Backside Illuminated Sensor Vertical Stacking Reduces Crosstalk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Backside illuminated imaging sensors face challenges with optical crosstalk and limited fill factor due to the thinning of silicon wafers, which affects sensitivity and signal quality, and existing stacked pixel designs struggle with color separation and process complexities.
Innovation Solution
The use of vertically stacked silicon layers with different depths for blue, green, and red light detection, along with an anti-reflective coating and amorphous polysilicon for enhanced red light detection, reduces optical and electronic cross-talk and improves fill factor by positioning blue light detection closest to the backside and red light detection deepest, with amorphous silicon and oxide layers optimizing light absorption and separation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the silicon wafer is thinned to improve sensitivity to light, then sensitivity is improved, but optical crosstalk increases
Solution Approach 1:
The patent transitions from planar pixel arrangement to a three-dimensional stacked architecture where multiple pixel layers are vertically arranged. This dimensional change allows light to be detected at different depths, improving sensitivity while the vertical stacking with isolation structures reduces optical crosstalk by confining light paths in the depth dimension.
Solution Approach 2:
The pixel array is segmented into multiple stacked layers, with each layer containing isolated pixel elements. The isolation structures divide the continuous silicon into discrete segments, preventing optical crosstalk between adjacent pixels while maintaining high light sensitivity in each segment.
2Measurement precision
If color filters and micro-lenses are added to improve sensitivity, then sensitivity is improved, but device complexity increases
Solution Approach 1:
Instead of adding complex color filters and micro-lenses on the surface, the patent uses the depth dimension by stacking multiple pixel layers at different depths. Each layer detects different wavelengths naturally through selective absorption, eliminating the need for additional optical components and reducing structural complexity.
Solution Approach 2:
The patent changes the detection parameter from surface-level optical filtering to depth-based wavelength separation. By varying the depth position of pixel layers, different wavelengths are detected at different depths, replacing complex filtering mechanisms with a simpler depth-parameter approach.
3Area of stationary object
If vertically stacked pixel sensors are used to improve fill factor, then fill factor is improved, but manufacturing precision requirements increase
Solution Approach 1:
The patent utilizes the vertical dimension for stacking pixels, which increases the effective light-sensitive area (fill factor) without requiring precise lateral alignment. The vertical stacking approach shifts the alignment challenge to the depth dimension, where standard semiconductor fabrication processes can achieve sufficient precision more easily.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This configuration enhances color separation, reduces cross-talk, and improves quantum efficiency by optimizing light absorption and detection, leading to improved sensitivity and signal quality in backside illuminated imaging sensors.
Implementation Method 1
vertically stacked silicon layers with different depths for blue, green, and red light detection... optimizing light absorption and detection
Implementation Method 2
anti-reflective coating and amorphous polysilicon for enhanced red light detection, reduces optical and electronic cross-talk
Data Source
AI summary
A backside illuminated imaging sensor includes a vertical stacked sensor that reduces cross talk by using different silicon layers to form photodiodes at separate levels within a stack (or separate stacks) to detect different colors. Blue light-, green light-, and red light-detection silicon layers are formed, with the blue light detection layer positioned closest to the backside of the sensor and the red light detection layer positioned farthest from the backside of the sensor. An anti-reflective coating (ARC) layer can be inserted in between the red and green light detection layers to reduce the optical cross talk captured by the red light detection layer. Amorphous polysilicon can be used to form the red light detection layer to boost the efficiency of detecting red light.