CMP Eddy Current Monitoring for Thin-Layer Edge Thickness Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In-situ monitoring systems, such as eddy current systems, face challenges in accurately measuring substrate thickness near the edge during chemical mechanical polishing due to signal distortions and the unreliability of neural networks trained on thick calibration substrates for thin conductive layers, leading to errors in endpoint detection and polishing parameter adjustments.
Innovation Solution
The use of a neural network trained with calibration substrates of lower conductivity materials to generate adjusted thickness values, compensating for signal distortions and improving accuracy in measuring thin conductive layers by scaling measurements based on conductivity ratios, and integrating this system with a CMP apparatus for real-time profile control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If neural networks are trained on thick calibration substrates for in-situ monitoring, then the system can provide thickness measurements, but the measurements become unreliable for thin conductive layers near substrate edges
Solution Approach 1:
The patent changes the training parameters of the neural network by using calibration substrates with varying conductivities (including lower conductivity materials) and varying layer thicknesses (including thin layers). This allows the network to learn accurate thickness measurements across the full range of conditions, particularly improving reliability for thin conductive layers near substrate edges where previous systems failed.
Solution Approach 2:
The patent creates virtual training data by scaling measurements from calibration substrates with different conductivities. By using lower conductivity calibration substrates and applying conductivity-based scaling factors, the system generates equivalent training data for thin high-conductivity layers without physically creating all possible calibration samples, thus improving measurement reliability across different material types.
2Productivity
If in-situ monitoring is performed during CMP, then real-time thickness data is available, but signal distortions occur near substrate edges leading to incorrect endpoint detection
Solution Approach 1:
The patent applies parameter changes by training the neural network on calibration data that specifically includes edge region measurements from substrates with varying conductivities. This enables the network to compensate for signal distortions that occur near substrate edges during CMP, maintaining measurement precision while preserving real-time monitoring capabilities for productivity.
3Ease of manufacture
If standard calibration methods are used with high conductivity materials, then calibration is straightforward, but the system cannot accurately measure thin layers of the same material type
Solution Approach 1:
The patent uses lower conductivity calibration substrates as proxies to generate training data for measuring high conductivity materials. By scaling the measurements based on conductivity ratios, the system creates virtual calibration data for thin high-conductivity layers without requiring physically thin calibration substrates, thus maintaining calibration simplicity while achieving thin layer measurement precision.
Solution Approach 2:
The patent changes the conductivity parameter of calibration materials to lower conductivity values, making calibration substrates easier to manufacture with sufficient thickness for reliable measurements. The neural network then learns to compensate for this conductivity difference through scaling, enabling accurate measurement of thin high-conductivity layers without requiring difficult-to-manufacture thin calibration substrates.
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
Enhances the accuracy of thickness measurements at the substrate edge, improving within-wafer and wafer-to-wafer non-uniformity and enabling more precise endpoint control and polishing parameter adjustments, especially during the clearing step of the CMP process.
Implementation Method 1
an in-situ eddy current monitoring system to generate a plurality of measured signals values
Implementation Method 2
processing the initial thickness values through a neural network that was trained using training data acquired by measuring calibration substrates having a conductive layer formed of a second material having a second conductivity
Data Source
AI summary
A method of polishing a substrate includes polishing a conductive layer on the substrate at a polishing station, monitoring the layer with an in-situ eddy current monitoring system to generate a plurality of measured signals values for a plurality of different locations on the layer, generating thickness measurements the locations, and detecting a polishing endpoint or modifying a polishing parameter based on the thickness measurements. The conductive layer is formed of a first material having a first conductivity. Generating includes calculating initial thickness values based on the plurality of measured signals values and processing the initial thickness values through a neural network that was trained using training data acquired by measuring calibration substrates having a conductive layer formed of a second material having a second conductivity that is lower than the first conductivity to generated adjusted thickness values.