Conclusions

The simulations show that a vertex detector like the one described here will be able to measure the charged particle multiplicity except for very low multiplicity events, where statistical effects in the sampling of the distribution limit the measurement. The detector will be able to measure in all cases studied as long as the noise is understood. The noise will be an important factor in the multiplicity and measurements for p+p and p+Au.

On-line vertex finding with high efficiency is possible using the correlation method for central Au+Au events. Offline, the pseudo-tracking method should have 100% efficiency in this case. The correlation method can not determine the vertex on-line for p+Au or p+p because the multiplicities are too low but the pseudo-tracking method finds the vertex in 90% or more of the p+Au events and 70% or more of the p+p events. The pseudo-tracking method breaks down for charged particle multiplicity less than 10, and is affected by the noise in the detector.

A vertex detector with reduced coverage⁷ would make the already marginal measurement of the multiplicity in p+p and p+Au worse, but a trigger-level multiplicity may not be needed for these collisions. For an easily interpretable multiplicity measurement, at least one segment of strips parallel to the beam should remain in the system, as the angle of incidence of the particle makes it difficult to extract meaningful multiplicities (or ) from the perpendicular strips. The last 14% in vertex finding efficiency that comes with the more complete vertex detector for p+p and 8% for p+Au (compare tables 3 and 4) may not be worth the extra cost of an extra azimuthal segment. However, it is the coverage out to 3 and the correlation method for finding the vertex which constrain the length of the detector. Reducing the length of the detector will make the correlation method fail more often, eliminating the possibility of a high-resolution high-efficiency on-line vertex reconstruction. For comparison, a 100cm long detector covers 97.7% and a 32cm detector covers 85.4% of the Au+Au interaction diamond. If the inefficiency at the ends is acceptable, a shorter length for the azimuthal segments with perpendicular strips could be used. For 100cm long azimuthal segments with parallel strips, the coverage is maintained.

Another potentially valuable feature of the parallel strips is finding the vertex position in the transverse direction. The transverse size of the beam should be very small (0.45mm⁵,9 for Au+Au), but the ability to measure the beam position in the transverse direction could be extremely useful in the alignment of the detector.

There are still some problems with the vertex detector that require further study. Thermal expansion and contraction of the detector (and its support structure) are important and related to the power dissipation by the electronics. These effects are being studied and will constrain the electronics design. A study of electronics components with low power dissipation is underway. Some sample detectors and electronic components have been acquired and are in the process of being tested as a system.