Phase Difference Pixel Sub-Lens Misalignment for Crosstalk Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image sensors face challenges in enhancing sensitivity and preventing optical crosstalk in phase difference detection pixels, particularly in high integration density devices like digital cameras and medical micro-cameras.

Innovation Solution

The image sensor design incorporates first and second sub-lenses with vertices misaligned from the central axis of photoelectric conversion elements, a microlens overlapping these sub-lenses, and an isolation pattern with a lower refractive index, which improves sensitivity and prevents optical crosstalk by optimizing lens shapes and positions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a conventional microlens structure is used in phase difference detection pixels, then the device structure is simple, but optical crosstalk occurs and sensitivity is reduced

Engineering Contradiction:
Improveoptical crosstalk preventionVSAvoidlens structure complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The single microlens is divided into multiple sub-lenses (first sub-lens and second sub-lens) that are positioned over different photoelectric conversion elements. Each sub-lens has a specific shape and position designed to direct light from its corresponding photoelectric conversion element to reduce optical crosstalk while maintaining detection sensitivity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Each sub-lens is designed with specific local characteristics including asymmetric convex lens shapes in one direction and bar or hemispherical shapes in the perpendicular direction. The vertices of the sub-lenses are positioned at specific locations relative to the photoelectric conversion elements and microlens central axis to optimize light direction and prevent crosstalk locally at each pixel region.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If the vertex of sub-lens is aligned with the central axis of photoelectric conversion element, then the lens structure is simple, but sensitivity of phase difference detection is reduced

Engineering Contradiction:
Improvephase difference detection sensitivityVSAvoidlens positioning complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The sub-lenses are designed with asymmetric shapes where the vertices are deliberately misaligned from the central axes of the underlying photoelectric conversion elements. The asymmetric convex lens shape in one direction and bar or hemispherical shape in the perpendicular direction create specific light path deviations that enhance phase difference detection sensitivity by better separating overlapping light paths from adjacent pixels.

Inventive Principle:
Principle #4Asymmetry

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 enhances the sensitivity of phase difference detection pixels while effectively preventing optical crosstalk, improving the overall performance of image sensors in various electronic devices.

Implementation Method 1

a first sub-lens formed over the first photoelectric conversion element, and having a vertex misaligned from a central axis of the first photoelectric conversion element; a second sub-lens formed over the second photoelectric conversion element

Methodology Applied
Scientific EffectRefraction: Refraction

Implementation Method 2

a microlens formed over the photoelectric conversion element so as to overlap the first and second sub-lenses

Methodology Applied
Scientific EffectFocusing: Lens

Data Source

PatentUS10446599B2Image sensor with phase difference detection pixel
Publication Date: 2019.10.15 SK HYNIX INC
  • US10446599B2 patent drawing
  • US10446599B2 patent drawing
  • US10446599B2 patent drawing

AI summary

An image sensor includes a pixel array having a plurality of pixels arranged therein. At least any one of the plurality of pixels include: a photoelectric conversion unit including first and second photoelectric conversion elements; a first sub-lens formed over the first photoelectric conversion element, and having a vertex out of a central axis of the first photoelectric conversion element; a second sub-lens formed over the second photoelectric conversion element, and having a vertex out of a central axis of the second photoelectric conversion element; and a microlens formed over the photoelectric conversion element so as to overlap the first and second sub-lenses. The first sub-lens is symmetrical with the second sub-lens, based on a boundary surface between the first and second photoelectric conversion elements.