Grating Design Optimization for Overlay Metrology
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional bright-field metrology systems face challenges in meeting precision requirements for overlay measurement due to limitations in image resolution, and existing numerical algorithms fail to accurately account for noise and process variations in diffraction spectra, leading to measurement errors.
Innovation Solution
An optimization algorithm is used to select structural parameters for grating design, minimizing the influence of process variations on diffraction spectrum intensity by iteratively calculating and updating difference values between simulated and nominal spectra, thereby reducing measurement errors without modifying hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional bright-field metrology systems are used for overlay measurement, then the measurement process is simple, but the measurement precision cannot meet the requirement of 3.5 nanometers for 130 nanometer node
Solution Approach 1:
The patent replaces conventional bright-field metrology systems with scatterometry-based measurement systems. This substitution enables achieving 3.5 nanometer precision for 130 nanometer node overlay measurement by using diffraction-based optical measurement instead of traditional imaging-based mechanical/optical systems that are limited by resolution constraints.
2Productivity
If numerical algorithms such as principle component analysis or neural network analysis are used to process diffraction spectra, then data processing speed is improved, but measurement accuracy deteriorates because these algorithms treat measured data as signal without noise
Solution Approach 1:
The patent extracts and separately processes signal components and noise components from diffraction spectra. By identifying and removing noise elements before applying numerical algorithms like principle component analysis or neural network analysis, the system maintains both high data processing speed and high measurement accuracy, avoiding the trade-off present in conventional approaches.
3Reliability
If scatterometry is used for overlay measurement on next-generation process, then measurement repeatability and reproducibility are improved, but measurement sensitivity to process variation deteriorates due to large variation in diffraction spectrum intensity caused by refractive index changes
Solution Approach 1:
The patent converts the harmful effect of refractive index variation (which causes large intensity variation in diffraction spectra) into a beneficial factor. By using the measured intensity variation information in conjunction with process variation data, the system compensates for sensitivity losses and maintains high measurement precision while enjoying the repeatability and reproducibility benefits of scatterometry.
4Adaptability or versatility
If fabrication process conditions are varied, then manufacturing flexibility is improved, but measurement accuracy deteriorates because refractive index of fabrication material deeply depends on fabrication condition variation
Solution Approach 1:
The patent implements a feedback mechanism where refractive index values are measured and monitored during fabrication process variations. These measured refractive index values are then fed back into the overlay measurement algorithm to compensate for intensity variations in diffraction spectra, enabling both fabrication flexibility and measurement accuracy to be maintained simultaneously.
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
This approach decreases measurement errors caused by process variations, enhancing measurement sensitivity and accuracy while maintaining existing hardware, by identifying optimal structural parameters that minimize the impact of refractive index variations on diffraction light intensity.
Implementation Method 1
a first diffraction spectrum is generated based on calculation values of a plurality of structural parameters
Data Source
AI summary
A method for designing a grating comprises steps of (a1) generating a first diffraction spectrum based on calculation values of a plurality of structural parameters, (a2) calculating a first difference value between the first diffraction spectrum and a first nominal spectrum, (a3) setting a default difference value with the first difference value and default structural parameter values with the structural parameter values, (b1) changing one of the structural parameter values to generate a second diffraction spectrum, (b2) calculating a second difference value between the second diffraction spectrum and a second nominal spectrum, and (c) comparing the default difference value and the second difference value, updating a default difference value with the smaller one, and updating the default structural parameter values with the structural parameter values corresponding to the smaller one.