|Annu. Rev. Astron. Astrophys. 1997. 35:
Copyright © 1997 by . All rights reserved
In this review I refer to heavy elements as those elements beyond the iron peak, with nuclear charge Z 31. The reader is directed to Meyer (1994), Busso et al (1995), Lambert et al (1995b), Käppeler et al (1989) for more detailed discussions of heavy element synthesis. Elements beyond the iron peak cannot be efficiently produced by charged-particle interactions owing to the large Coulomb repulsion between nuclei; temperatures high enough to overcome the Coulomb barrier tend to photodisintegrate even the iron-peak nuclei (e.g. Woosley & Weaver 1995). Burbidge et al (1957) showed that heavy elements can be synthesized by successive neutron captures onto iron-peak nuclei, followed by decays.
The neutron captures can occur on a time scale long enough for all decays to occur, which is called the s-process (for slow neutron capture), or on a time scale that is short compared to decay, called the r-process (for rapid neutron capture); these two processes lead to two characteristic abundance patterns.
In a steady flow of neutrons the abundance of each isotope is inversely proportional to its neutron capture cross section. The closed neutron shells with 50, 82, and 126 neutrons have small neutron capture cross sections, leading to abundance peaks for these nuclei. Similarly, even-numbered nuclei have smaller neutron capture cross sections than odd-numbered nuclei, resulting in higher abundances for the even nuclei; this is called the odd-even effect. The s-process abundance pattern is characterized by abundance peaks near mass numbers 87, 138, and 208 neutrons and a strong odd-even effect. The r-process abundance pattern is characterized by the abundance peaks shifted to mass numbers near 80, 130, and 195 with no odd-even effect.
Seeger et al (1965) showed that observed abundances of s-process-only isotopes can be represented by an exponential distribution of neutron exposures. For the Solar System material, the heavy element abundance pattern is best fit by a combination of two s-process exponentials (e.g. Käppeler et al 1989): (a) the weak component, which corresponds to the light elements A 85 (thought to occur in the cores of massive stars, M 10 M; see Raiteri et al 1991, 1992, 1993), and (b) the main component, which fits the region approximately between Rb and Pb.
The s-process main component is thought to occur during the thermal pulse stage of low-mass (1-3 M) AGB stars at neutron densities of 107-109 cm-3. Quantitative calculations of AGB (asymptotic giant branch) nucleosynthesis were first performed by Iben (1975), Truran & Iben (1977). Iben & Truran (1978) estimated that AGB nucleosynthesis in intermediate mass AGB stars could account for a significant fraction of the Galactic abundances of carbon and s-process elements. Since that time, much observational and theoretical work has converged on the idea that the s-process occurs during the AGB phase of low-mass stars (e.g. see Busso et al 1995, Lambert et al 1995b), between the H and He burning shells with neutrons liberated by the 13C(, n)16O reaction.
Smith & Lambert (1990) showed that the observed s-process abundances in M, MS, and S stars indicate a mean neutron exposure of 0 ~ 0.3 at 30 keV. Because this is equal to the Solar System neutron exposure, it is consistent with AGB s-process nucleosynthesis as a major supply of the main component of the Solar System s-process elements. Recent observational information on the conditions of AGB s-process nucleosynthesis has come from neutron densities inferred from measurements of Rb and Zr isotopic abundances (Lambert et al 1995b).
Another exciting area of AGB nucleosynthesis research involves the study of presolar grains, embedded in meteorites (see Zinner 1996). Anomalous isotopic abundances of carbon and s-process elements in SiC grains indicate a carbon star origin (e.g. Anders & Zinner 1993). Boothroyd et al (1994), Wasserburg et al (1995) infer the presence of deep circulation currents in AGB stars from the 12C / 13C and 18O / 16O ratios in these grains.
The site of the r-process is still in debate, although SN have been suspected from the beginning (Burbidge et al 1957). The most popular model is due to Meyer et al (1992), who suggested that the r-process occurs in the hot high-entropy bubble surrounding the nascent neutron star during the SN explosion. In this region, the high photon-to-baryon ratio favors photodissociation, thus keeping the number of free neutrons high and the number of nuclei low. Once the material cools, the nuclei are exposed to a sea of neutrons, at neutron densities of ~ 1020 cm-3, which drives the r-process.