Wafer Defect Detection Using Selective Fluorescence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor wafer inspection systems lack flexibility in light wavelengths for excitation and emission detection, struggle with patterned wafers that scatter light, and have difficulty distinguishing defects from background noise, leading to inefficient defect detection and characterization.
Innovation Solution
A system and method that utilize a customizable illumination subsystem to direct specific wavelengths of light to wafers, causing fluorescence only from target materials while avoiding other materials, combined with a detection subsystem that selectively captures fluorescence or non-fluorescent light, and a computer subsystem for defect information determination, enhancing defect detection and characterization on patterned wafers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single wavelength illumination is used, then the inspection system is simpler to operate, but the flexibility in detecting different materials and defects is reduced
Solution Approach 1:
The illumination system dynamically adjusts wavelength by switching between multiple light sources (UV, DUV, blue, green, red lasers) based on the inspection requirements. This allows the system to adapt to different materials and defect types while maintaining ease of operation through automated wavelength selection controlled by the computer subsystem.
Solution Approach 2:
The system incorporates multiple illumination sources covering a broad spectrum (UV to red wavelengths) that can detect various materials including organic, inorganic, metal, and semiconductor materials. This multi-functional approach enables a single system to handle diverse inspection tasks without requiring separate specialized equipment.
2Measurement precision
If fluorescence detection is used for all materials, then defect detection sensitivity is improved, but noise from patterned wafer backgrounds increases
Solution Approach 1:
The system applies different detection methods to different materials based on their local properties. Fluorescence detection is used selectively for materials that exhibit fluorescence (organic materials, oxides, nitrides), while non-fluorescence detection is used for materials that do not fluoresce (metals, silicon). This localized approach maximizes sensitivity for fluorescent materials while avoiding noise from non-fluorescent patterned backgrounds.
Solution Approach 2:
The system changes the detection parameter by offering two distinct detection modes: fluorescence detection for enhanced sensitivity to specific materials, and non-fluorescence detection for reducing background noise from patterned wafers. The computer subsystem automatically selects the appropriate detection mode based on the material being inspected, optimizing the signal-to-noise ratio for each case.
3Device complexity
If limited spectral filters are used, then the device complexity is reduced, but the ability to distinguish different materials and defects is limited
Solution Approach 1:
Instead of relying solely on spectral filtering in the wavelength dimension, the system uses the time dimension and material-specific fluorescence properties to achieve material discrimination. By exciting materials at specific wavelengths and detecting their characteristic fluorescence emissions, the system can distinguish between different materials without requiring complex multi-layer spectral filters for each wavelength band.
4Device complexity
If conventional inspection systems are used for patterned wafers, then the inspection process is simpler, but the signal-to-noise ratio deteriorates due to light scattering from patterns
Solution Approach 1:
The system converts the harmful light scattering from patterned wafer backgrounds into a beneficial effect by using fluorescence detection. When UV or DUV light illuminates the wafer, patterned features that would normally scatter light and create noise instead emit characteristic fluorescence. This fluorescence emission occurs at different wavelengths than the incident light, allowing the system to detect defects while the patterned features appear as uniform background fluorescence rather than scattering noise.
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 improves defect capture rate, sensitivity, and material discrimination, allowing for accurate defect classification and composition analysis by selectively illuminating and detecting fluorescence, reducing noise from patterned wafer backgrounds and enhancing signal-to-noise ratio.
Implementation Method 1
The one or more illumination wavelengths are selected to cause fluorescence from one or more materials on the wafer without causing fluorescence from one or more other materials on the wafer
Implementation Method 2
a detection subsystem configured to detect only the fluorescence from the one or more materials or to detect non-fluorescent light from the wafer without detecting the fluorescence from the one or more materials
Data Source
AI summary
Systems and methods for determining information for defects on a wafer are provided. One system includes an illumination subsystem configured to direct light having one or more illumination wavelengths to a wafer. The one or more illumination wavelengths are selected to cause fluorescence from one or more materials on the wafer without causing fluorescence from one or more other materials on the wafer. The system also includes a detection subsystem configured to detect only the fluorescence from the one or more materials or to detect non-fluorescent light from the wafer without detecting the fluorescence from the one or more materials. In addition, the system includes a computer subsystem configured to determine information for defects on the wafer using output generated by the detection subsystem responsive to the detected fluorescence or the detected non-fluorescent light.