Electron energy loss at interfaces

To understand materials and apply this knowledge to new technological challenges, one has to know and comprehend the materials` composition. The main approach to study materials on the atomic scale is to use a transmission electron microscope (TEM) combined with electron energy loss spec- troscopy (EELS). The latest generation aberration corrected TEMs make it possible to focus an elec- tron probe to Angström size and send it through the sample. Via the sample`s interaction with the probe beam, it is possible to deduce many of the material`s properties such as the sample`s crystal structure, the chemical composition, or the bonding of its atoms. To this end, however, it is neces- sary to interpret the measured data correctly, which is difficult due to the complexity of the interac- tion between the probe beam and the sample. Common approaches to this interpretation are based on the ideal, perfectly ordered, crystalline case which exhibits translational symmetries; these ap- proaches have shown to give good results in bulk materials. Completely new challenges arise when studying interfaces; the common approaches cannot cater for the drastic changes of the above-mentioned properties over very short distances close to an inter- face. Nevertheless, interfaces, ultra-thin films, and quantum wells are key research topics across many fields: they are of major interest to fundamental research, applied research, and industrial ap- plications alike. This project will focus on finding ways to overcome the challenges interfaces pose for the interpreta- tion of electron microscopic data, both theoretically and experimentally: by improving the accuracy of the simulation algorithms and software, new insights will be gained on how the interface interacts with the electron beam. This expertise will then be applied first to standard model systems to verify the findings and then to technologically important materials such as Heusler alloys or the AlN-GaN interface, which is of paramount importance for the semiconductor industry. Our understanding of interfaces and their technological application will greatly profit from solutions on how to best acquire, interpret and simulate the experimental data. As research in this area is still in its beginnings, the project is expected to yield novel approaches with high applicability for industry and research alike.

While many material properties are well understood in the theoretical, ideal case of infinitely large, perfectly periodical crystals, the equivalent statement is not always correct in the vicinity of interfaces and other defects. However, it is precisely those interfaces and defects that become increasingly important in todays material applications. One well-suited technique for studying such structures is transmission electron microscopy (TEM) with electron energy-loss spectrometry (EELS). In this technique, an electron beam is sent through the material and the energy transferred from the beam to the material is measured. Because electron beams can be focused to less than the size of a single atom using the latest instruments, TEM and EELS are the ideal approach to study interfaces and defects with atomic precision.In the course of this project, the dependence of the obtainable information and achievable resolution on several common experimental parameters was studied. Surprisingly, it was found that in scanning-TEM (STEM), the largest convergence angles which are commonly associated with the best resolution in STEM as they give rise to the sharpest spot do not necessarily give the best spatial resolution in STEM-EELS. In addition, it was found that imperfections in the TEM itself lead to incoherent broadening which can also significantly decrease the obtainable resolution.The results obtained in this project increase our understanding of how technologically important materials such as magnetic materials, protection coatings and superconductors can be studied more efficiently with TEM and EELS. This is expected to lead to new advances in material science such as new data storage, better wear protection and more energy efficiency. In addition, the newly obtained insights can potentially be of great use for other applied as well as fundamental research.