Study of Ds Decays

A limit for the absolute branching ratio of BR(Ds+@c)c4.1% at 90% C.L. and evidence for the decay Ds+ f&375@ is presented from the Mark III. I n the spectator model, the lowest lying bound state of the charm-strange quarks, the DS . meson, is predicted to decay via W emission from the charm quark and form states with SE hidden strangeness. The conventional hadronic modes will contain K’i?, o and the f,(975), which is considered to be the Jpc=O* isoscalar ss meson. Hadronic decays of the Ds meson have been measured in relatively few modesf’] and all modes are measured relative to the ox mode. In this paper, an upper limit for the absolute branching ratio of the DS decay into the @rc mode and evidence for the the decay, DS -+f,(975)x is presented from the Mark III. Data Set and Mark III detector The data for the Ds analysis was taken at a center mass energy of 4.14 GeV with a total integrated luminosity of the 6.3kO.46 inverse picobarns. The main features of the Mark III detecto@) have been discussed elsewhere. This analysis used tracking information from the drift chamberf31, timing measurements from the time-of-flight (TOF) countersf4] and shower tracks detected in the barrelt51 and endcap(61 shower counters. Invited taIk presented at the KEK Topical Conference on e+eCollision Physics at KEK National Laboratory for Physics, Tsukuba. Japan, May 17-19,1989 *Work supported in part by the National Science Foundation and by the Department of Energy contracts DEAC03-76SF00515,DE-AC02-76ER01195,DE-AC03-81ER40050, DE-AC02-87ER40318 and DE-Ah+0376SFOOO324 Studv of Ds Decavs 2 i Analvsis of the Absolute Branchina Ratio uDDer limit The production of the Ds decays into final states i and j occurs in the reaction e+e+DsDs*, Ds+final state i, D*s +yDs, Ds+final stute j. The produced rate of this exclusive reaction will be, B(Ds+final state i).B(Ds+finul stutej)a(e+e-+DsDs*).JLdt. By measuring several exclusive reactions ( called double tags ) and forming the various ratios, we may divide out the cross section and obtain a measurement of the absolute branching ratios. In this analysis the final states i and j were investigated in the modes; on+, K”K+, f,(975)n+, K*(892)‘K+, K*‘K*+, @c+Ic+~c-, and oz+$ and this results in a limit on the branching ratio B(Ds +$z+). A total of twenty-eight possible double tag final states were considered.t7s8] . For the individual final states, combinations of shower and drift chamber tracks are selected and the charged tracks assigned masses. The charged tracks were used as a rc: or a K unless the measured TOF was more than 50 away from the predicted time of its mass hypothesis. Kinematic fits are applied to the hypothesis e+e-+DsDs*+yDs+Dsand the Ds+ and Dscandidates are constrained to have equal but unspecified mass M(X). The DsDs* events will produce a peak in the M(X) distribution at the Ds mass. In the region of the M(X) distribution which would contain 95% of the Ds signal as estimated by Monte Carlo, no -. events were observed and this results in an upper limit for the absolute branching ratio. In Fig. 1 is shown the M(X) distribution where in the DS region no events are detected. A Monte Carlo estimate of 2.8 produced DS events is drawn in. : ‘An upper limit for B(Ds +@c+) is obtained by computing the likelihood of observing zero events as a function of B(Ds +@I+). The expected number of double-tag events is [o(e+e. +DsDs*). J titOB(Ds+$r+)].B(Ds+o~+).~b(i)b(j)c.. ‘I’ where b(i) is the ratio of B(Ds+fifinul state i) divided by B(Ds+@+) and Eij is the Monte Carlo determined efficiency of reconstructing the event of double i and j. The measured values used in the likelihood calculation are: [cr(e+e-+DsDs*)~JLdt.B(Ds-xpc+)] =156+38fg], b(K°K+)=.92f.38,t10] b(f,z+)=.28f.10,f11] b(K*“K+)=.93f.l 2,[12-151 b(K*“K*+)=2.3fl.4,[‘6] b(+@=.41&1 0,f16S15V171 and b(@c+ 2.4f 1 .1.[181 In the likelihood function, Poisson statistical errors is used for the observed number of events and Gaussian statistical errors is used for the measured quantities. The correlations between the measurements of Mark III results are included. The marginal likelihood is deterStudy of Ds Decays 3 f, mined by integrating the likelihood over the o(e’e-~DsDs’).ILdt.B(Ds~~~‘) and the b(i)‘s. This results in an upper limit of B(Ds -+@c+)<3.8% at 90% C.L. and this is shown in Fig. 2. This limit corresponds to 2.8 observed events which is drawn in Fig. 1. The uncertainties on the number of signal events includes the errors on the track reconstruction (7%), Monte Carlo generation (2%), and the efficiency of the signal cuts (2%). This totals in quadrature to 8% and enlarges our final value on the upper limit to; B(Ds +$I?) c 4.1% at 90% C.L. Evidence for the decav Ds+ f-(975)x ..