Pixel Isolation Trench Layout for Low-Crosstalk Image Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image sensor manufacturing processes face challenges in achieving high yield and precision in forming patterns on substrates, particularly in defining pixel regions and preventing crosstalk between adjacent pixels.
Innovation Solution
The image sensor design incorporates a deep trench and shallow trench structure on a substrate with specific corner radii of curvature to define pixel regions and prevent crosstalk, using a pixel separation structure with insulating, conductive, and capping patterns to enhance pattern definition and isolation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional pattern formation processes are used, then manufacturing simplicity is maintained, but manufacturing precision and yield are insufficient
Solution Approach 1:
The isolation structure is divided into two distinct components: deep trenches that extend through the substrate to define pixel regions, and shallow trenches that extend partially into the substrate to define patterns within pixel regions. This segmentation allows each trench type to perform its specific function optimally, with deep trenches providing complete isolation between pixels and shallow trenches providing pattern definition, thereby achieving high manufacturing precision without requiring overly complex single-structure solutions.
Solution Approach 2:
The patent introduces a depth dimension differentiation between deep trenches (extending through the substrate) and shallow trenches (extending partially into the substrate). This dimensional variation allows the same basic trench structure to serve multiple functions at different depth levels, improving pattern formation precision while maintaining reasonable structural complexity.
2Reliability
If uniform corner radii are used in patterns, then manufacturing simplicity is maintained, but crosstalk prevention between adjacent pixels is insufficient
Solution Approach 1:
Different corner radii are assigned to different corners of the same pattern based on their spatial location and functional requirements. Specifically, first and second corners have a first radius of curvature, while third and fourth corners have a second radius of curvature. This local differentiation optimizes crosstalk prevention at critical locations while maintaining manufacturability, as the varying corner radii can be controlled through standard photolithography processes without requiring excessive precision.
3Reliability
If deep trenches extend through the entire substrate, then pixel region isolation is improved, but manufacturing complexity and cost increase
Solution Approach 1:
The isolation function is segmented between deep trenches (providing complete pixel region isolation) and shallow trenches (providing pattern-level isolation). This segmentation allows deep trenches to be optimized for their specific through-substrate isolation function while shallow trenches handle pattern definition, making the overall manufacturing process more manageable and cost-effective than using a single complex deep trench structure for all isolation requirements.
Solution Approach 2:
Shallow trenches provide partial isolation by extending only partway into the substrate, which is sufficient for their specific function of defining patterns within pixel regions. This partial action approach reduces manufacturing complexity compared to extending all trenches through the entire substrate, while still achieving the necessary isolation效果 for pattern definition.
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 design increases the yield of pattern formation and reduces crosstalk between pixels, enabling higher precision and efficiency in image sensor manufacturing while maintaining compatibility with existing CMOS process technology.
Implementation Method 1
A photoelectric conversion region is disposed within the pixel region
Data Source
AI summary
An image sensor includes a substrate having a first surface and a second surface facing each other. A deep trench extends from the first surface to the second surface of the substrate and defines a pixel region within the substrate. A photoelectric conversion region is disposed within the pixel region. A shallow trench extends from the first surface of the substrate into the substrate and at least partially defines a pattern within the pixel region. In a plan view, the pattern has a first corner, a second corner facing the first corner in a first diagonal direction, a third corner, and a fourth corner facing the third corner in a second diagonal direction that crosses the first diagonal direction. In the plan view, a radius of curvature of the third corner is less than a radius of curvature of the first corner.