The 2.5D vs 3D process and device simulations are compared based on Synopsys Sentaurus simulator. Also Lumerical FDTD simulator was used for the optical part. The 1.4um FSI pixel simulations show 3D Qsat of 4200e- while it is 5800e- in both measurements and 2.5D simulations.
The discrepancy exists also between the simulated and measured QE:
The paper conclusion is that "further simulations calibration adjustments are required to match experimental Qsat and QE".
Another interesting paper is "Modeling Statistical Distribution of Random Telegraph Noise Magnitude" by Ken’ichiro Sonoda, Motoaki Tanizawa, Kiyoshi Ishikawa, and Yasuo Inoue from
Renesas.
