Atomic Vibration Mapping via Super-Resolved Microscopy Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current optical techniques lack the spatial resolution to map atomic vibrations with atomic precision, and existing electron microscopy methods, like electron energy loss spectroscopy, suffer from long data acquisition times and potential sample damage, failing to provide sufficient resolution for measuring atomic vibrations and temperature at the scale of a single atom.
Innovation Solution
A method involving super-resolution transformation and temporal alignment of multidimensional microscopy data, including steps such as convolution with HBSG filters, segmentation using machine learning, and spectral filtering, to extract centroids and calculate atomic temperature and vibration maps with picometric spatial resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If electron energy loss spectroscopy is used to map atomic vibrations, then spatial resolution can be improved, but data acquisition time increases and sample damage occurs
Solution Approach 1:
The method segments the multidimensional data into multiple 2D slices along different directions (e.g., horizontal, vertical, diagonal). Each slice is processed independently to extract atomic column positions, which are then integrated to reconstruct the atomic vibration map. This segmentation allows parallel processing and reduces the time required compared to conventional full-spectrum analysis.
Solution Approach 2:
The invention creates multiple copies of the data by acquiring images at different focus settings (focus series) or different convergence angles (angular series). These copies provide redundant information about the same atomic structure from different perspectives, enabling robust extraction of atomic positions without requiring excessively long acquisition times at a single setting.
2Measurement precision
If electron energy loss spectroscopy is used to map atomic vibrations, then spatial resolution can be improved, but sample integrity deteriorates
Solution Approach 1:
Instead of performing complete energy loss spectroscopy on the entire sample, the method applies a partial action by selecting only specific 2D slices from the multidimensional data that contain the necessary information for atomic vibration mapping. This reduces the total electron dose required, minimizing sample damage while maintaining sufficient spatial resolution.
Solution Approach 2:
Multiple copies of the sample information are obtained through focus series or angular series acquisitions. These copies allow the extraction of atomic position information from geometric relationships in the projected images rather than requiring high-energy electron scattering measurements, thereby reducing the harmful electron bombardment on the sample.
3Speed
If conventional optical methods are used to map atomic vibrations, then spatial resolution is limited by diffraction, but measurement speed can be maintained
Solution Approach 1:
The invention replaces conventional optical detection methods with electron microscopy-based detection. By using electron waves with much shorter wavelengths than visible light, the diffraction limit is overcome, enabling atomic-scale spatial resolution. The mechanical/electronic image processing and super-resolution algorithms further enhance the precision of atomic vibration mapping beyond what optical methods can achieve.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to a method for processing multidimensional microscopy data for mapping atomic vibrations and/or an atomic temperature in a sample of at least one material, comprising, after acquisition (40) of a temporal succession of input data blocks, for each input data block of said succession, an application (44) of a homogeneous component block super-resolution transformation of an input data block, to obtain a super-resolved drop block, comprising drops representative of an atomic structure of a part of the observed sample; a temporal alignment (80) of the super-resolved drop blocks of said temporal succession;an extraction of centroids (84) from said drop tiles obtained after temporal alignment, and a calculation of a temporal average (86) of said centroids making it possible to obtain an atomic temperature map and/or projection (88) of displacement vectors of said centroids making it possible to obtain an atomic vibration map.;