Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The focal surface of a curve γ in the Euclidean 3-space is defined as the envelope of the normal planes of γ. The focal surface of γ is singular along a curve Cγ, called the focal curve or generalized evolute. This curve is given by the centers of the osculating spheres of γ. In this work we study the geometry of the focal surface, focusing on the properties of the focal curve. These concepts can be generalized for curves in Rm+1. The focal curve may be parametrized in terms of the Frenet frame of γ. Through this parametrization, we obtain coefficients called focal curvatures. It is then obtained a formula relating the Euclidean curvatures of γ with its focal curvatures. Defining a vertex of a curve in Rm+1 as a point at which the curve has at least (m+3)-point contact with its osculating hypersphere, we give necessary and sufficient conditions for a point of γ to be a vertex. In such points the focal surface is locally diffeomorphic to the swallowtail surface. / A superfície focal de uma curva γ no espaço euclidiano tridimensional é definida como o envelope dos planos normais a γ. A superfície focal de γ é singular ao longo de uma curva Cγ, chamada curva focal ou evoluta generalizada. Esta curva é dada pelos centros das esferas osculatrizes de γ. Neste trabalho estudamos a geometria da superfície focal, dando ênfase nas propriedades da curva focal. Estes conceitos podem ser generalizados para curvas em Rm+1. A curva focal pode ser parametrizada em termos do referencial de Frenet da curva γ. Através desta parametrização, obtemos coeficientes chamados curvaturas focais. Obtemos então uma expressão relacionando as curvaturas euclidianas de γ com suas curvaturas focais. Definindo vértice de uma curva em Rm+1 como um ponto em que a curva tem contato de ordem pelo menos m+3 com sua hiperesfera osculatriz, são dadas condições necessárias e suficientes para um ponto de γ ser um vértice. Em tais pontos a superfície focal é localmente difeomorfa à superfície rabo de andorinha.

In recent years, it has been shown that microlensing is a powerful tool for examining the atmospheres of stars in the Galactic bulge and Magellanic Clouds. The high gradient of magnification across the source during both small impact parameter events and caustic crossings offers a unique opportunity for determining the surface brightness profile of the source. Furthermore, models indicate that these events can also provide an appreciable polarization signal: arising from differential magnification across the otherwise symmetric source. Earlier work has addressed the signal from a scattering photosphere for both point mass lenses and caustic crossings. In a previous paper, polarimetric variations from point lensing of a circumstellar envelope were considered, as would be suitable for an extended envelope around a red giant. In this work, we examine the polarization in the context of caustic crossing events, the scenario that represents the most easily accessible situation for actually observing a polarization signal in Galactic microlensing. Furthermore, we present an analysis of the effectiveness of using the polarimetric data to determine the envelope properties, illustrating the potential of employing polarimetry in addition to photometry and spectroscopy with microlensing follow-up campaigns.

Background: Caustic ingestions are rare but potentially life-threatening events requiring multidisciplinary emergency approaches. Although particularly respiratory functions may be impaired after caustic ingestions, studies involving acute emergency care are scarce. The goal of this study was to explore acute emergency care with respect to airway management and emergency department (ED) infrastructures. Methods: We retrospectively evaluated adult patients after caustic ingestions admitted to our university hospital over a 10-year period (2005–2014). Prognostic analysis included age, morbidity, ingested agent, airway management, interventions (endoscopy findings, computed tomography (CT), surgical procedures), intensive care unit (ICU) admission, length of stay in hospital and hospital mortality. Results: Twenty-eight patients with caustic ingestions were included in the analysis of which 18 (64 %) had suicidal intentions. Ingested agents were caustic alkalis (n = 22; 79 %) and acids (n = 6; 21 %). ICU admission was required in 20 patients (71 %). Fourteen patients (50 %) underwent tracheal intubation and mechanical ventilation, of which 3 (21 %) presented with difficult airways. Seven patients (25 %) underwent tracheotomy including one requiring awake tracheotomy due to progressive upper airway obstruction. Esophagogastroduodenoscopy (EGD) was performed in 21 patients (75 %) and 11 (39 %) underwent CT examination. Five patients (18 %) required emergency surgery with a mortality of 60 %. Overall hospital mortality was 18 % whereas the need for tracheal intubation (P = 0.012), CT-diagnostic (P = 0.001), higher EGD score (P = 0.006), tracheotomy (P = 0.048), and surgical interventions (P = 0.005) were significantly associated with mortality. Conclusions: Caustic ingestions in adult patients require an ED infrastructure providing 24/7-availability of expertise in establishing emergent airway safety, endoscopic examination (EGD and bronchoscopy), and CT diagnostic, intensive care and emergency esophageal surgery. We recommend that - even in patients with apparently stable clinical conditions - careful monitoring of respiratory functions should be considered as long as diagnostic work-up is completed.

Στην συγκεκριμένη διατριβή διερευνήθηκαν οι συνθήκες γένεσης του μικροκρυσταλλικού μαγνησίτη στις περιοχές Βάβδου και Γερακινής-Ορμύλιας της Χαλκιδικής. Η πετρογραφική μελέτη του μαγνησίτη και των ξενιστών του αποκάλυψαν δύο δομές δημιουργίας του μαγνησίτη: in situ εξαλλοίωση μέσα σε μανδυακά πετρώματα ξενιστές και ex situ αποθέσεις μέσα στους μανδυακούς σχηματισμούς. Επιπλέον, πιθανά υπάρχει στενή σχέση μεταξύ της κοιτασματογένεσης του μαγνησίτη και των περιβαλλόντων πυριτιωμένων πετρωμάτων (λιστβανίτες). Επίσης, οι γεωχημικές παρατηρήσεις συγχρόνως με τις ορυκτοχημικές αναλύσεις, δίνουν νέα στοιχεία όσον αφορά στην γένεση του κοιτάσματος. Παράλληλα, έγινε εργαστηριακή παραγωγή καυστικής μαγνησίας και παράχθησαν, με προσθήκη αυτής συγχρόνως με χρήση ερυθράς ιλύς και τέφρας πυθμένα, νέα κεραμικά υψηλού πορώδους. Τα αποτελέσματα υπήρξαν ιδιαίτερα ικανοποιητικά, καθώς παρατηρήθηκε ότι με προσθήκη μαγνησίας σε μικρό ποσοστό βελτιώνονται οι φυσικομηχανικές ιδιότητές τους. Κεραμικά υψηλού πορώδους είναι κατάλληλα για χρήση σε φίλτρα και σε απορροφητικά υλικά. Τέλος, προτείνεται νέος τρόπος αξιοποίησης στείρων υλικών μαγνησίτη από τις εξορυκτικές διεργασίες, καθώς αποδεικνύεται ότι ο μαγνησίτης έχει πολύ θετικά αποτελέσματα με προσθήκη του σε οξινισμένα ύδατα, δίνοντας νέες προοπτικές αξιοποίησης του σε περιβαλλοντικές εφαρμογές. Το πειραματικό σκέλος της δοκιμής σε οξινισμένα νερά, πραγματοποιήθηκε σε σύγκριση με πετρώματα που χρησιμοποιούνται για την αντιμετώπιση όξινων απορροών μεταλλείων. / In the present thesis, the microcrystalline magnesite from Central Chalkidiki (Vavdos and Gerakini areas) are investigated in order to clarify the conditions of its genesis. Petrographic study of the magnesite and its host rocks revealed two types of formation: a) in situ alteration in the mantle host rocks, due to metasomatic process, and b) ex situ deposit within mantle formations, related to hydrothermal fluids. Furthermore, petrographic observations suggest a close association between the genesis of the microcrystalline magnesite with the surrounding high silica-carbonate rocks, also termed as listwaenites. Moreover, scanning electron microscopy (S.E.M.) observations along with geochemical analysis provide new evidence for the formation of the magnesite. Caustic magnesia produced in the laboratory was mixed with red mud and bottom ash from industrial wastes, in order to make new, high porosity ceramics. It was observed that the addition of small amounts of caustic magnesia improved the physicomechanical properties of the briquettes. High porosity ceramics are suitable for their use in filters and absorbent materials for environmental applications. Finally, a new way for utilizing waste materials from magnesite mining processes in the disposal of acidified waters is proposed. Addition of magnesite improved the alkalinity of high acidic waters, offering new perspectives for environmental applications. The experimental results from magnesite were compared with similar results from other rocks used to treat acid mine drainage.

Increased equipment failures and the resultant increase in unplanned downtime as the result of process optimization programs continue to plague pulp mills. The failures are a result of a lack of understanding of corrosion in the different pulping liquors, specifically the parameters responsible for its adjustment such as the role and identification of inorganic and organic species. The current work investigates the role of inorganic species, namely sodium hydroxide and sodium sulfide, on liquor corrosivity at a range of process conditions beyond those currently experienced in literature. The role of sulfur species, in the activation of corrosion and the ability of hydroxide to passivate carbon steel A516-Gr70, is evaluated with gravimetric and electrochemical methods. The impact of wood chip weathering on process corrosion was also evaluated. Results were used to identify black liquor components, depending on the wood species, which play a significant role in the activation and inhibition of corrosion for carbon steel A516-Gr70 process equipment. Further, the effect of black liquor oxidation on liquor corrosivity was evaluated. Corrosion and stress corrosion cracking performance of selected materials provided information on classes of materials that may be reliably used in aggressive pulping environments.

Submitted by Emanuel Varela Cardoso (emanuel.varela@ufcg.edu.br) on 2018-09-27T18:51:38Z No. of bitstreams: 1 KALINE MELO DE SOUTO VIANA – TESE (PPGEP) 2006.pdf: 3746935 bytes, checksum: 4a030ab6637f715855808519f3e7825e (MD5) / Made available in DSpace on 2018-09-27T18:51:38Z (GMT). No. of bitstreams: 1 KALINE MELO DE SOUTO VIANA – TESE (PPGEP) 2006.pdf: 3746935 bytes, checksum: 4a030ab6637f715855808519f3e7825e (MD5) Previous issue date: 2009-02-06 / Para a avaliação de diafragmas alternativos ao de amianto para produção de cloro e soda cáustica a partir da eletrólise da salmoura, foi utilizada uma célula eletroquímica em escala de bancada pelo processo diafragma. O objetivo foi investigar a possível substituição dos diafragmas de amianto por diafragmas de PEUAPM. Desta forma, este trabalho foi dividido em duas etapas. Na primeira etapa, foi realizada a investigação da resistência química do PEUAPM na forma de fibras, em meio ácido, básico e salino, numa temperatura de 90ºC, simulando-se as condições de operação das células eletroquímicas para produção cloro-soda. Para isto, foram utilizados os ensaios de infravermelho (FTIR), calorimetria diferencial exploratória (DSC), tração e microscopia eletrônica de varredura (MEV). Na segunda etapa, foram realizadas eletrólises e para isto foram confeccionados diafragmas com diferentes composições, variando de 100% amianto (diafragma utilizado como referência), passando pelo balanço de 5:95 a 90:10% de PEUAPM:amianto, até 100% PEUAPM. Para caracterização eletroquímica utilizou-se a determinação do número de MacMullim, bem como determinação dos parâmetros operacionais após 72 horas de eletrólise, numa temperatura de aproximadamente 85ºC, onde foram quantificados os seguintes parâmetros: eficiência de corrente, eficiência de energia, teor de cloretos e teor de cloratos, com o objetivo de correlacionar os dados obtidos com o novo diafragma aos dados obtidos com o diafragma de amianto (utilizado industrialmente). As condições operacionais utilizadas foram: diafragma de amianto, PEUAPM: amianto e PEUAPM, alimentação contínua de uma salmoura ácida (pH entre 3,5 e 4,0) na concentração de 320 g/L, densidade de corrente anódica de 2,63 KA/m2 com um tempo total de operação da célula eletrolítica na intensidade de corrente de 13,1 A de aproximadamente 72 horas, até atingir a concentração de 12% de NaOH (de acordo com os parâmetros industriais). Os resultados obtidos na etapa I, evidenciaram que as fibras de PEUAPM apresentam boa resistência ao ataque químico das soluções de NaCl, HCl e salmoura na temperatura de 90ºC, durante as 96 horas de exposição, apresentando apenas um início de oxidação, detectada pela presença de carbonilas, além de uma redução de 15% na cristalinidade das fibras de PEUAPM. Já por meio dos resultados obtidos na etapa II, observou-se que os diafragmas amianto/PEUAPM estudados não apresentaram bom empacotamento de fibras e que há uma tendência de redução da eficiência de corrente (E.C.) e eficiência de energia (E.E.) com o aumento do percentual de PEUAPM. Assim, conclui-se que as fibras de PEUAPM são um substituto em potencial para as fibras de amianto na produção cloro-soda. / To evaluate the alternatives diaphragm in relation to the asbestos in the chloro-alkali production from the electrolysis of brine, it was used an electrochemical cell in a laboratory scale by diaphragm process. The aim of this work is to study the substitution of asbestos diaphragm by UHMWPE. So, this work was divided in two steps. Firstly, it was done an investigation related to the chemical resistance of the UHMWPE in a fiber shape and in acid, base and salty solutions and 90oC of temperature. This condition was similar to the electrochemical cell operation in the chloro-alkali industry. It was used test of infrared (FTIR), differential scanning calorimetry (DSC), tensile strength and scanning electron microscopy (SEM) to evaluate the UHMWPE fibers. In the second step it was done the electrolysis with different compositions of diaphragm, changing from 100% of asbestos (used as reference) to 5:95, 90:10% of UHMWPE, until 100% of UHMWPE. To chemical characterization it was estimated the MacMullin number, as well the operational cell parameters, such as, current and energy efficiency, amounts of chlorates and chlorides. The aim of to study these parameters is the correlate the new diaphragm with the asbestos (used currently in the industry). The operational conditions used in this study were: continue feed of acid brine (pH from 3.5 to 4.0), concentration of 320 g/L, anodic current density of 2.63 KA/m2, and total time operation in the electrolytic cell of approximately 74 ho urs (current intensity of 13.1A) until reach 12% of NaOH concentration (similar to the industrial conditions). The results obtained in the first step showed that the UHMWPE fibers presented good resistance to the chemical attack of NaCl, HCl and brine solution at 90oC and in the period of 96 hours of exposition. It was observed only an start of oxidation, with presence of carbonyl groups, also it was observed an decrease in 15% in the crystallinity of the UHMWPE fibers. From the results of the second step it was observed that the asbestos/UHMWPE diaphragms studied presented good fiber packing and a tendency to decrease current and energy efficiencies with the increase of the UHMWPE content. So, it was concluded that with the conditions of preparation of asbestos/UHMWPE diaphragm done in this study the use of these in the choro-alkali production still limited due to the low surface energy and physical interaction presented by the UHMWPE fibers and it was necessary an previous surface treatment of the fibers.

We investigate in this thesis the generic properties of curves in the Minkowski plane R2 1 and of smooth Lorentzian surfaces. The generic properties of curves in R2 1 are obtained by studying the contacts of curves in R2 1 with lines and pseudo-circles. These contacts are captured by the singularities of the families of height and distancesquared functions on the curves. On the other hand, the generic properties of smooth Lorentzian surfaces are obtained by studying certain Binary Differential Equations defined on the surfaces.

We investigate the propagation of light in different conditions that lead to exotic propagation of photons and use near-resonant light-matter interactions to enhance these effects. First, we study the propagation of light in a moving highly dispersive medium, namely rubidium atoms. Based on the special relativity the speed of light changes with the speed of the medium. However, this drag effect in a non-dispersive medium is very small and thus difficult to measure. We show that the drag effect is enhanced significantly when the moving medium is highly dispersive. Thus, with this enhancement even a slow motion can be detected. Next, we employ the large nonlinear response of rubidium atoms to accentuate the formation of optical caustics. Caustics are important as nature uses caustics to concentrate the energy of waves. Moreover, caustics can be formed in many physical systems such as water waves in oceans to amplify tsunamis or generate rogue waves. The connection of our study to these giant water waves is discussed. Finally, we explore light-matter interactions in plasmonic systems. We show that photons experience a significant phase jump as they couple into and out of a plasmonic structure. This coupling phase, also known as the scattering phase shift, is generic to all scattering events. We measure this coupling phase with a triple-slit plasmonic structure. Moreover, we use the near-field enhancement of the plasmonic structure to enhance the coupling between the slits. Consequently, the photons can take non-trivial trajectories that pass through all three slits. We measure such exotic trajectories for the first time that are seemingly in violation of the superposition principle. The application of the superposition principle and the validity of Born’s rule is discussed.

[pt] Neste trabalho se analisam as características de colapso de um solo laterítico submetido à percolação de um licor cáustico a diferentes concentrações. O solo é uma areia quartzosa com uma pequena fração de finos cauliníticos. O licor é um efluente industrial, resultante da produção de alumínio, a base de soda cáustica. Mediante o presente estudo, determinou-se o efeito da infiltração do licor cáustico sobre três aspetos: colapsividade, transporte de solutos e características de sucção. Isto foi feito através de um programa experimental que incluiu ensaios de caracterização física-química do solo e do licor, ensaios edométricos, em edômetros especiais, para avaliar as características de colapso e de transporte de solutos, e ensaios de sucção pelo método do papel filtro. Os resultados do programa experimental mostraram que o solo é colapsível em presença de água, sendo sua colapsividade aumentada em presença do licor cáustico. Também indicaram que os parâmetros de transporte e sucção são fortemente afetados pela percolação do licor. A mudança destes parâmetros é função da concentração do licor. / [en] In this work, characteristics of collapse of a laterítico soil submitted to percolation of a caustic liqueur to different concentrations are analysed. The soil comprises a quartzitic sand with a small fraction of kaolinitic clay. The liqueur is a caustic fluid, from an aluminium industry. In the present study, effects of the caustic liqueur infiltration were observed considering three aspects: collapse, solute transport and soil suction characteristics. This was done through an experimental program that included physical and chemical characterization of the soil and liqueur, special oedometer tests to evaluate simultaneously collapse characteristics and solute transport, and suction tests using the filter paper method. The results of the experimental program showed that the soil suffers volume change in presence of water, and its colapsivity is increased in presence of the caustic liqueur. They also indicated that the transport and suction parameters are affected strongly by the percolation of the liqueur.

Das Ziel dieser Arbeit ist es, die Strukturen im äußeren Erdkern zu untersuchen und Rückschlüsse auf die sich daraus ergebenden Konsequenzen für geodynamische Modellvorstellungen zu ziehen. Die Untersuchung der Kernphasenkaustik B mit Hilfe einer kumulierten Amplituden-Entfernungskurve ist Gegenstand des ersten Teils. Dazu werden die absoluten Amplituden der PKP-Phasen im Entfernungsbereich von 142 &#176; bis 147 &#176; bestimmt und mit den Amplituden synthetischer Seismogramme verglichen. Als Datenmaterial dienen die Breitbandregistrierungen des Deutschen Seismologischen Re-gionalnetzes (GRSN 1 ) und des Arrays Gräfenberg (GRF). Die verwendeten Wellen-formen werden im WWSSN-SP-Frequenzbereich gefiltert. Als Datenbasis dienen vier Tiefherdbeben der Subduktionszone der Neuen Hebriden (Vanuatu Island) und vier Nuklearexplosionen, die auf dem Mururoa und Fangataufa Atoll im Südpazifik stattgefunden haben. Beide Regionen befinden sich vom Regionalnetz aus gesehen in einer Epizentraldistanz von ungefähr 145 &#176;. Die Verwendung eines homogen instrumentierten Netzes von Detektoren und die Anwendung von Stations- und Magnitudenkorrekturen verringern den Hauptteil der Streuung bei den Amplitudenwerten. Dies gilt auch im Vergleich zu Untersuchungen von langperiodischen Amplituden im Bereich der Kernphasenkaustik (Häge, 1981). Ein weiterer Grund für die geringe Streuung ist die ausschließliche Verwendung von Ereignissen mit kurzer impulsiver Herdzeitfunktion. Erst die geringe Streuung der Amplitudenwerte ermöglicht eine Interpretation der Daten. Die theoretischen Amplitudenkurven der untersuchten Erdmodelle zeigen im Bereich der Kaustik B einen gleichartigen Kurvenverlauf. Bei allen Berechnungen wird ein einheitliches Modell für die Güte der P- und S-Wellen verwendet, das sich aus den Q-Werten der Modelle CIT112 und PREM 2 zusammensetzt. Die mit diesem Q-Modell berechneten Amplituden liegen in geringem Maße oberhalb der gemessenen Amplituden. Dies braucht nicht berücksichtigt zu werden, da die kumulierte Amplituden-Entfernungskurve anhand der Lage des Maximums auf der Entfernungsachse ausgewertet wird. Folglich wird darauf verzichtet, ein alternatives Q-Modell zu entwickeln.<br /> Hinsichtlich der Lage des Kaustikmaximums lassen sich die untersuchten Erdmodelle in zwei Kategorien einteilen. Eine Gruppe besteht aus den Modellen IASP91 und 1066B, deren Maxima bei 144.6 &#176; und 144.7 &#176; liegen. Zur zweiten Gruppe von Modellen zählen AK135, PREM und SP6 mit den Maxima bei 145.1 &#176; und 145.2 &#176; (SP6). Die gemessene Amplitudenkurve hat ihr Maximum bei 145 &#176;. Alle Entfernungsangaben beziehen sich auf eine Herdtiefe von 200 km. Die Kaustikentfernung für einen Oberflächenherd ist jeweils um 0.454 &#176; größer als die angegeben Werte. Damit liegen die Maxima der Modelle AK135 und PREM nur 0.1 &#176; neben dem der gemessenen kumulierten Amplitudenkurve. Daher wird auf die Erstellung eines eigenen Modells verzichtet, da dieses eine unwesentlich verbesserte Amplitudenkurve aufweisen würde. Das Ergebnis der Untersuchung ist die Erstellung einer gemessenen kumulierten Amplituden-Entfernungskurve für die Kaustik B. Die Kurve legt die Position der Kaustik B für kurzperiodische Daten auf &#177; 0.15 &#176; fest und bestimmt damit, welche Erdmodelle für die Beschreibung der Amplituden im Entfernungsbereich der Kaustik B besonders geeignet sind. Die Erdmodelle AK135 und PREM, ergänzt durch ein einheitliches Q-Modell, geben den Verlauf der Amplituden am besten wieder. Da die Amplitudenkurven beider Modelle nahe beieinander liegen, sind sie als gleichwertig zu bezeichnen.<br /> Im zweiten Teil der Arbeit wird die Struktur der Übergangszone in den inneren Erdkern anhand des spektralen Abklingens der Phase PKP(BC)diff am Punkt C der Laufzeitkurve untersucht. Der physikalische Prozeß der Beugung ist für die starke Abnahme der Amplituden dieser Phase verantwortlich. Die Diffraktion beeinflußt das Abklingverhalten verschiedener Frequenzanteile des seismischen Signals auf unterschiedliche Weise. Eine Deutung des Verhaltens erfordert die Berechnung von Abklingspektren. Dabei wird die Abschwächung des PKP(BC)diff Signals für acht Frequenzen zwischen 6.4 s und 1.25 Hz ermittelt und als Spektrum dargestellt. Die Form des Abklingspektrums ist charakteristisch für die Beschaffenheit der Geschwindigkeitsstruktur direkt oberhalb der Grenze zum inneren Erdkern (GIK). Die Beben, deren Kernphasen im Regionalnetz als diffraktierte Kernphasen BCdiff registriert werden, liegen in einem Entfernungsbereich jenseits von 150 &#176;. In dieser Distanz befinden sich die Erdbebenherde der Tonga-Fidschi-Subduktionszone, deren Breitbandaufzeichnungen verwendet werden. Die Auswertung unkorrigierter Wellenformen ergibt Abklingspektren, die mit plausiblen Erdmodellen nicht in Einklang zu bringen sind. Aus diesem Grund werden die Daten einer spektralen Stationskorrektur unterzogen, die eigens zu diesem Zweck ermittelt wird. Am Beginn der Auswertung steht eine Prüfung bekannter Erdmodelle mit unterschiedlichen Geschwindigkeitsstrukturen oberhalb der GIK. Zu den untersuchten Modellen zählen PREM, IASP91, AK135Q, PREM2, SP6, OICM2 und eine Variante des PREM. Die Untersuchung ergibt, daß Modelle, die einen verringerten Gradienten oberhalb der GIK aufweisen, eine bessere Übereinstimmung mit den gemessenen Daten zeigen als Modelle ohne diese Übergangszone. Zur Verifikation dieser These wird ein Erdmodell, das keinen verringerten Gradienten oberhalb der GIK besitzt (PREM), durch eine Reihe unterschiedlicher Geschwindigkeitsverläufe in diesem Bereich ergänzt und deren synthetische Seismogramme berechnet. Das Resultat der Untersuchung sind zwei Varianten des PREM, deren Frequenzanalyse eine gute Übereinstimmung mit den Daten zeigt. Das Abklingspektrum des Erdmodells PD47, das in einer 380 km mächtigen Schicht einen negativen Gradienten besitzt, zeigt eine große Ähnlichkeit mit den gemessenen Spektren. Dennoch kann es nicht als realistisches Modell angesehen werden, da der Punkt C in einer zu großen Entfernung liegt. Darüber hinaus müßte die zu kurze Differenzlaufzeit zwischen PKP(AB) und PKP(DF) beziehungsweise PKIKP durch eine größere Änderung der Geschwindigkeitsstruktur im inneren Kern kompensiert werden. Es wird deshalb das Modell PD27a favorisiert, das diese Nachteile nicht aufweist. PD27a besitzt eine Schicht konstanter Geschwindigkeit oberhalb der GIK mit einer Mächtigkeit von 150 km. Die Art des Geschwindigkeitsverlaufs steht im Einklang mit der geodynamischen Modellvorstellung, nach der eine Anreicherung leichter Elemente oberhalb der GIK vorliegt, die als Ursache für die Konvektion im äußeren Erdkern anzusehen ist. / In this thesis the structure of the outer core is investigated with PKP core phases. The knowledge of the physical properties of the earth&prime;s deep interior in this region is important for the understanding of geodynamical processes like the convective flow in the liquid outer core and the differential rotation of the earth&prime;s inner core.<br /> The first part of this thesis describes the investigation of the PKP caustic point B near 145 &#176;. For this purpose a cumulative amplitude distance curve is determined and compared with theoretical amplitude distance curves of different standard earth models. The data are broadband seismograms of the German Regional Seismic Network (GRSN) and the Gräfenberg Array (GRF). In order to measure the absolute amplitudes of the PKP phases, a WWSSN-SP filter is applied to the seismograms. The source regions are located in the South Pacific near Vanuatu Island (4 earthquakes) and on the French atolls Mururoa and Fangataufa (4 explosions). The advantage of a standardized network of seismic stations and the usage of station and magnitude corrections is a reduction of the scatter of the amplitude data. There is even less scatter than in studies with long period amplitude data (Häge, 1981). Another reason for the reduced scattering is the use of events with an impulsive source time function. Only the low scattering of the amplitude values makes it possible to interpret the data. More scattering of the data would have prevented an interpretation. The theoretical amplitude curves are similar in the caustic B distance range. The Q depth distribution for P and S waves used for calculating the synthetic seismograms is a combination of the values of the models CIT112 and PREM. The amplitudes determined with the help of this kind of model are slightly higher than the actually measured amplitudes. However, this needs not be taken into account because the interpretation is based on the position of the caustic peak. Therefore I rejected the computation of an improved Q model.<br /> Regarding the position of the caustic point there are two categories of earth models. The first group consists of the models IASP91 and 1066B with their maxima at 144.6 &#176; and 144.7 &#176; respectively. AK135, PREM and SP6 belong to a second group of models with caustic peaks at 145.1 &#176; and 145.2 &#176; (SP6). The measured curve has its maximum at 145 &#176;. All distances refer to a source depth of 200 km. For a surface focus the increase in distance is 0.454 &#176;. Therefore the peaks of the models AK135 and PREM are only 0.1 &#176; beside the maximum of the measured amplitude curve. The main result of this investigation is the amplitude distance curve in the vicinity of the cusp B. The curve determines the position of this point with an accuracy of &#177; 0.15 &#176; and points to earth models which would be good for modeling the amplitudes in the distance range of the PKP caustic B. The synthetic seismograms calculated for AK135 and PREM together with a standardized Q model fit the measured amplitude curve equally well.<br /> In the second part of this study the structure of the transition zone to the earth&prime;s inner core is investigated by using the spectral decay of the diffracted wave PKP(BC)diff at point C of the travel time curve. The physical process of diffraction is responsible for the strong reduction in amplitude of this wave. The influence of the diffraction on the seismic signal strongly depends on frequency. The interpretation of this phenomenon requires a calculation of decay spectra. In practice the attenuation of the PKP(BC)diff signal for eight frequencies between 6.4 s and 1.25 Hz are measured and visualized as a decay spectrum. The shape of a spectrum is characteristic of the velocity gradient above the inner core boundary (ICB). Those earthquakes whose core phases are recorded as diffracted core phases BCdiff lie beyond 150 &#176;. In this distance range there are the epicenters of the Tonga-Fiji slab. The broadband waveform data of the earthquakes in this region is used in this study. Decay spectra of waveform data which are not corrected for station site effects are incompatible with standard earth models. Therefore a spectral station correction is applied, which was especially determined for this purpose. The investigation starts with a review of a number of well-known earth models like PREM, IASP91, AK135Q, PREM2, SP6, OICM2 and a version of PREM. All these models have different velocity structures at the ICB. It is shown that models with a reduced velocity gradient above the ICB agree with the data rather than models without such a transition zone. For verification purposes a model without such a reduced gradient (PREM) is completed with different kinds of gradient zones to calculate synthetic seismograms. Two variants of the PREM correspond with the measured decay constants. The decay constants of model PD47 are very close to the measured ones. This model has a 380 km thick negative gradient above the ICB. Nevertheless it is not a realistic model because point C lies in a unrealistic great distance. As a result of the low velocity zone above the inner core there is a differential travel time between the PKP(AB) and the PKP(DF) phase (also PKIKP) which is too short. This would have to be compensated by a correction of the velocities in the inner core. Thus PD27a is the most suitable model which does not have the above mentioned disadvantages. PD27a has a 150 km thick layer of constant velocity above the ICB. This kind of velocity model is compatible with geodynamical theories according to which an enrichment of light elements above the ICB is present and powers the convection in the outer earth core by its buoyancy.