The post-BBN evolution of 3He is much more complex than that of D. Indeed, when D is incorporated into a star it is rapidly burned to 3He, increasing the 3He abundance. The more tightly bound 3He, with a larger coulomb barrier, is more robust than D to nuclear burning. Nonetheless, in the hotter interiors of most stars 3He is burned to 4He and beyond. However, in the cooler, outer layers of most stars, and throughout most of the volume of the cooler, lower mass stars, 3He is preserved (Iben 1967, Rood, Steigman, & Tinsley 1976; Iben & Truran 1978, Dearborn, Schramm, & Steigman 1986; Dearborn, Steigman, & Tosi 1996). As a result, prestellar 3He is enhanced by the burning of prestellar D, and some, but not all, of this 3He survives further stellar processing. However, there's more to the story. As stars burn hydrogen to helium and beyond, some of their newly synthesized 3He will avoid further processing so that the cooler, lower mass stars should be significant post-BBN sources of 3He.
Aside from studies of meteorites and in samples of the lunar soil (Reeves et al. 1973, Geiss & Gloeckler 1998, Gloeckler & Geiss 2000), 3He is only observed via its hyperfine line (of singly-ionized 3He) in interstellar HII regions in the Galaxy. It is, therefore, unavoidable that models of stellar yields and Galactic chemical evolution are required in order to go from here and now (ISM) to there and then (BBN). It has been clear since the early work of Rood, Steigman and Tinsley (1976) that according to such models, 3He should have increased from the big bang and, indeed, since the formation of the solar system (see, e.g. Dearborn, Steigman & Tosi 1996 and further references therein). For an element whose abundance increases with stellar processing, there should also be a clear gradient in abundance with galactocentric distance. Neither of these expectations is borne out by the data (Rood, Bania & Wilson 1992; Balser et al. 1994, 1997, 1999; Bania, Rood & Balser 2002) which shows no increase from the time of the formation of the solar system, nor any gradient within the Galaxy. The most likely explanation is that before the low mass stars can return their newly processed 3He to the interstellar medium, it is mixed to the hotter interior and destroyed (Charbonnel 1995, Hogan 1995). Whatever the explanation, the data suggest that for 3He there is a delicate balance between production and destruction. As a result, the model-dependent uncertainties in extrapolating from the present data to the primordial abundances are large, limiting the value of 3He as a baryometer. For this reason I will not dwell further on 3He in these lectures; for further discussion and references, the interested reader is referred to the excellent review by Tosi (2000).