Test of the Flavor Independence Of

We compare the ratio of the strong coupling Q, measured in Z” --) b8 events to that measured in Z” -+ qq (q=u,d,a,c) in the SLD experiment at SLAC. Z” -* b6 events are identified with the aid of a precision silicon vertex detector by requiring that at least thiee charged tracks pass more than 3.017 from the Z” decay vertex. This method has an estimated efficieticy of 54% with a purity of 71%. Jets are identified and counted by clustering charged tracks according to the JADE algorithm. By comparing the 3-jet fraction in tagged events to that in all hadronic events, we extract a preliminary result of cr,(b)/o,(udx) = 1.18 & O.ll(stat) & O.O5(sys). 1. Detector Fnd Data Sample The SLAC Large Detector’ (SLD) analyzes the decays of Z” bosons produced in e+ecollisions at the SLC. The analysis presented here takes advantage of two tracking components, the Central Drift Chamber (CDC) and the Silicon Vertex Detector (VXD)2. The CDC consists of 80 layers of sense wires, 48 of which are stereo layers, in an axial magnetic field of 0.6 T. Tracks with polar angle Icos(8)1 < 0.71 traverse all 80 layers, with good reconstruction extending to Icos(8)1 < 0.80. The VXD consists of 480 silicon CCD chips with a total of 120 million 22x22 pm pixels. Charge division between neighboring pixels affords an intrinsic position resolution of 6 pm. The CCD hi c ps are arranged in four layers, ranging from 29.5 mm to 41.5 mm in radius. Track acceptance for the VXD extends to Icos(8)1 < 0.75. The VXD is described in more detail in these proceedings.394 Well-measured tracks are required to pass within a cylinder of radius 5 cm and half-length 10 cm along the beam axis, centered on the interaction point. In addition, they must be contained within Icos(8)l < 0.8 and have a momentum transverse to the beam axis of at least 150 MeV/c. For a Z” decay to be selected for use in this analysis, it must have at least 5 well-measured tracks, at least 20% of the total center-of-mass energy must be visible in well-measured tracks, and ICOS(~ thrrrt)l < 0.71. These requirements ensure good containment in the barrel of the detector. Of 11659 ZO candidates, 5602 satisfy “these criteria and are used in this analysis. * Work Supported by the U.S. Department of Energy under Contract DEAC03-76SF00515. Presented at Particles & Fields 92: 7th Meeting of the Division of Particles and Fields of the APS (DPF 92), Batavia, II, November IO-14,1992 2. Tagging Z” -+ bi Events Events with B hadron decays are tagged using the x-y distance of closest approach (DOCA) of charged t racks to the interaction point (IP). In addition to the normal track selection cuts, further restrictions on tracks contributing to the tag are imposed to ensure good measurements of their DOCAs. Tracks must have at least 2 clusters of hits in different CCDs of the VXD, they must have at least 40 out of a possible 80 hits in the CDC, and they must have x2/DOF < 5.0 in a track fit to the CDC hits. They must not pair with other charged tracks to form K”, ho, or photon conversion candidates. The error on the DOCA measurement arises from the intrinsic resolution of the VXD and multiple scattering terms which become dominant at low momentum. The average error on the DOCA of a track is given by < UDOCA >= 13Ccrn e 7Opm/(P&FfF)), where P is the track momentum in GeV. Tracks with uDoCA > 250pm are not used in the tag. . The interaction point (IP) ’ f 1s ound by combining tracks from several events and fitting for a common origin. The SLC IP measures 2 pm in x and y, and 650 I.crn in z. The location of the IP is stable within 20 pm over periods of more than 200 hours. As only the x-y position of the IP is necessary for this analysis, the errors in its measured location may be neglected. For each event, tracks passing the above selection criteria are extrapolated past the IP and their DOCA to the IP is measured. A sign is applied to the DOCA by comparing the track to the nearest jet axis. Jets used for the purpose of signing the DOCA are found using the JADE algorithm5 with Y,,~ = 0.02. If the x-y projection of a track intersects the projection of the jet axis on the opposite side of the IP in which the track travels, the DOCA is negative. The normalized DOCA is formed by dividing the DOCA by its measurement error. To identify an event as a ~0 -+ bi event, at least ~~~~ charged tracks must have normalized DOCAs exceeding Naig. Fig. 1 illustrates the estimated efficiency and purity of the b tag for various values of ~*,k and N,ig, using JETSET 6.3 with the SLD heavy quark decay package ,‘I along with full GEANT detector simulation and SLD event reconstruction. The values of Nf,k = 3 and N,ig = 3.0 are used in this analysis. Using this identification procedure, we estimate an efficiency of 54.0%*1.6% with a purity of 71.0%02.5%.~ Of the 5602 hadronic events used in this analysis, 919 are tagged as Z” -+ bg decays. 3. Measuring a,(b)/a,(wfm) A jet rates analysis6 was performed on the b tagged sample and on the hadronic dataset using the JADE algorithm. The ratio of the Q-jet rates in these samples, R mea, = f+tag)/f3(hdr), is shown as a function of ycrf in fig. 2. Because each event in the sample contributes to each point, the values of R,,,~,,, at different ycuf are correlated. To quote a result with proper errors, one point on fig. 2 is selected. We adopt the conventions of the L3 collaboration and quote our results at Y,,~ at which the measured ratio is R,,,~,,, = 1.11 f o.w(stat). = 0.05, s