Report from Charm WG White Paper TDR Running

Report from Charm WG (White Paper TDR) Running at “Charm Threshold” and Rare Decays D. Asner, Carleton U, I. Bigi, U. Notre Dame, R. Andreassen, B. Meadows, M. Sokoloff U. Cincinnati for the charm team Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

2 Outline q Running at charm threshold § General § Improved precision in mixing parameters § Semi-leptonic asymmetries – a. SL § Time-Dependent Quantum Correlations (TDQC)? q Rare decays § D 0 + § D 0 hl+l- q Summary Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

Threshold Running Scenario q The run envisaged here is a 500 fb-1 exposure at (3770) § About 1000 x CLEO-c and (50 -100) x BES III § Just above DD threshold § At one-tenth nominal luminosity, Super B can complete this in a few months § Set up and tuning of final focus may take another few months Perhaps a year altogether – maybe less? q Possible schedules (in order of preference) § § Plan this in as a run after 25 ab-1 at (4 S) Run because of a discovery at (4 S) Start-up Super B with this run Do this only at the end of 10 years at (4 S)

General Reasons for a (3770) Run q It is possible (likely? ) that an unusual effect – perhaps due to NP – will be found in the charm sector from (4 S) running § A true confirmation in the entirely different charm scenario at (3770) will trump a simple repeat performance with more luminosity at (4 S) q q A 500 fb-1 run will improve both our own, and also LHCb’s CKM measurement by a factor ~ 3. Decays of (3770) D 0 D 0 produce coherent (C=-1) pairs of D 0’s. Quantum correlations in their subsequent decays allow measurements of strong phases § Required for improved measurement of CKM § Also required for D 0 mixing studies

General Reasons for a (3770) Run q A record exists of several cases where CLEO-c, sometimes with just 281 pb-1 of their 818 pb-1 sample at (3770), has successfully competed with Ba. Bar’s ~0. 5 ab-1 and Belle’s ~1 ab-1 at (4 S). Examples: § Measurement of decay constants – f. D and f. Ds § Studies of leptonic and semi-leptonic decays, § Measurements of absolute hadronic branching fractions q q Other runs for CLEO-c at other CM masses have produced useful and novel results for Ds as well as D decays from (3770). Is a program at (4 S) complete without the “service” of one at (3770), etc. ?

CLEO-c Hadronic Charm Decays at (3770) Single Tags Only E ~ 0 in signal E sidebands in yellow

Semi-Leptonic Decays q q Example of D 0 K-e+ e decay from CLEO-c’s 818 pb-1 exposure at (3770). Quantity plotted is U = Emiss – c Pmiss (Note background level ~0)

8 Quantum-Correlations (QC) q Running at (3770) CLEO c were able to measure K- + strong phase: from 281 pb-1 q q Including other D mixing results: ALSO, they were able to measure bin-by-bin strong phase variations in, for example, the Ks + - Dalitz plot § Of value to improved (model-independent) measurement of CKM § Also of value in D 0 mixing measurements Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

9 Time-Integrated QC in Multi-body Channels Relative phases “bin-by-bin” in multi-body phase-space. X Quantum correlations remove necessity for model describing phase space. Probably ONLY way to obtain a true PWA fit. For higher multiplicities, we could measure “coherence” and a single strong phase. m-2 (Ge. V 2/c 4) • X m 02 (Ge. V 2/c 4) For 3 -body final states this provides model-independent measurement of strong phase variation over the Dalitz plot Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

10 “Double-Dalitz” Plot Fit: An example from CLEO c e+e- (3770) (Ks + -) q There are two Dalitz plots correlated in a time-dependent way: § If you select CP-odd Ks 0 in one, you see “no” Ks 0 in the other. § If you select RS K*- + in one, you see only (opp. sign) RS K*+ - in the other. EXCEPT: § There will be a small signal arising from MIXING. § This has a definite time-dependence, making it more identifiable above background. § Backgrounds at (3770) are small ! CLEO data: - It works !! J. Napolitano http: //indico. ihep. ac. cn/conference. Display. py? conf. Id=176 Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

DD Threshold Measurements q q Data from DD threshold provide measurement of strong phases such as K and < K¼ 0 >. They also provide measured values of in Dalitz plot bins This can be used (with a model for the values for r) to significantly reduce uncertainties in the Dalitz plot model used in the golden channel analyses. q As a basis for projection, we take results from CLEO-c: N. Lowrey et al, PRD 80, 031105 (2009), 0903. 4853 q We assume that new data from threshold will reduce the uncertainties in model uncertainty: § BES III – ~factor 3 improvement in model uncertainty § Super B 500 fb-1 DD threshold run – ~factor 10 improvement.

Value of Strong Phase Measurements q Two improvements in mixing precision come from threshold data: q Dalitz plot model uncertainty shrinks q Information on overall strong phase is added Uncertainty in x. D improves more than that of y. D

Origin of CPV from a. SL q Wrong Sign (WS) lepton asymmetry measures CPV that can arise ONLY from mixing: No CPV See Talk by I. Bigi here ? a. SL |q/p| Current Measurement ! If measurement lies on curve, CPV is ALL in MIXING

Precision of a. SL Uncertainty from a signal S over background B is

Measurement of a. SL – Needs (3770) Data! q Decays to wrong-sign (WS) leptons proceed through mixing § Rate is very small RM=(x 2+y 2)/2 x 10 -4 ~ 5 x 10 -5 and not yet observed: Right-sign (RS) Decays 0 K D -e+ e+ c. c. 344 fb-1 Another D In event 0. 82 fb-1 492 fb-1 RM< 6 x 10 -4 RM< 13 x 10 -4 (4 S): D 0 tagged by D*+ D 0 + e not reconstructed Background limits WS data (asl) for 75 ab-1 ~ 0. 20 *At (asl) for 75 ab-1 ~ 0. 80 (3770): D 0 tagged e fully reconstructed Include Hadronic decays* (asl) for 500 fb-1 ~ 0. 20 (3770). We need CLEO-c data to refine our knowledge of backgrounds we can expect.

16 a. SL from (3770) Data (Needs evaluation from simulation) q Two un-ambiguous possibilities. K+ - (3770) K+e- e K+ - K+e- e Bose statistics requires that one D 0 had to mix q One ambiguous possibility. K+ - (3770) K+e- e D 0 K+ - 5 x more likely to be DCS than D 0 ->K+e- e is from mixing q Could resolve ambiguity using difference in timedependence of mixing vs. direct DCS decay. Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

17 Summary of CPV Sensitivity from Mixing Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

18 Time-dependent QC Decays – “Super D” ? The moving CMS means we could measure time-dependent (TD) strong phases resulting from D 0 mixing. X Boost is ~same as for Y(4 S) but D 0 ~ 0. 4 B 0 Samples are large. E. g. Ks + - (on both sides) will have about 500 K events. m-2 (Ge. V 2/c 4) • X m 02 (Ge. V 2/c 4) Is this possible or useful ? Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

19 “Super D” ? q q The time-dependent “Double-Dalitz” analysis can be made on a wide variety of double-tagged events: § (Ks + -) - (Ks + -) § (Ks + -) – (Ks K+K-) + -) – (K- + 0) § (Ks + -) – (K- +) § (Ks + -) – (K-K+) § … etc. • Large samples. • Highly constrained (model-dependent) • Measure |q/p| and =Arg{q/p} direct CPV, … We need a simulation to learn how well this might work in face of reduced time-resolution wrt “Super B”. . Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

20 Rare Decay Super B, Elba, Italy, 5/30 -6/05/2010 0 D + Brian Meadows, U. Cincinnati

21 Search for D 0 + q q There is considerable theoretical interest in the FCNC decay D 0 + (see I. Bigi’s talk, Perugia, Friday am) The SM estimates a lower limit BF > 4 x 10 -13 Long-range mechanism q (Dominant in KL + - - NOT necessarily so in D 0 case) Estimates would be improved by measurement of D 0 § Only estimate so far - BF[D 0 ] < 1. 5 x 10 -5 [Recall that, for KL + -, SM BF = 6. 4 x 10 -8 – right on the money!] Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

22 Interest in D 0 + q R-parity SUSY could increase the decay rate by up to to 7 orders to BF~3. 5 x 10 -6 c 12 k + qk u q q 22 k Opens a 7 orders of magnitude search window for NP - Current experimental limits are: Super B should be able to arrive at upper limits much better than this and also measure BF[D 0 ] to better understand the SM limit. Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

23 Interest in D 0 + q Ikaros: “ 10 -8 would be an interesting goal” § At D threshold, the + - (or ) are “back-to-back” in transverse momentum and should present an excellent signal-to-noise ratio. [BESIII claim they will use 20 fb-1 to give them a 90% limit at 3 x 10 -8 – apparently a mistake] http: //ar. Xiv. org/pdf/0801. 1833 and Eur. Phys. J. C 57: 309 -492, 2008. q Further study of CLEO-c data at threshold will show if a better limit can be reached. Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati

D 0 q q q Best limit so far comes from CLEO 2. 9 x 10 -5 using 13. 8 fb-1 at (4 S) Ba. Bar “expects” 2. 5 x 10 -6 from 481 fb-1 at (4 S) which implies, for Super B ~2 x 10 -7 from 75 ab-1 at (4 S) (limits gluino enhancement to factor 70) A recent CLEO-c study (private communication from D. Asner) has shown that an efficiency of 5% has been achieved with no background events in the full 818 pb-1 sample. This implies a limit of about ~few x 10 -8 from 500 fb-1 at (3770) !! David Asner will follow up on this interesting analysis

Summary q q A strong case for a 500 fb-1 run at (3770), based upon a single study of NP is difficult to make. The best candidate(s) may be a clean search for D 0 + decays with a limit near or below 10 -8. § This requires study of CLEO-c data and a better idea of PID performance at Super B q q AND a much improved limit on D 0 A great breadth of excellent results to support other findings at Super B related to charm physics or spectroscopy could, however, be made. Though a convincing argument for this run is hard to make based on a single result, it is also hard to allow the opportunities it opens up to pass un-defended.

26 David Asner … our young man, heading West. Expert at running at DD threshold. Ikaros. Bigi … willing to grow a beard if CPV is not observed in D decays by 2017! Get well soon, Ikaros !! Super B, Elba, Italy, 5/30 -6/05/2010 Brian Meadows, U. Cincinnati