Stellar parents: how many and what kind?

14 Thanks to astronomical observations, we know that most dust in the Galaxy (constituting about 1% of the mass of a typical giant molecular cloud, or 10³-10⁴ M_Q per cloud, where MO is the Sun's mass) is produced by red giant stars—specifically, red giants in the thermally pulsing asymptotic giant branch (AGB) phase of their evolution.

15 In the early AGB phase, the outer envelopes of these stars are rich in oxygen. Some of this oxygen ends up in silicate and oxide dust, which condenses into material that flows from the outer envelopes into the interstellar me­dium. However, as the stars age, the envelopes of the low-mass (M ≤ 4 MO) AGB stars become rich in carbon due to the convective dredge-up of material produced by nuclear reactions in the hot interior. Once the envelopes' carbon-to-oxygen ratio exceeds unity, carbonaceous dust like SiC can condense.

16 Several lines of evidence confirm that most presolar SiC comes from low-mass AGB stars. First, such stars are believed to be the sites of s-process nucleosynthesis. This is consistent with measurements of the minor ele­ments barium, strontium, neodymium and samarium in ensembles of presolar SiC grains that show characteristic enrichments in s-process isotopes. (See box 1 and figure 2.) The evidence for s-process nucleosynthesis has been nicely extended by recent analyses of molybdenum and zirconium isotopes in individual presolar SiC grains at Argonne by Gunther Nicolussi, Andrew Davis and Roy Lewis, who used resonant ionization mass spectrometry to study these elements separately.

17 But perhaps the most compelling evidence for the AGB origin of most grains is the close match between two distributions of ¹²C / ¹³C ratios—namely, that of presolar SiC grains and that measured remotely in carbon stars today. The similarity of these distributions, which are shown in figure 3, further emphasizes that the SiC grains are not simply from a single presolar star. Rather, they constitute a sample from at least several tens of AGB stars that injected this matter into the molecular cloud from which our Solar System formed. Input from many AGB stars is also implied by the compositional trends of Si isotopes revealed by ion microprobe analyses of hun­dreds of presolar SiC grains.(Figure 4).

18 Detailed analysis of the oxygen isotopic compositions of presolar corundum (A1₂O₃) grains indicates that most of these grains also originated in red giant stars. Oxides pose a formidable experimental problem because, unlike either graphite or SiC grains, most oxide grains were formed within the Solar System and therefore have a normal isotopic composition. The first task of the experi­menter is therefore to find the rare presolar oxide grains in a sea of isotopically normal material.

19 The oxygen isotopes in corundum grains are consis­tent with those predicted to result from the first dredge-up. (See figure 5.) At this stage of its evolution, the star becomes a red giant, and deep convection mixes the ashes of main sequence nucleosynthesis into its outer envelope. As shown in figure 5, the variation in the isotopic com­position of oxygen among these grains indicates that they formed in at least several tens of red giant parents, each having a different initial mass and metallicity. (Metallicity is the mass fraction of atoms of elements heavier than and including carbon, which astronomers call metals.)

20 An interesting outgrowth of these studies is the realization that the oxygen compositions can be used to determine the age of the Galaxy. The basic idea, as devised by Nittler and Ramanath Cowsik  is that one can estimate the time required for the average metallicity of the Galaxy to evolve to the stage of that inferred for the parent stars, as well as use stellar evolution models to determine the life span of parent stars having the inferred masses. Since grains are expelled from their parent stars at the end of the stars' lives, the Galaxy's age is given by summing the known age of the Solar System (4.6 Ga), the stellar life span and the metallicity evolution time. The Galactic age derived in this way (14-15 Ga) is nicely consistent with

the ages that have been obtained from various astronomi­cal studies (for example, of globular clusters). Inadequa­cies in the theoretical models make the age uncertain by several billion years, but the reliability of the estimate will improve if better models of stellar and Galactic chemi­cal evolution can be constructed.

21 Presolar grains provide direct evidence that superno­vae also contributed to the solar mix. About 1% of the presolar SiC grains were originally given the noncommit­tal name X grains because of their extremely exotic silicon isotopes—specifically, an excess of ²⁸Si, which is produced when oxygen burns deep in the interior of massive stars. One way to remove this silicon to where it can form grains is by blasting the star apart in a supernova.  Large excesses of ⁴⁴Ca have been detected in X grains. Some graphite grains also appear to have been produced in supernova explosions.

The X grains provide unique insights into the hydro-dynamical behavior of exploding stars. Isotopic measure­ments of different elements in single grains show that nuclear products from the deep stellar interior (for exam­ple, ²⁸Si and ⁴⁴Ca) must have been joined with carbon-rich material from outer stellar layers, reflecting extensive mixing at all length scales

22 Various isotopically distinct subgroups of presolar grains have been found in addition to those considered above. The result is what some refer to as the alphabet soup of grain studies—for example, SiC types X, Y, Z, A, B; group 4 oxide grains and so on. Indeed, there may be almost as many subtypes of grains as there are subtypes of stars in the Galaxy. Relating the various grain subtypes to specific kinds of stellar objects such as novae and the very hot, very luminous Wolf—Rayet stars is a continuing enterprise that should deepen the understanding of a wide variety of astronomical objects.