Fluid Bath Bubble Tracking for Wafer Recipe Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor processing, correlating differences in wafer processing results to varying parameters in wafer baths is difficult without a fiducial marker, and there is a need for an apparatus and method to detect and monitor objects in the fluid bath to adjust processing recipes based on measured object properties.
Innovation Solution
An apparatus and method that uses video data processing circuitry to detect and track objects, such as bubbles, in a fluid bath, determining their metrics and positions across frames, and adjusts processing recipes based on these measurements to improve wafer quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video data processing is used to detect and track objects in fluid bath, then measurement precision of fluid parameters is improved, but device complexity increases
Solution Approach 1:
The patent replaces traditional mechanical measurement systems with video data processing and image analysis. Objects in the fluid bath are tracked using computer vision algorithms that analyze video frames to determine position, size, and movement of bubbles, eliminating the need for complex mechanical sensors while achieving high measurement precision.
Solution Approach 2:
The patent uses optical copying through video imaging to capture and analyze the physical state of objects in the fluid bath. By creating digital copies (video frames) of the physical system, the measurement process becomes non-intrusive and can be performed without physical contact, thereby improving precision while keeping the actual measurement apparatus simple.
2Manufacturing precision
If object detection and tracking is implemented to monitor fluid bath, then manufacturing precision of wafer processing is improved, but ease of operation deteriorates
Solution Approach 1:
The system performs self-service through automated object detection and tracking algorithms that continuously monitor the fluid bath without human intervention. The processing circuitry automatically analyzes video data, tracks object positions and sizes, and correlates these measurements with wafer processing parameters, eliminating the need for manual operation while maintaining high manufacturing precision.
Solution Approach 2:
The patent implements feedback mechanisms where detected object properties (position, size, movement) are continuously fed back to the processing system. This feedback loop allows the system to automatically adjust and correlate fluid bath conditions with wafer processing outcomes, improving manufacturing precision through data-driven insights without requiring complex manual operation.
3Reliability
If multiple frames are analyzed to track object movement, then reliability of fluid parameter measurement is improved, but loss of time increases
Solution Approach 1:
The patent applies partial action by analyzing only the necessary portions of video data to achieve reliable measurements. Instead of processing every single frame in detail, the system selectively analyzes frames where objects are visible and relevant measurements can be obtained, reducing overall processing time while maintaining measurement reliability through strategic sampling of critical moments.
Solution Approach 2:
The system performs preliminary action by pre-processing video frames to identify and flag frames containing relevant objects before detailed analysis. This preliminary filtering step allows the system to quickly scan through multiple frames, identify those worth analyzing in detail, and thus maintain reliability through comprehensive monitoring while minimizing the time spent on exhaustive frame-by-frame processing.
Data Source
AI summary
Techniques herein include an apparatus and method for measuring and monitoring properties of objects, such as bubbles, detected in a fluid in a semiconductor processing apparatus. The method can track the detected objects over time through multiple frames of video data and determine metrics for the detected objects to improve tracking accuracy as well as correlate recipe parameters to resulting wafer quality processed via the recipe parameters. Based on the determined correlation between the wafer quality data and the determined object metrics, the recipe parameters can be adjusted to improve wafer quality further.