Metallic and non-metallic nanoparticles, usually supported on non-metallic substrates have attracted much interest concerning their application in the field of electrocatalysis. To characterize catalysts with respect to size, morphology, structure and composition (alloys or core-shell) of nanoparticles and their associated electrocatalytic activity, transmission electron microscopy (TEM) is the state of the art method. This investigation shows the advantages of advanced image processing using the local adaptive threshold (LAT) routine.

Three-dimensional electron diffraction data was collected with our recently developed module for automated diffraction tomography and used to solve inorganic as well as organic crystal structures ab initio. The diffraction data, which covers nearly the full relevant reciprocal space, was collected in the standard nano electron diffraction mode as well as in combination with the precession technique and was subsequently processed with a newly developed automated diffraction analysis and processing software package. Non-precessed data turned out to be sufficient for ab initio structure solution by direct methods for simple crystal structures only, while precessed data allowed structure solution and refinement in all of the studied cases.

A variety of tomographic experiments and modes for electron tomography of nanostructures are introduced, derived from the general concepts of quantitative computed tomography, binarised geometric tomography, including shape-from-silhouette, and spectroscopic chemical mapping. Our emphasis is on working out concepts of combining at least two of these tomography modes in order to share their respective advantages and improve the overall reconstruction quality. In this work, the following three hybrid modes are presented: (i) ADF-STEM tomography and EDX tomography into high-resolution 3D chemical mapping, (ii) geometric tomography and lattice-resolved backprojection into HREM-tomography for convex bodies, and (iii) geometric tomography and e-beam nanosculpting into “tomographic nanofabrication”.

Strains in GaN nanowires with InGaN quantum wells (QW) were measured from transmission electron microscope (TEM) images. The nanowires, all from a single growth run, are single crystals of the wurtzite structure that grow along the <0001> direction, and are approximately 1000 nm long and 60 nm to 130 nm wide with hexagonal cross-sections. The In concentration in the QWs ranges from 12 to 15 at %, as determined by energy dispersive spectroscopy in both the transmission and scanning electron microscopes. Fourier transform (FT) analyses of <0002> and <1100> lattice images of the QW region show a 4 to 10 % increase of the c-axis lattice spacing, across the full specimen width, and essentially no change in the a-axis value. The magnitude of the changes in the c-axis lattice spacing far exceeds values that would be expected by using a linear Vegard's law for GaN – InN with the measured In concentration. Therefore the increases are considered to represent tensile strains in the <0001> direction. Visual representations of the location and extent of the strained regions were produced by constructing inverse FT (IFT) images from selected regions in the FT covering the range of c-axis lattice parameters in and near the QW. The present strain values for InGaN QW in nanowires are larger than any found in the literature to date for other forms of InxGa1-xN (QW)/GaN.

In order to continuously improve the performances of microelectronics devices through scaling, SiO2 is being replaced by high-k materials as gate dielectric; metal gates are replacing poly-Si. This leads to increasingly more complex stacks. For future generations, the replacement of Si as a substrate by Ge and/or III/V material is also considered. This also increases the demand on the metrology tools as a thorough characterization, including composition and thickness is thus needed. Many different techniques exist for composition analysis. They usually require however large area for the analysis, complex instrumentation and can be time consuming. EDS (Energy Dispersive Spectroscopy) when coupled to Scanning Electron Microscopy (SEM) has the potential to allow fast analysis on small scale areas.

In this work, we evaluate the possibilities of EDS for thin film analysis based on an intercomparison of composition analysis with different techniques. We show that using proper modeling, high quality quantitative composition and thickness of multilayers can be achieved.