Design Of Cells And Readout Channel Circuit For Black Silicon CMOS Image Sensor

Compared with non-silicon-based image sensors,silicon-based image sensors have the advantages of simple device structure,more mature manufacturing process,ability to work at room temperature,low cost and strong anti-environmental interference ability,and have been widely used as imaging components in people’s daily lives.However,monocrystalline silicon materials are limited by their band gap,and the optical bands that can be detected are mainly concentrated in the visible light band,which limits the development of their detection in the near-infrared band.In order to expand the detection band of silicon-based image sensors for near-infrared detection,"black silicon" can be etched on the surface of silicon-based image sensors by means such as femtosecond lasers.The so-called "black silicon" refers to a material with a microstructure surface layer and impurity energy level,so black silicon can achieve high absorption of visible light and near-infrared light.In addition,CMOS image sensors have the advantages of random reading,low cost,low power consumption and small size,and have a wide range of applications in modern life.Therefore,based on the black silicon CMOS image sensor,this thesis uses Silvaco TCAD simulation software to design and simulate the device structure of the photosensitive pixel unit,which improves its quantum efficiency and full well capacity.Based on the SMIC 0.18 μm process,the cell and readout channel circuits are designed using Cadence simulation software.The main contents of this thesis are as follows:(1)A photosensitive pixel cell structure with high quantum efficiency,full well capacity and charger transfer efficiency is designed.The main function of photosensitive pixel pixels is to complete the conversion of optical signals to electrical signals.Based on the principle of photoelectric detection,a pixel structure with P-type doped insertion and transfer of N buried layer is designed,the quantum efficiency of the 500 nm band is increased by 5.5%,the full well capacity is expanded from 1269 e-to 7209 e-of the traditional structure,and the structural optimization also improves the charger transfer efficiency,which greatly optimizes the image tailing problem.(2)A 9-tube pixel circuit structure with high dynamic range is designed.The photocurrent output by the photosensitive pixel unit is very weak and easy to be annihilated by the noise signal,so the pixel circuit is required to integrate and amplify it.In this paper,a 9-tube pixel circuit structure is designed based on the 8-tube pixel circuit structure,and the extended capacitor Cext is connected,which improves the dynamic range by 11 d B.(3)A readout channel circuit with adjustable gain and wide swing is designed.To dynamically adjust the detection range of the photoelectric signal,a gain-adjustable amplifier(PGA)with four adjustable gears is designed,and finally an output buffer with wide swing is designed.