Semester Projects | Fall 2025
(can be adapted to Bachelor or Master level)
Assessing the performance of segmentation models for identifying the material components from scanning electron images of gas diffusion electrodes used in CO2 electrolyzers
AvailableGas diffusion electrodes (GDEs) are used in various electrochemical systems, including CO2 electrolyzers, for the controllable transport of the gaseous reactants. They consist of a porous layer on which a catalyst layer is deposited. FIB-SEM tomography is a technique that enables 3D reconstruction of GDE microstructures, allowing the extraction of key material properties. A critical step in this workflow is segmentation, i.e. distinguishing based on the contrast the solid material from the pore space. This project aims to quantitatively assess various segmentation models by comparing different methods, tools, and parameter sets to improve accuracy and reliability. Prior experience with image processing or Python scripting would be helpful, but it is not a requirement.
Contact: Ms. Richa RajadhyaxQuantifying molecular oxygen evolved from various iridium-based catalysts
AvailableThe oxygen evolution reaction (OER) is a key bottleneck in water splitting and remains a major focus in catalyst research. In situ techniques such as in situ scanning transmission electron microscopy with electron energy loss spectroscopy (STEM-EELS) can provide valuable insights into catalyst dynamics and molecular oxygen evolution, but validation with conventional methods is crucial. In this project, the student will optimize bench-top conditions to trigger oxygen release from iridium-based catalysts under electrochemical OER conditions and quantify the evolved O2 using gas chromatography-mass spectrometry (GC-MS). Prior experience with electrochemistry or GC-MS would be beneficial.
Contact: Ms. Elizaveta ShcherbachevaDesign of XPS-compatible MEMS chips using finite element modelling
AvailableIn situ techniques such as X-ray photoelectron spectroscopy (XPS) and liquid-phase transmission electron microscopy (LCTEM) offer powerful tools for studying catalyst evolution under operating conditions. However, implementing such methods requires specialized electrochemical cells that permit electron or X-ray transmission without significant signal distortion. While microelectromechanical systems (MEMS) chips with silicon nitride membranes are widely used for liquid-phase TEM, they are unsuitable for XPS. This project aims to optimize the design of new MEMS chips incorporating membranes with graphene windows for in situ XPS measurements. The student will design and simulate the membrane design (for example the size and shape) using finite element modelling within COMSOL Multiphysics software. Prior experience in simulation would be an advantage, but it is not necessary.
Contact: Ms. Elizaveta ShcherbachevaCalibration and optimization of liquid-phase electron microscopy parameters for geothermal hydrogen production experiments
AvailableGeothermal hydrogen production relies on complex chemical reactions involving minerals that capture CO2 and evolve hydrogen, making it a promising area of research for renewable energy. However, the specific morphological, structural and chemical transformations that these minerals undergo remain poorly understood. This is why experiments using high-temperature liquid-phase electron microscopy are being developed to study it. The objective of this project is to apply a systematic experimental design approach, such as the Taguchi method, to calibrate and optimize key microscope settings for in-situ electron microscopy of geothermal hydrogen evolution. The student will design and execute a calibration matrix to identify optimal conditions for: electron beam parameters, imaging sequence strategies, sample temperature control protocols, spacer thickness selection, etc. By analyzing the results of each trial, the student will establish robust operating windows that balance image quality with minimal beam-induced artefacts.
Contact: Dr. Louis-Marie LebasAutomated image analysis pipeline for in-situ geothermal hydrogen production experiments
AvailableTo better understand the electron microscopy results from in-situ experiments of geothermmal hydrogen production (a reaction described above), an efficient method of converting raw electron microscopy data into clear, quantifiable insights is necessary. The goal of this project is to develop a reproducible, end-to-end image analysis pipeline for in situ electron microscopy datasets. The pipeline will automate the extraction of metrics such as bubble size distributions and dynamic surface area changes. The student will implement preprocessing routines, benchmark and fine-tune machine-learning denoising models to suppress noise while preserving morphological features, configure and train segmentation networks to identify and track evolving phases, and automate the generation of histograms of bubble size distributions, time-resolved surface area curves and related plots. The student will also package the entire workflow into a user-friendly Jupyter notebook or simple graphical interface. Python coding skills and Git knowledge are preferred, though not required. It is also possible to work with ImageJ plugins/macros.
Contact: Dr. Louis-Marie LebasCycling of cathode materials in lithium-ion battery microcells
AvailableRechargeable lithium-ion batteries (LiBs) are currently largely utilized in portable electronic devices and electric vehicles (EVs), with potential future applications to enhance energy sustainability. The cathode is particularly significant because it constitutes the most considerable weight fraction in LiB cells, but degrades during operation. We propose to investigate it using liquid-phase electron microscopy, which could help in obtaining insights into their degradation mechanism. Thus, this project aims to prepare lithium-ion battery microcells and perform charge-discharge cycles within a liquid-phase electron microscopy holder. The microcell preparation includes ink preparation and its drop-casting on dedicated electrochemical chips. The student will then be trained to mount the microcell on a dedicated liquid-phase holder and perform charge-discharge cycling in a glovebox. Previous experience in electrochemical or battery systems is preferred but not mandatory.
Contact: Dr. Morgan BinggeliAssessing denoising pipelines for liquid-phase scanning electron microscopy images
AvailableLiquid-phase electron microscopy is a technique that enables imaging of liquid phenomena within an electron microscope. Limited electron irradiation is preferred when imaging in the liquid phase to avoid beam-induced artifacts. Consequently, the generated images often have a poor signal-to-noise ratio, resulting in noisy images. Therefore, denoising pipelines are a crucial part of post-processing the acquired data. This project aims to assess the efficiency of different denoising pipelines on various liquid-phase scanning electron microscopy datasets. The student will need to propose metrics for quantifying the denoising efficiency of different pipelines and comparing them across various datasets. Previous experience in image processing is preferred but not mandatory; however, ease in programming and large dataset analysis is expected.
Contact: Dr. Morgan Binggeli