Copper structures in modern power electronics based on wide bandgap semiconductors are exposed to extreme heat pulses with very high heating rates. To study the materials behaviour of such copper structures, test samples placed on poly-Silicon heater structures have been designed to study their materials behaviour as well as potential fatigue phenomena due to heat pulsing. These test structures are designed in a way that lab-based nano-X-ray tomography as well as Synchrotron-based methods can be correlated one and the same sample. This enables a deeper insight in the microstructural evolution and the related micro-strain fields in copper, both ultimately contributing to the understanding of emerging fatigue mechanisms. This correlation also helps to better understand and model copper sub-systems in modern power electronics. The research presented was done in the frame of the HEU project AddMorePower.eu.

Light Optical Microscopy (LOM) is a traditional tool in metallography. Despite tailoring sample preparation, limitations of LOM make analysis of fine-tuned microstructure of modern steels difficult. Scanning electron microscopes (SEM) provide higher resolution at the cost of being much more expensive and sensitive devices, hence not widespread in the field. We use correlative microscopy to devise a machine learning image transformation to help in the analysis of LOM images. Early results have been published and we have begun preliminary exploration of commercialisation of this and our other tools.

At high-brightness synchrotron sources, a wide range of X-ray Diffracton Microscopy techniques enable mapping of the strain tensor in epitaxial layers with submicron resolution. However, the elastic deformation is typically governed by a multitude of overlapping effects, especially in functional devices. Here, we demonstrate how a detailed comparison to complementary methods and theoretical modelling allows to extract physical insights and interpret even the most complex strain landscape. 

Hyperspectral Vision combines the advantages of spectroscopy and imaging together with algorithm-based data evaluation, enabling comprehensive analysis of a wide variety of surface and thin film properties across an entire surface. By integrating Hyperspectral Vision technology into production environments, limitations of state-of-the-Art random inspection can be overcome, and quality and functionality can be ensured by a 100% inspection. This will be shown on an exempel of an ALD-based process.

This work presents SEM and AFM-based near-field characterizations (KPFM, sMIM, SCM) for SiC devices. For both SiC-MOSFET and SiC-JFET structures, SEM (BED, LED, EDX) provides access to chemical composition and SiC-pn junctions, while AFM electrical modes simultaneously probe surface topography, and potential or local electrical properties. The comparison highlights complementary detection capabilities, underscoring the strong potential of SCM and sMIM for advanced SiC device analysis.

Combining multiple microscopy methods within one instrument can have great advantages for user but often poses big challenges for instrument designers and engineers. Methods that independently work very well, can become very difficult when combined. In our work we combine ex-citu optical microscopy with in-situ correlative AFM and EM imaging. This workflow results in Correlative Light-, Electron-, and Atomic force-Microscopy - CLEAM. In this presentation I will discuss technical developmentswhich enables such correlative instruments, how they cab be used for advanced SPM analysis of electronic materials and will show first results of 3D AFM/EM tomography of zebrafish retina.

The talk will focus on data handling possibilies for correlative datasets which are available in the Gwyddion open source software. This includes aspects of use of multiple channel information when processing the data from SPM methods (e.g. data coming from different channels in a single SPM measurement) and when processing data coming from different instruments (e.g. combining SPM data with optical measurements). The aspects of handling different types of data (images, volume data, spectroscopy) and aspects of uncertainty use and its propagation in data fusion methods will be addressed as well.

Scanning Probe Microscopy is a multi-faceted tool, which has pushed nano-scale science forward in many fields. Despite its versatility, working stand-alone it cannot provide unequivocal and definitive results. Correlating SPM results with other imaging modalities, as well as spectroscopic, diffraction-based, and additional techniques then provides an essential boost, even when such techniques are applied at the macro- or meso-scale. I will discuss how cross-correlating SPM studies with complementary techniques can unravel complex scientific enquiries in a fashion unattainable when used alone.

Focusing X-rays from a synchrotron offers tremendous opportunities for in situ and operando experiments with diverse imaging modalities at highest spatial resolution. Correlative concepts bridging the X-ray results and lab-based microscopy can provide valuable complementary structural and chemical information. We utilize hierarchically arranged markers created by ion beam induced deposition, of e.g., a Pt precursor material to assist in the re-localization of pre-selected regions of interest.

X-ray CT provides the 3D data essential for linking a material's internal architecture to its macroscopic performance. This talk shows how the quantitative analysis of CT data can reveal relationships between structure and function in diverse materials. For example, it can link electrospun fiber 3D morphology with filtration efficiency and glass foam pore networks with strength to guide high-performance material design.

Cryo-volume electron microscopy (CVEM) enables 3D imaging of cells and tissues in a near-native state, bridging molecular to cellular scales. We present an automated CVEM workflow achieving sub-20 nm isotropic resolution via optimized milling, orthogonal imaging, and adaptive control. The method yields artifact-free volumetric data, providing ultrastructural insight across tissues and cells while supporting high-resolution structural analysis.

Machine learning (ML) is increasingly used across sciences for automating complex analyses and reducing manual parameter tuning. In this talk, I show how ML techniques can be applied to 4D Scanning Transmission Electron Microscopy (4DSTEM) datasets to minimize user input in segmentation and classification. Using real data, I demonstrate how ML improves efficiency, consistency, and reproducibility in analysis.

Correlative microscopy integrates multiple imaging techniques to reveal a sample's chemical and structural properties. This talk presents our work advancing LIBS (Laser-Induced Breakdown Spectroscopy) by combining it with LA-ICP-MS, XRF, SEM-EDX, Raman spectroscopy, and XCT for comprehensive multi-modal analysis. We address challenges in sample prep and precise image registration and showcase case studies, including human tooth ankylosis, highlighting the power of this integrated approach.

Raman spectroscopy can be treated as a very interesting alternative for the conventional methods such as XRD, SEM+EDS/EPMA, EBSD or TEM, when investigating materials at high temperatures. This method, especially Raman Confocal Imaging, is very efficient, time- and cost-effective tool that enables to observe the phase distribution in 2D or 3D scale in ex-situ conditions after long-term exposures or to unravel corrosion mechanisms during early stages of exposures using in-situ testing.

Mathematical modeling provides a powerful tool to explore how tissues grow and adapt. I will present two case studies. First, a 3D vertex model of organoids where cell growth and division are coupled with reaction–diffusion signaling, producing diverse morphologies in collaboration with biologists. Second, experiments and models of continuously growing mouse incisors, revealing natural design principles that preserve strength even under accelerated regeneration.

Integrating different types of microscopies helps us to better characterize complex biological systems. The combination of topographical and mechanical descriptions is a proven benefit of using AFM. Adding optical information provides further data on structure, while a multi-electrode array adds data on electrophysiology. This comprehensive approach will be demonstrated using examples of biomolecules, cells, and tissues.