Thesis (Ph. D.)--University of Chicago, Dept. of Physics, June 1997. / Includes bibliographical references. Also available on the Internet.

Collisions of black holes and neutron stars, named mixed binaries in the following, are interesting because of at least two reasons. Firstly, it is expected that they emit a large amount of energy as gravitational waves, which could be measured by new detectors. The form of those waves is expected to carry information about the internal structure of such systems. Secondly, collisions of such objects are the prime suspects of short gamma ray bursts. The exact mechanism for the energy emission is unknown so far. <br><br> In the past, Newtonian theory of gravitation and modifications to it were often used for numerical simulations of collisions of mixed binary systems. However, near to such objects, the gravitational forces are so strong, that the use of General Relativity is necessary for accurate predictions. <br><br> There are a lot of problems in general relativistic simulations. However, systems of two neutron stars and systems of two black holes have been studies extensively in the past and a lot of those problems have been solved. One of the remaining problems so far has been the use of hydrodynamic on excision boundaries. Inside excision regions, no evolution is carried out. Such regions are often used inside black holes to circumvent instabilities of the numerical methods near the singularity. Methods to handle hydrodynamics at such boundaries have been described and tests are shown in this work. <br><br> One important test and the first application of those methods has been the simulation of a collapsing neutron star to a black hole. The success of these simulations and in particular the performance of the excision methods was an important step towards simulations of mixed binaries. <br><br> Initial data are necessary for every numerical simulation. However, the creation of such initial data for general relativistic situations is in general very complicated. In this work it is shown how to obtain initial data for mixed binary systems using an already existing method for initial data of two black holes. <br><br> These initial data have been used for evolutions of such systems and problems encountered are discussed in this work. One of the problems are instabilities due to different methods, which could be solved by dissipation of appropriate strength. Another problem is the expected drift of the black hole towards the neutron star. It is shown, that this can be solved by using special gauge conditions, which prevent the black hole from moving on the computational grid. <br><br> The methods and simulations shown in this work are only the starting step for a much more detailed study of mixed binary system. Better methods, models and simulations with higher resolution and even better gauge conditions will be focus of future work. <br><br> It is expected that such detailed studies can give information about the emitted gravitational waves, which is important in view of the newly built gravitational wave detectors. In addition, these simulations could give insight into the processes responsible for short gamma ray bursts. / Zusammenstöße eines schwarzen Lochs und eines Neutronensterns, im Folgenden "gemischte Zusammenstöße" genannt, sind aus wenigstens zwei Gründen interessant. Erstens wird erwartet, dass dabei große Mengen Energie als Gravitationswellen freigesetzt werden und diese mit neuen Detektoren gemessen werden können. Die Form dieser Wellen verrät viel über die Beschaffenheit eines solchen Systems und stellt neben elektromagnetischen Wellen eine wichtige Informationsquelle dar. Zweitens sind Zusammenstöße von kompakten Objekten wie Neutronensternen und schwarze Löchern sehr wahrscheinlich die Ursache sogenannter kurzer Gammastrahlungsblitze. Deren genauer Mechanismus für die Umwandlung der gewaltigen Energiemengen, die bei diesen Blitzen ausgesandt werden, ist jedoch bisher unbekannt. <br><br> Computersimulationen von Zusammenstößen eines gemischten Systems wurden bisher oft unter Benutzung der Newtonschen Gravitationstheorie, bzw. Korrekturen dazu, durchgeführt. In der Nähe so kompakte Objekte wie schwarzer Löcher oder Neutronensterne ist jedoch die Gravitationswirkung so stark, dass Näherungen wie die erwähnten Korrekturen der Newtonschen Gravitationstheorie zu ungenau sind. Eine Benutzung der allgemeinen Relativitätstheorie ist daher für dieses Problem unumgänglich. <br><br> Die Probleme allgemein-relativistischer Simulationen sind vielfältig. Jedoch wurden Binärsysteme zweier schwarzer Löcher und zweier Neutronensterne schon eingehend untersucht, und so viele Probleme, die auch Simulationen gemischter Systeme betreffen, gelöst. Eins der bisher ausstehenden Probleme war die Behandlung der Hydrodynamik an Ausschneiderändern; Rändern zu Gebieten, die in der Zeitentwicklung der Simulation ignoriert werden. Solche Ränder werden zum Beispiel innerhalb eines schwarzen Lochs benutzt, um Instabilitäten des Programms in der Nähe der Singularität zu vermeiden. Methoden, solche Ränder zu behandeln wurden in der Arbeit entwickelt, getestet und gezeigt, dass sie verlässlich arbeiten. <br><br> Ein wichtiger Test für diese Methoden, der gleichzeitig der Gewinnung neuer Erkenntnisse diente, war deren Anwendung auf Simulationen von zu schwarzen Löchern kollabierenden, rotierenden Sternen. Der Erfolg, diese Simulationen ohne Probleme mit den erwähnten Methoden durchzuführen, war ein wichtiger Schritt zu Simulationen gemischter Binärsysteme. <br><br> Für Computersimulationen sind Anfangsdaten notwendig, die das gewünschte Problem beschreiben. Die Erstellung solcher Anfangsdaten ist jedoch unter Benutzung der allgemeinen Relativitätstheorie ausser in Spezialfällen sehr komplex. Wir zeigen, wie man einen schon vorhandenen Algorithmus für Anfangsdaten für zwei schwarze Löcher ändern kann, um Anfangsdaten für ein gemischtes Binärsystem zu erhalten. <br><br> Diese Anfangsdaten wurden für Simulationen eines gemischten Binärsystems benutzt. Während dieser Simulationen traten mehrere Probleme auf. Zwei dieser Probleme waren numerische Instabilitäten unterschiedlicher Herkunft. Beide konnten jedoch mit angepasst starker Dissipation (der künstliche Entnahme von hochfrequenter Energie aus dem System) unterdrückt werden. Ein weiteres Problem war die erwartete Bewegung des schwarzen Lochs in Richtung des Neutronensterns. Da ein Teil des Simulationsgebietes innerhalb des schwarzen Lochs ausgeschnitten wird und das verwendete Programm bewegte Ausschneidegebiete nicht behandeln kann, darf sich das schwarze Loch jedoch auf dem Gitter kaum bewegen. Wir haben dieses Problem durch eine an das Problem angepasste Eichbedingung gelöst, die auf Bewegungen des scheinbaren Horizons reagiert und die Position des schwarzen Lochs auf diese Weise nahezu konstant hält. <br><br> Die Methoden und Simulationen dieser Arbeit sind nur der Anfangspunkt einer ausführlichen Studie von Binärsystemen eines schwarzen Lochs und eines Neutronensterns. Bessere Methoden, Modelle und Simulationen mit höherer Auflösung und besser an das System angepassten Koordinaten werden Mittelpunkt zukünftiger Arbeit sein. <br><br> Es wird erwartet, dass solche detailierten Studien Erkenntnisse über die abgestrahlten Gravitationswellen liefern, die gerade in Hinblick auf die neuen Gravitationswellendetektoren wichtig sind. Weiterhin könnten diese Simulationen dabei helfen, die Prozesse, die kurze Gammastrahlungsblitze hervorrufen, und über die im Moment kaum etwas bekannt ist, aufzuklären.

Neutron stars are some of the most fascinating objects in Nature. Essentially all aspects of physics seems to be represented inside them. Their cores are likely to contain deconfined quarks, hyperons and other exotic phases of matter in which the strong interaction is the dominant force. The inner region of their solid crust is penetrated by superfluid neutrons and their magnetic fields may reach well over 1012 Gauss. Moreover, their extreme mean densities, well above the densities of nuclei, and their rapid rotation rates makes them truly relativistic both in the special as well as in the general sense. This thesis deals with a small subset of these phenomena. In particular the exciting possibility of trapping of gravita-tional waves is examined from a theoretical point of view. It is shown that the standard condition R &lt; 3M is not essential to the trapping mechanism. This point is illustrated using the elegant tool provided by the optical geometry. It is also shown that a realistic equation of state proposed in the literature allows stable neutron star models with closed circular null orbits, something which is closely related to trapped gravitational waves. Furthermore, the general relativistic theory of elasticity is reviewed and applied to stellar models. Both static equilibrium as well as radially oscillating configurations with elasticsources are examined. Finally, Killing tensors are considered and their applicability to modeling of stars is discussed

No description available.

Prior to this thesis no serious attempt has been made within binary system population synthesis research to model the selection effects of observational surveys. Conversely, many pulsar population models have accounted for radio survey selection effects but not detailed binary evolution. Such modelling becomes especially important when comparing theory directly to observations. In examining the factors that influence pulsar evolution, both in binary systems and as single stellar objects, we have bridged this existing gap between these two research fields. This thesis populates a model Galaxy with binary systems and evolves the population forward in time. A prediction of the Galactic pulsar population characteristics is produced, at the assumed age of the Galaxy, after we have accounted for detailed changes in stellar and binary evolution and Galactic kinematics. Synthetic observational surveys mimicking a variety of radio pulsar surveys are then performed on this population. The population synthesis synthetic survey (PS3) package is comprised of three components: stellar/binary evolution (binpop), Galactic kinematics (binkin), and survey selection effects (binsfx). The resultant pulsar populations, assuming the magnetic-dipole decay and accretion induced magnetic decay models, can compare well to many of the detected pulsar population characteristics. The comparisons between models and observations have lead to the conclusions described below. The models exclude short (∼ 5 Myr) timescales for standard pulsar exponential field decay and find that ablation of low-mass millisecond pulsar companions can redress both the lack of synthetic isolated pulsars and their excessive distances in height from the Galactic plane. Coalescing double neutron star and collapsar Galactic populations, evolved owing to standard binary evolutionary assumptions, are too centrally concentrated owing to the typical merger timescale of double neutron stars being a few million years. Dwarf galaxy models of coalescing double neutron stars and collapsars produce equally good agreement with long gamma-ray burst projected distances. Therefore our models cannot provide any distinction between which of these populations (coalescing double neutron stars or collapsars) are the progenitor of long gamma-ray bursts. The Galactic birth rate of double neutron star binaries in our model is 8.2 Myr−1 and the merger rate is 6.8 Myr−1. Scintillation is found to be an important aspect in the detection of low flux density pulsars. The assumed pulsar luminosity law is found to require an inverse trend with spin period and our favoured models suggest that there are one million radio active pulsars within the Galaxy.

Small amplitude rotational irregularities are observed in a number of rotation-powered pulsars. They fall into two classes: (i) glitches, defined as abrupt increases in the angular velocity of a pulsar (accompanied sometimes by changes in the angular acceleration Ω), of which 286 have been observed in 101 objects; and (ii) timing noise, a continuous stochastic fluctuation in phase, or, which is observed mostly in young and adolescent pulsars (with ages ≥ 10 4 yr). Both classes of irregularity seem to arise from some mechanism that couples the angular momentum of the solid crust and superfluid core of the star, which is activated suddenly when differential rotation exceeds a threshold. Coupling mechanisms proposed to date include catastrophic vortex unpinning in the inner crust, triggered by starquakes; vortex creep, due to thermally activated quantum tunnelling; superfluid-superconductor interactions in the core; and superfluid instabilities. The associated theories are phenomenological, not predictive.

This thesis considers the various morphologies of radio supernova remnants (SNRs), and attempts to determine whether their appearance results from the properties of the progenitor star and its supernova explosion, or from the structure of the interstellar medium (ISM) and ambient magnetic field into which a SNR consequently expands. High-resolution observations of Supernova 1987A show a young remnant whose appearance and evolution are completely dominated by the structure of its progenitor wind. A statistical study of the Galactic population of bilateral SNRs demonstrates that the symmetry axes of these remnants run parallel to the Galactic Plane. This result can be explained by the interaction of main sequence stellar wind-bubbles with the ambient magnetic field; expansion of SNRs into the resulting elongated cavities results in a bilateral appearance with the observed alignment. Radio observations of SNR G296.8-00.3 show a double-ringed morphology which is best explained by expansion either into an anisotropic main-sequence progenitor wind or into multiple cavities in the ISM. Data on SNRs G309.2-00.6 and G320.4-01.2 (MSH 15-52) make a strong case that the appearance of both remnants is significantly affected by collimated outflows from a central source; for G309.2-00.6 the source itself is not detected, but for G320.4-01.2 there is now compelling evidence that the remnant is associated with and is interacting with the young pulsar PSR B1509-58. I conclude that, while the youngest SNRs are shaped by their progenitor's circumstellar material, the appearance of most SNRs reflects the properties of the local ISM and magnetic field. Remnants which interact with an associated pulsar or binary system appear to be rare, and are easily distinguished by their unusual and distorted morphologies.

Thesis (Ph. D.)--University of Washington, 2002. / Vita. Includes bibliographical references (p. 131-138).

Thesis (MSc)--University of Stellenbosch, 2008. / Bibliography. Also available via the Internet.

M.Sc. / The content of this dissertation is directed at a clearer understanding of pulsar behaviour. Pulsars are neutron stars in relatively early stages of their evolution. The intervals between observed arrival times of pulsar beams have been observed to suddenly decrease implying an increase in the star’s angular velocity Ω. This phenomenon is called a glitch. The first observed glitch occurred in the Vela pulsar (PSR 0833 -45) in 1969. The largest observed glitch also occurred in the Vela pulsar and exhibited a relative spin-up, Ω∕Ω~10−6 [1]. These relative pulsar spin-ups generally fall within the range 10−9 to 10−6 [2]. Pulsar glitches occur as a result of the processes in the interior of a neutron star [3]. Thus, a rigorous study of neutron star interiors is justified. Neutron star interiors are composed of hadronic matter: mostly neutrons, with a few per cent of protons (and, of course, non-hadronic electrons to conserve electric charge). The density of a neutron star interior increases with increasing depth as illustrated in 1.1.