SEM Observation Condition Search Using ML Image Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In scanning electron microscopy (SEM), finding an optimal observation condition is challenging, especially for samples that are easily charged, leading to inefficiencies in sample inspection and measurement.
Innovation Solution
A charged particle beam inspection system utilizing an image prediction model obtained through machine learning of simulation results to derive optimal observation conditions, which includes a charged particle beam irradiation device and an observation condition search device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If simulation calculation is performed to reproduce actual measurement results, then measurement precision is improved, but loss of time increases due to the lengthy calculation process
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing simulation results for various device conditions in a database before actual measurement. When measurement is needed, the system directly retrieves pre-computed results instead of performing time-consuming calculations in real-time, thus maintaining high measurement precision while significantly reducing time loss.
Solution Approach 2:
The patent creates a copy of the simulation calculation results and stores them in a database. Instead of repeatedly performing the same complex calculations, the system copies and retrieves pre-computed data, which preserves the accuracy of simulation results while eliminating the time penalty of repeated calculations.
2Measurement precision
If material parameters are matched to actual samples in simulation, then measurement precision is improved, but device complexity increases due to the need for accurate material characterization
Solution Approach 1:
The patent applies self-service by enabling the system to automatically acquire material parameters from the sample manufacturer or database without requiring manual intervention. The system autonomously matches material parameters to actual samples, reducing the complexity burden on users while maintaining high measurement precision through accurate parameter correspondence.
3Measurement precision
If optimal observation condition is searched through repeated actual measurement, then measurement precision is improved, but productivity decreases due to the time-consuming search process
Solution Approach 1:
The patent applies preliminary action by pre-calculating simulation results for multiple device conditions and storing them in advance. When optimal observation conditions are needed, the system retrieves pre-computed results from the database and compares them with actual measurements, eliminating the need for repeated trial measurements and significantly improving productivity while maintaining image quality.
Solution Approach 2:
The patent creates copies of simulation results for various observation conditions and stores them in a database. Instead of performing repeated actual measurements to find optimal conditions, the system copies and compares pre-computed simulation data with actual measurements, thereby maintaining measurement precision while dramatically increasing inspection throughput.
4Extent of automation
If machine learning model is trained with simulation results, then extent of automation is improved, but device complexity increases due to the need for model training and validation
Solution Approach 1:
The patent introduces a machine learning model as an intermediary between the simulation database and the observation condition determination. The model learns the mapping from device conditions to simulation results and automatically predicts optimal conditions for actual measurements, thereby achieving high automation while managing complexity through the use of a trained predictive model rather than complex real-time calculations.
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 significantly shortens the time required to search for optimal observation conditions, increases inspection throughput, and automates the determination of observation conditions, reducing reliance on user experience and know-how.
Implementation Method 1
primary electrons generated from an electron gun are accelerated
Implementation Method 2
a condenser lens and an object lens using a plurality of electric fields or a magnetic field are used to transport an electron beam
Implementation Method 3
a trajectory of a primary electron beam is bent by using an electric field and a magnetic field to irradiate a place where the electron beam is applied while being shifted on the sample
Implementation Method 4
When the sample is irradiated with an electron beam, secondary electrons and scattered and reflected electrons are generated
Implementation Method 5
When the sample is charged, an electric field is formed on the sample, and the trajectory of the primary electrons and the secondary electrons is bent
Data Source
AI summary
Provided is a charged particle beam inspection system that can derive an optimal observation condition using an image prediction model obtained by machine learning of simulation results. The charged particle beam inspection system includes: a charged particle beam irradiation device configured to acquire an image of a sample; and an observation condition search device configured to search for an observation condition of the charged particle beam irradiation device and control image acquisition performed by the charged particle beam irradiation device. The observation condition search device acquires a module including a learning device subjected to training using labeled training data, which includes a plurality of simulation images obtained by inputting image generation condition including a plurality of first device conditions and a plurality of first sample conditions into a simulator and the first image generation condition, sets a plurality of second device conditions in an image generation tool to acquire a plurality of output images output by the image generation tool, collates the plurality of output images with an image obtained by inputting the first sample condition and the second device condition to the learning device, and generates a second sample condition based on the collation result.