Back-Illuminated Sensor Deep Trench Isolation Crosstalk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Back-illuminated image sensors face challenges with reduced modulation transfer function (MTF) performance and increased electrical crosstalk due to lateral diffusion of photocarriers, especially as pixel size decreases, and issues with deep n-well contact connectivity affecting depletion region depth and charge collection efficiency.
Innovation Solution
The implementation of deep trench isolations starting at the frontside and extending beyond the depletion region, with a thin n-type dopant lining to reduce crosstalk and ensure continuous electrical connection between shallow and deep n-wells, and the use of fill trench isolation to completely isolate pixels and reduce optical crosstalk.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If back-illuminated image sensors are used to improve fill factor and quantum efficiency, then light detection capability is improved, but electrical crosstalk between pixels increases due to lateral diffusion of photocarriers
Solution Approach 1:
The sensor is divided into isolated pixel regions using deep trench isolations that extend beyond the depletion region. These trenches physically segment the semiconductor substrate, preventing lateral diffusion of photocarriers between adjacent pixels while maintaining individual pixel functionality for efficient light detection.
Solution Approach 2:
Different regions of the sensor have different conductivity types created through ion implantation. The deep n-well regions are selectively formed in specific areas to create localized electrical fields that confine photocarriers within their respective pixels, reducing crosstalk while preserving quantum efficiency in the photodetector regions.
2Productivity
If pixel size is decreased to increase resolution, then the number of pixels is increased, but modulation transfer function performance deteriorates due to increased crosstalk
Solution Approach 1:
Deep trench isolations create physical barriers between pixels, enabling higher pixel density without compromising MTF performance. The segmentation prevents photocarrier diffusion to adjacent pixels even as pixel dimensions decrease, maintaining image sharpness and resolution.
Solution Approach 2:
The isolation trenches extend in the vertical dimension beyond the depletion region depth, creating a three-dimensional barrier structure. This vertical extension provides effective crosstalk prevention without occupying horizontal space that would reduce pixel density.
3Object-generated harmful factors
If deep n-well contacts are implemented to reduce crosstalk, then electrical isolation is improved, but connectivity issues arise affecting depletion region depth and charge collection
Solution Approach 1:
The n-type dopant is applied selectively to line the sidewalls of deep trench isolations where needed for electrical connection. This localized doping creates continuous n-type pathways between shallow and deep n-wells in specific regions, ensuring reliable connectivity while maintaining the isolation function in other areas.
Solution Approach 2:
The n-type dopant lining acts as an intermediary conductive pathway between the shallow n-well contacts and the deep n-well regions. This intermediate n-type layer ensures continuous electrical connection through the isolation trenches, resolving connectivity issues while preserving the deep n-well's crosstalk reduction function.
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 significantly reduces electrical and optical crosstalk, maintains deep n-well connectivity, and enhances image quality by preventing photocarrier diffusion, thereby improving MTF performance and charge collection efficiency.
Implementation Method 1
lateral diffusion of photocarriers
Implementation Method 2
light-sensitive photodetectors that convert incident light into electrical signals
Implementation Method 3
Biasing both the shallow n-well 29 and deep n-well 21 to a known potential with respect to ground steers the photo-generated charges 31 into the photodetector 91
Data Source
AI summary
A back-illuminated image sensor includes a sensor layer having a frontside and a backside opposite the frontside. An insulating layer is situated adjacent the backside and a circuit layer is adjacent the frontside. A plurality of photodetectors of a first type conductivity convert light incident on the backside into photo-generated charges. The photodetectors are disposed in the sensor layer adjacent the frontside. A region of a second type conductivity is formed in at least a portion of the sensor layer adjacent the frontside and is connected to a voltage terminal for biasing the second type conductivity region at a predetermined voltage. A well of the second type conductivity is formed in the sensor layer adjacent the backside. Trench isolations in the sensor layer start at the frontside and extend beyond the depletion region of the photodiodes.