The persistent discrepancy between the predicted primordial lithium-7 abundance from standard Big Bang nucleosynthesis (SBBN) and observations of metal-poor halo stars has remained unresolved for over four decades, currently exceeding 10 sigma. A resolution has been derived from M-theory compactified on a G2-holonomy manifold with Betti numbers (b₂, b₃) = (27, 451). Spectral zeta function regularisation of the vacuum energy on the compact manifold yields a one-loop correction factor Delta = 2*gammaEM + 1 = 2. 154, where gammaEM denotes the Euler-Mascheroni constant. This topological invariant enhances the effective nuclear cross-section for all Be-7 destruction channels through wavefunction renormalisation at the interaction vertex, yielding an enhancement factor of 3. 154. The detailed balance theorem guarantees that reactions in chemical equilibrium (deuterium, helium-4) remain unaffected, while the marginally frozen Be-7 network is selectively suppressed. The predicted lithium-to-hydrogen ratio (Li/H) QGU = (1. 48 +/- 0. 35) x 10^-10 agrees with the Spite plateau observations (1. 58 +/- 0. 31) x 10^-10 at the 0. 3 sigma level, reducing the tension from 10 sigma to below 0. 3 sigma. Bayesian model comparison yields a Bayes factor of 10²0 in favour of the geometric framework over SBBN. No adjustable parameters or hypothetical particle species have been introduced.
Moustafa Amin Radwan (2026) studied this question.