Stacked CMOS Image Sensor Pixel Binning via Common Floating Diffusion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

As pixel size in CMOS image sensors decreases to increase the number of pixels, the illuminated area of photodetectors also shrinks, leading to reduced signal levels and degraded performance.

Innovation Solution

A CMOS image sensor structure with multiple sensor layers and interlayer connectors that allow pixel binning onto a common floating diffusion, enabling improved light collection and signal processing without the need for additional wavelength selective filters, thereby enhancing quantum efficiency and sensitivity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If pixel size is decreased to increase the number of pixels, then the number of pixels is improved, but the illuminated area of photodetectors is reduced and signal level decreases

Engineering Contradiction:
Improvenumber of pixelsVSAvoidsignal level
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

Multiple photodetectors (e.g., four photodetectors in a 2x2 array) are merged to share a common floating diffusion node. This combining of multiple small photodetectors allows the sensor to maintain high pixel density while improving signal collection capability, as the shared floating diffusion integrates signals from multiple photodetectors effectively increasing the functional light-sensitive area.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent transitions from a planar single-layer pixel structure to a three-dimensional stacked structure with multiple sensor layers vertically integrated. By stacking multiple layers of photodetectors above common floating diffusion nodes, the design exploits the vertical dimension to increase the effective light collection volume without increasing the horizontal pixel footprint, thereby maintaining high pixel density while improving signal levels.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Reliability

If multiple sensor layers are stacked to improve light collection, then quantum efficiency is improved, but device complexity increases

Engineering Contradiction:
Improvequantum efficiencyVSAvoidstructure complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The common floating diffusion nodes serve multiple functions: they act as charge collection nodes for multiple photodetectors across different layers, serve as charge-to-voltage conversion nodes, and provide a shared readout path. This multi-functionality reduces the need for separate circuitry for each photodetector layer, thereby limiting the increase in device complexity despite the stacked multi-layer structure.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

By stacking sensor layers in the vertical dimension, the patent increases light collection capability without expanding the horizontal footprint. The vertical integration allows multiple photodetector layers to share common circuitry (floating diffusions, readout circuitry) located in the same horizontal plane, thus improving quantum efficiency while containing structural complexity through three-dimensional packaging rather than planar expansion.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

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

The solution enhances quantum efficiency and sensitivity by allowing multiple photodiodes to share a common floating diffusion, reducing optical interference and eliminating the need for certain color filter layers, thus improving image capture performance.

Implementation Method 1

A first sensor layer 101 having a first array of pixels 111 is provided. A second sensor layer 102 is situated over the first sensor layer 101, which has a second array of pixels 112. The first sensor layer 101 has a first thickness T1 to collect light with a first preselected range of wavelengths and the second sensor layer has a second thickness T2 to collect light with a second preselected range of wavelengths

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentEP2313927B1Image sensor having multiple sensing layers
Publication Date: 2015.12.02 OMNIVISION TECHNOLOGIES INC
  • EP2313927B1 patent drawingFigure 1
  • EP2313927B1 patent drawingFigure 2~4B
  • EP2313927B1 patent drawingFigure 5

AI summary

An image sensor includes a first sensor layer (101) having a first array of pixels and a second sensor (102) layer having a second array of pixels. Each pixel of the first and second arrays has a photodetector (140) for collecting charge in response to incident light, a charge-to- voltage conversion mechanism (144), and a transfer gate (142) for selectively transferring charge from the photodetector to the charge- to- voltage mechanism. The first and second sensor layers each have a thickness to collect light with a first and second preselected ranges of wavelengths, respectively. A circuit layer (120) is situated below the first sensor layer and has support circuitry (122) for the pixels of the first and second sensor layers, and interlayer connectors (130) are between the pixels of the first and second layers and the support circuitry.