Graded SiGe Image Sensor for Low Noise NIR Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
SiGe photodiodes in image sensors suffer from high dark noise and random noise due to lattice mismatch, leading to increased image lag and reduced sensitivity, especially in near-infrared applications.
Innovation Solution
A gradual change in the Ge concentration along the thickness of the SiGe layer (Si1-xGex) is implemented to reduce lattice mismatch, positioning the deepest potential point shallowly within the Si or Si1-xGex layer, facilitating easier charge transfer and enhancing sensitivity to near-infrared light.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SiGe photodiodes are used to increase sensitivity to near infrared light, then sensitivity to near infrared light is improved, but dark noise and random noise increase
Solution Approach 1:
The patent applies local quality by creating a non-uniform Ge concentration distribution within the SiGe layer. The Ge concentration varies spatially, being higher in regions that benefit from enhanced NIR sensitivity while lower in regions where noise generation is problematic. This localized variation in material composition allows different parts of the photodiode to optimize for different functions, resolving the contradiction between sensitivity and noise.
Solution Approach 2:
The patent changes the compositional parameter of the SiGe layer by varying the Ge concentration from a uniform to a graded distribution. This parameter change transforms the material properties throughout the layer, enabling regions with optimal Ge content for NIR absorption while minimizing regions that generate excessive noise, thus resolving the technical contradiction.
2Measurement precision
If SiGe photodiodes are used to increase sensitivity to near infrared light, then sensitivity to near infrared light is improved, but image lag increases
Solution Approach 1:
The graded Ge concentration creates local variations in charge carrier mobility and recombination characteristics. Regions with lower Ge content facilitate faster charge extraction and reduced trapping, thereby decreasing image lag in critical areas while maintaining high NIR sensitivity in regions with higher Ge content.
Solution Approach 2:
By varying the Ge concentration parameter through the layer thickness, the patent optimizes the balance between sensitivity and response time. The parameter change enables faster charge collection in certain regions, reducing the time delay that causes image lag while preserving the enhanced NIR sensitivity where needed.
3Ease of manufacture
If uniform Ge concentration is used in SiGe layer, then manufacturing is simplified, but lattice mismatch increases causing noise
Solution Approach 1:
Instead of a uniform Ge concentration, the patent implements a graded Ge concentration where each local region has an optimized composition. This gradual variation in material quality reduces lattice mismatch at interfaces and dislocations, minimizing noise generation while remaining compatible with existing manufacturing processes through controlled deposition techniques.
Solution Approach 2:
The patent introduces dynamic variation in the Ge concentration profile rather than a static uniform composition. This dynamic approach, where the Ge content changes continuously or in steps through the layer, allows the material structure to adapt and reduce lattice strain, thereby reducing noise while maintaining manufacturability through established graded structure fabrication methods.
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 approach effectively decreases dark noise and image lag while maintaining high sensitivity to near-infrared light, improving the overall performance of image sensors by reducing lattice mismatch and optimizing charge transfer efficiency.
Implementation Method 1
SiGe photodiodes suffer from relatively high dark noise and random noise due to lattice mismatch
Implementation Method 2
photodiodes of the image sensor are made of silicon-germanium (SiGe) semiconductor, because of a relatively high sensitivity of the SiGe photodiodes to near infrared rays (NIR)
Implementation Method 3
The charge generation region extends across an interface between a first Si layer and a second Si1-xGex layer, wherein x has a maximum ratio in the charge generation region
Data Source
AI summary
Low noise silicon-germanium (SiGe) image sensor. In one embodiment, an image sensor includes a plurality of pixels arranged in rows and columns of a pixel array disposed in a semiconductor substrate. The photodiodes of an individual pixel are configured to receive an incoming light through an illuminated surface of the semiconductor substrate. The semiconductor substrate includes a first layer of semiconductor material having silicon (Si); and a second layer of semiconductor material having silicon germanium (Si1-xGex). A concentration x of Ge changes gradually through at least a portion of thickness of the second layer. Each photodiode includes a first doped region extending through the first layer of semiconductor material and the second layer of semiconductor material; and a second doped region extending through the first layer of semiconductor material and the second layer of semiconductor material.