Stacked CMOS Image Sensor Dual-Layer Photoelectric Conversion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing imaging technologies face challenges in providing a wide variety of photoelectric conversion outputs for different wavelength regions with one pixel, and the manufacturing processes often require ultra-high energy ion implantation and specialized facilities, increasing costs and complexity.

Innovation Solution

The implementation of a CMOS image sensor with a dual-layer structure, where one layer is responsible for visible light and the other for near-infrared light photoelectric conversion, using a support substrate with integrated photo diodes to achieve efficient conversion without the need for ultra-high energy ion implantation, allowing for independent charge accumulation and reading from each layer.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If ultra-high energy ion implantation is used to form deep photo diodes for infrared light detection, then photoelectric conversion efficiency for infrared light is improved, but manufacturing cost and device complexity considerably increase

Engineering Contradiction:
Improvephotoelectric conversion efficiencyVSAvoidmanufacturing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The invention divides the single substrate into two separate substrates: a first substrate containing photo diodes for visible light detection and a second substrate containing photo diodes for infrared light detection. Each substrate can be manufactured using conventional ion implantation energies suitable for its specific detection depth requirements, avoiding the need for ultra-high energy implantation on a single substrate.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention transitions from a single-substrate vertical structure to a multi-substrate stacked structure. By adding the dimension of multiple substrates stacked together, each substrate can be optimized independently for its wavelength range with appropriate photo diode depths, eliminating the need for ultra-high energy ion implantation to achieve deep photo diodes on a single substrate.

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

2Adaptability or versatility

If ultra-high energy ion implantation is used to form deep photo diodes, then infrared light detection capability is improved, but development cost and manufacturing cost increase

Engineering Contradiction:
Improveinfrared detection capabilityVSAvoidmanufacturing cost
Core Design Contradiction:
Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The invention segments the detection function into two separate substrates: one optimized for visible light and another for infrared light. Each substrate uses conventional ion implantation energies appropriate for its detection depth, avoiding the need for expensive ultra-high energy implantation facilities and reducing development costs.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention introduces an optical member (such as a dichroic mirror or beam splitter) as an intermediary between the light source and the two substrates. This optical member directs different wavelength ranges to the appropriate substrate, enabling both visible and infrared detection using conventional manufacturing processes.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Length of stationary object

If ion is implanted from both front and rear surfaces to form deep photo diodes, then photoelectric conversion area depth is doubled, but manufacturing process complexity and activation processing requirements increase

Engineering Contradiction:
Improvephotoelectric conversion depthVSAvoidmanufacturing process complexity
Core Design Contradiction:
Length of stationary objectVSDevice complexity

Solution Approach 1:

Instead of attempting to create deep photo diodes by implanting from both surfaces of a single substrate, the invention segments the function into two separate substrates. Each substrate undergoes ion implantation from only one surface (front or rear) to create photo diodes at the appropriate depth for its wavelength range, simplifying the manufacturing process.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention resolves the depth complexity issue by moving from a single-substrate approach requiring bidirectional implantation to a multi-substrate stacked approach. Each substrate is processed independently with unidirectional ion implantation, and the combined stack achieves the equivalent of deep photoelectric conversion without the manufacturing complexity.

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

This configuration enables efficient photoelectric conversion in multiple wavelength regions for one pixel, reducing manufacturing costs and complexity while maintaining high image quality, and allows for flexible control over charge accumulation times and image synthesis.

Implementation Method 1

a photoelectric conversion element layer including a photoelectric conversion element and a support substrate including another photoelectric conversion element, wherein incident light is photoelectrically converted

Methodology Applied
Scientific EffectPhotoelectric conversion: Photoelectric Effect

Data Source

PatentEP2879181B1Image pickup element, electronic device, and information processing device
Publication Date: 2022.04.06 SONY GROUP CORP
  • EP2879181B1 patent drawingFigure 1
  • EP2879181B1 patent drawingFigure 2
  • EP2879181B1 patent drawingFigure 3

AI summary

The present disclosure relates to an imaging element, an electronic device, and an information processing device capable of more easily providing a wider variety of photoelectric conversion outputs. An imaging element of the present disclosure includes: a photoelectric conversion element layer containing a photoelectric conversion element that photoelectrically converts incident light; a wiring layer formed in the photoelectric conversion element layer on the side opposite to a light entering plane of the incident light, and containing a wire for reading charges from the photoelectric conversion element; and a support substrate laminated on the photoelectric conversion element layer and the wiring layer, and containing another photoelectric conversion element. The present disclosure is applicable to an imaging element, an electronic device, and an information processing device.