Stacked Photoelectric Conversion Layer Imaging Device
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional single-plate color solid-state imaging devices have low light utilization efficiency and sensitivity due to mosaic color filters, leading to low resolution and false colors, while multi-layer photoelectric conversion layer structures are difficult to manufacture and costly, and require complex signal processing to separate color signals.
Innovation Solution
A hybrid photoelectric conversion layer stack type solid-state imaging device with a green-detecting photoelectric conversion layer on the surface and blue- and red-detecting photodiodes in the semiconductor substrate, using 3-transistor signal reading circuits for the green pixels and 4-transistor circuits for blue and red pixels, with correlation double sampling to reduce noise and afterimages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If mosaic color filters are used in single-plate color solid-state imaging devices, then color signals can be obtained, but light utilization efficiency and sensitivity are reduced because 2/3 of incident light is absorbed by the color filters
Solution Approach 1:
The patent transitions from a two-dimensional mosaic arrangement of color filters to a three-dimensional stacked structure with photoelectric conversion layers at different depths. This vertical stacking allows multiple color detections within the same pixel area, eliminating the need to block 2/3 of light with color filters while maintaining color signal accuracy.
Solution Approach 2:
The patent divides the photoelectric conversion function into multiple separate layers, each optimized for detecting specific color wavelengths. The first photoelectric conversion layer detects blue light, the second layer detects green light, and the third layer detects red light, allowing each layer to operate at peak efficiency without interference from other color filters.
2Measurement precision
If mosaic color filters are arranged on photoelectric conversion pixels, then color signals are produced, but resolution is reduced and false colors appear noticeably
Solution Approach 1:
The patent moves color detection from the horizontal plane to the vertical dimension by stacking photoelectric conversion layers at different depths. Each layer is positioned to detect specific wavelengths, allowing full-resolution color imaging without the aliasing and false colors that occur in mosaic filter arrangements.
Solution Approach 2:
The patent assigns different spectral detection characteristics to different layers based on their depth and material properties. The first layer (shallowest) is optimized for blue light detection, the second layer for green light, and the third layer (deepest) for red light, with each layer having locally optimized properties for its specific color detection task.
3Loss of energy
If triple wells (photodiodes) are formed at different depths to detect different colors, then high sensitivity and resolution are achieved, but spectra of R, G, and B signals are not separated sufficiently and heavy computation load is required for addition/subtraction processing
Solution Approach 1:
The patent enhances spectral separation by optimizing the local properties of each photoelectric conversion layer, including bandgap energy, thickness, and material composition. This creates natural spectral filtering effects that reduce cross-contamination between color signals, minimizing the need for complex post-processing subtraction operations.
Solution Approach 2:
The patent varies key parameters such as layer thickness, material bandgap, and doping concentrations across the three layers to achieve optimal spectral separation. By adjusting these parameters, the patent creates distinct detection windows for each color, reducing signal overlap and computation requirements.
4Loss of energy
If photoelectric conversion layers are formed in three layers with vertical interconnections, then light utilization efficiency approaches 100%, but manufacturing difficulty increases and production yields decrease
Solution Approach 1:
The patent designs a unified semiconductor processing platform that handles all three photoelectric conversion layers and their interconnections through standardized fabrication steps. This universal process architecture, using consistent photolithography, etching, and doping techniques across all layers, maintains manufacturing simplicity while achieving high light utilization efficiency.
Solution Approach 2:
The patent combines the formation of multiple photoelectric conversion layers and their interconnections into an integrated fabrication sequence. By merging the processing steps for creating distinct layers with the steps for forming vertical interconnections, the patent achieves complex three-dimensional structures through streamlined manufacturing processes rather than separate assembly operations.
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 image quality by increasing light utilization efficiency, reducing noise and afterimages, and simplifying the manufacturing process, while maintaining high sensitivity and resolution.
Implementation Method 1
a first photoelectric conversion layer 43 sensitive to green light
Implementation Method 2
second photoelectric conversion elements (photodiodes; 36) and third photoelectric conversion elements (photodiodes; 37) which are sensitive to blue light and red light, respectively
Data Source
AI summary
A solid-state imaging device as defined herein, in which each of the signal reading circuits for reading the detection signals of the first-color pixels includes three transistors which are a reset transistor, a row selection transistor, and an output transistor; and each of the signal reading circuits for reading the detection signals of the second-color pixels and each of the signal reading circuits for reading the detection signals of the third-color pixels include four transistors which are a read transistor, a reset transistor, a row selection transistor, and an output transistor.