Lithography Process Window Detection Under Measurement Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for measuring roughness in pattern structures, particularly in semiconductor manufacturing, face challenges due to noise contamination in scanning electron microscope images, leading to biased and inaccurate measurements, and require improved techniques to differentiate between noise and actual roughness without relying on image filtering that alters feature measurements.
Innovation Solution
The implementation of a physics-based inverse linescan model that processes SEM images to separate noise from actual roughness, allowing for unbiased measurement of roughness parameters by subtracting noise from the power spectral density, thereby providing accurate and precise edge detection and feature characterization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If image filtering is applied to reduce noise in SEM images, then noise contamination is reduced, but measurement accuracy deteriorates because filtering alters feature measurements
Solution Approach 1:
The patent converts the harmful noise in SEM images into a beneficial component by measuring the noise power spectral density and subtracting it from the total power spectral density. This allows the noise to be identified and removed mathematically rather than through filtering, thereby eliminating noise contamination while preserving the true roughness signal without distortion
Solution Approach 2:
The patent replaces the mechanical/image filtering approach with a mathematical signal processing approach. Instead of applying filters that physically alter the image data, the method uses power spectral density analysis and subtraction to separate noise from the true roughness signal, achieving noise reduction without measurement distortion
2Device complexity
If conventional roughness measurement methods are used, then measurement process is simple, but measurement accuracy deteriorates due to biased measurements
Solution Approach 1:
The patent segments the total power spectral density into two distinct components: noise power spectral density and true roughness power spectral density. By measuring and separating these components through mathematical operations, the method achieves unbiased roughness measurements while maintaining a relatively simple process that builds upon conventional measurement techniques
3Device complexity
If measurement uncertainty is not accounted for, then process window determination is straightforward, but manufacturing precision deteriorates
Solution Approach 1:
The patent incorporates measurement uncertainty into process window determination by calculating probabilities that account for uncertainty ranges. This feedback mechanism allows the process window analysis to consider measurement limitations, thereby improving lithography process precision without requiring overly complex analysis methods
Data Source
AI summary
Methods, systems, and computer-readable mediums for configuring a lithography tool to manufacture a semiconductor device. The method includes selecting a first variable, selecting a second variable, selecting at least one response variable that is a function of the first variable and second variable, determining a measurement uncertainty for each response variable, determining, based on a measurement of the response variable, and the measurement uncertainty for the response variable, a plurality of probabilities representing a plurality of indications of whether a plurality of points associated with a lithography process meet a specification requirement for each response variable, wherein the plurality of probabilities represent a process window, and configuring, based on the process window, a lithography tool to manufacture a semiconductor device.