Segmented Electron Detector Layout for Low-Current SEM Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional charged particle detectors in SEM tools face challenges with poor signal-to-noise ratio (SNR) at low electron beam currents, leading to image degradation and reduced throughput, especially in semiconductor manufacturing where high accuracy and resolution are crucial.
Innovation Solution
A method of calibrating a charged particle detector by scanning a beam across a calibration sample, receiving electrical signals from sensing elements, and selecting subsets based on the distribution of secondary and backscattered particles to optimize detection, using an array of sensing elements with selective activation and deactivation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional detectors are used to detect electrons from the sample, then the detection system can operate with simple structure, but the signal-to-noise ratio deteriorates at low electron beam currents
Solution Approach 1:
The detector is divided into multiple sensing elements arranged in an array, where each element independently detects electrons. This segmentation allows selective activation of only those elements receiving returning particles, reducing noise from inactive elements while maintaining detection capability at low beam currents.
Solution Approach 2:
The detector dynamically adjusts which sensing elements are active based on the distribution of returning particles. By selectively activating elements in regions where secondary electrons are detected while deactivating elements in regions dominated by backscattered electrons, the system adapts to optimize signal-to-noise ratio in real-time.
2Reliability
If image averaging or extended integration time is used to improve SNR at low beam currents, then the signal-to-noise ratio improves, but the electron dose on the sample increases causing surface charging artifacts
Solution Approach 1:
The system extracts and separates the signal from secondary electrons from the noise of backscattered electrons by using spatial distribution information. By identifying and selecting only those sensing elements that receive secondary electrons (based on their characteristic angular distribution), the system isolates the useful signal without requiring increased electron dose.
3Area of stationary object
If the entire array of sensing elements is activated to maximize detection area, then the detection coverage is maximized, but the noise level increases reducing signal-to-noise ratio
Solution Approach 1:
Different regions of the detector array are treated differently based on their function. The system identifies regions where sensing elements receive primarily secondary electrons and activates only those elements, while deactivating elements in regions receiving primarily backscattered electrons. This local quality differentiation maintains effective detection area while eliminating noisy regions.
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
Improves SNR and contrast-to-noise ratio, enhancing defect detection accuracy and throughput in semiconductor manufacturing by distinguishing between secondary and backscattered electrons.
Implementation Method 1
an array of sensing elements configured to generate electrical signals in response to incident secondary particles or backscattered particles from the sample
Data Source
AI summary
A method of configuring a detector of a charged particle assessment system, the detector having an array of sensing elements configured to generate electrical signals in response to incident secondary particles or backscattered particles from a sample, the method comprising:selecting a first subset of the set of sensing elements for activation based on data derived from a predicted distribution of secondary particles or backscattered particles; and selecting a second subset of the set of sensing elements for deactivation based on the predicted distribution;wherein the first subset has a different predicted ratio of incident secondary particles to incident backscattered particles than the second subset.