SEM Tomography Reconstruction for Non-Destructive 3D Specimen Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electron tomography techniques struggle to produce high-quality 3D reconstructions of non-thin specimens due to multiple scattering events, which violate the assumption of single scattering, necessitating destructive methods like lamella extraction or slice shaving.
Innovation Solution
A scanning electron microscopy-based method that accounts for e-beam expansion and attenuation within specimens by using prior knowledge of e-beam behavior in similar designs, allowing non-destructive 3D reconstruction through algebraic reconstruction techniques, reducing angular increments, and relying on prior knowledge to quantify deviations from intended designs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If transmission electron microscopy is used to collect data from thin specimens, then tomographic reconstruction can be achieved, but the method is destructive and requires lamella extraction or slice shaving
Solution Approach 1:
The patent creates a computational model (copy) of the expected sinogram from a nominal specimen design and compares it with the actual measured sinogram. This allows non-destructive testing by detecting deviations between the measured data and the predicted data, eliminating the need for destructive physical copying through lamella extraction.
Solution Approach 2:
The patent performs preliminary computational work by calculating the expected sinogram from the nominal specimen design before actual measurement. This preliminary model is then used to guide the interpretation of measured data, enabling non-destructive reconstruction by comparing actual measurements against the pre-computed reference.
2Measurement precision
If multiple scattering events are present in non-thin specimens, then conventional tomography assumptions are violated, but the patent accounts for e-beam expansion and attenuation to enable reconstruction
Solution Approach 1:
The patent uses feedback by comparing the measured sinogram with the predicted sinogram from the nominal design. The difference between measured and predicted values provides feedback that guides the reconstruction process, allowing the algorithm to account for multiple scattering effects and e-beam expansion/attenuation without requiring overly complex physical models.
Solution Approach 2:
The patent changes the approach from trying to directly model complex physical interactions to changing parameters by comparing measured sinograms against predicted sinograms generated from nominal design parameters. This parameter-based approach simplifies the handling of multiple scattering and e-beam effects.
3Productivity
If scanning electron microscopy is used for non-destructive tomography, then high-volume mapping is enabled, but e-beam expansion and attenuation must be accounted for
Solution Approach 1:
The patent creates a computational copy of the expected measurement data from the nominal specimen design. This predicted sinogram serves as a reference that accounts for e-beam expansion and attenuation effects, enabling accurate interpretation of measured data while maintaining high productivity through non-destructive SEM-based measurement.
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
Enables high-volume, non-destructive 3D mapping of semiconductor structures and other specimens by accurately reconstructing tomographic maps, suitable for in-line tools in semiconductor manufacturing.
Implementation Method 1
measuring a respective intensity of electrons returned from the tested specimen (due to the penetration thereinto the e-beam)
Implementation Method 2
The matrix W accounts for e-beam expansion and/or attenuation with depth within the nominal specimen
Data Source
AI summary
Disclosed herein is a system for non-destructive tomography of specimens. The system includes a scanning electron microscope (SEM) and a processor(s). The SEM is configured to obtain a sinogram of a tested specimen, parameterized by a vector {right arrow over (s)}, by projecting e-beams on the tested specimen, at each of a plurality of projection directions and offsets, and. for each e-beam, measuring a respective intensity of electrons returned from the tested specimen, The processor(s) is configured to obtain a tomographic map, pertaining to the tested specimen, by determining values indicative of components of a vector {right arrow over (t)} defined by an equation W{right arrow over (t)}={right arrow over (s)}. W is a matrix with components wij specifying a contribution of a j-th voxel in a nominal specimen to an i-th element of a nominal sinogram of the nominal specimen. The matrix W accounts for e-beam expansion and attenuation with depth within the nominal specimen.