In-Situ Spectral Thickness Monitoring for CMP Endpoint Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Chemical mechanical polishing (CMP) processes face challenges in achieving consistent material removal rates due to variations in initial substrate thickness, slurry distribution, polishing pad condition, and load, leading to within-wafer and wafer-to-wafer non-uniformity, making it difficult to determine the polishing endpoint accurately.

Innovation Solution

A method involving in-situ spectrographic monitoring with an artificial neural network trained using labeled training spectra and non-optical monitoring system data to predict substrate thickness, reducing dimensionality, and adjusting polishing parameters for uniformity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If in-situ spectrographic monitoring is used to measure substrate thickness during polishing, then measurement speed is improved, but measurement precision deteriorates due to variations in polishing conditions

Engineering Contradiction:

Improvemeasurement timeVSAvoidthickness measurement accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The system uses in-situ spectrographic monitoring to continuously measure substrate thickness during polishing and feeds this information back to the control system. The measured thickness values are compared with target specifications, and the polishing parameters (pressure, speed, slurry flow) are automatically adjusted to maintain measurement precision despite varying conditions

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system dynamically changes polishing parameters such as carrier head pressure, platen speed, and slurry flow rate based on real-time thickness measurements. By adjusting these parameters, the system compensates for variations in material removal rate and maintains consistent measurement accuracy throughout the polishing process

Inventive Principle:

Principle #35Parameter changes

2Ease of operation

If constant pressure is applied during polishing, then operation simplicity is improved, but manufacturing precision deteriorates due to within-wafer and wafer-to-wafer non-uniformity

Engineering Contradiction:

Improvepolishing control simplicityVSAvoidsubstrate flatness uniformity

Core Design Contradiction:

Ease of operationVSManufacturing precision

Solution Approach 1:

The system transitions from static constant pressure to dynamic pressure control. The carrier head pressure is continuously adjusted during polishing based on real-time thickness measurements from the spectrographic monitoring system. This dynamic adjustment compensates for variations in material removal rate across different wafer locations and between wafers, achieving uniform flatness while maintaining operational simplicity through automated control

Inventive Principle:

Principle #15Dynamics

3Manufacturing precision

If polishing parameters are adjusted based on real-time monitoring, then manufacturing precision is improved, but device complexity increases

Engineering Contradiction:

Improveendpoint detection accuracyVSAvoidmonitoring system complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system replaces complex mechanical measurement and control mechanisms with optical spectrographic monitoring and computational analysis. Instead of using multiple physical sensors and manual adjustment mechanisms, the system uses light interaction with the substrate to obtain thickness information and employs algorithms to translate spectral data into actionable polishing control signals, reducing mechanical complexity while improving precision

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

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

Improves the accuracy and speed of thickness measurement, reduces within-wafer and wafer-to-wafer non-uniformity, and enhances the reliability of endpoint detection in CMP processes.

Implementation Method 1

measuring by an in-situ spectrographic monitoring system a sequence of test spectra of light reflected from the substrate during polishing of the test substrate

Methodology Applied

Scientific EffectLight reflection: Reflection

Data Source

PatentEP4301548B1In-situ monitoring to label training spectra for machine learning system for spectrographic monitoring

Publication Date: 2026.01.21 APPLIED MATERIALS INC

Figure 1
Figure 2~3
Figure 4

AI summary

Method of training a neural network for spectrographic monitoring includes polishing a test substrate, measuring by an in-situ spectrographic monitoring system a sequence of test spectra of light reflected from the substrate and measuring by an in-situ non-optical monitoring system a sequence of test values from the substrate during polishing of the test substrate, measuring at least one of an initial characterizing value for the substrate before polishing or a final characterizing value for the substrate after polishing, inputting the sequence of test values and the initial characterizing value and/or final characterizing value into a thickness predictive model that outputs a sequence of training values with each respective training value in the sequence of training values associated with a respective test spectrum from the sequence of test spectra, and training an artificial neural network using the plurality of training spectra and the plurality of training values.