Abundance Stratification in Type Ia Supernovae

Type Ia Supernovae are among the most powerful events in the Universe, which makes them very appealing for astronomers. Because of their enormous brightness and their standardizable luminosity they are highly suitable to measure cosmological distances. In order to achieve reliable results it is necessary to understand the physical processes that drive the Supernova explosion. Synthetic spectra from which the chemical abundance distribution can be derived are one of the most important keys to understand these fascinating events. They deliver the link between theoretical explosion models and the observations.

Fig. 1: The Type Ia Supernova SN 2002bo, discovered on March 9, 2002 in the galaxy NGC 3190 by Paulo Cacella and Yoji Hirose.

Fig. 2: Spectrum of SN 2002bo observed on March 23, 2002 near the optical maximum. The observed spectrum (black line) covers wavelengths up to the near infrared. A synthetic spectrum is shown in red.

Fig. 3: Chemical abundance distribution in SN 2002bo. The abundances are plotted against expansion velocity. Lower velocity means that these nuclei were produced closer to the center of the Supernova. Higher velocities are associated with the outer layers.

New computer models to simulate Type Ia Supernova explosions are currently leading to a breakthrough in this field. Friedrich Röpke and Wolfgang Hillebrandt from MPA recently presented new results of a three dimensional thermonuclear explosion simulation leading to the complete disruption of a White Dwarf. The White Dwarf originally contained mostly Carbon and Oxygen. These atomic nuclei are converted into heavier elements — up to Nickel and Iron — during the violent explosion. Only a few seconds after the ignition the thermonuclear fusion is finished, and the Supernova ejecta expand homologously, i.e. with a constant velocity whose value is proportional to the distance from the center of the Supernova. The efficiency of the nuclear fusion depends on the conditions of the environment, such as temperature and density, so that the determination of the chemical abundances of the Supernova ejecta allows us to draw conclusions on the processes that lead to the explosion.

Therefore, extensive observations of nearby Supernovae are also necessary. The observational campaign is part of the "Research Training Network" (RTN) The Physics of Type Ia Supernovae, funded by the European Union. On the one hand these data are used to analyze the physical properties of Type Ia Supernovae from the observational point of view. On the other hand explosion models can be tested by means of the observations. Synthetic spectra play a key role in this context. In particular by reconstructing the chemical abundances layer by layer, explosion models can be checked quantitatively for the very first time.

This method is based on the computational simulation of a series of spectra obtained in steps of a few days between the explosion and about one year later. For each of these epochs, a "photosphere" is defined, above which the ejecta is mostly transparent ("optically thin") whereas the photons below this layer are trapped due to the high optical depth. The photosphere moves inwards with time, i.e. towards lower velocities, due to the expansion of the Supernova until a few weeks after maximum light the photosphere eventually disappears and the Supernova ejecta become completely transparent.

Chemical abundances above the photosphere of the earliest spectrum in the series are derived from the corresponding model. They are stored for further calculations. Only the abundances in the layer between the first and the second photosphere are determined through the next synthetic spectrum in the time sequence. Continuing this procedure throughout the entire set of available spectra delivers a detailed abundance stratification of the Supernova ejecta. This result can easily be compared with theoretical predictions from the explosion models.

Supernova SN 2002bo (Fig. 1) is perfectly suited for this kind of analysis because a good time series of observations, with spectra taken almost every other night, between about 13 days before maximum light and about 10 days after, together with two observed spectra in the so-called nebular phase, were obtained. Figure 2 shows an observed spectrum near the maximum of optical light (in black) together with the corresponding model (in red) as an example for the 13 synthetic spectra that were calculated. The observed spectra cover wavelengths from the ultraviolet to the near infrared. Since the chemical ingredients in the synthetic spectrum are well known, the abundances can be derived from the absorption lines and their depths.

The analysis of all synthetic spectra delivers the abundances of all chemical elements in the Supernova ejecta against expansion velocity (Fig. 3). Most obvious is the distribution of heavy nuclei near the center of the Supernova (Iron, Nickel, Titanium, and Chromium) followed by the intermediate mass elements (IME) (Silicon, Calcium, Magnesium) at larger radii. Unburned material (mainly Oxygen) is located in the outermost parts of the ejecta. However, the individual layers are not completely separated, but clearly overlap. This effect is a consequence of the explosion mechanism where the different zones are mixed during the burning process. If mixing was ignored the spectra could not be modeled with such high quality. Meanwhile there are many indications that mixing is not a global effect but a local phenomenon in the Supernova ejecta. This substantiates the three dimensional character of the Supernova explosions. The spectra, and therefore also the chemical abundances, should look slightly different depending on the viewing angle to the Supernova.

In order to draw more detailed conclusions on the accuracy of the explosion models it is necessary to determine absolute abundances and their exact distribution. It is of particular interest to know about the amount and distribution of radioactive 56Ni, as this affects the explosion energy as well as the maximum luminosity.

These questions will soon be answered when a larger sample of individual Supernovae will be analyzed in the same way as SN 2002bo. Further steps towards obtaining a complete picture of these objects will be made through computer models which simulate spectra in three dimensions. With this knowledge the uncertainties in measuring distances with Type Ia Supernovae can be reduced significantly, and cosmological consequences will be confirmed properly.


M. Stehle, P.A. Mazzali, W. Hillebrandt


Further information:

M. Stehle, P.A. Mazzali, S. Benetti, W. Hillebrandt, Abundance stratification in Type Ia Supernovae: I. The case of SN 2002bo, 2005, MNRAS, 360, 1231