CD-SEM Model Convolution for Nanometric Dimension Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current CD-SEM techniques face challenges in accurately measuring critical dimensions of nanometric structures due to arbitrary threshold choices in image interpretation and the inability to account for artefacts like dark marks in raster scan mode, leading to unreliable measurements, especially for objects smaller than 100 nm.
Innovation Solution
A method that modifies parametric CD-SEM models by adding a convolution product with a given kernel to account for artefacts, allowing for precise determination of critical dimensions by adjusting parameters to minimize differences between theoretical and experimental images, effectively addressing the limitations of existing techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a fixed empirical threshold is applied to any type of pattern, then the measurement process is simplified, but the measurement precision deteriorates because there is no direct physical link between the percentage of collected secondary electrons and the real height of the pattern
Solution Approach 1:
The invention changes the parameter used for measurement from a fixed empirical threshold (e.g., 80% of secondary electrons) to a variable threshold that is physically linked to the actual pattern height. The threshold is determined by analyzing the intensity profile shape and identifying the position corresponding to the top surface of the pattern, thereby establishing a direct physical relationship between the measurement parameter and the real geometric dimension.
2Measurement precision
If Monte Carlo type physical simulations are used to overcome thresholding problems, then the measurement precision improves, but the device complexity and computing requirements increase significantly
Solution Approach 1:
The invention replaces expensive and time-consuming Monte Carlo simulations with a simpler, computationally efficient intensity profile analysis method. Instead of performing complex physical simulations that require great computing power and very long execution times, the method uses direct analysis of the measured intensity profile shape to determine the critical dimension, achieving comparable precision with minimal computational resources.
3Extent of automation
If thresholding algorithms are used for image analysis, then the measurement process is automated, but the reliability deteriorates due to arbitrary threshold choices that distort measurements especially for objects smaller than 100 nm
Solution Approach 1:
The invention introduces feedback by using the shape of the intensity profile itself to determine the appropriate threshold. The method analyzes the curvature and derivatives of the intensity profile to identify the position corresponding to the pattern top surface, creating a self-adjusting mechanism that adapts to each specific pattern geometry. This feedback loop eliminates arbitrary threshold selection and ensures reliable measurements for nanometric structures.
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 enables more precise and reliable measurement of critical dimensions by accounting for artefacts such as dark marks, improving the accuracy of CD-SEM techniques, particularly for nanometric structures.
Implementation Method 1
Collisions between primary electrons and atoms of the sample produce several physical phenomena which may be very complex. Among these phenomena there is the emission of secondary electrons
Implementation Method 2
Collisions between primary electrons and atoms of the sample produce several physical phenomena which may be very complex. Among these phenomena there is the emission of backscattered electrons
Data Source
AI summary
A method for implementing a scanning electron microscopy characterisation technique for the determination of at least one critical dimension of the structure of a sample in the field of dimensional metrology, known as CD-SEM technique, the method including producing an experimental image representative of the structure of the sample and derived from a scanning electron microscope, from a first theoretical model based on parametric mathematical functions, calculating a second theoretical model obtained by algebraic summation of a corrective term, the corrective term being the convolution product between a given convolution kernel and the first theoretical model, the second theoretical model comprising a set of parameters to determine, and determining the set of parameters present in the second theoretical model by means of an adjustment between the second theoretical model and the experimental image.