3.1. Big Bang Nucleosynthesis (BBN)

It is reassuring that the same baryon abundance _b h² = 0.0214 ± 0.0020 implied by the deuterium abundance D/H in low-metallicity Lyman limit systems in quasar spectra [21] agrees with that implied by the relative heights of the first two peaks in the WMAP angular power spectrum [4], giving _b h² = 0.0224 ± 0.0009. This is also in agreement with the baryon abundance deduced from clusters [22] and with the lower limit from the opacity of the Lyman alpha forest [23]. However, there are uncertainties in the measured deuterium abundance evolution [24], and potential problems - or perhaps clues to new physics - in discrepancies between the observed helium and lithium abundances and the predictions of BBN.

The abundance by mass of ⁴He measured in low-metallicity ionized regions in nearby galaxies implies an extrapolated primordial abundance Y_p = 0.2421 ± 0.0021 according to the latest published data [25], which corresponds according to standard BBN to a baryon abundance _b h² = 0.012^+0.003_-0.002, lower by about 3 than the value just mentioned from D/H and CMB measurements. It remains to be seen whether this could be remedied by improved analyses (for example, based on more realistic models of these low-metallicity galaxies or of their HII regions), or alternatively whether it is perhaps an indication of problems with standard BBN.

The D/H [21] _b in standard BBN implies a primordial ⁷Li/H 3.3 - 6.0 × 10^-10, in serious disagreement with the value of ⁷Li/H = 1.23^+0.34_-0.16 × 10^-10 measured in atmospheres of galactic halo metal-poor stars in the "Spite plateu" (i.e. with metallicity [Fe/H] less than about -2) [26]. It disagrees even with the puzzlingly higher value ⁷Li/H = 2.19^+0.30_-0.27 × 10^-10 from a sample of globular cluster stars [27]. It is possible that some of the ⁷Li in such stars is destroyed by astration, consistent with the small range of ⁷L abundance in the Spite plateau stars [28], but this may not resolve the discrepancy.

These disagreements call into question the usual assumptions of standard BBN, for example, the assumption of no significant electron neutrino asymmetry and N = 3 light neutrino species [29]. However, an alternative possibility that might neatly account for the ⁷Li discrepancy is the injection of energetic nucleons around 1000 s after the big bang, for example due to decay of the next-to-lightest supersymmetric partner particle into the lightest one [30].