Thesis Topics - Interdisciplinary
This section offers topics within cooperation between departments at FNSPE, where the supervisor may be from another department and KMAT provides guidance for the experimental part of the topic. Alternatively, these are topics that require deeper study of a subject not taught at KMAT.
If there is interest, analogous topics can also be created for KMAT students.
Topic M1/2025: Computer simulations in the mechanics and elastodynamics of ferroelastic alloys (BP/DP)
The Department of Ultrasonic Methods of the Institute of Thermomechanics of the Czech Academy of Sciences is looking for students of bachelor's, master's and doctoral programmes to work on topics addressed within the FerrMion project. The range of work includes topics in theoretical continuum mechanics and computer simulations; by agreement, the theoretical / modelling work can be extended by participation in experiments – the department has world-unique apparatus for non-contact resonant ultrasound spectroscopy (RUS) and transient grating spectroscopy (TGS), and in the coming years an atom-probe tomography (3D APT) facility will also be built there.
More infoTopic M2/2025: Quaternions – description of misorientations and the misorientation distribution function (BP/DP)
In polycrystalline materials, misorientation – the local relative rotation of two crystal lattices (grains) – is one of the important parameters, because statistical data on the character of grain boundaries often govern the properties of materials. So-called grain-boundary engineering uses many parametrizations of misorientation; however, most of them are unintuitive and, moreover, contain a number of mathematical problems. Quaternions are an interesting option for representing misorientation, including the description of an entire sample by means of the so-called misorientation distribution function.
More infoTopic M3/2025: Mathematical modelling of dislocation dynamics in materials (BP/DP)
Dislocations are microscopic defects in the crystal lattice of solid materials, especially metals. These microscopic defects subsequently influence the macroscopic physical properties as well, and their propagation can lead all the way to plastic deformation of the material. While working on the topic, the student will become acquainted with the physics of dislocations, their mathematical description and various approaches to modelling dislocation dynamics. Within the bachelor's thesis, the student will then become thoroughly acquainted with working with the simulation code ParaDIS, developed at Stanford University, currently regarded as one of the standard tools in the dislocation community. This interdisciplinary topic is suitable for students interested in computational simulations of physical problems who are keen to become familiar with modern computational tools. The topic is addressed in cooperation with the Department of Materials, FNSPE (Ing. Karel Tesař, Ph.D.), and within a follow-up research task / diploma thesis a specific simulation assignment will be planned, based on research into the properties of magnesium–zinc alloys.
Topic on the KM websiteTopic M4/2025: Local thickness of layers and features – dimensional measurement in 2D and 3D based on ImageJ2 (BP/DP)
Measuring the thickness of various layers is a typical task not only in materials and corrosion engineering, but also in medicine (e.g. the dimensions of trabeculae in spongy bone tissue). For image analysis there already exists a wide range of scripts in the widespread, freely available program ImageJ2, which uses its own scripting language, the ImageJ Macro language (IJM). Although these tools provide interesting initial estimates, they need to be adapted to specific applications. This can be done either by modifying the freely available ImageJ2 scripts or by rewriting these algorithms into another environment, including the use of machine learning for the segmentation and reconstruction of imperfect measured data.
The aim of the work is to become acquainted with the available options for analysing the thickness of corrosion products on 2D electron-microscopy images and with the options for image segmentation. The findings obtained will be applied to the corrosion of Mg-0.4Zn wires in a simulated body environment, where the currently available methods, for various reasons, do not provide accurate results. The obtained distributions of corrosion-layer thickness will be used to describe the localization of corrosion as one of the main factors in the loss of mechanical properties of wires made of bioresorbable metals. The topic is suitable for students at all levels of study.
More infoTopic M5/2025: Image analysis – correlation of images from microscopy techniques with 3D data from µCT tomography (BP/DP)
In X-ray microtomography, it is often necessary, in both scientific research and healthcare, to locate within 3D data, with great precision, the 2D slice that corresponds to a 2D representation of the data obtained by another technique. An example might be a section of a histological sample or an image from an electron microscope. In implant development this is still done manually in a fair number of cases, with the tomograph operator searching for the most suitable slice visually. Automatically finding the best-matching slice in 3D data will be a complex task requiring an understanding of the artefacts and distortions of experimental data from various techniques. The resulting method would, however, have considerable scientific and application potential.
The aim of the work is to become acquainted with methods for the automated search for 2D slices within 3D data using conventional algorithms and machine learning, and then to implement these methods for finding a user-selected 2D slice from tomographic data in a 3D reconstruction. After creating a working procedure on the same data set, methods will be developed that find the 2D slice of the 3D data most similar to images from other techniques, especially histological sections. The topic is suitable for students at all levels of study.
More info