Author

Kimberly Kolb

Abstract

Geiger-mode avalanche photodiode (GM APD) detectors are capable of counting single photons, measuring arrival times with high resolution, and generating zero read noise (when operated with a CMOS digital readout circuit) due to their unique internal gain characteristics. These capabilities make them exceptionally suited to tasks that require precise arrival time measurements or characterization of faint signals (low photon flux). Laser ranging systems use their arrival time measurement capabilities to build three-dimensional images, while adaptive optics applications have recently begun to capitalize on their low noise and high-speed operation for correcting wavefront imperfections due to atmospheric interference. There is now growing interest in using GM APDs for imaging applications where accurate measurements of faint signals are necessary, such as in astronomy. MIT Lincoln Laboratory and the RIT Center for Detectors have developed silicon GM APDs with unique architecture, utilizing scupper regions to minimize detector noise. This thesis investigates the performance of these detectors in terms of dark count rate (DCR). There are a number of mechanisms that produce dark counts, the most prominent being thermal excitation of carriers. Thermal carrier generation rates are generally only dependent on the temperature of the diode and may be constant under certain controlled conditions. Afterpulsing results from the release of carriers trapped in intermediate energy states (states with energy in the band gap of the material). Unlike thermal carrier generation, afterpulsing is dependent on the quenching time of the device (during which the device is unable to detect a carrier). Another mechanism, called self re-triggering, occurs when relaxing carriers emit photons during an avalanche. These photons can be absorbed in the substrate and generate dark carriers. Self-retriggering is also dependent on the quenching time of the device. Theories for afterpulsing and self-retriggering are discussed. Specialized test circuitry is used with a customized data acquisition technique, and the author develops a method for parameter extraction from the raw data. Device characteristics derived from experimental results are examined. The author also develops a simulation program to approximate the dark count rate (among other parameters) of a device based on semiconductor characteristics and testing conditions. This thesis makes conclusions about the dependence of DCR on device architecture and how individual carrier generation mechanisms affect device performance.

Library of Congress Subject Headings

Avalanche photodiodes; Photon detectors

Publication Date

7-15-2011

Document Type

Thesis

Department, Program, or Center

Chester F. Carlson Center for Imaging Science (COS)

Advisor

Figer, Donald

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works. Physical copy available through RIT's The Wallace Library at: TK8312 .K65 2011

Campus

RIT – Main Campus

Share

COinS