Polishing Endpoint Detection via Dynamic Zone Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semiconductor substrate polishing technologies face challenges in accurately detecting the polishing endpoint, leading to potential insufficient or over-polishing, which can adversely affect device performance due to non-uniform film removal across the substrate.
Innovation Solution
A method and program that utilize a film-thickness sensor to create a profile of the substrate, divide it into measurement zones, and dynamically select a monitoring zone based on film thickness differences or variations, using a learned model to determine the polishing endpoint by adjusting the zones according to predetermined thresholds or inflection points, ensuring accurate endpoint detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single fixed monitoring zone is used for polishing endpoint detection, then the detection process is simple, but the accuracy deteriorates due to non-uniform film removal across the substrate
Solution Approach 1:
The substrate surface is divided into multiple measurement zones (first measurement zone, second measurement zone, etc.) based on radial distance from the center. Each zone is monitored independently, allowing the system to detect different polishing states in different regions. This segmentation enables accurate endpoint detection by identifying the zone with the smallest film thickness, thereby resolving the contradiction between detection accuracy and system complexity.
Solution Approach 2:
The monitoring zone is dynamically selected based on real-time film thickness measurements. The system identifies which measurement zone has the smallest film thickness at each polishing stage and switches monitoring focus to that zone. This dynamic adaptation allows the system to track the evolving polishing state accurately without requiring complex fixed-zone analyses, improving detection accuracy while managing complexity through algorithmic selection rather than physical complexity.
2Manufacturing precision
If film thickness is measured at multiple points to create a profile, then the manufacturing precision improves, but the measurement time increases
Solution Approach 1:
The measurement process is segmented into multiple zones rather than continuous scanning. By dividing the substrate into discrete measurement zones and selecting only the critical zone (with smallest film thickness) for detailed monitoring, the system achieves high manufacturing precision without requiring exhaustive measurement of every point, thereby reducing total measurement time.
Solution Approach 2:
The system performs preliminary measurements across multiple zones to identify which zone has the smallest film thickness, then focuses subsequent detailed monitoring on that specific zone. This preliminary action allows the system to efficiently allocate measurement resources to the most critical area, improving manufacturing precision while minimizing time spent on less critical regions.
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 enables precise detection of the polishing endpoint, reducing the risk of insufficient or over-polishing and improving the uniformity of the substrate surface, thus enhancing the performance and reliability of semiconductor devices.
Implementation Method 1
a film-thickness sensor mounted to the polishing table moves across the substrate, thereby creating a film-thickness profile of the substrate
Implementation Method 2
a polishing apparatus for performing CMP includes a polishing table that supports a polishing pad having a polishing surface... The surface of the substrate is brought into sliding contact with the polishing surface in the presence of the polishing liquid, so that the surface of the substrate is polished to a flat and mirror finish
Data Source
AI summary
A polishing endpoint detecting method capable of accurately detecting a polishing endpoint of a substrate is disclosed. The polishing endpoint detecting method includes: polishing a substrate W; creating a film-thickness profile of the substrate by used of a film-thickness sensor 7 mounted to a polishing table 2; dividing the film-thickness profile into a plurality of measurement zones and selecting a monitoring zone from the plurality of measurement zones for determining a polishing endpoint of the substrate; determining whether or not the monitoring zone is changed to another measurement zone based on changes in the film-thickness profile obtained from start to end of polishing of the substrate; and determining a polishing endpoint of the substrate based on the signals in the monitoring zone.