Binary star population synthesis within a galactic framework:

Scriptie/masterproef: Doctoral Thesis

Uittreksel

The present study focuses on the evolution of binary star systems and their contributions to
the chemical evolution of galaxies. A first aspect concerns the detailed evolutionary
calculation of interacting binaries. The code used for this purpose has been in development
and constantly updated for almost four decades. The results of these detailed evolutionary
calculations, mainly concerning the amounts of chemical elements that are returned by the
binary to the interstellar medium and at what time, are then used in a galactic setting. For this
purpose, a binary population number synthesis code was re-written and upgraded. This code
uses as input thousands of evolutionary calculations performed with the actual binary
evolutionary code (these calculations were not part of this investigation). These data enable to
calculate events (e.g. different types of supernovae or mergers) and outputs (e.g. chemical
yields) of a starburst with a certain binary fraction and metallicity. The output of this
population code is then introduced into a third code, the chemical evolution model. This
convolves the pre-mentioned output with a galaxy and star formation model to study the
chemical evolution (star formation rate, metallicity, star and gas densities, chemical
abundances, supernova rates, etc.) of an evolving galaxy (e.g. the Milky Way) as a function of
time. The combination of these three codes allows to study various aspects of the influence of
binary stars on the chemical evolution of galaxies, all from first principles. By investigating
how the properties or modeling of any (binary) stellar or galactic process affect the
macroscopic, observationally testable properties of galaxies, these processes and the
parameters that describe them can be constrained. This concerns topics such as the type Ia
supernova delay time distribution and Galactic rate, as well as the influence of these events on
the metallicity distribution of G-type dwarfs. We find that the 'single degenerate' scenario for
type Ia supernova formation alone cannot explain observations, while a combination with the
'double degenerate' scenario can. The method also allows to study the importance of mass
return by intermediate mass close binaries for the chemical enrichment in globular clusters,
and of merger products created by massive close binaries, such as blue-type core collapse
supernova progenitors. Furthermore, the merger rate of double compact objects (neutron stars
and/or black holes), their expected gravitational wave radiation signal, and their importance
for the enrichment in Galactic r-process elements can be studied. A common topic in all of the
mentioned population-wide studies, which links them to detailed underlying calculations, is
the uncertainty in some binary evolutionary processes, and the way they are parameterized in
population synthesis studies. The most uncertain aspects in all of our studies involve mass and
angular momentum transfer between binary components themselves and with their
environment. The most important ones, all critical to the outcome of such evolutionary phase,
are (1) the stable mass transfer efficiency ?, the fraction of mass lost by the mass donor in a
binary that is accreted by the mass gainer; (2) in the case of ?efficiency, determining how much angular momentum is lost with a fixed amount of mass;
and (3) in the case of unstable mass transfer, resulting in both stellar cores rotating within one
"common envelope", the efficiency with which rotational energy is converted into kinetic
energy used to expel this envelope. Given the fact that neither intense stable or unstable mass
transfer are sufficiently long-lasting processes to ever have been observed, many different
models exist for them in detailed evolutionary computations, and many different
parameterizations in global population studies. In a study comparing four different population
synthesis codes, we have shown that when these and other parameterizations in the codes are
equalized, this leads to a significant convergence of the codes' predictions concerning single
and double white dwarf systems.
Datum Prijs26 mrt 2014
TaalEnglish
BegeleiderJean De Greve (Promotor) & Dany Vanbeveren (Promotor)

Citeer dit

'