Sweep Photogate: Optimized Photosensors for Optical Spectrometry in CMOS

C photosensors principally consist of a zero or reverse biased photodiode connected to an amplifying transistor as shown in fig.1. Impinging photons generate electron-hole pairs in the photodiode which are separated due to the electric field in the space charge region of the pn-junction. The associated charge transport results in a signal dependent photocurrent iph = q/ t. The transistor either serves as a signal amplifier for the voltage caused by the collected charge on the capacitance at its gate, or it is used as a sense amplifier for a regulation feedback which produces a constant photodiode voltage and transports the collected charge to a separate conversion node. High resolution and high dynamic range requirements at the typical low light intensities in optical spectrometry demand large photosensitive areas because high levels of signal charge have to be reached in order to avoid shot noise limitations. Large photodiodes imply large junction capacitances yielding a large CD in the equivalent circuit fig.1. System performance is then limited by the thermal noise of the transistor, whose noise equivalent input voltage vn causes a noise equivalent charge qneq at the conversion node: qneq = √ 4αkTB