| There exists an overwhelming body of observational evidence that most of the matter in the universe is dark. In order to have avoided detection thus far, this dark matter must interact only through the weak and gravitational forces. A well motivated candidate for this Weakly Interacting Massive Particle (WIMP) is a relic partner predicted from super-symmetric (SUSY) theories.; If our Milky Way halo were composed of WIMPs, a detector on the earth could detect the recoil energy imparted by a WIMP scattering with one of its nuclei. Characteristic energies in this interaction would be ∼10 keV for WIMP masses of ∼10 GeV/c2. A suitable dark matter detector would have to have a threshold of ∼1 keV to be viable. The density of dark matter in the solar neighborhood, combined with an upper limit on the WIMP-nucleon cross section, predicts extremely low event rates of ∼1 per kilogram per day. Dark matter detectors must therefore have a high degree of background rejection capability. This thesis describes the development of such a detector capable of detecting relic dark matter particles.; In order to achieve these goals, we have developed a cryogenic athermal-phonon-based dark matter detector. By cooling a semiconductor absorber to millikelvin temperatures, phonons generated by particle events have lifetimes of ∼100 mus and propagate many centimeters. We present evidence that the phonon readout in this dark matter detector is sufficiently fast to distinguish between ballistic and quasi-diffusively propagating phonons in silicon. Taking advantage of the down-conversion of phonons in metallic layers at the detector surface we use rise time to reject a soft electron background at better than 1 in 30. We present data from this detector in the Stanford Underground Facility (SUF) and discuss their implication for possible WIMP-nucleon cross section upper limits. In a series of calibration experiments we show that phonons generated by drifting electron-holes have a different spectrum than those generated in particle events. A first attempt to exploit this effect in the development of a phonon-based discrimination technique is discussed. Finally, we discuss theoretical limits to this detector technology and speculate on progress in the near future. |
