Astron. Astrophys. 346, 798-804 (1999)

2. Th cosmochronometry and the r-process

Assuming that the whole r-abundance distribution observed in CS 22892-052 and HD115444 is essentially solar, it is straighforward to relate the star age to the Th abundances,

where is the characteristic -decay timescale of Th and the subscripts obs and r refer to the observed and universal r-process abundance ratios, respectively. As classically done, the Th abundance is here expressed relative to the spectroscopically relevant Eu r-dominant element. The recent accurate observation of Th at the surface of CS 22892-052 amounts to (Sneden et al. 1996; Cowan et al. 1997). Th has also been observed at the surface of HD 115444, but its precise abundance remains to be determined (Pfeiffer et al. 1998). Assuming that a solar-like mix of the r-elements ingested in these halo stars originates from a small number of nucleosynthetic events that took place just before the formation of the stars, the age of the star can be estimated from Eq. (1) without calling for a complex model of the chemical evolution of the Galaxy. The only difficulty of the methodology is therefore related to the theoretical estimate of the r-production ratio .

Unfortunately, the r-process remains the most complicated nucleosynthetic process to model from the astrophysics as well as nuclear physics point of view (for a review see Arnould & Takahashi 1999). On the nuclear physics side, the nuclear structure properties (such as the nuclear masses, deformation, ...) of thousands of nuclei located between the valley of -stability and the neutron drip line have to be known, as well as their interaction properties, i.e the () and () rates, - and -decay half-lives and the fission probabilities. Despite much recent experimental effort, those quantities for most of the nuclei involved in the r-process remain unknown, so that they have to be extracted on theoretical grounds and are subject to the associated uncertainties. On top of these nuclear difficulties, the question of the astrophysical conditions under which the r-process can develop is far from being settled. The site(s) of the r-process is (are) not identified yet, all the proposed scenarios facing serious problems. For this reason, only parametric approaches, such as the so-called canonical model (Seeger et al. 1965) can be used to estimate the Th production. We use in the present study the multi-event model ¹ (Bouquelle et al. 1996; Goriely & Arnould 1996) in which the best fit to the solar abundances is derived from a superposition of canonical events with the aid of an iterative inversion procedure. Compared with other treatments of the canonical model (e.g Pfeiffer et al. 1998), a major advantage of the multi-event approach is to provide an efficient tool for a systematic study of the various uncertainties affecting the model (Goriely 1999). The iterative inversion method works in such a way that the modification of a given (nuclear or astrophysics) input in the r-process model leads to an automatic renormalization of the thermodynamic conditions necessary to optimize the fit to the solar r-abundance distribution. Therefore, the uncertainties affecting the input data of the parametric model, as well as their impact on the Th production can be studied systematically within the multi-event approach, as shown in the next section.

Our standard calculation is performed under the following thermodynamic conditions: , and (where T is the temperature, the neutron density and the number of neutrons captured per seed nucleus). Note that the r-process calculations are performed making use of the waiting point approximation, since under the thermodynamic conditions considered here, an almost complete equilibrium is established (Goriely & Arnould 1996). When not available experimentally, the nuclear data are taken from the ETFSI nuclear masses of Aboussir et al. (1995) and from the gross theory (GT2) of decay (and -delayed neutron emission) of Tachibana et al. (1990). In addition, -decay and fission processes are also considered (before and after the neutron irradiation freeze-out). The fission processes include spontaneous, -delayed and neutron-induced fission, the probabilities of which are calculated according to the prescriptions of Kodoma & Takahashi (1975) with the ETFSI fission barriers (Mamdouh et al. 1998). The procedure used to fit the solar r-abundance distribution is similar to the one described in Bouquelle et al. (1996) though each isotope is now given a weight inversely proportional to the error affecting its solar r-abundance (Goriely, 1999). Since we are mainly concerned with Th cosmochronometry, no details are given for the representative thermodynamic conditions required to fit the solar system r-abundance distribution (Such details can be found in Goriely & Arnould, 1996).

© European Southern Observatory (ESO) 1999

Online publication: June 17, 1999
helpdesk.link@springer.de