Backside Illuminated Image Sensor Reducing Electrical Cross-Talk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As image sensors and pixels become smaller, they face challenges in efficiently capturing incident light due to reduced sensitivity and increased electrical cross-talk, which degrades image quality and color fidelity.
Innovation Solution
The implementation of a substrate with a backside illuminated design, featuring a photodiode, frontside P-wells, and an N-type region formed deeper than the P-wells, along with optimized P-type and N-type implantation and diffusion regions to reduce cross-talk and enhance sensitivity, including the use of backside processing for improved alignment and reduced straggle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If image sensors and pixels are made smaller to reduce component size, then device area is reduced, but light capture efficiency and sensitivity deteriorate
Solution Approach 1:
The patent inverts the conventional frontside illumination approach by implementing backside illumination. Light enters through the backside of the substrate where no circuitry blocks it, allowing photons to directly reach the photodiode without passing through metal interconnects and transistors. This inversion resolves the contradiction by maintaining high light capture efficiency in smaller pixels while still achieving compact sensor design.
Solution Approach 2:
The patent extends the light path into the depth dimension by creating a through-substrate structure where light travels through the entire substrate thickness to reach the photodiode. This dimensional approach allows smaller surface area pixels to maintain adequate light collection volume, resolving the trade-off between miniaturization and light capture efficiency.
2Area of stationary object
If pixel size is reduced to increase sensor integration, then sensor area is reduced, but electrical cross-talk between adjacent pixels increases
Solution Approach 1:
The patent introduces deep P-type isolation regions as intermediary structures between adjacent photodiodes. These isolation regions extend through the substrate and act as electrical barriers that prevent charge carrier diffusion between neighboring pixels. This mediator structure effectively blocks electrical cross-talk while allowing the sensor to maintain high pixel density and small overall area.
3Reliability
If substrate thickness is increased to improve light absorption, then light capture efficiency is improved, but electrical cross-talk between pixels increases
Solution Approach 1:
The patent segments the substrate into electrically isolated regions using deep P-type isolation structures that divide the continuous substrate into discrete pixel compartments. This segmentation allows each pixel to have adequate substrate thickness for light absorption while preventing electrical cross-talk through the isolation barriers, thus resolving the contradiction between thickness requirements for light capture versus cross-talk prevention.
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 pixel sensitivity and reduces electrical cross-talk, leading to improved signal-to-noise ratio and image quality by retaining photo-generated electrons within the intended pixel.
Implementation Method 1
Each pixel has a substrate, a photodiode formed in the substrate, frontside P-wells formed using frontside processing that are adjacent to the photosensitive region
Implementation Method 2
optimized P-type and N-type implantation and diffusion regions to reduce cross-talk and enhance sensitivity
Data Source
AI summary
An array of pixels is formed using a substrate, where each pixel has a substrate having a backside and a frontside that includes metalization layers, a photodiode formed in the substrate, frontside P-wells formed using frontside processing that are adjacent to the photosensitive region, and an N-type region formed in the substrate below the photodiode. The N-type region is formed in a region of the substrate below the photodiode and is formed at least in part in a region of the substrate that is deeper than the depth of the frontside P-wells.