@phdthesis{Kissner2022,
  author    = {Kißner, Katharina},
  title     = {Manipulation of electronic properties in strongly correlated Cerium-based surface alloys},
  doi       = {10.25972/OPUS-27306},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-273067},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2022},
  abstract  = {Photoelectron spectroscopy proves as a versatile tool for investigating various aspects of the electronic structure in strongly correlated electron systems. Influencing the manifestation of strong correlation in Ce-based surface alloys is the main task of this work. It is shown, that the manifestation of the Kondo ground state is influenced by a multitude of parameters such as the choice of the metal binding partner in binary Ce compounds, the surface alloy layer thickness and accompanying variations in the lattice structure as well as the interfaces to substrate or vacuum. Gaining access to these parameters allows to directly influence essential state variables, such as the f level occupancy nf or the Kondo temperature TK. The center of this work are the intermetallic thin films of CePt5/Pt(111) and CeAgx/Ag(111). By utilizing different excitation energies, photoemission spectroscopy provides access to characteristic features of Kondo physics in the valence band, such as the Kondo resonance and its spin-orbit partner at the Fermi level, as well as the multiplet structure of the Ce 3d core levels. In this work both approaches are applied to CePt5/Pt(111) to determine nf and TK for a variety of surface alloy layer thicknesses. A temperature dependent study of the Ce 3d core levels allows to determine the systems TK for the different layer thicknesses. This leads to TK ≈200-270K in the thin layer thickness regime and TK >280K for larger layer thicknesses. These results are confirmed by fitting the Ce 3d multiplet based on the Gunnarsson-Sch{\"o}nhammer formalism for core level spectroscopy and additionally by valence band photoemission spectra of the respective Kondo resonances. The influence of varying layer thickness on the manifestation of strong correlation is subsequently studied for the surface alloy CeAgx/Ag(111). Furthermore, the heavy element Bi is added, to investigate the effects of strong spin-orbit coupling on the electronic structure of the surface alloy.},
  subject      = {Korrelation},
  language  = {en}
}
@phdthesis{Lewandowska2015,
  author    = {Lewandowska, Natalia Ewelina},
  title     = {A Correlation Study of Radio Giant Pulses and Very High Energy Photons from the Crab Pulsar},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-123533},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2015},
  abstract  = {Pulsars (in short for Pulsating Stars) are magnetized, fast rotating neutron stars. The basic picture of a pulsar describes it as a neutron star which has a rotation axis that is not aligned with its magnetic field axis. The emission is assumed to be generated near the magnetic poles of the neutron star and emitted along the open magnetic field lines. Consequently, the corresponding beam of photons is emitted along the magnetic field line axis. The non-alignment of both, the rotation and the magnetic field axis, results in the effect that the emission of the pulsar is only seen if its beam points towards the observer. The emission from a pulsar is therefore perceived as being pulsed although its generation is not. This rather simple geometrical model is commonly referred to as Lighthouse Model and has been widely accepted. However, it does not deliver an explanation of the precise mechanisms behind the emission from pulsars (see below for more details). Nowadays more than 2000 pulsars are known. They are observed at various wavelengths. Multiwavelength studies have shown that some pulsars are visible only at certain wavelengths while the emission from others can be observed throughout large parts of the electromagnetic spectrum. An example of the latter case is the Crab pulsar which is also the main object of interest in this thesis. Originating from a supernova explosion observed in 1054 A.D. and discovered in 1968, the Crab pulsar has been the central subject of numerous studies. Its pulsed emission is visible throughout the whole electromagnetic spectrum which makes it a key figure in understanding the possible mechanisms of multiwavelength emission from pulsars. The Crab pulsar is also well known for its radio emission strongly varying on long as well as on short time scales. While long time scale behaviour from a pulsar is usually examined through the use of its average profile (a profile resulting from averaging of a large number of individual pulses resulting from single rotations), short time scale behaviour is examined via its single pulses. The short time scale anomalous behaviour of its radio emission is commonly referred to as Giant Pulses and represents the central topic of this thesis. While current theoretical approaches place the origin of the radio emission from a pulsar like the Crab near its magnetic poles (Polar Cap Model) as already indicated by the Lighthouse model, its emission at higher frequencies, especially its gamma-ray emission, is assumed to originate further away in the geometrical region surrounding a pulsar which is commonly referred to as a pulsar magnetosphere (Outer Gap Model). Consequently, the respective emission regions are usually assumed not to be connected. However, past observational results from the Crab pulsar represent a contradiction to this assumption. Radio giant pulses from the Crab pulsar have been observed to emit large amounts of energy on very short time scales implying small emission regions on the surface of the pulsar. Such energetic events might also leave a trace in the gamma-ray emission of the Crab pulsar. The aim of this thesis is to search for this connection in the form of a correlation study between radio giant pulses and gamma-photons from the Crab pulsar. To make such a study possible, a multiwavelength observational campaign was organized for which radio observations were independently applied for, coordinated and carried out with the Effelsberg radio telescope and the Westerbork Synthesis Radio Telescope and gamma-ray observations with the Major Atmospheric Imaging Cherenkov telescopes. The corresponding radio and gamma-ray data sets were reduced and the correlation analysis thereafter consisted of three different approaches: 1) The search for a clustering in the differences of the times of arrival of radio giant pulses and gamma-photons; 2) The search for a linear correlation between radio giant pulses and gamma-photons using the Pearson correlation approach; 3) A search for an increase of the gamma-ray flux around occurring radio giant pulses. In the last part of the correlation study an increase of the number of gamma-photons centered on a radio giant pulse by about 17\% (in contrast with the number of gamma-photons when no radio giant pulse occurs in the same time window) was discovered. This finding suggests that a new theoretical approach for the emission of young pulsars like the Crab pulsar, is necessary.},
  subject      = {Pulsar},
  language  = {en}
}
@phdthesis{Homola2011,
  author    = {Homola, Gy{\"o}rgy {\´A}d{\´a}m},
  title     = {Functional and Microstructural MRI of the Human Brain Revealing a Cerebral Network Processing the Age of Faces},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-56740},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2011},
  abstract  = {Although age is one of the most salient and fundamental aspects of human faces, its processing in the brain has not yet been studied by any neuroimaging experiment. Automatic assessment of temporal changes across faces is a prerequisite to identifying persons over their life-span, and age per se is of biological and social relevance. Using a combination of evocative face morphs controlled for global optical flow and functional magnetic resonance imaging (fMRI), we segregate two areas that process changes of facial age in both hemispheres. These areas extend beyond the previously established face-sensitive network and are centered on the posterior inferior temporal sulcus (pITS) and the posterior angular gyrus (pANG), an evolutionarily new formation of the human brain. Using probabilistic tractography and by calculating spatial cross-correlations as well as creating minimum intersection maps between activation and connectivity patterns we demonstrate a hitherto unrecognized link between structure and function in the human brain on the basis of cognitive age processing. According to our results, implicit age processing involves the inferior temporal sulci and is, at the same time, closely tied to quantity decoding by the presumed neural systems devoted to magnitudes in the human parietal lobes. The ventral portion of Wernicke's largely forgotten perpendicular association fasciculus is shown not only to interconnect these two areas but to relate to their activations, i.e. to transmit age-relevant information. In particular, post-hoc age-rating competence is shown to be associated with high response levels in the left angular gyrus. Cortical activation patterns related to changes of facial age differ from those previously elicited by other fixed as well as changeable face aspects such as gender (used for comparison), ethnicity and identity as well as eye gaze or facial expressions. We argue that this may be due to the fact that individual changes of facial age occur ontogenetically, unlike the instant changes of gaze direction or expressive content in faces that can be "mirrored" and require constant cognitive monitoring to follow. Discussing the ample evidence for distinct representations of quantitative age as opposed to categorical gender varied over continuous androgyny levels, we suggest that particular face-sensitive regions interact with additional object-unselective quantification modules to obtain individual estimates of facial age.},
  subject      = {Gesicht},
  language  = {en}
}
@phdthesis{Wissel2009,
  author    = {Wissel, Julia},
  title     = {A new biased estimator for multivariate regression models with highly collinear variables},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-36383},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2009},
  abstract  = {Es ist wohlbekannt, dass der Kleinste-Quadrate-Sch{\"a}tzer im Falle vorhandener Multikollinearit{\"a}t eine große Varianz besitzt. Eine M{\"o}glichkeit dieses Problem zu umgehen, besteht in der Verwendung von verzerrten Sch{\"a}tzern, z.B den Ridge-Sch{\"a}tzer. In dieser Arbeit wird ein neues Sch{\"a}tzverfahren vorgestellt, dass auf Addition einer kleinen Konstanten omega auf die Regressoren beruht. Der dadurch erzeugte Sch{\"a}tzer wird in Abh{\"a}ngigkeit von omega beschrieben und es wird gezeigt, dass dessen Mean Squared Error kleiner ist als der des Kleinste-Quadrate-Sch{\"a}tzers im Falle von stark korrelierten Regressoren.},
  subject      = {Starke Kopplung},
  language  = {en}
}