Spectral Filter Selection for Low-Latency Semiconductor Process Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current optical monitoring systems for semiconductor processes face challenges in achieving lower latency, increased repeatability, and improved signal detectability due to the mixing of continuous and discrete optical signals from OES and IEP applications, which complicates the detection of small changes and requires wider optical bandwidth and greater signal-to-noise ratios, especially for advanced processes like FINFET and 3D NAND structures.
Innovation Solution
The proposed solution involves a method for processing spectral data that includes collecting time-ordered optical emission spectroscopy data, extracting attributes, analyzing characteristics, conditioning the data, and selecting an optimal filter configuration to minimize latency and enhance signal processing, allowing for real-time control of semiconductor processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical monitoring systems collect and analyze mixed continuous and discrete optical signals from OES and IEP applications, then the system can monitor semiconductor processes, but the detection of small changes becomes complicated and requires wider optical bandwidth and greater signal-to-noise ratios
Solution Approach 1:
The patent segments the optical signal processing by separating continuous OES signals from discrete IEP signals into distinct processing streams. Each signal type undergoes specialized processing tailored to its characteristics, with OES signals processed through continuous monitoring pathways and IEP signals processed through discrete event detection pathways, thereby simplifying the detection of small changes in each modality
Solution Approach 2:
The system dynamically adapts its processing mode based on the type of optical signal being received. The control system automatically switches between continuous processing modes for OES and discrete processing modes for IEP, optimizing the detection algorithms and filtering parameters according to the specific signal characteristics and process requirements
2Measurement precision
If the system processes optical signals with wider optical bandwidth to detect small changes, then measurement precision improves, but device complexity and processing requirements increase
Solution Approach 1:
The patent applies local quality by using wavelength-specific processing for different signal types. OES signals are analyzed at specific emission wavelengths characteristic of process chemistry, while IEP signals are processed at wavelengths corresponding to film reflectance features. This localized processing approach enhances detectability of small changes without requiring uniformly wide bandwidth across all wavelengths
3Productivity
If the system reduces latency in endpoint detection, then productivity improves, but measurement precision may be compromised
Solution Approach 1:
The system performs preliminary actions by pre-calculating endpoint criteria and processing thresholds before the actual endpoint is reached. The control system establishes prediction models and decision boundaries in advance, allowing for faster endpoint determination without compromising accuracy, as the evaluation framework is already prepared and optimized for the specific process conditions
Data Source
AI summary
The disclosure provides processing of optical data with improvements in latency, repeatability, stability, signal detectability, and other benefits. The improved processing can be used to more accurately and consistently monitor and control semiconductor processes. In one example, a method of processing spectral data includes: (1) collecting a time-ordered sequence of optical emission spectroscopy data over one or more wavelengths, (2) extracting one or more attributes from the time-ordered sequence of optical emission spectroscopy data, (3) analyzing characteristics of the one or more attributes, (4) determining conditioning of the one or more attributes, (5) processing the one or more attributes according to a predetermined set of filters, the conditioning, and the characteristics, and (6) selecting a filter configuration for processing the spectral data based upon the processing of the one or more attributes.