Pixel-to-Pixel Variance in Graphene-Silicon Photodetector Arrays

Understanding pixel-to-pixel (P-P) variance of dark current and photoresponse from individual pixels in image sensors is crucial to address artifacts, noise, and photoresponse inaccuracies in the overall imaging process. Specifically, well-calibrated pixel-to-pixel variance and specific photoresponse statistics of the raw output data help to identify the unavoidable error contributions from individual pixels, thus enabling compensation through computational image processing or readout circuits. This is particularly vital for emerging image sensors based on 2D materials integrated with silicon, which often utilize readout circuits distinct from traditional CCD and CMOS sensors. This work investigates pixel-to-pixel variance in graphene-silicon Schottky photodiodes, assessing the feasibility of large-scale fabrication, uniformity of dark current, and photoresponse in small-scale arrays. The array-based pixel statistics investigated in this study have potential implications for designing in-sensor processing and developing accurate statistical estimation methods for imaging in astronomy, biomedical, and spectroscopy.