Thesis (M.S.)--Brigham Young University. Dept. of Mechanical Engineering, 2006. / Includes bibliographical references (p. 105-106).

Universiẗat, Diss., 2000--Tübingen. / Gedr. Ausg. in Selbstverl. Stefan Kunze, Tübingen.

Michael Triantafyllou, Michael Athans. / "May 1982" "Reprinted from Proc. OCEANS' 81, Boston, Mass., September 1981." / Bibliography: leaf 1094. / "Grant NGL-22-009-124"

Michael S. Triantafyllou, Marc Bodson. / "May 1982." "Reprinted from the Proceeding of the 14th Annual Offshore Technology Conference, Houston, Texas, May 1982." / Includes bibliography. / Grant NGL-22-009-124

Thesis (M.S.)--Oregon State University, 2001. / Typescript (photocopy). Includes bibliographical references (leaves 97-103). Also available via the World Wide Web.

No description available.

Giant stars are believed to lose significant fractions of their total mass over their lifetimes, but the mechanisms responsible for this are ill-understood. One possible mechanism is dynamical mass loss - a hydrodynamical process in which matter is ejected from the stellar surface in ballistic outflows. In this thesis, dynamical mass loss is studied in three stellar regimes: common-envelope objects, asymptotic giant branch stars, and red supergiants. Using hydrodynamical simulations performed with the stellar evolution code MESA, we examine the dynamical behaviour and stability of stars in each of these regimes. We examine the dynamical properties of common-envelope objects during the slow spiral-in phase using a parameterised 1-dimensional model of orbital dissipatory heating. We find that the envelope becomes unstable to high-amplitude dynamical pulsations that can lead to repeated mass-ejection events capable of removing the entire envelope and terminating the common-envelope phase. We estimate this process's α efficiency value and suggest how these results might be employed in parameterised common-envelope models. We employ coupled evolutionary and hydrodynamical simulations of AGB stars to study their dynamical properties as they traverse the TP-AGB and examine their dependence on basic stellar properties and on the thermal pulse cycle. We find that these models experience large amounts of dynamical mass loss, and we construct a parameterised model to estimate its strength. We find that this model is successful at locating the termination of the AGB. We apply a similar approach to a study of RSGs, and find that dynamical mass loss also emerges in this regime. We estimate the conditions under which this occurs and discuss how this mechanism may resolve theoretical problems relating to the Humphreys-Davidson limit and the progenitors of SNe IIn. We conclude that dynamical mass loss is likely to form a vital part of the mass-loss histories of cool giant stars.

xvii, 371 p. : col. ill. / An important problem in astrophysics involves understanding the formation of planetary systems. When a star-forming cloud collapses under gravity its rotation causes it to flatten into a disk. Only a small percentage of the matter near the rotation axis falls inward to create the central object, yet our Sun contains over 99% of the matter of our Solar System. We examine how global hydrodynamic instabilities transport angular momentum through the disk causing material to accrete onto the central star. We analyze the stability of polytropic disks in the linear regime. A power law angular velocity of power q is imposed, and the equilibrium disk structure is found through solution of the time-independent hydrodynamic equations via the Hachisu self-consistent field method. The disk is perturbed, and the time-dependent linearized hydrodynamic equations are used to evolve it. If the system is unstable, the characteristic growth rate and frequency of the perturbation are calculated. We consider modes with azimuthal e im[varphi] dependence, where m is an integer and [varphi] is the azimuthal angle. We map trends across a wide parameter space by varying m , q and the ratios of the star-to-disk mass M * /M d and inner-to-outer disk radius r - /r + . We find that low m modes dominate for small r - /r + , increasing to higher r - /r + as M * /M d increases, independent of q . Three main realms of behavior are identified, for M * << M d , M * [approximate] M d and M * >> M d , and analyzed with respect to the I, J and P mode types as discussed in the literature. Analysis shows that for M * << M d , small r - /r + disks are dominated by low m I modes, which give way to high m J modes at high r - /r + . Low m J modes dominate M * [approximate] M d disks for small r - /r + , while higher m I modes dominate for high r - /r + . Behavior diverges with q for M * >> M d systems with high q models approximating M * [approximate] M d characteristics, while low q models exhibit m = 2 I modes dominating where r - /r + < 0.60. / Committee in charge: Raymond Frey, Chairperson; James Imamura, Advisor; Robert Zimmerman, Member; Paul Csonka, Member; Alan Rempel, Outside Member

abstract: I combine, compare, and contrast the results from two different numerical techniques (grid vs. particle methods) studying multi-scale processes in galaxy and structure formation. I produce a method for recreating identical initial conditions for one method from those of the other, and explore methodologies necessary for making these two methods as consistent as possible. With this, I first study the impact of streaming velocities of baryons with respect to dark matter, present at the epoch of reionization, on the ability for small halos to accrete gas at high redshift. With the inclusion of this stream velocity, I find the central density profile of halos is reduced, overall gas condensation is delayed, and infer a delay in the inevitable creation of stars. I then combine the two numerical methods to study starburst outflows as they interact with satellite halos. This process leads to shocks catalyzing the formation of molecular coolants that lead to bursts in star formation, a process that is better captured in grid methods. The resultant clumps of stars are removed from their initial dark matter halo, resemble precursors to modern-day globular clusters, and their formation may be observable with upcoming telescopes. Finally, I perform two simulation suites, comparing each numerical method's ability to model the impact of energetic feedback from accreting black holes at the core of giant clusters. With these comparisons I show that black hole feedback can maintain a hot diffuse medium while limiting the amount of gas that can condense into the interstellar medium, reducing the central star formation by up to an order of magnitude. / Dissertation/Thesis / Doctoral Dissertation Astrophysics 2014

In this thesis we study the interaction of supernova ejecta in the environment of young massive clusters. It has been already shown that winds of massive stars can be thermalized by mutual interactions inside the cluster and drive the strong star cluster wind. The SNe are, as discrete and extremely energetic events, in all ways diferent from the continuous stellar winds. This triggers the question under which parameter and if at all can the SNe ejecta interaction from a smooth star cluster wind. Therefore we at first parametrize the SNe explossions and based on the 3D simulations in FLASH we show for the first time that the convergence of the SNe ejecta interaction to a smooth star cluster wind is controlled by a single parameter ΠSN . The paramater ΠSN estimates the mean number of interacting SN ejecta based on a comparison of supernova rate and crossing time of SN ejecta in a cluster. For high enough values ΠSN > 1 the cluster is able to build up smooth a star cluster wind. This allows us to use a 1D semi-analytic code WINDCALC to calculate the cooling of the hot gas due to dust and estimate under which conditions the SNe-inserted matter is captured. This may explain the origin of so-called anomalous globular clusters. 1