Micro-LED Photoluminescence Sorting for Dense uASSEMBLER Chiplets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional machine vision systems struggle to identify and differentiate micro-LEDs with special ID marks due to the increasing miniaturization and density of chiplets, leading to reduced optical resolution and impaired manufacturing efficiency.
Innovation Solution
A machine vision system utilizing excitation light to induce photoluminescence in micro-LEDs, combined with image processing and high-resolution cameras, enables precise identification and orientation of micro-LEDs by detecting luminescence light signals and employing image stitching techniques to enhance overall field-of-view and resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional machine vision systems are used to identify micro-LEDs with special ID marks, then the system can identify large chiplets with ID marks, but the optical resolution becomes insufficient as chip size decreases to low tens of microns and component density increases
Solution Approach 1:
The patent uses photoluminescence color changes to identify different types of micro-LEDs. Each micro-LED type (Red, Green, Blue) emits a characteristic color when excited by UV light, enabling the machine vision system to distinguish between different chiplet types without requiring visible ID marks. This color-based identification method maintains measurement precision even as component size decreases and density increases.
Solution Approach 2:
The system changes the excitation parameter by using UV light instead of visible light for illumination. This parameter change enables the detection of photoluminescence emissions from micro-LEDs, providing a new contrast mechanism that works effectively at high component densities and small sizes where traditional visual ID mark detection fails.
2Quantity of substance
If the number of heterogeneous chiplets in a defined real estate increases for high-resolution displays, then the display pixel count increases, but the ability to identify individual micro-components deteriorates
Solution Approach 1:
The patent exploits the inherent color emission properties of different micro-LED types (Red, Green, Blue) when excited by UV light. This allows the machine vision system to identify individual micro-components and distinguish between different types based on their photoluminescence color, maintaining identification ability even when the number of heterogeneous chiplets increases to fill high-resolution display requirements.
Solution Approach 2:
The system introduces UV excitation light as an intermediary that enables the micro-LEDs to emit photoluminescence. This intermediary mechanism provides a new detection pathway that allows identification of individual micro-components at high densities, where direct visual inspection of ID marks becomes impossible.
3Area of moving object
If chip size is continuously miniaturized to increase component density, then the number of components per unit area increases, but the optical resolution required to see ID markings becomes unattainable
Solution Approach 1:
The patent uses photoluminescence color emission as an alternative to visual ID mark detection. When micro-LEDs are excited by UV light, they emit characteristic colors (Red, Green, Blue) that can be detected by the machine vision system. This color-based identification method works effectively at the miniaturized chip sizes required for high component density, where traditional ID mark visibility becomes impossible.
Solution Approach 2:
The system replaces the mechanical/visual inspection method (reading ID marks) with an optical detection method (detecting photoluminescence emissions). This substitution enables identification of miniaturized components without relying on visible ID markings, maintaining measurement precision even as chip size decreases to low tens of microns.
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 system effectively identifies and orients micro-LEDs with high accuracy, improving manufacturing efficiency by providing optical feedback for precise assembly, even in densely packed environments.
Implementation Method 1
The excitation light wavelength range is selected to cause certain micro-components on the planar working surface to emit photoluminescence light signals in response to being illuminated by the incident excitation light within the excitation light wavelength range.
Data Source
AI summary
A machine vision system and method uses photoluminescence light response of micro-LEDs to identify types of micro-LEDs (e.g., red, green, or blue) that are used to assemble a micro-LED display. Excitation light (e.g., ultraviolet excitation light) in a certain wavelength range is illuminated on a random pool of heterogeneous micro-LEDs consisting of materials, for example, that photoluminesce in three different colors-red, green, or blue. The micro-LED is optically excited and will emit either red, green, or blue, photoluminescence light based on the type of the micro-LED. The machine vision system uses a camera device that includes color response sensors to differentiate the type of micro-LED. The orientation of the micro-LED can also be detected. The machine vision system, based on the type, location, and orientation of the heterogeneous micro-LEDs, provides image-data based optical feedback to a microassembler system to move the micro-LEDs on a planar working surface according to an electrostatic template.