Leptonic, semi-leptonic and rare decays at B-Factories Gabriele - PowerPoint PPT Presentation

Leptonic, semi-leptonic and rare decays at B-Factories Gabriele Simi University and INFN Padova On behalf of BaBar and BELLE collaborations

Outline ● Introduction ● D 0 → gg NEW – [Belle PRD 93 (2016) 051102] – [BaBar PRD 85, 091107(R) (2012)] ● D s → n l – [Belle JHEP09 (2013) 139] – [BaBar PRD 82, 091103(R) (2010)] ● D 0 → p l n – [Belle PRL 97, 061804 (2006)] – [BaBar PRD 91, 052022 (2015)] 6/9/2016 G. Simi - CHARM 2016 2

Introduction ● Semileptonic and leptonic decays have a are very clean signature from an experimental point of view: a single lepton in the final state ● Are also relatively simple from a theoretical perspective provide a mean of both measuring fundamental standard model parameters (CKM, lepton universality) and perform detailed studies of the decay dynamics (QCD) ● Deviations of the parameters (for instance lepton universality) from the expectation can be a clear sign of new physics 6/9/2016 G. Simi - CHARM 2016 3

Introduction ● Semileptonic and leptonic decays have a are very clean signature from an experimental point of view: a single lepton in the final state ● Are also relatively simple from a theoretical perspective provide a mean of both measuring fundamental standard model parameters (CKM, lepton universality) and perform detailed studies of the decay dynamics (QCD form factors, decay constants) ● Deviations of the parameters (for instance lepton universality) from the expectation can be a clear sign of new physics ● Rare and forbidden modes can be a window in new physics trough the effect of virtual particles in the loop diagrams 6/9/2016 G. Simi - CHARM 2016 4

→ gg D 0 ● Flavour changing neutral current forbidden at tree level ● Sensitive probe for NP because highly GIM suppressed in SM ● Short distance contribution including QCD corrections – BF~O(10 -11 ) [Fajfer et. al. (2001), Burdman et. al. (2002)] ● Inluding long distance effects – BF~(1-3) 10 -8 (ex. S. W. Bosch and G. Buchalla (2002)) ● Possible BF~10 -6 – due to NP (ex. exchange of gluinos in S. Prelovsek and D. Wyler 2001) 6/9/2016 G. Simi - CHARM 2016 5

→ gg D 0 ● BaBar (470.5 fb -1 ) and Belle (832 fb -1 ) analysis use similar strategies ● To reduce large combinatorial 0 p - background require D *- → D ● Reduce QED background requiring > 4 tracks and >4 photons ● Largest remaining background from D 0 → p 0 p 0 , p 0 h,hh – Measure p 0 p 0 (with respect to Ks p 0 ) and use the p 0 p 0 shape in gg background – Implement p 0 , h veto ● Measure gg with respect to k s0 p 0 to remove D * production uncertainty and cancel systematics 6/9/2016 G. Simi - CHARM 2016 6

→ gg [BaBar PRD 85, 091107(R) (2012)] D 0 ● BaBar (470.5 fb -1 ) and Belle (832 fb -1 ) D 0 → gg signal + background analysis use similar strategies p 0 p 0 background ● To reduce large combinatorial 0 p - background require D *- → D ● Reduce QED background requiring > 4 combinatoric tracks and >4 photons ● Largest remaining background from D 0 → p 0 p 0 – Measure p 0 p 0 (with respect to Ks p 0 ) and use the p 0 p 0 shape in gg background – Implement p 0 , h veto ● Measure gg with respect to k s0 p 0 to remove D * production uncertainty and cancel systematics 6/9/2016 G. Simi - CHARM 2016 7

→ gg [Belle PRD 93 (2016) 051102] D 0 combinatoric ● BaBar (470.5 fb -1 ) and Belle (832 fb -1 ) analysis use similar strategies ● To reduce large combinatorial signal 0 p - background require D *- → D peaking ● Reduce QED background requiring > 4 M gg tracks and >4 photons and in-time sig. combinatoric ● Largest remaining background from D 0 → p 0 p 0 , p 0 h,hh peaking – Measure p 0 p 0 (with respect to Ks p 0 ) and use signal the p 0 p 0 shape in gg background – Implement p 0 , h veto ΔM ● Measure gg with respect to k s0 p 0 to remove D * production uncertainty and cancel systematics 6/9/2016 G. Simi - CHARM 2016 8

→ gg [Belle PRD 93 (2016) 051102 D 0 BaBar PRD 85, 091107(R) (2012)] ● Results for 90% CL upper limit – BaBar: N sig =-6 +/-15; BF<2.2 10 -6 – Belle: N sig = 4 +/-15; BF< 8.5 10 -7 CLEO ● No statistically significant signal BES III ● Belle result more stringent than BaBar Belle S. W. Bosch and G. Buchalla, previous limits ● Already posing contraints in new physics models ● Will be significantly improved in the next generation Belle-II experiment 6/9/2016 G. Simi - CHARM 2016 9

Purely leptonic decays c l + W+ D+ (S) d(s) n ● Helicity suppressed ● f Ds can be compare to lattice QCD predictions ● f Ds feeds into Bs mixing by affecting the calculation of f Bs ( and therefore | V ts |) ● Search for new physics – Some new physics scenarios could affect differently D and Ds leptonic decays [for example A.G. Akeroyd, hep-ph/0308260] 6/9/2016 G. Simi - CHARM 2016 10

- n - Belle D s → l JHEP09 (2013) 139 l - g ● Method: fully reconstruct K frag D* - n s D -- s cc X frag ● First inclusively select events D tag based on system recoiling against D tag K frag X frag g – – X frag =nothing,n p (n p <4, n p 0 <2) ● N inc ~94k events 6/9/2016 G. Simi - CHARM 2016 11

- n - Belle D s → l JHEP09 (2013) 139 ● Reconstruction of D s → m - n D s + → m + n – Add single track identified as m – Select events based on missing (neutrino) mass – N(D s → m - n ) = 492 ± 26 – BF(D s → m - n ) = (0.531 ± 0.028 ± 0.020)% ● Reconstruction of D s → t - n – Add single track identified as e, m or p – Select events based on missing D s + → t + n (neutrino) mass – N(D s → t - n ) = 2217 ± 83 – BF(D s → t - n ) = (5.70 ± 0.21 ±0.30)% ● Decay constant results: 6/9/2016 G. Simi - CHARM 2016 12

- n - BaBar D s → l PRD 82, 091103(R) (2010) ● Using an inclusive reconstruction method similar to Belle e+e- → c → DKXD s*- , D s * - c D s + → m + n s g → D ● Reconstruction of D s → m - n – Add single track identified as m with E extra <1GeV/c 2 – Select events based on missing (neutrino) mass m r (DKX gm ) – N(D s → m - n ) = 275±17 – BF(D s → m - n ) = (0.602 ± 0.038 ± 0.034)% Muonic mode ● Reconstruction of D s → t - n + → t + n D s – Add single track identified as e, m – Select events based on missing (neutrino) mass M r2 (DKX µ)>0.5 GeV γ 2 /c 4 (Remove µ + ν µ ) – BF(D s → t - n ) = (5.00 ± 0.35 ±0.49)% ● f Ds+ = (258.6 ± 6.4(stat) ± 7.5(syst)) MeV 6/9/2016 G. Simi - CHARM 2016 13

Comparison of decay constants ● HFAG averages f Ds+ |V cs | and extracts f Ds+ assuming |V cs |; f Ds+ =257.4 ± 4.6 MeV ● Value is in agreement with lattice QCD calculation f Ds+ =(248.6 ± 2.7) http://www.slac.stanford.edu/xorg/hfag/charm/Vcs/december14/results.html 6/9/2016 G. Simi - CHARM 2016 14

Why D semileptonics decays ● D n decay rate can be used to: → Pl ∝ |V cx | – Determine Vcx using the form factors from lattice QCD: d Γ 2 FF 2 – Determine V ub and V cb and validate LQCD using the form factors from CKM fits n can give information on the hadronic system ● D → PP'l without additional hadrons in the final state ● Decay rate = Short distance x (form factor) 2 x |CKM| 2 Use unitarity of Use theory to Experiment CKM predict form Ds→ lnu Known Martrix to extract factors and D→ K(p)lnu Decay constant extract Vcd and D→ l + l - Form factors Vcs Test QCD 6/9/2016 G. Simi - CHARM 2016 15

D semileptonic decays reconstruction Off peak On peak ● @Y(4S) charm cross section is large (1.3nb) ● Jet like events from off-peak and continuum provide better reconstruction of missing neutrino than secondary D from B ● Divide event in two hemispheres ● Use partial reconstruction on tag side – Large statistics – Non negligible background – Poor resolution on kinematical variables ● Background and resolutions can be derived Opposite Signal from data side side 6/9/2016 G. Simi - CHARM 2016 16

Belle approach: e + e - (*) tag D * → D sig X ● Signal D 0 is tagged using reconstruction of all other particles – D (*)tag reconstructed in D *+ → D 0 , D π + π 0 and D *0 → D 0 π 0 , D 0 with γ D 0/+ → K(n ) [n=1,2,3] π – high resolution on decay kinematic variables (∆q 2 ~0.015 GeV 2 ), allows absolute BR measurements, low background – negative : low efficiency (~1%) Mass spectrum for D sig candidates For ~ 56k tagged D 0 6/9/2016 G. Simi - CHARM 2016 17

Belle approach: e + e - (*) tag D * → D sig X [Belle PRL 97, 061804 (2006)] ● BF measured normalizing to the total number of D 0 tags – BF(D0 → k - l + n)=3.45+/-0.07+/- 0.02)% D→ k l n – BF(D0 → p - l + n)=0.255+/- ~2500 signal 0.019+/-0.016)% ● For D → p l n the unfolded q 2 distribution can be described D→ p l n by simple pole model ~230 signal 6/9/2016 G. Simi - CHARM 2016 18

→ p e n signal D 0 [BaBar PRD 91, 052022 (2015)] ● Partial reconstruction from 347.2 fb -1 of e + e - → cc events at the (4S) Υ ● D *+ → D 0 π + with the D 0 →π - e + ν ● q 2 =(p e +p n ) 2 =(p D -p p ) 2 ● Since The ν e momentum is unmeasured two kinematic fits are performed, imposing in turn the D 0 and D *+ mass constraint . Signal 5.3k ● Use Fisher discriminant to suppress S/B~0.5 background from B events and other semileptonic decays from continuum ● Signal events selected in δ m=m D*+ -m D0 <0.150 GeV/c 2 6/9/2016 G. Simi - CHARM 2016 19