Design and Construction of a Vertex Chamber and Measuremement of the Average B-hadron Lifetime*

Four parameters describe the mixing of the three quark generations in the Standard Model of the weak charged current interaction. These four parameters are experimental inputs to the model. A measurement of the mean lifetime of hadrons containing b-quarks, or B-Hadrons, constrains the magnitudes of two of these parameters. Measurement of the B-Hadron lifetime requires a device that can measure the locations of the stable particles that result from B-Hadron decay. This device must function reliably in an inaccessible location, and survive high radiation levels. We describe the design and construction of such a device, a gaseous drift chamber. Tubes of 6.9 mm diameter, having aluminized mylar walls of 100 pm thickness are utilized in this Vertex Chamber. It achieves a spatial resolution of 45pm, and a resolution in extrapolation to the B-Hadron decay location of 87pm. Its inner layer is 4.6 cm from e+ecolliding beams. The Vertex Chamber is situated within the MAC detector at PEP. We have analyzed both the 94pb -’ of integrated luminosity accumulated at fi = 29 GeV with the Vertex Chamber in place as well as the 2lOpb-1 accumulated previously. We require a lepton with large momentum transverse to the event thrust axis to obtain a sample of events enriched in B-Hadron decays. The distribution of signed impact parameters of all tracks in these events is used to measure the B-Hadron flight distance, and hence lifetime. . The trimmed mean signed impact parameters are 130 & 19pm for data accumulated with the Vertex Chamber, and 162 & 25pm for previous data. Together these indicate an average B-Hadron lifetime of [ +0.22 70 = 1.37 -0.19 stat.hO.11 sys. I x (1 310.15 sys.) psec. We separate additive and multiplicative systematic errors because the second does not degrade the statistical significance of the difference of the result from 0. If &c dominates b-quark decay the corresponding weak mixing matrix element lVc.l = 0.047 3~ 0.006 rt 0.005, where the first error is from this experiment, and the second theoretical uncertainty. If &.L dominates, lVubl = 0.033 -+ 0.004 -+ 0.12.