The use of the quaternary semiconductor alloy Ga_xIn_1-xAs_yP_1-y for the development of new electronic, optoelectronic or high speed microwave devices is of great technological interest e.g. in telecommunication. Under certain manufacturing conditions unwanted variations in the chemical composition of these materials can occur, which can be attributed to the existence of a miscibility gap. These decomposition phenomena occur within the nanometer and subnanometer scale. Therefore it is necessary to use characterisation methods of high sensitivity and at the same time highest spatial resolution to investigate independently key parameters such as layer thickness, chemical composition or crystalline structure. The Scanning Transmission Electron Microscope (STEM) is suited for such material analyses since it combines illustrating and analytic characterisation methods together with high spatial resolution. The goal of this work was a comprehensive qualitative and quantitativ e investigation of decomposition in GaxIn1-xAsyP1-y using characterisation techniques like bright-field and Z-contrast imaging as well as electron energy loss spectroscopy (EELS) and convergent beam electron diffraction (CBED), performed in a STEM. For the first time the chemical decomposition process were quantified on the nanometer scale. The course of the decomposition and the predicted expansion of the miscibility gap could be acknowledged in the experiment. Additionally it was shown that by optimising growth parameters (e.g. pressure) of strain compensated superlattices the decomposition process could be inhibited or even stopped. At the same time for the improvement of the evaluation of high resolution Z-contrast images the maximum entropy method (MEM) was applied. Due to the use of the MEM the high resolution Z-contrast images permits the investigation of defect structures and for the first time using a STEM at 100 keV the dumb bells of GaSb was resolved in maximum entropy reconstruction

This paper analyses the transfer of standardized products by FDI between industrial countries, newly industrialized countries and developing countries. One focus of this study is the question whether local entrepreneurs can acquire knowledge as joint venture partners and utilize this knowlegde for an outgoing investment in a third country. This requires that the local entrepreneurs have access to product, production process and management know how. This paper proposses the hypothesis, that the transfer of simple, standardized products can be explained with the catching-up product cycle. In the empirical part, Japanese FDI in Taiwan are compared with Taiwanese FDI in China in the branches electric & electronic and automobile parts. Case studies supplement this research. There two major findings. Taiwanese-Japanese joint ventures are operating successfully in China. The Taiwanese side has acquired at least management know-how in co-operation with the Japanese.

In the theoretical part of this work it is shown that the Lamm´s Differential Equation for polymer solutions can be extended to swollen polymer networks, if the phenomenological equations and the equation for the chemical potential of the polymer component in the network are taken into account. All differences for a transition from a polymer solution to a swollen polymer network can be assigned a "Elastically active coefficient E", which describes a deformation velocity with the same unit as the diffusion coefficient. In the practical part of this thesis some new improvements/developments regarding the Schlieren optical system are introduced and discussed. In the part "Results and Discussion" the results of the centrifugal measurements of gelatine gels are presented and discussed. The following evaluation of the centrifugal runs of polyacrylat gels (chemical networks) and the comparison with the results of the gelatine gels (physical networks) shows clearly that the new theory is suitable to describe the move ment of the polymer component in the network in a centrifugal field quantitatively as well as qualitatively.

Restoration of electrical power systems after severe disturbances is characterised by high time pressure and low experience of the operating personnel. The thesis describes a flexible approach for training and/or guidance of operators in system restoration, for the first time realised as a generic solution applicable to any given disturbance situation in any power system. Under consideration of the actual states of network and power units the system step by step suggests operational measures to be per-formed. The implementation also comprises components for knowledge acquisition, consistency check and explanation of reasoning; optionally an acoustic user dialogue can be activated. Coupling of the entire system with an operator training simulator gave the opportunity for verification with a large variety of scenarios on different power systems replicated under operational realism.

Host and guest compounds self-assemble to supramolecular complexes. Noncovalent bonding donor-acceptor-interactions of supramolecular complexes like that of J. F. Stoddart et al. can be weakened by redox reactions, but the complexes desintegrate. Here, it is described the reversed way. The initial reaction mixture contains a host and a potential guest compound not interacting with each other. The potential guest compound will be transferred in a real guest compound by electrochemical reduction and the supramolecular complexation is induced. Tetracationic, electronwithdrawing cyclophanes with two redoxactive units of 4,4-bipyridindiium respectively 2,7-diazapyrendiium act as host compounds, electronwithdrawing quinones as potential guest compounds. The host and guest compounds and their interaction in complex are characterized and studied by cyclic voltammetry and spectroelectrochemistry. It can be demonstrated that the reduction potentials are shifted, that the cyclophane and the guest compound form an inclusion complex and that the inclusion complexes are low soluble, exclusion complexes well soluble.

Nearly 80% of the worldwide semiconductor market volume is covered by standard-CMOS-technology products. Due to the increasing demand on highly flexible, reliable and robust image sensors from many different industry parts, CMOS-technology has become more and more attractive, since it offers several features in contrast to the CCD-technology. The focus of this work lies on the development of high dynamic range CMOS image sensors suitable for industrial, automotive and consumer applications. The developed sensors should yield more powerful and cost-saving system solutions compared with CCD-sensor systems that are available from the market. The work concentrates on mainly linear CMOS-sensors offering a linear dependency between illumination and electronic signal representation. The developed sensors achieve a dynamic range of more than 90dB, which is much more than CCD´s have with 50-60dB. Two sensors are presented with the first one using a mixed logarithmic/linear characteristic and the second sensor working with a multi-illumination scheme using different exposure/integration times, resp. The work finishes with a discussion of the presented sensors and their underlying signal processing algorithms and a short outlook on future developments.

Fiber-optic Bragg-Grating sensors are often the only possibility to perform measurements under adverse environmental conditions, ionizing radiation among them. However, up to now no basic research has been carried out to determine the influence of ionizing radiation on the operativeness of Bragg-Grating-Sensors. Within this work gratings of type I have been written into 3 different singlemode-fibers, with different Ge-content, one of them hydrogen-loaded. Various kinds of ionizing radiation (electrons, photons, He-nuclei) of various energy have been applied to the gratings and grating-temperature-sensors with doses up to 90 MGy. Changees of grating-properties have been monitored before and after irradiation. Experiments show only slight increase of the Bragg-wavelength within the measurement-precision of 1E-03 nm. According to these changes which were also estimated by theoretical considerations, typical applications for grating-sensors with exposure to ionizing radiation have been examinded to tell whether the sensors will meet the specific demands of the respective application.

In inductively coupled plasma spectrometry, signal suppression by acids has been recognized for a long time. It is always explained as the physical effects on the sampling processes, plasma conditions, and aerosol properties. In some cases, signal suppression apparently relates to the chemical properties of element. However, there was no publication discussed this phenomenon. In this work, it has been proven that mercury signal suppression depends on the plasma properties and nitric acid concentration. It relates to the atomization yield of mercury compounds, which is governed by the red-ox equilibrium in the central channel of plasma. Mercury complex formation equilibrium has been proven to exert in the heated droplets in plasma. The equilibrium state is governed by the complex formation with ammonium nitrate, the complex dissociation by nitric acid, and the heat obtained from plasma. Mercury halogen complex is not sensitive to the changes in acidity. Therefore, it is able to keep mercury free from nitric acid interferences and unify mercury atomization in the central channel of plasma. Similar to mercury, the uncertain signal and memory effect of platinum group elements and gold mainly arises from the chemical interference on the atomization efficiency. It not only depends on nitric acid concentration, but also on the standard reduction potential (Eo) of precious metal species in solution. Chemical interference on the spectrometric signals is governed by the robustness of the plasma, the capabilities to ingest aerosol and to protect foreign disturbance. The robustness of plasma is not equal to the excitation conditions. It has been observed that nitric acid has no significant interferences on mercury signal intensity of MIP-OES. When wet aerosol enters into the resonant cavity, it is completely desolvated in the powerful electromagnetic field. Liquid phase instantly disappears from the plasma observation zone. No doubt, the chemical equilibrium does not exist. In conclusion, chemical reactions exist in the plasma and interference the spectrometric signal. The best way to overcome the chemical matrix interferences in the plasma source spectrometry is to realize the controllable atomization.

For every natural number m there exists a ring R with a completely prime ideal P so that there are exactly m non-isomorphic indecomposable injective right R-modules with P as associated prime ideal.

For two-slanted matrices, there is shown the close connection between their spectral properties and the zeros of their corresponding symbols. The results are applied to two-scale difference equations.