Patterned Structure Measurement Library for QOR-Driven Parameter Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semiconductor process control methods face challenges in achieving accurate and efficient data interpretation for patterned structures, particularly in high-end applications, due to the dependence on manual trial and error for library matching and the need for finer control of wafer-to-wafer, within-wafer, and die-to-die variations, which are time-consuming and impractical.
Innovation Solution
A computer system and method for constructing an optimized library that automatically adjusts to user-defined quality of results (QOR) criteria, using a library optimizer with internal degrees of freedom to iteratively refine theoretical data, ensuring both best fit and quality of results, thus enabling robust and efficient interpretation of measured data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual trial and error library matching is used, then flexibility in adjusting to user-defined quality criteria is improved, but time consumption and operational efficiency deteriorate
Solution Approach 1:
The system performs automatic library optimization where the computer system autonomously adjusts theoretical data to meet user-defined quality of results criteria without requiring manual trial and error intervention. The system serves itself by automatically iterating through parameter adjustments and evaluating quality metrics.
Solution Approach 2:
The library optimization is performed offline before actual production measurements, preparing an optimized library in advance. This preliminary action ensures that when real measurements are taken, the interpretation can proceed rapidly using the pre-optimized library, thus reducing time consumption during critical measurement phases.
2Measurement precision
If grid interval is decreased to improve measurement accuracy, then parameter extraction accuracy is improved, but processing time increases dramatically
Solution Approach 1:
The system performs library optimization offline before production use, pre-calculating and storing optimized theoretical data at appropriate grid intervals. This preliminary computation allows rapid querying during actual measurements without experiencing the dramatic time increase that would occur if high-resolution grid searches were performed in real-time.
Solution Approach 2:
The system dynamically adjusts the effective grid resolution based on the optimization needs. During offline optimization, finer grid intervals can be used to achieve high accuracy, while during online measurement interpretation, the pre-optimized library provides accurate results without requiring fine grid searches, thus adapting the computational effort to the operational context.
3Manufacturing precision
If tighter process control is implemented for wafer-to-wafer and die-to-die variations, then manufacturing precision is improved, but system complexity and operational difficulty increase
Solution Approach 1:
The system incorporates user-defined quality of results criteria that can specify different control requirements for different parameter types (wafer-to-wafer, within-wafer, die-to-die variations). By programmatically defining these control parameters and their acceptable variation ranges, the system achieves tight process control through parameter specification rather than through complex hardware or procedural changes.
Solution Approach 2:
The optimized library system serves multiple control functions simultaneously - it can evaluate wafer-to-wafer variations, within-wafer variations, and die-to-die variations using the same underlying optimization framework. This multi-functionality reduces system complexity compared to having separate control systems for each type of variation analysis.
Data Source
AI summary
A computer system is presented configured and operable as a library constructor for use in extracting one or more parameters of a patterned structure from real time measured data obtained on said structure. The system comprises: data input utility for receiving input data comprising preliminary measured data obtained from at least a part of a structure, and comprising data indicative of user-defined quality of measurement results (QOR); and a data processor.