Sub-pixel Mineral Analysis via Spectral Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SEM-EDS systems face challenges in accurately analyzing fine-grained mineral samples due to the limitations of the x-ray interaction volume, leading to inaccurate or incomplete mineral identification when grain sizes are smaller than the pixel size, and difficulties in resolving contributions of different minerals with overlapping peaks.
Innovation Solution
A computer-implemented method that illuminates a sample with a charged particle beam, generates a sample emission spectrum, and fits it with candidate emission spectra from combinations of minerals using least squares analysis to identify the best-fitting mineral composition, allowing for sub-pixel mineral identification and accurate proportion determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the pixel size is reduced by increasing the resolution of the SEM, then the measurement precision of mineral composition is improved, but the device complexity and operational difficulty increase when approaching the x-ray interaction volume limit
Solution Approach 1:
The invention changes the analytical approach from physical resolution (pixel size) to spectral resolution (energy channel integration). By integrating photon counts across energy channels corresponding to specific elements, the method achieves accurate mineral identification without requiring pixel sizes smaller than the x-ray interaction volume, thus avoiding the need for excessively high SEM resolution
Solution Approach 2:
The invention replaces the mechanical/optical resolution system (SEM pixel scanning) with a spectral analysis system (EDS energy channel integration). Instead of relying on physical pixel resolution to distinguish minerals, the method uses energy-dispersive spectral fingerprinting to identify mineral compositions, substituting mechanical resolution requirements with spectral discrimination capabilities
2Measurement precision
If the beam voltage is reduced to minimize the x-ray interaction volume, then the measurement precision is improved, but the beam energy becomes insufficient to excite x-ray emissions of the elements in the sample
Solution Approach 1:
The invention changes the approach from controlling signal origin volume through beam voltage to controlling spectral detection through energy channel integration. By using higher beam voltages (15-20 keV) combined with selective energy channel integration for specific elements, the method maintains sufficient x-ray excitation while achieving element-specific mineral identification without being constrained by interaction volume size
3Productivity
If the least squares method is used to fit a combination of mineral spectra to the sample spectrum, then the productivity is improved by enabling sub-pixel identification, but the measurement precision deteriorates due to overlapping peaks making it difficult to resolve contributions of different minerals
Solution Approach 1:
The invention extracts specific element signals from the complex mineral spectra by integrating photon counts in energy channels corresponding to characteristic x-ray emissions of individual elements. This extraction of elemental fingerprints from overlapping mineral spectra enables accurate mineral identification and proportion determination without being confounded by spectral overlap, as each element serves as a unique identifier regardless of which mineral it originates from
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 accurate and detailed mineral composition analysis of fine-grained samples by blending colors corresponding to identified minerals based on their proportions, providing a rich and complex image representation of mineral distributions, improving upon the limitations of conventional methods.
Implementation Method 1
A portion of the sample is illuminated with a charged particle beam. A sample emission spectrum is generated from detected emissions from the sample. The sample emission spectrum can be an x-ray emission spectrum from detected x-ray emissions from the sample.
Implementation Method 2
The sample x-ray emission spectrum is fit to a linear combination of elemental x-ray emission spectra stored in a library. A least squares algorithm is used to determine the percentage of each element in the combination and how well the spectrum of the linear combination fits the sample spectrum.
Data Source
AI summary
Method and apparatus for analysis and display of fine grained mineral samples. A portion of the sample is illuminated with a charged particle beam. Emitted radiation is detected, and a sample emission spectrum is generated and fit with a plurality of standard emission spectra of minerals in a candidate mineral composition. A mineral composition whose emission spectrum best fits the sample emission spectrum is selected from a plurality of candidate mineral compositions. An assigned color is received for each mineral in the selected mineral composition, and the assigned colors are blended according to the proportion of each mineral in the selected mineral composition. An image pixel corresponding to the portion of the sample is rendered for display.