Modeling And Analysis Of Full Well Capacity Of Pinned Photodiode In CMOS Image Sensor Based On Energy Level Distribution Of Interface State Trap

Driven by low-noise applications,the full well capacity(FWC)has been widely used to clarify the imaging characteristics of the pinned photodiode(PPD)of the complementary metal oxide semiconductor image sensor(CIS),which affects the signalto-noise ratio,sensitivity and dynamic range of CIS.Therefore,it is necessary to establish an accurate physical model for the full well capacity of the PPD in order to fully understand the PPD characteristics.However,the existing studies only established the model that the full well capacity depends on temperature,N-well doping concentration and the wavelength of light,ignoring the influence of interface state trap Level distribution on full well capacity.Moreover,the existing models directly assume that the trap Level is a single fixed value when analyzing effect of interface states,which will lead to a single fixed value of full well capacity.However,due to the fluctuation of CMOS process,the location and distribution of trap energy Levels are different,and the value of full well capacity is also different,which means that the value of full well capacity is dynamic.In addition,the existing studies only model the full well capacity,but have not analyzed the stability of the model.Therefore,the existing models are still unable to accurately predict the variation of full well capacity.In order to overcome the shortcomings of the existing models,a dynamic model of full well capacity of the pinned photodiode for CMOS image sensor is proposed,and the stability of the model is analyzed.The main contributions are as follows:(1)A dynamic model of the full well capacity of PPD is established.In this paper,a two-dimensional explicit function of the full well capacity dependent on the trap energy Level and the trap density of states is established.Combining the trap energy Level and the trap state density depend on the characteristics of the trap energy Level distribution of the interface state,the dynamic changes of the full well capacity under the Level,Gaussian and Exponential distributions are analyzed.(2)The simulation verification of full well capacity model of PPD is given.In this paper,Sentaurus Tcad 2016 software and Matlab R2018 a software are used to obtain the actual simulation results and theoretical calculation results of the full well capacity respectively when the trap energy Level follows the Level,Gaussian and Exponential distribution.Origin 2017 software is used to fit the two results obtained above with the formula.The high consistency of the two results shows that the full well capacity is not a single fixed value in the existing model,but a series of dynamic values consistent with the distribution of trap energy Levels.(3)The stability analyses of full well capacity model of PPD are given.In this paper,a method to verify the stability of the model by using the existing model parameters is firstly proposed,and the related simulation experiment is innovatively designed.By simulating the distributions of full well capacity under different temperature,N-well doping concentration and wavelength with the different trap energy Level distributions,it is found that the dynamic model of the full well capacity dependent on the trap Level distribution of the interface state is always stable.