3D Stacked BSI Sensor NIR Quantum Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image sensors, particularly back-side illuminated (BSI) sensors, face challenges in fully realizing near-infrared (NIR) signals due to the longer wavelength of NIR light, leading to inefficiencies in converting this light into current, resulting in reduced quantum efficiency and incomplete signal capture.
Innovation Solution
The implementation of 3D stacked BSI sensors with metal reflectors and reflective gratings that cause NIR light to reflect back through the photodetector region, effectively doubling the thickness of the photodetector region and enhancing the conversion of NIR light to current, thereby improving spectral response and quantum efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If conventional BSI sensors are used, then the device size is reduced, but the quantum efficiency for NIR detection deteriorates
Solution Approach 1:
The patent introduces a vertical stacking architecture where multiple sensor layers are arranged in the depth dimension. The NIR-optimized layer is positioned below the visible light layer, allowing both spectral ranges to be detected within a compact footprint. This vertical dimensionality resolves the contradiction by enabling specialized NIR detection without increasing the device's planar size.
Solution Approach 2:
The sensor is segmented into functionally distinct layers: a visible light-optimized photodetector layer and a separate NIR-optimized photodetector layer. Each layer is independently designed with appropriate photodetector materials and structures tailored to its specific spectral range, allowing both to coexist in a single device without compromising either performance.
2Reliability
If the photodetector region thickness is increased to improve NIR absorption, then the quantum efficiency for NIR wavelengths improves, but the device complexity increases
Solution Approach 1:
Instead of increasing the thickness of a single photodetector layer, the patent distributes the detection function across multiple layers in the vertical dimension. The NIR-optimized layer maintains a thickness sufficient for NIR absorption, while the overall device complexity is managed through systematic layering rather than monolithic thickening.
Solution Approach 2:
Each photodetector layer is locally optimized for its specific spectral range. The NIR layer uses materials and structures specifically tailored for NIR wavelengths, while the visible light layer uses different optimizations. This local quality approach allows each layer to achieve high quantum efficiency for its target range without requiring the entire device to be overly complex.
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 significantly increases the quantum efficiency of BSI sensors, allowing them to capture a broader range of NIR wavelengths, enhancing depth sensing capabilities and image quality in applications where NIR light is utilized.
Implementation Method 1
metal reflectors and reflective gratings that cause NIR light to reflect back through the photodetector region
Implementation Method 2
converting this light into current
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
One innovation includes an IR sensor (400) having an array of sensor pixels (401a-d) to convert light into current, each sensor pixel of the array including a photodetector region (404, 412), a lens (402) configured to focus light into the photodetector region, the lens adjacent to the photodetector region so light propagates through the lens and into the photodetector region, and a substrate (421) disposed with photodetector region between the substrate and the lens, the substrate having one or more transistors (410) formed therein. The sensor also includes reflective structures (408) positioned between at least a portion of the substrate and at least a portion of the photodetector region and such that at least a portion of the photodetector region is between the one or more reflective structures and the lens, the one or more reflective structures configured to reflect the light that has passed through at least a portion of the photodetector region into the photodetector region.