Pattern Edge Measurement Correction for SEM Directional Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pattern measuring methods using charged particle beam devices, such as scanning electron microscopes, suffer from measurement errors due to beam scanning direction, which affect the accuracy of edge roughness evaluation, particularly in microfabrication processes like ArF lithography, where edge roughness significantly influences semiconductor device performance and yield.
Innovation Solution
A pattern measuring method and apparatus that calculates power spectral densities for both sides of a pattern edge using a charged particle beam scanning in a direction intersecting the edge, calculates the difference between these curves, and corrects one curve using the difference value to reduce measurement errors inherent to beam scanning devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If beam scanning is performed in a specific direction to obtain measurement data, then measurement efficiency is improved, but measurement precision deteriorates due to directional errors
Solution Approach 1:
The patent applies local quality by performing asymmetric correction on only one of the two edges (second edge) based on the difference in power spectral densities. Instead of uniformly processing both edges, the correction is locally applied where the scanning direction error manifests, preserving the original data of the first edge while correcting the second edge to compensate for the directional scanning bias.
Solution Approach 2:
The patent changes the parameter of power spectral density by calculating and comparing PSD values for both edges, then using the difference to correct one edge. This transforms the raw measurement data into a corrected form that eliminates the scanning direction artifact, effectively changing the data representation to remove the systematic error.
2Device complexity
If single-edge measurement is performed to simplify the process, then device complexity is reduced, but measurement precision deteriorates due to uncorrected scanning errors
Solution Approach 1:
The patent transforms the measurement approach by calculating power spectral densities for both edges and using their difference as a correction parameter. This parameter transformation allows the system to eliminate scanning direction errors through mathematical correction rather than complex hardware modifications, maintaining simplicity while improving precision.
Solution Approach 2:
The patent implements a feedback mechanism where the power spectral density difference between the two edges is calculated and then fed back to correct the measurement of one edge. This closed-loop approach uses the information from both edges to improve the accuracy of the final result, creating a self-correcting measurement 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
This approach enables high-accuracy pattern measurement by effectively mitigating measurement errors caused by beam scanning direction, improving the precision of edge roughness evaluation and reducing the risk of product defects in semiconductor manufacturing.
Implementation Method 1
a first power spectral density with respect to an edge of one side and a second power spectral density with respect to an edge of the other side of a pattern are calculated based upon a signal that is obtained when scanning a charged particle beam
Data Source
AI summary
A pattern measuring method and a pattern measuring apparatus that efficiently prevent a measurement error inherent to a device that performs beam scanning in a specific direction such as a scanning electron microscope are provided. The invention is directed to a pattern measuring method and a pattern measuring apparatus in which a first curve with respect to an edge of one side and a second curve with respect to an edge of the other side are obtained by calculating a first power spectral density with respect to the edge of one side of a pattern and a second power spectral density with respect to the edge of the other side of the pattern based upon a signal that is obtained when a charged particle beam is scanned in a direction intersecting the edge of the pattern; a difference value between the first curve and the second curve is calculated; and one of the first curve and the second curve is corrected by using the difference value.