Masters semester project: picometre analysis of atomic resolution STEM data
A masters semester project for materials science or physics students is offered in the field of atomic resolution imaging using electron microscopy.
Aberration-corrected scanning transmission electron microscopy (STEM) has the potential to identify atomic positions within a crystalline sample with a precision of a few picometres. Such precision allows the correlation of physical properties to materials structure on the atomic scale, for instance by measuring charge ordering in ferroelectric and antiferroelectric oxides across domain boundaries or epitaxial interfaces. Towards this, in recent years, a number of software tools have been developed by different laboratories for identifying atom column locations in atomic resolution STEM images. In this semester project, various of these tools will be used and tested, with aims of both comparing their performance and precision, and making the optimum analysis of real experimental data. Primarily the tools will be applied to data recorded on the state-of-the-art Titan Themis microscope installed in CIME, EPFL, in studies of vanadate-based functional oxide thin films made by the Triscone lab at UNIGE. Key goals will be to identify the limits of precision, and ability of the tools for identifying the positions of light oxygen atom columns, or closely-separated cation columns. To benchmark the tools, they can also be tested on synthetic data generated for the task. The tools run in different environments, such as Matlab, Python, Digital Micrograph. The project is linked to the course MSE-450 on Electron Microscopy: Advanced Methods.
Masters semester project: spectral deconvolution of atomic chemical signatures
A masters semester project for materials science or physics students is offered in the field of chemical analysis by electron microscopy.
Electron energy-loss spectroscopy in scanning transmission electron microscopy (STEM) is a powerful technique for spatially mapping the chemical composition of materials, at resolutions down to the atomic scale. However, the nature of the signal, in which each element may have an ionisation signature of a combination of peaks and an extended edge, all superimposed on a decaying background, frequently gives challenges of deconvolving overlapped signals from each other. Without good deconvolution it is not possible to generate faithful chemical maps from the data. The subject of this project will be to help develop and apply appropriate strategies for the signal separation. The project will use data which have been acquired using aberration-corrected STEM EELS on the state-of-the-art Titan Themis at CIME, EPFL. The samples in question are functional oxide thin films constituted of multi-layers of LaVO3 and PrVO3, which were made by pulsed laser deposition by the Triscone lab at UNIGE. The student can work with and adapt existing analysis codes in environments of, for example, Python or Digital Micrograph. The project is linked to the course MSE-450 on Electron Microscopy: Advanced Methods.