The high momentum end of the particle ID requirements for B physics is determined by the spectra of the two-body decay modes, such as and the background process . A TOF system will not be an effective tool to separate such decay modes. Nevertheless, to illustrate how high in momentum the optimal particle ID system would tag kaons, Figure 7 [4] shows the mass distribution for the combination of , , and assuming all K's to be 's. The solid histogram is the combination and the dashed histograms show the four separate components. The CDF mass resolution is not sufficient to separate the different background sources in . Furthermore, the mode lies directly under the peak. As Figure 8 indicates, the momenta from two-body decay modes are substantially higher than decays from multi-body modes. From this we conclude that eventually it would be desirable to build a particle identification system capable of tagging kaons up to or above 5 GeV/c. However, such a system is non-trivial. It would require a sophisticated ring imaging detector or new technology. In either case long development times and lots of money will be required to make it work in the Tevatron collider environment. In the mean time, it appears that a TOF system is an attractive and practical alternative for a broad range of other B physics studies. However, as all the figures in this section indicate, we should strive to design such a system so that it will separate kaons from pions at the highest possible momenta.

  
Figure 7: Mass distribution for the combination of , , and assuming all K's to be 's. The solid histogram is the combination and the dashed histograms show the four separate components.

  
Figure 8: Momentum distribution for the kaons from , and , using the HERWIG Monte Carlo Generator. A RICH system or equivalent will be necessary to identify a significant fraction of these modes