Low noise, high sensitivity active pixel sensors

In this dissertation the author argues that low noise and high sensitivity can be achieved in active pixel sensors (APS) by both circuit and device approaches. Low noise and high sensitivity are the major goals in low signal level applications such as minimum ionizing particle detection in vertex detectors. This dissertation presents three circuit approaches and one device approach to achieve these goals.; The first circuit implements column level active reset. It suppresses kT/C reset noise via two negative feedback mechanisms, namely the control of the reset current via the transconductance of the reset switch, and the Miller effect. In tests, it reduces measured noise by up to a factor of 7.6, to an estimated 8.3 input-referred electrons, rms. The technique also dramatically reduces fixed pattern noise: by up to a factor of 21, as measured.; The second circuit implements a simplified version of the above column level active reset. By comparison, it requires one fewer transistor per pixel, has improved dynamic range, and is simpler to operate.; The third circuit implements a per-pixel correlated double sampling (CDS) reset noise reduction technique. It requires only one read, as compared to two for typical correlated double sampling in active pixel sensors, and no off-pixel storage or subtraction is needed. This technique, as measured, reduced input-referred temporal noise by a factor of 2.5, to an estimated input referred noise of 15.6 electrons, rms.; Taking a device approach, a large-area native epitaxial silicon photogate sensor was designed to increase the charge collected per hit pixel and to reduce charge diffusion to neighboring pixels. High charge to voltage conversion gain is maintained by subsequent charge transfer to a low capacitance readout node.; Lastly, the tradeoffs of CDS are discussed, using a high speed CMOS image sensor with in-situ frame storage as an example.