研究目的
To model a photodiode for accurate quantum efficiency simulation and develop a guideline for cross-talk suppression using Monte-Carlo simulation based on Boltzmann transport equation.
研究成果
The new method of modeling an electrical cross-talk in a PD by tracing the position of electron using BTE provides very good match to the measured QE spectrum curves, as well as the CCMs and SNR10s. A new cross-talk index is developed, indicating that the product of the distance and electric field has to be more than ~0.02 [V] to ensure Pc of >90%.
研究不足
The study is limited by the computational resources and time required for the Monte-Carlo BTE simulation, which typically takes ~100 hours with 78K voxels, and 100 electron injections per voxel.
1：Experimental Design and Method Selection:
The study uses Monte-Carlo BTE simulation to model individual photo-generated particles in a photodiode.
2：Sample Selection and Data Sources:
Two measurement samples, "Control" and "Experiment", fabricated with different implant conditions based on tsmc-internal
3：1um Bayer patterned color CMOS image sensors. List of Experimental Equipment and Materials:
In-house Monte-Carlo BTE simulation software built on C-language, commercial software employing FDTD method for optical simulation, workstation (Xeon X5690,
4：47GHz, 24cores). Experimental Procedures and Operational Workflow:
Simulation space is sub-divided into voxels, one electron is injected to each voxel and its motion is calculated using Monte-Carlo BTE.
5：Data Analysis Methods:
The electron generation result from optical simulation is multiplied with the Pc at each voxels, and integrated over one PD to calculate its QE.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
