in charm decays on threedifferent Cabibbo levels, differing in rates by close tothree orders of magnitude. With the Standard Modelproviding one amplitude, observable CP asymmetriescan be linear in a New Physics amplitude, thus greatlyenhancing their sensitivity. Finally, as stated repeat-edly, the goal has to be to identify salient features ofthe anticipated New Physics beyond “merely” ascer-taining its existence. This will require probing chan-nels with one or even two neutral mesons in the fi-nal state – something that is possible only in an e + e − production environment. CLEO- c and BESIII are un-likely to find CP asymmetries in charm decays, andthe B Factory results will still be statistics limited.2
Proceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007
A Super Flavour Factory would allow conclusive mea-surements. Super B , with data taken at the Υ (4 S )and near threshold, will complete the charm programdown to the Standard Model level. References [1] M. Bona et al. , “SuperB: A High-LuminosityAsymmetric e+ e- Super Flavor Factory. Concep-tual Design Report,” arXiv:0709.0451 [hep-ex].[2] PEP-II Conceptual Design Report, SLAC-372, LBL-PUB-5303, CALT-68-1715, UCRL-ID-106426, UC-IIRPA-91-01 (1991); J. Seeman etal. , SLAC-PUB-12023, Contributed to EuropeanParticle Accelerator Conference (EPAC 06), Ed-inburgh, Scotland, 26-30 Jun 2006.[3] S. Kurokawa and E. Kikutani, Nucl. Instrum.Meth.
A 499 , 1 (2003), and other papers thatvolume.[4] B. Aubert et al. , B A B AR Collaboration, Nucl. In-strum. Methods
A 479 , 1 (2002).[5] A. Abashian et al. , (Belle Collaboration), Nucl.Instrum. Meth.
A 479 , 117 (2002).[6] The Physics Performance of LHC b . N. Harnew et al. A 408 , 137, (1998).[7] The New Snowmass Year is an updating of theconvention that multiplying peak luminosity bya “year” containing 10 seconds provides a goodmeasure of actual running time, the effects of ac-celerator and detector down time, dead time ef-fects and the difference between peak and averageluminosity. PEP-II/ B A B AR experience has shownthat a “New Snowmass Year” with 1 . × sec-onds is a more precise estimator of actual perfor-mance at a B Factory.[8] Design for a 10 Super B Factory at PEP-II.J. Seeman et al.
Proceedings of the 9th EuropeanParticle Accelerator Conference (EPAC 2004),Lucerne (2004).[9] Letter of Intent for KEK Super B Factory.S. Hashimoto ed., KEK-REPORT-2004-4 (2004).[10] The starting point of this effort was an attempt toleverage the active development effort in supportof a high energy linear collider that has been go-ing on for the past two decades. The idea, whichhas antecedents dating to the mid 1980’s [11], wasto achieve the high luminosity by using very lowemittance beams with high disruption, and to re-capture at least the positron beam and recirculateit, to minimize power consumption. It is, however,a substantial challenge to produce luminosities ofthe order of 10 cm − s − while having center-of-mass energy spread less than 10 MeV and keepingthe power consumption to a tolerable level. Thisproved to be a difficult problem [12]. [11] A Beauty Factory Using an SRF Linac anda Storage Ring. J.J. Bisognano, J.R. Boyce.,D. Douglas., S. Heifetz, J. Kewisch., G. Krafft.and R. Rossmanith. CEBAF-PR-89-017 (1989).Conceptual Design of a Multipurpose BeautyFactory Based on Superconducting Cavities.U. Amaldi and G. Coignet Nucl. Instr. Meth. A260 , 7 (1987). Proceedings of the Workshop onHeavy-Quark Factory and Nuclear-Physics Facil-ity with Superconducting Linacs. E. De Sanctis,M. Greco, M. Piccolo. and S. Tazzari eds., Cour-mayeur (1987). A High-Luminosity ¯ BB FactoryLinear Collider Using Positron Recirculation andRecovery. D. Cline and C. Pellegrini Nucl. Instr.Meth.
A290 , 297 (1990).[12] Super B : a linear high-luminosity B Factory.J. Albert et al.
INFN-AE 05-08 (2005).[13] Q. He et al. [CLEO Collaboration], Phys. Rev.Lett. , 121801 (2005) [Erratum-ibid. ,199903 (2006)].[14] S. Dobbs et al. [CLEO Collaboration],arXiv:0709.3783 [hep-ex] - submitted to Phys.Rev. D.[15] M. Gronau, Y. Grossman and J. L. Rosner, Phys.Lett. B , 37 (2001).[16] I. I. Y. Bigi and A. I. Sanda, Camb. Monogr. Part.Phys. Nucl. Phys. Cosmol. , 1 (2000).[17] S. Bianco et al. , Riv. Nuovo Cim. , 1 (2003)[18] I.I. Bigi, in: Proceed. of the Tau-Charm Work-shop , L.V. Beers (ed.), SLAC-Report-343, 1989,p. 169;[19] D. M. Asner and W. M. Sun, Phys. Rev. D ,034024 (2006).[20] D. Silverman and H. Yao, Phys. Rev. D , 214(1988).[21] A. G. Akeroyd, Prog. Theor. Phys. , 295(2004) [arXiv:hep-ph/0308260].[22] M. Artuso et al. [CLEO Collaboration], Phys.Rev. Lett. , 251801 (2005).[23] G. Bonvicini et al. [CLEO Collaboration], Phys.Rev. D , 112004 (2004).[24] P. Rubin et al. [CLEO Collaboration], Phys. Rev.D , 112005 (2006).[25] T. K. Pedlar et al. [CLEO Collaboration], Phys.Rev. D , 072002 (2007).[26] M. Artuso et al. [CLEO Collaboration], Phys.Rev. Lett. , 071802 (2007).[27] S. Stone, arXiv:0709.2706 [hep-ex] - Presentedat International Europhysics Conference on HighEnergy Physics (EPS-HEP2007), Manchester,England, 19-25 Jul 2007..[28] B. Aubert et al. [BABAR Collaboration], Phys.Rev. Lett. , 141801 (2007).[29] N. E. Adam et al. [CLEO Collaboration], Phys.Rev. Lett. , 251801 (2006)[30] S. Pislak et al. [BNL-E865 Collaboration], Phys.Rev. D , 010001 (2002).[31] J. R. Batley et al. [NA48 Collaboration], Phys. roceedings of the CHARM 2007 Workshop, Ithaca, NY, August 5-8, 2007 , 75 (2004).[32] D. Becirevic and A. B. Kaidalov, Phys. Lett. B ,417 (2000).[33] D. Asner, “Impact of Charm Physics on CKM”,presented at 4 th International Workshop on theCKM Unitarity Triangle, Nagoya, Japan, Decem-ber 12-16, 2006.[34] L. Widhalm et al. , Phys. Rev. Lett. , 061804(2006) [arXiv:hep-ex/0604049].[35] C. Aubin et al. , Phys. Rev. Lett. , 122002(2005).[36] M. Gronau and D. Wyler, Phys. Lett. B , 172(1991).[37] M. Gronau and D. London, Phys. Lett. B ,483 (1991).[38] D. Atwood, I. Dunietz and A. Soni, Phys. Rev.Lett. , 3257 (1997)[39] D. Atwood, I. Dunietz and A. Soni, Phys. Rev. D , 036005 (2001).[40] A. Giri, Y. Grossman, A. Soffer and J. Zupan,Phys. Rev. D , 054018 (2003).[41] A. Bondar and A. Poluektov, Eur. Phys. J. C ,347 (2006).[42] E. White and Q. He, arXiv:0711.2285 [hep-ex] -To be published in the proceedings of CHARM07,Ithaca, NY, August 2007, eConf C070805.[43] W. M. Yao et al. [Particle Data Group], J. Phys.G , 1 (2006) http://pdg.lbl.gov .[44] I.I. Bigi, Y. Dokshitzer, V. Khoze, J. K¨uhn,P. Zerwas, Phys. Lett. B181 , 157 (1986).[45] G. Burdman, E. Golowich, J. Hewett and S. Pak-vasa, Phys. Rev. D , 014009 (2002).[46] G. Burdman and I. Shipsey, Ann. Rev. Nucl.Part. Sci. , 431 (2003). [47] B. Aubert [ B A B AR Collaboration], [arXiv:hep-ex/0703020].[48] K. Abe [Belle Collaboration], arXiv:hep-ex/0703036. See also talk by M. Staric atRencontres de Moriond EW 2007 http://belle.kek.jp/belle/talks/moriondEW07/Staric.pdf [49] R. Godang et al. [CLEO Collaboration],Phys. Rev. Lett. , 5038 (2000) [arXiv:hep-ex/0001060].[50] K. Abe et al. [Belle Collaboration], [arXiv:hep-ex/0308034].[51] L. M. Zhang et al. [Belle Collaboration],Phys. Rev. Lett. , 151801 (2006) [arXiv:hep-ex/0601029].[52] B. Aubert et al. [ B A B AR Collaboration], Phys.Rev. Lett. , 221803 (2006) [arXiv:hep-ex/0608006].[53] D. M. Asner et al. [CLEO Collaboration],Phys. Rev. D , 012001 (2005) [arXiv:hep-ex/0503045].[54] [55] For a pioneering study see: J. M. Link et al. [FOCUS Collaboration], Phys. Lett. B , 239(2005) [arXiv:hep-ex/0506012].[56] For a very recent and detailed analysis see:Y. Grossman, A. L. Kagan and Y. Nir,[arXiv:hep-ph/0609178].[57] K. Agashe, G. Perez, A. Soni, Phys.Rev.
D71 ,016002 (2005).[58] I. Shipsey, Int. J. Mod. Phys. A , 5381 (2006)[arXiv:hep-ex/0607070].[59] D. M. Asner et al. [CLEO Collaboration], Phys.Rev. D70