Refine
Has Fulltext
- yes (107) (remove)
Year of publication
- 2018 (107) (remove)
Document Type
- Doctoral Thesis (107) (remove)
Language
- English (107) (remove)
Keywords
- Fernerkundung (3)
- Magnetismus (3)
- magnetism (3)
- remote sensing (3)
- uncertainty (3)
- Angriffserkennung (2)
- Bakterien (2)
- Big Data (2)
- Biodiversität (2)
- DNA Origami (2)
Institute
- Institut für Geowissenschaften (20)
- Institut für Physik und Astronomie (20)
- Institut für Chemie (19)
- Institut für Biochemie und Biologie (13)
- Hasso-Plattner-Institut für Digital Engineering GmbH (12)
- Extern (11)
- Wirtschaftswissenschaften (6)
- Department Linguistik (5)
- Department Psychologie (3)
- Institut für Umweltwissenschaften und Geographie (3)
Ferroic materials have attracted a lot of attention over the years due to their wide range of applications in sensors, actuators, and memory devices. Their technological applications originate from their unique properties such as ferroelectricity and piezoelectricity. In order to optimize these materials, it is necessary to understand the coupling between their nanoscale structure and transient response, which are related to the atomic structure of the unit cell.
In this thesis, synchrotron X-ray diffraction is used to investigate the structure of ferroelectric thin film capacitors during application of a periodic electric field. Combining electrical measurements with time-resolved X-ray diffraction on a working device allows for visualization of the interplay between charge flow and structural motion. This constitutes the core of this work. The first part of this thesis discusses the electrical and structural dynamics of a ferroelectric Pt/Pb(Zr0.2,Ti0.8)O3/SrRuO3 heterostructure during charging, discharging, and polarization reversal. After polarization reversal a non-linear piezoelectric response develops on a much longer time scale than the RC time constant of the device. The reversal process is inhomogeneous and induces a transient disordered domain state. The structural dynamics under sub-coercive field conditions show that this disordered domain state can be remanent and can be erased with an appropriate voltage pulse sequence. The frequency-dependent dynamic characterization of a Pb(Zr0.52,Ti0.48)O3 layer, at the morphotropic phase boundary, shows that at high frequency, the limited domain wall velocity causes a phase lag between the applied field and both the structural and electrical responses. An external modification of the RC time constant of the measurement delays the switching current and widens the electromechanical hysteresis loop while achieving a higher compressive piezoelectric strain within the crystal.
In the second part of this thesis, time-resolved reciprocal space maps of multiferroic BiFeO3 thin films were measured to identify the domain structure and investigate the development of an inhomogeneous piezoelectric response during the polarization reversal. The presence of 109° domains is evidenced by the splitting of the Bragg peak.
The last part of this work investigates the effect of an optically excited ultrafast strain or heat pulse propagating through a ferroelectric BaTiO3 layer, where we observed an additional current response due to the laser pulse excitation of the metallic bottom electrode of the heterostructure.
Signals stored in sediment
(2018)
Tectonic and climatic boundary conditions determine the amount and the characteristics (size distribution and composition) of sediment that is generated and exported from mountain regions. On millennial timescales, rivers adjust their morphology such that the incoming sediment (Qs,in) can be transported downstream by the available water discharge (Qw). Changes in climatic and tectonic boundary conditions thus trigger an adjustment of the downstream river morphology. Understanding the sensitivity of river morphology to perturbations in boundary conditions is therefore of major importance, for example, for flood assessments, infrastructure and habitats. Although we have a general understanding of how rivers evolve over longer timescales, the prediction of channel response to changes in boundary conditions on a more local scale and over shorter timescales remains a major challenge. To better predict morphological channel evolution, we need to test (i) how channels respond to perturbations in boundary conditions and (ii) how signals reflecting the persisting conditions are preserved in sediment characteristics. This information can then be applied to reconstruct how local river systems have evolved over time.
In this thesis, I address those questions by combining targeted field data collection in the Quebrada del Toro (Southern Central Andes of NW Argentina) with cosmogenic nuclide analysis and remote sensing data. In particular, I (1) investigate how information on hillslope processes is preserved in the 10Be concentration (geochemical composition) of fluvial sediments and how those signals are altered during downstream transport. I complement the field-based approach with physical experiments in the laboratory, in which I (2) explore how changes in sediment supply (Qs,in) or water discharge (Qw) generate distinct signals in the amount of sediment discharge at the basin outlet (Qs,out). With the same set of experiments, I (3) study the adjustments of alluvial channel morphology to changes in Qw and Qs,in, with a particular focus in fill-terrace formation. I transfer the findings from the experiments to the field to (4) reconstruct the evolution of a several-hundred meter thick fluvial fill-terrace sequence in the Quebrada del Toro. I create a detailed terrace chronology and perform reconstructions of paleo-Qs and Qw from the terrace deposits. In the following paragraphs, I summarize my findings on each of these four topics.
First, I sampled detrital sediment at the outlet of tributaries and along the main stem in the Quebrada del Toro, analyzed their 10Be concentration ([10Be]) and compared the data to a detailed hillslope-process inventory. The often observed non-linear increase in catchment-mean denudation rate (inferred from [10Be] in fluvial sediment) with catchment-median slope, which has commonly been explained by an adjustment in landslide-frequency, coincided with a shift in the main type of hillslope processes. In addition, the [10Be] in fluvial sediments varied with grain-size. I defined the normalized sand-gravel-index (NSGI) as the 10Be-concentration difference between sand and gravel fractions divided by their summed concentrations. The NSGI increased with median catchment slope and coincided with a shift in the prevailing hillslope processes active in the catchments, thus making the NSGI a potential proxy for the evolution of hillslope processes over time from sedimentary deposits. However, the NSGI recorded hillslope-processes less well in regions of reduced hillslope-channel connectivity and, in addition, has the potential to be altered during downstream transport due to lateral sediment input, size-selective sediment transport and abrasion.
Second, my physical experiments revealed that sediment discharge at the basin outlet (Qs,out) varied in response to changes in Qs,in or Qw. While changes in Qw caused a distinct signal in Qs,out during the transient adjustment phase of the channel to new boundary conditions, signals related to changes in Qs,in were buffered during the transient phase and likely only become apparent once the channel is adjusted to the new conditions. The temporal buffering is related to the negative feedback between Qs,in and channel-slope adjustments. In addition, I inferred from this result that signals extracted from the geochemical composition of sediments (e.g., [10Be]) are more likely to represent modern-day conditions during times of aggradation, whereas the signal will be temporally buffered due to mixing with older, remobilized sediment during times of channel incision.
Third, the same set of experiments revealed that river incision, channel-width narrowing and terrace cutting were initiated by either an increase in Qw, a decrease in Qs,in or a drop in base level. The lag-time between the external perturbation and the terrace cutting determined (1) how well terrace surfaces preserved the channel profile prior to perturbation and (2) the degree of reworking of terrace-surface material. Short lag-times and well preserved profiles occurred in cases with a rapid onset of incision. Also, lag-times were synchronous along the entire channel after upstream perturbations (Qw, Qs,in), whereas base-level fall triggered an upstream migrating knickzone, such that lag-times increased with distance upstream. Terraces formed after upstream perturbations (Qw, Qs,in) were always steeper when compared to the active channel in new equilibrium conditions. In the base-level fall experiment, the slope of the terrace-surfaces and the modern channel were similar. Hence, slope comparisons between the terrace surface and the modern channel can give insights into the mechanism of terrace formation.
Fourth, my detailed terrace-formation chronology indicated that cut-and-fill episodes in the Quebrada del Toro followed a ~100-kyr cyclicity, with the oldest terraces ~ 500 kyr old. The terraces were formed due to variability in upstream Qw and Qs. Reconstructions of paleo-Qs over the last 500 kyr, which were restricted to times of sediment deposition, indicated only minor (up to four-fold) variations in paleo-denudation rates. Reconstructions of paleo-Qw were limited to the times around the onset of river incision and revealed enhanced discharge from 10 to 85% compared to today. Such increases in Qw are in agreement with other quantitative paleo-hydrological reconstructions from the Eastern Andes, but have the advantage of dating further back in time.
The movement of organisms has formed our planet like few other processes. Movements shape populations, communities, entire ecosystems, and guarantee fundamental ecosystem functions and services, like seed dispersal and pollination. Global, regional and local anthropogenic impacts influence animal movements across ecosystems all around the world. In particular, land-use modification, like habitat loss and fragmentation disrupt movements between habitats with profound consequences, from increased disease transmissions to reduced species richness and abundance. However, neither the influence of anthropogenic change on animal movement processes nor the resulting effects on ecosystems are well understood. Therefore, we need a coherent understanding of organismal movement processes and their underlying mechanisms to predict and prevent altered animal movements and their consequences for ecosystem functions.
In this thesis I aim at understanding the influence of anthropogenically caused land-use change on animal movement processes and their underlying mechanisms. In particular, I am interested in the synergistic influence of large-scale landscape structure and fine-scale habitat features on basic-level movement behaviours (e.g. the daily amount of time spend running, foraging, and resting) and their emerging higher-level movements (home range formation). Based on my findings, I identify the likely consequences of altered animal movements that lead to the loss of species richness and abundances.
The study system of my thesis are hares in agricultural landscapes. European brown hares (Lepus europaeus) are perfectly suited to study animal movements in agricultural landscapes, as hares are hermerophiles and prefer open habitats. They have historically thrived in agricultural landscapes, but their numbers are in decline. Agricultural areas are undergoing strong land-use changes due to increasing food demand and fast developing agricultural technologies. They are already the largest land-use class, covering 38% of the world’s terrestrial surface. To consider the relevance of a given landscape structure for animal movement behaviour I selected two differently structured agricultural landscapes – a simple landscape in Northern Germany with large fields and few landscape elements (e.g. hedges and tree stands), and a complex landscape in Southern Germany with small fields and many landscape elements.
I applied GPS devices (hourly fixes) with internal high-resolution accelerometers (4 min samples) to track hares, receiving an almost continuous observation of the animals’ behaviours via acceleration analyses. I used the spatial and behavioural information in combination with remote sensing data (normalized difference vegetation index, or NDVI, a proxy for resource availability), generating an almost complete idea of what the animal was doing when, why and where. Apart from landscape structure (represented by the two differently structured study areas), I specifically tested whether the following fine-scale habitat features influence animal movements: resource, agricultural management events, habitat diversity, and habitat structure.
My results show that, irrespective of the movement process or mechanism and the type of fine-scale habitat features, landscape structure was the overarching variable influencing hare movement behaviour. High resource variability forces hares to enlarge their home ranges, but only in the simple and not in the complex landscape. Agricultural management events result in home range shifts in both landscapes, but force hares to increase their home ranges only in the simple landscape. Also the preference of habitat patches with low vegetation and the avoidance of high vegetation, was stronger in the simple landscape. High and dense crop fields restricted hare movements temporarily to very local and small habitat patch remnants. Such insuperable barriers can separate habitat patches that were previously connected by mobile links. Hence, the transport of nutrients and genetic material is temporarily disrupted. This mechanism is also working on a global scale, as human induced changes from habitat loss and fragmentation to expanding monocultures cause a reduction in animal movements worldwide.
The mechanisms behind those findings show that higher-level movements, like increasing home ranges, emerge from underlying basic-level movements, like the behavioural modes. An increasing landscape simplicity first acts on the behavioural modes, i.e. hares run and forage more, but have less time to rest. Hence, the emergence of increased home range sizes in simple landscapes is based on an increased proportion of time running and foraging, largely due to longer travelling times between distant habitats and scarce resource items in the landscape. This relationship was especially strong during the reproductive phase, demonstrating the importance of high-quality habitat for reproduction and the need to keep up self-maintenance first, in low quality areas. These changes in movement behaviour may release a cascade of processes that start with more time being allocated to running and foraging, resulting into an increased energy expenditure and may lead to a decline in individual fitness. A decrease in individual fitness and reproductive output will ultimately affect population viability leading to local extinctions.
In conclusion, I show that landscape structure has one of the most important effects on hare movement behaviour. Synergistic effects of landscape structure, and fine-scale habitat features, first affect and modify basic-level movement behaviours, that can scales up to altered higher-level movements and may even lead to the decline of species richness and abundances, and the disruption of ecosystem functions. Understanding the connection between movement mechanisms and processes can help to predict and prevent anthropogenically induced changes in movement behaviour. With regard to the paramount importance of landscape structure, I strongly recommend to decrease the size of agricultural fields and increase crop diversity. On the small-scale, conservation policies should assure the year round provision of areas with low vegetation height and high quality forage. This could be done by generating wildflower strips and additional (semi-) natural habitat patches. This will not only help to increase the populations of European brown hares and other farmland species, but also ensure and protects the continuity of mobile links and their intrinsic value for sustaining important ecosystem functions and services.
The solar activity and its consequences affect space weather and Earth’s climate. The solar activity exhibits a cyclic behaviour with a period of about 11 years. The solar cycle properties are governed by the dynamo taking place in the interior of the Sun, and they are distinctive. Extending the knowledge about solar cycle properties into the past is essential for understanding the solar dynamo and forecasting space weather. It can be acquired through the analysis of historical sunspot drawings. Sunspots are the dark areas, which are associated with strong magnetic fields, on the solar surface. Sunspots are the oldest and longest available observed features of solar activity.
One of the longest available records of sunspot drawings is the collection by Samuel Heinrich Schwabe during 1825–1867. The sunspot sizes measured from digitized Schwabe drawings are not to scale and need to be converted into physical sunspot areas. We employed a statistical approach assuming that the area distribution of sunspots was the same in the 19th century as it was in the 20th century. Umbral areas for about 130 000 sunspots observed by Schwabe were obtained. The annually averaged sunspot areas correlate reasonably well with the sunspot number. Tilt angles and polarity separations of sunspot groups were calculated assuming them to be bipolar. There is, of course, no polarity information in the observations. We derived an average tilt angle by attempting to exclude unipolar groups with a minimum separation of the two surmised polarities and an outlier rejection method, which follows the evolution of each group and detects the moment, when it turns unipolar as it decays. As a result, the tilt angles, although displaying considerable natural scatter, are on average 5.85° ± 0.25°, with the leading
polarity located closer to the equator, in good agreement with tilt angles obtained from 20th century data sets. Sources of uncertainties in the tilt angle determination are discussed and need to be addressed whenever different data sets are combined.
Digital images of observations printed in the books Rosa Ursina and Prodromus pro sole mobili by Christoph Scheiner, as well as the drawings from Scheiner’s letters to Marcus Welser, are analyzed to obtain information on the positions and sizes of sunspots that appeared before the Maunder minimum. In most cases, the given orientation of the ecliptic is used to set up the heliographic coordinate system for the drawings. Positions and sizes are measured manually displaying the drawings on a computer screen. Very early drawings have no indication of the solar orientation. A rotational matching using common spots of adjacent days is used in some cases, while in other cases, the assumption that images were aligned with a zenith–horizon coordinate system appeared to be the most likely. In total, 8167 sunspots were measured. A distribution of sunspot latitudes versus time (butterfly diagram) is obtained for Scheiner’s observations. The observations of 1611 are very inaccurate, but the drawings of 1612 have at least an indication of the solar orientation, while the remaining part of the spot positions from 1618–1631 have good to very good accuracy. We also computed 697 tilt angles of apparent bipolar sunspot groups, which were observed in the period 1618–1631. We find that the average tilt angle of nearly 4° does not significantly differ from the 20th century values.
The solar cycle properties seem to be related to the tilt angles of sunspot groups, and it is an important parameter in the surface flux transport models. The tilt angles of bipolar sunspot groups from various historical sets of solar drawings including from Schwabe and Scheiner are analyzed. Data by Scheiner, Hevelius, Staudacher, Zucconi, Schwabe, and Spörer deliver a series of average tilt angles spanning a period of 270 years, in addition to previously found values for 20th-century data obtained by other authors. We find that the average tilt angles before the Maunder minimum were not significantly different from modern values. However, the average tilt angles of a period 50 years after the Maunder minimum, namely for cycles 0 and 1, were much lower and near zero. The typical tilt angles before the Maunder minimum suggest that abnormally low tilt angles were not responsible for driving the solar cycle into a grand minimum.
With the Schwabe (1826–1867) and Spörer (1866–1880) sunspot data, the butterfly diagram of sunspot groups extends back till 1826. A recently developed method, which separates the wings of the butterfly diagram based on the long gaps present in sunspot group occurrences at different latitudinal bands, is used to separate the wings of the butterfly diagram. The cycle-to-cycle variation in the start (F), end (L), and highest (H) latitudes of the wings with respect to the strength of the wings are analyzed. On the whole, the wings of the stronger cycles tend to start at higher latitudes and have a greater extent. The time spans of the wings and the time difference between the wings in the northern hemisphere display a quasi-periodicity of 5–6 cycles. The average wing overlap is zero in the southern hemisphere, whereas it is 2–3 months in the north. A marginally significant oscillation of about 10 solar cycles is found in the asymmetry of the L latitudes. This latest, extended database of butterfly wings provides new observational constraints, regarding the spatio-temporal distribution of sunspot occurrences over the solar cycle, to solar dynamo models.
The aim of this doctoral thesis was to establish a technique for the analysis of biomolecules with infrared matrix-assisted laser dispersion (IR-MALDI) ion mobility (IM) spectrometry. The main components of the work were the characterization of the IR-MALDI process, the development and characterization of different ion mobility spectrometers, the use of IR-MALDI-IM spectrometry as a robust, standalone spectrometer and the development of a collision cross-section estimation approach for peptides based on molecular dynamics and thermodynamic reweighting.
First, the IR-MALDI source was studied with atmospheric pressure ion mobility spectrometry and shadowgraphy. It consisted of a metal capillary, at the tip of which a self-renewing droplet of analyte solution was met by an IR laser beam. A relationship between peak shape, ion desolvation, diffusion and extraction pulse delay time (pulse delay) was established. First order desolvation kinetics were observed and related to peak broadening by diffusion, both influenced by the pulse delay. The transport mechanisms in IR-MALDI were then studied by relating different laser impact positions on the droplet surface to the corresponding ion mobility spectra. Two different transport mechanisms were determined: phase explosion due to the laser pulse and electrical transport due to delayed ion extraction. The velocity of the ions stemming from the phase explosion was then measured by ion mobility and shadowgraphy at different time scales and distances from the source capillary, showing an initially very high but rapidly decaying velocity. Finally, the anatomy of the dispersion plume was observed in detail with shadowgraphy and general conclusions over the process were drawn.
Understanding the IR-MALDI process enabled the optimization of the different IM spectrometers at atmospheric and reduced pressure (AP and RP, respectively). At reduced pressure, both an AP and an RP IR-MALDI source were used. The influence of the pulsed ion extraction parameters (pulse delay, width and amplitude) on peak shape, resolution and area was systematically studied in both AP and RP IM spectrometers and discussed in the context of the IR-MALDI process. Under RP conditions, the influence of the closing field and of the pressure was also examined for both AP and RP sources. For the AP ionization RP IM spectrometer, the influence of the inlet field (IF) in the source region was also examined. All of these studies led to the determination of the optimal analytical parameters as well as to a better understanding of the initial ion cloud anatomy.
The analytical performance of the spectrometer was then studied. Limits of detection (LOD) and linear ranges were determined under static and pulsed ion injection conditions and interpreted in the context of the IR-MALDI mechanism. Applications in the separation of simple mixtures were also illustrated, demonstrating good isomer separation capabilities and the advantages of singly charged peaks. The possibility to couple high performance liquid chromatography (HPLC) to IR-MALDI-IM spectrometry was also demonstrated. Finally, the reduced pressure spectrometer was used to study the effect of high reduced field strength on the mobility of polyatomic ions in polyatomic gases.
The last focus point was on the study of peptide ions. A dataset obtained with electrospray IM spectrometry was characterized and used for the calibration of a collision cross-section (CCS) determination method based on molecular dynamics (MD) simulations at high temperature. Instead of producing candidate structures which are evaluated one by one, this semi-automated method uses the simulation as a whole to determine a single average collision cross-section value by reweighting the CCS of a few representative structures. The method was compared to the intrinsic size parameter (ISP) method and to experimental results. Additional MD data obtained from the simulations was also used to further analyze the peptides and understand the experimental results, an advantage with regard to the ISP method. Finally, the CCS of peptide ions analyzed by IR-MALDI were also evaluated with both ISP and MD methods and the results compared to experiment, resulting in a first validation of the MD method. Thus, this thesis brings together the soft ionization technique that is IR-MALDI, which produces mostly singly charged peaks, with ion mobility spectrometry, which can distinguish between isomers, and a collision cross-section determination method which also provides structural information on the analyte at hand.
Deoxyribonucleic acid (DNA) is the carrier of human genetic information and is exposed to environmental influences such as the ultraviolet (UV) fraction of sunlight every day. The photostability of the DNA against UV light is astonishing. Even if the DNA bases have a strong absorption maximum at around 260 nm/4.77 eV, their quantum yield of photoproducts remains very low 1. If the photon energies exceed the ionization energy (IE) of the nucleobases ( ̴ 8-9 eV) 2, the DNA can be severely damaged. Photoexcitation and -ionization reactions occur, which can induce strand breaks in the DNA. The efficiency of the excitation and ionization induced strand breaks in the target DNA sequences are represented by cross sections. If Si as a substrate material is used in the VUV irradiation experiments, secondary electrons with an energy below 3.6 eV are generated from the substrate. This low energy electrons (LEE) are known to induce dissociative electron attachment (DEA) in DNA and with it DNA strand breakage very efficiently. LEEs play an important role in cancer radiation therapy, since they are generated secondarily along the radiation track of ionizing radiation.
In the framework of this thesis, different single stranded DNA sequences were irradiated with 8.44 eV vacuum UV (VUV) light and cross sections for single strand breaks (SSB) were determined. Several sequences were also exposed to secondary LEEs, which additionally contributed to the SSBs. First, the cross sections for SSBs depending on the type of nucleobases were determined. Both types of DNA sequences, mono-nucleobase and mixed sequences showed very similar results upon VUV radiation. The additional influence of secondarily generated LEEs resulted in contrast in a clear trend for the SSB cross sections. In this, the polythymine sequence had the highest cross section for SSBs, which can be explained by strong anionic resonances in this energy range. Furthermore, SSB cross sections were determined as a function of sequence length. This resulted in an increase in the strand breaks to the same extent as the increase in the geometrical cross section. The longest DNA sequence (20 nucleotides) investigated in this series, however, showed smaller cross section values for SSBs, which can be explained by conformational changes in the DNA. Moreover, several DNA sequences that included the radiosensitizers 5-Bromouracil (5BrU) and 8-Bromoadenine (8BrA) were investigated and the corresponding SSB cross sections were determined. It was shown that 5BrU reacts very strongly to VUV radiation leading to high strand break yields, which showed in turn a strong sequence-dependency. 8BrA, on the other hand, showed no sensitization to the applied VUV radiation, since almost no increase in strand breakage yield was observed in comparison to non-modified DNA sequences.
In order to be able to identify the mechanisms of radiation damage by photons, the IEs of certain DNA sequences were further explored using photoionization tandem mass spectrometry. By varying the DNA sequence, both the IEs depending on the type of nucleobase as well as on the DNA strand length could be identified and correlated to the SSB cross sections. The influence of the IE on the photoinduced reaction in the brominated DNA sequences could be excluded.
The development of self-adaptive software requires the engineering of an adaptation engine that controls the underlying adaptable software by a feedback loop. State-of-the-art approaches prescribe the feedback loop in terms of numbers, how the activities (e.g., monitor, analyze, plan, and execute (MAPE)) and the knowledge are structured to a feedback loop, and the type of knowledge. Moreover, the feedback loop is usually hidden in the implementation or framework and therefore not visible in the architectural design. Additionally, an adaptation engine often employs runtime models that either represent the adaptable software or capture strategic knowledge such as reconfiguration strategies. State-of-the-art approaches do not systematically address the interplay of such runtime models, which would otherwise allow developers to freely design the entire feedback loop.
This thesis presents ExecUtable RuntimE MegAmodels (EUREMA), an integrated model-driven engineering (MDE) solution that rigorously uses models for engineering feedback loops. EUREMA provides a domain-specific modeling language to specify and an interpreter to execute feedback loops. The language allows developers to freely design a feedback loop concerning the activities and runtime models (knowledge) as well as the number of feedback loops. It further supports structuring the feedback loops in the adaptation engine that follows a layered architectural style. Thus, EUREMA makes the feedback loops explicit in the design and enables developers to reason about design decisions.
To address the interplay of runtime models, we propose the concept of a runtime megamodel, which is a runtime model that contains other runtime models as well as activities (e.g., MAPE) working on the contained models. This concept is the underlying principle of EUREMA. The resulting EUREMA (mega)models are kept alive at runtime and they are directly executed by the EUREMA interpreter to run the feedback loops. Interpretation provides the flexibility to dynamically adapt a feedback loop. In this context, EUREMA supports engineering self-adaptive software in which feedback loops run independently or in a coordinated fashion within the same layer as well as on top of each other in different layers of the adaptation engine. Moreover, we consider preliminary means to evolve self-adaptive software by providing a maintenance interface to the adaptation engine.
This thesis discusses in detail EUREMA by applying it to different scenarios such as single, multiple, and stacked feedback loops for self-repairing and self-optimizing the mRUBiS application. Moreover, it investigates the design and expressiveness of EUREMA, reports on experiments with a running system (mRUBiS) and with alternative solutions, and assesses EUREMA with respect to quality attributes such as performance and scalability.
The conducted evaluation provides evidence that EUREMA as an integrated and open MDE approach for engineering self-adaptive software seamlessly integrates the development and runtime environments using the same formalism to specify and execute feedback loops, supports the dynamic adaptation of feedback loops in layered architectures, and achieves an efficient execution of feedback loops by leveraging incrementality.
Poly(vinylidene fluoride-trifluoroethylene) (P(VDF-TrFE)) ferroelectric thin films of different molar ratio have been studied with regard to data memory applications. Therefore, films with thicknesses of 200 nm and less have been spin coated from solution. Observations gained from single layers have been extended to multilayer capacitors and three terminal transistor devices.
Besides conventional hysteresis measurements, the measurement of dielectric non-linearities has been used as a main tool of characterisation. Being a very sensitive and non-destructive method, non-linearity measurements are well suited for polarisation readout and property studies. Samples have been excited using a high quality, single-frequency sinusoidal voltage with an amplitude significantly smaller than the coercive field of the samples. The response was then measured at the excitation frequency and its higher harmonics. Using the measurement results, the linear and non-linear dielectric permittivities ɛ₁, ɛ₂ and ɛ₃ have been determined. The permittivities have been used to derive the temperature-dependent polarisation behaviour as well as the polarisation state and the order of the phase transitions.
The coercive field in VDF-TrFE copolymers is high if compared to their ceramic competitors. Therefore, the film thickness had to be reduced significantly. Considering a switching voltage of 5 V and a coercive field of 50 MV/m, the film thickness has to be 100 nm and below. If the thickness becomes substantially smaller than the other dimensions, surface and interface layer effects become more pronounced. For thicker films of P(VDF-TrFE) with a molar fraction of 56/44 a second-order phase transition without a thermal hysteresis for an ɛ₁(T) temperature cycle has been predicted and observed. This however, could not be confirmed by the measurements of thinner films. A shift of transition temperatures as well as a temperature independent, non-switchable polarisation and a thermal hysteresis for P(VDF-TrFE) 56/44 have been observed. The impact of static electric fields on the polarisation and the phase transition has therefore been studied and simulated, showing that all aforementioned phenomena including a linear temperature dependence of the polarisation might originate from intrinsic electric fields.
In further experiments the knowledge gained from single layer capacitors has been extended to bilayer copolymer thin films of different molar composition. Bilayers have been deposited by succeeding cycles of spin coating from solution. Single layers and their bilayer combination have been studied individually in order to prove the layers stability. The individual layers have been found to be physically stable. But while the bilayers reproduced the main ɛ₁(T) properties of the single layers qualitatively, quantitative numbers could not be explained by a simple serial connection of capacitors. Furthermore, a linear behaviour of the polarisation throughout the measured temperature range has been observed. This was found to match the behaviour predicted considering a constant electric field.
Retention time is an important quantity for memory applications. Hence, the retention behaviour of VDF-TrFE copolymer thin films has been determined using dielectric non-linearities. The polarisation loss in P(VDF-TrFE) poled samples has been found to be less than 20% if recorded over several days. The loss increases significantly if the samples have been poled with lower amplitudes, causing an unsaturated polarisation. The main loss was attributed to injected charges. Additionally, measurements of dielectric non-linearities have been proven to be a sensitive and non-destructive tool to measure the retention behaviour.
Finally, a ferroelectric field effect transistor using mainly organic materials (FerrOFET) has been successfully studied. DiNaphtho[2,3-b:2',3'-f]Thieno[3,2-b]Thiophene (DNTT) has proven to be a stable, suitable organic semiconductor to build up ferroelectric memory devices. Furthermore, an oxidised aluminium bottom electrode and additional dielectric layers, i.e. parylene C, have proven to reduce the leakage current and therefore enhance the performance significantly.
Three Essays on EFRAG
(2018)
This cumulative doctoral thesis consists of three papers that deal with the role of one specific European accounting player in the international accounting standard-setting, namely the European Financial Reporting Advisory Group (EFRAG). The first paper examines whether and how EFRAG generally fulfills its role in articulating Europe’s interests toward the International Accounting Standards Board (IASB). The qualitative data from the conducted interviews reveal that EFRAG influences the IASB’s decision making at a very early stage, long before other constituents are officially asked to comment on the IASB’s proposals. The second paper uses quantitative data and investigates the formal participation behavior of European constituents that seek to determine EFRAG’s voice. More precisely, this paper analyzes the nature of the constituents’ participation in EFRAG’s due process in terms of representation (constituent groups and geographical distribution) and the drivers of their participation behavior. EFRAG’s official decision making process is dominated by some specific constituent groups (such as preparers and the accounting profession) and by constituents from some specific countries (e.g. those with effective enforcement regimes). The third paper investigates in a first step who of the European constituents choose which lobbying channel (participation only at IASB, only at EFRAG, or at both institutions) and unveils in a second step possible reasons for their lobbying choices. The paper comprises quantitative and qualitative data. It reveals that English skills, time issues, the size of the constituent, and the country of origin are factors that can explain why the majority participates only in the IASB’s due process.
This paper introduces a novel measure to assess similarity between event hydrographs. It is based on Cross Recurrence Plots and Recurrence Quantification Analysis which have recently gained attention in a range of disciplines when dealing with complex systems. The method attempts to quantify the event runoff dynamics and is based on the time delay embedded phase space representation of discharge hydrographs. A phase space trajectory is reconstructed from the event hydrograph, and pairs of hydrographs are compared to each other based on the distance of their phase space trajectories. Time delay embedding allows considering the multi-dimensional relationships between different points in time within the event. Hence, the temporal succession of discharge values is taken into account, such as the impact of the initial conditions on the runoff event. We provide an introduction to Cross Recurrence Plots and discuss their parameterization. An application example based on flood time series demonstrates how the method can be used to measure the similarity or dissimilarity of events, and how it can be used to detect events with rare runoff dynamics. It is argued that this methods provides a more comprehensive approach to quantify hydrograph similarity compared to conventional hydrological signatures.