CMP Pad Thickness Compensation for Accurate Endpoint Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing chemical mechanical polishing (CMP) systems face challenges in accurately determining substrate thickness and uniformity due to signal distortions near the substrate edge caused by changes in polishing pad thickness, which affect endpoint detection and process control.
Innovation Solution
An in-situ monitoring system using an eddy current sensor and machine learning, specifically a neural network, compensates for signal distortions by adjusting measured values based on polishing pad thickness, enabling precise thickness measurements and real-time process control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional in-situ monitoring systems are used without pad thickness compensation, then the system structure remains simple, but measurement precision deteriorates due to signal distortions near substrate edge
Solution Approach 1:
The system performs preliminary determination of polishing pad thickness before conducting thickness measurements of the conductive layer. This preliminary action allows the system to compensate for pad thickness variations during subsequent measurements, eliminating signal distortions without requiring complex hardware modifications
Solution Approach 2:
The system introduces an intermediary compensation mechanism that uses the determined pad thickness value to adjust and correct the raw sensor signals. This intermediary step transforms the distorted measurements into accurate thickness values, resolving the contradiction between measurement precision and system complexity
2Manufacturing precision
If pad thickness changes are not compensated, then the system operation remains simple, but manufacturing precision deteriorates due to inaccurate thickness control
Solution Approach 1:
The system implements a feedback mechanism where the determined pad thickness is continuously used to compensate and adjust the thickness measurements during the polishing process. This feedback loop ensures accurate thickness control and uniformity across the substrate, maintaining manufacturing precision while keeping the operation straightforward
3Reliability
If signal distortions are not corrected, then the data processing remains simple, but reliability deteriorates due to inaccurate endpoint detection
Solution Approach 1:
The system performs preliminary determination of pad thickness characteristics before endpoint detection. This preliminary action enables the subsequent endpoint detection algorithm to compensate for signal distortions, significantly improving detection accuracy without requiring overly complex processing during the critical endpoint detection phase
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 provides improved within-wafer and wafer-to-wafer non-uniformity by accurately detecting polishing endpoints and adjusting parameters, ensuring consistent layer thickness across the substrate.
Implementation Method 1
an eddy current sensing system may be used to induce eddy currents in a conductive region on the substrate to determine parameters such as the local thickness of the conductive region
Data Source
AI summary
Data received from an in-situ monitoring system includes, for each scan of a sensor, a plurality of measured signal values for a plurality of different locations on a layer. A thickness of a polishing pad is determined based on the data from the in-situ monitoring system. For each scan, a portion of the measured signal values are adjusted based on the thickness of the polishing pad. For each scan of the plurality of scans and each location of the plurality of different locations, a value is generated representing a thickness of the layer at the location. This includes processing the adjusted signal values using one or more processors configured by machine learning. A polishing endpoint is detected or a polishing parameter is modified based on the values representing the thicknesses at the plurality of different locations.