A Bitter Pill: The Primordial Lithium Problem Worsens

TL;DR

The paper revisits the primordial lithium problem in light of updated nuclear reaction rates, the precise baryon density from WMAP, and improved observations of metal-poor stars. Using the revised ${}^{3}{ m He}( abla o){^{7}{ m Be}}$ rate and the WMAP5 baryon density, the predicted $7Li/H$ rises to $5.24^{+0.71}_{-0.62} × 10^{-10}$, widening the gap with observed values near $1.2–2.3 × 10^{-10}$ and pushing the discrepancy to about $4.2$–$5.3σ$. The authors quantify the problem, discuss potential resolutions, and outline strategies spanning nuclear physics, stellar astrophysics, and new physics beyond the Standard Model. They conclude that nuclear-physics explanations are unlikely, while observational systematics and new physics scenarios remain plausible routes, with Planck, the LHC, and extragalactic Li measurements key to progress.

Abstract

The lithium problem arises from the significant discrepancy between the primordial 7Li abundance as predicted by BBN theory and the WMAP baryon density, and the pre-Galactic lithium abundance inferred from observations of metal-poor (Population II) stars. This problem has loomed for the past decade, with a persistent discrepancy of a factor of 2--3 in 7Li/H. Recent developments have sharpened all aspects of the Li problem. Namely: (1) BBN theory predictions have sharpened due to new nuclear data, particularly the uncertainty on 3He(alpha,gamma)7Be, has reduced to 7.4%, and with a central value shift of ~ +0.04 keV barn. (2) The WMAP 5-year data now yields a cosmic baryon density with an uncertainty reduced to 2.7%. (3) Observations of metal-poor stars have tested for systematic effects, and have reaped new lithium isotopic data. With these, we now find that the BBN+WMAP predicts 7Li/H = (5.24+0.71-0.67) 10^{-10}. The Li problem remains and indeed is exacerbated; the discrepancy is now a factor 2.4--4.3 or 4.2sigma (from globular cluster stars) to 5.3sigma (from halo field stars). Possible resolutions to the lithium problem are briefly reviewed, and key nuclear, particle, and astronomical measurements highlighted.

Figures (4)

• Figure 1: A comparison of the ${}^{7}{\rm Li}$/H abundances as a function of $\eta$ using the new ${}^{3}{\rm He}$($\alpha,\gamma$)${}^{7}{\rm Be}$ rate determined in cybdav (green band) to the older result from cyburt (red band). The thickness of the bands represents 1 $\sigma$ uncertainties in the calculated abundance. Central values are given by thick solid and dashed curves respectively. The thin upper curve (within the green shaded region) demarcates the border of the old result. The yellow vertical band is the WMAP value of $\eta$wmap.
• Figure 2: The fractional error $\sigma({\rm Li})/{\rm Li}$ in the BBN lithium prediction, shown as a function of $\eta$. The solid curve is our result; the broken curve gives the older result from cyburt as in Fig. \ref{['licomp']}. The yellow vertical band is the WMAP value of $\eta$wmap.
• Figure 3: The light element abundances of D, ${}^{3}{\rm He}$, ${}^{7}{\rm Li}$ by number with respect to H, and the mass fraction of ${}^{4}{\rm He}$ as a function of $\eta$. The thickness of the bands represents 1 $\sigma$ uncertainties in the calculated abundance. The yellow band gives the WMAP $\eta$wmap.
• Figure 4: The theoretical and observational likelihood functions for ${}^{4}{\rm He}$, D/H, ${}^{3}{\rm He}$/H, and ${}^{7}{\rm Li}$/H. BBN results have been convolved with the WMAP determination of $\eta$ and are shown as dark (blue) shaded area. The observational likelihoods are shown as light (yellow) shaded regions as well as alternative dotted curves. The data and distinctions are detailed in the text.