 Work supported by US Department of Energy under grant DE-FG03-95ER40938, and by Fermi National Accelerator Laboratory and Lawrence Berkeley National Laboratory under subcontracts 517872 and 6485560, respectively. E-mail address: alansill@tthepl.phys.ttu.edu (A. Sill). Nuclear Instruments and Methods in Physics Research A 447 (2000) 1}8 CDF Run II silicon tracking projects Alan Sill for the CDF Collaboration Department of Physics, Texas Tech University, Lubbock, TX 79409-1051, USA Abstract Design features, functionality, and expected performance are reviewed for the silicon charged particle track detectors to be used by the Collider Detector at Fermilab (CDF) during the upcoming Run II of the Fermilab Tevatron. The original design has been supplemented by addiition of a new layer of silicon mounted on the beam pipe that improves the vertexing performance of the combined assembly. Progress has been made in many areas of design and construction of the silicon sensors, readout electronics, and associated systems. The resulting detector array should provide substantial improvements in coverage and performance over those of previous CDF silicon vertex detectors.  2000 Published by Elsevier Science B.V. All rights reserved. 1. Introduction and goals The Collider Detector at Fermilab is a 5000 ton multi-purpose particle physics experiment [1] dedi- cated to the study of proton}antiproton collisions at the Fermilab Tevatron collider. It was designed, built and operated by a team of physicists, techni- cians and engineers that now spans 44 institutions and includes approximately 500 o$cial members. Previous versions of this detector have included among their complement of instrumentation silicon vertex track detectors that have added substantially to the overall physics capabilities of the experiment, especially for studies of top and bottom quarks. The earlier silicon vertex detectors [2,3] oper- ated for CDF during collider Run I between 1992 and 1996 were composed of four layers of single- sided sensors that covered interactions within $27 cm along the beam line of the nominal center of the experiment. These interactions were distrib- uted approximately as a Gaussian along the beam (z) direction with an average standard deviation of typically 30 cm. The relatively short length of the silicon sensors limited the geometric acceptance to about 60% for single tracks [4}6], averaged over the luminous region. The angular acceptance for tracks from any given interaction was also limited by the previous geometry. Although detection of some tracks in forward and backward directions was possible for interactions that were displaced along the z direction from the center of the detector [7], more complete geometric coverage of the inter- action region was clearly desirable. For the next operating period of the acceler- ator, to be known as collider Run II, the expected 0168-9002/00/$ - see front matter  2000 Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 9 0 0 2 ( 0 0 ) 0 0 1 6 6 - 2