Search

We present the results of an experimental study of a special type of CdZnTe detector of hard x and γ rays—a ring-drift detector. The device consists of a double ring electrode structure surrounding a central point anode with a guard plane surrounding the outer anode ring. The detector can be operated in two distinctively different modes of charge collection—pseudohemispherical and pseudodrift. We study the detector response profiles obtained by scanning the focused x-ray beam over the whole detector area, specifically the variations in count rate, peak position, and energy resolution for x rays from 10 to 100 keV. In addition, at 662 keV the energy resolution was shown to be 4.8 keV, more than a factor of 2 better than for CdZnTe coplanar grid detectors. To interpret the experimental data, we derive an analytical expression for the spatial distribution of the electric field inside the detector and neglecting carrier diffusion, and identify carrier collection patterns for both modes of operation within the drift model approximation. We show that this model provides a good understanding of measured profiles. [ABSTRACT FROM AUTHOR]

We report the development of a general analytic method for describing the responsivity and resolution for a pixellated semiconductor detector structure in terms of device and material properties. The method allows both drift and diffusive transport to be modelled, for which previously only Monte Carlo techniques have been available. We obtain a general solution, and show specific results for an array of square pixels, illustrating the device constraints required to optimize spatial and spectral resolution. [ABSTRACT FROM AUTHOR]

We describe a model for photon absorption by superconducting tunnel junctions in which the full energy dependence of all the quasiparticle dynamic processes is included. The model supersedes the well-known Rothwarf–Taylor approach, which becomes inadequate for a description of the small gap structures that are currently being developed for improved detector resolution and responsivity. In these junctions relaxation of excited quasiparticles is intrinsically slow so that the energy distribution remains very broad throughout the whole detection process. By solving the energy-dependent kinetic equations describing the distributions, we are able to model the temporal and spectral evolution of the distribution of quasiparticles initially generated in the photoabsorption process. Good agreement is obtained between the theory and experiment. © 2003 American Institute of Physics. [ABSTRACT FROM AUTHOR]

Pulses of nonequilibrium phonons (heat pulses) have been used to mimic the absorption of x-ray photons in superconducting niobium tunnel junctions. For device characterization, the technique provides a valuable alternative to photoabsorption with good time resolution and continuous variability of absorbed energy. In addition, excitation is uniform across the tunnel junction, so that the effects of quasiparticle diffusion can be neglected in the analysis, and hence, the Rothwarf–Taylor equations solved exactly. Consistency is obtained between device parameters obtained from phonon measurements and those inferred from modeling of photoabsorption. In addition, the quasiparticle recombination rate can be determined directly from the nonlinearity of the energy response. © 2003 American Institute of Physics. [ABSTRACT FROM AUTHOR]

Multiple tunneling of quasiparticle charge carriers in a superconducting tunnel junction (STJ) enhances the signal generated by a photon absorption event. It is also an additional source of noise, responsible for a substantial degradation of the energy resolution. Although tunneling is a binomial chance process, governed by a constant tunneling probability, the resulting cumulative statistics of tunnelled quasiparticles depend on time. In particular, the variance of the total number of tunneled quasiparticles reaches a minimum after a finite integration time, corresponding to a minimum in the spectral linewidth. Since the intrinsic energy resolution of the present generation of STJs is mainly limited by the scatter on the number of tunneled quasiparticles, the improvement of the tunnel noise can be experimentally tested by variation of the pulse integration time. An analytical theory is developed that describes the relation between the tunnel noise and the transfer function of the pulse integration hardware for an STJ characterized by a quasiparticle tunnel and loss time in each electrode. We present experiments that demonstrate that the noise contribution from multiple tunnelling is not constant during the time that the quasiparticles are present in the STJ, and that by proper filtering of the STJ pulses the tunnel noise can be optimized ata level which lies well below the canonical tunnel limit. [ABSTRACT FROM AUTHOR]