CMP Thickness Monitoring Using Neural Compensation for Pad Wear
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Chemical mechanical polishing (CMP) processes face challenges in accurately monitoring substrate thickness and compensating for signal distortions caused by varying polishing pad thickness, leading to inconsistencies in polishing endpoint and non-uniformity across wafers.
Innovation Solution
A neural network is trained using thickness measurements corresponding to different polishing pad thicknesses to generate corrected thickness values, which are used to compensate for signal distortions and improve endpoint control and within-wafer and wafer-to-wafer non-uniformity by adjusting polishing parameters in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If in-situ monitoring systems are used to measure substrate thickness during CMP, then real-time endpoint detection capability is improved, but measurement precision deteriorates due to signal distortions from varying polishing pad thickness
Solution Approach 1:
The patent introduces a neural network as an intermediary computational layer between the raw in-situ monitoring signals and the final thickness measurement. The neural network processes the distorted signals by learning the relationship between pad thickness variations and signal characteristics, effectively mediating the measurement process to compensate for pad wear effects and restore measurement precision.
Solution Approach 2:
The system dynamically adjusts measurement parameters by incorporating real-time polishing pad thickness measurements into the analysis. By changing the interpretation parameters based on actual pad conditions rather than assuming constant pad thickness, the system maintains measurement accuracy despite pad wear during the polishing process.
2Manufacturing precision
If polishing pad is conditioned frequently to maintain uniform abrasive condition, then polishing uniformity is improved, but productivity deteriorates due to increased process time
Solution Approach 1:
The patent replaces the mechanical conditioning process with a computational correction system. Instead of physically reconditioning the pad frequently to maintain performance, the system uses neural network-based signal processing to compensate for pad wear effects, substituting mechanical intervention with information processing to maintain polishing uniformity.
Solution Approach 2:
The system implements continuous feedback by monitoring both substrate thickness in real-time and polishing pad thickness, then using this feedback information to adjust the endpoint detection criteria dynamically. This allows the system to maintain polishing uniformity without requiring frequent physical pad conditioning interruptions.
3Measurement precision
If neural network is trained with multiple pad thickness conditions to compensate for signal distortions, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The neural network is designed as a universal correction system that handles multiple pad thickness conditions through a single integrated model. Rather than requiring separate measurement systems for different pad conditions, the universal neural network processes all conditions through one computational framework, maintaining precision while limiting complexity growth.
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 solution effectively compensates for signal distortions and variations in polishing pad thickness, enhancing the accuracy of substrate thickness measurement and achieving improved polishing uniformity and endpoint detection.
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
A method of training a neural network includes obtaining two ground truth thickness profiles a test substrate, obtaining two thickness profiles for the test substrate as measured by an in-situ monitoring system while the test substrate is on polishing pads of different thicknesses, generating an estimated thickness profile for another thickness value that is between the two thickness values by interpolating between the two profiles, and training a neural network using the estimated thickness profile.