Recent observational data on the type Ia supernova rates are in excellent agreement with the old prediction of the population synthesis of binary stars and confirm that the overwhelming majority of type Ia supernovas (~99%) in elliptical galaxies form via mergers of binary white dwarfs with the total mass exceeding the Chandrasekhar limit.
The interest in type Ia supernovas (used as standard candles for cosmology), which led the researchers to suspect the presence of dark energy in the Universe (Riess et al. 1998;Perlmutter et al. 1999), triggered mass discovery of supernovas, resulting in an almost 50-fold increase of the number of these stars studied in the last decade! The mass discovery of supernovas in recent years allowed the researchers to observe for the first time the dramatic evolution of the supernova rate in elliptical galaxies, which was predicted more than 10 years ago via population synthesis (Jorgensen et al. 1997).
Type Ia supernovas are now generally believed to be products of nuclear explosions of white dwarfs that have reached the Chandrasekhar limit (for a review see Livio et al. 2000).
No appreciable star formation goes on in elliptical galaxies. Only low-mass stars remain in these systems after the first billion years of their evolution. The evolution of all massive stars (with M > 8-10 M ) ends completely with the formation of neutron stars and black holes. Low-mass stars by themselves cannot produce supernova explosions, because their evolution ends with a soft formation of white dwarfs with masses below the stability limit (the Chandrasekhar limit). However, a delayed (by several billion years) accumulation of the Chandrasekhar mass may occur in binary systems -as a result of either the accretion of matter from a companion (the so-called SD-mechanism (Whelan & Iben 1973)), or merger (the DDmechanism (Iben & Tutukov 1984;Webbink 1984)).
The very first evolutionary computations of such processes in elliptical galaxies (population synthesis) performed using a special computer code -the Scenario Machine (Lipunov et al. 1996;Lipunov et al. 2009) -showed (Jorgensen et al. 1997) that the mechanism of white-dwarf merger outperforms accretion by two orders of magnitude already one billion years after the formation of the elliptical galaxy (see Figure 1).
The study of supernovas in recent years allowed the evolution of the supernova rates in elliptical galaxies to be observed for the first time (Totani et al. 2008) (Figure 1). These results were obtained by analyzing the observations of candidate type Ia supernovas based on Subaru/XMM-Newton Deep Survey (SXDS) data. The ages of elliptical galaxies were determined from nine-band photometry spanning from optical to mid-infrared wavelengths. The observed decrease of the SN Ia rate was found to be described by the
Supernova observers measure the supernova rates per unit K-band absolute magnitude 10 10 L K,0, and we therefore converted our old data into the new plot assuming that
] in accordance with modern data to find the results to be in excellent agreement (Figure 1) with the supernova rates predicted by the theory of the evolution of binary stars (Jorgensen et al. 1997) .
It goes without saying that population synthesis is a complex numerical process, which incorporates our knowledge and hypotheses about the evolution of binary stars, as well as the observed properties of binary stars (the initial mass function and the initial distribution of separations). However, we try to show that the results obtained 13 years ago, like those of more recent computations (Fedorova et al. 2004;Förster et al. 2006;Totani et al. 2008;Wang et al. 2010;Yungelson et al. 1996;Yungelson 2005), are extremely weakly sensitive to the «dark areas» of the evolution of binary stars.
Here the form of the initial distribution of separations of binary systems plays the crucial part (a).
It was shown (Popova et al 1982;Abt 1983) that the observed distribution of separations of binary stars in our Galaxy at the beginning of their main-sequence evolution can be described by the following law:
This distribution still remains a theoretical puzzle, which can be popularly formulated as follows -our Galaxy contains approximately equal numbers of wide and close binaries (i.e., equal logarithmic intervals -e.g., the decades -contain equal numbers of stars). We can assume, to a first approximation (as it is commonly done in population synthesis), that other galaxies must have had the same initial distribution of binary stars. There are no particular reasons to believe that binaries in other galaxies should form in a different way. After its formation a binary star undergoes a long and varied evolution accompanied by the change of the component separation. In low-mass binaries the most important and least understood evolutionary factor remains the so-called common-envelope phase, where one of the components is inside its companion star swollen to the red-giant state (stellar cannibalism). During the common-envelope stage the components approach each other catastrophically. However, because of the power-law form of distribution (1), a proportional approach by a certain factor that is almost independent of the component separation has no appreciable effect on the distribution function. Hence it would appear
This content is AI-processed based on open access ArXiv data.