Hiding relativistic degrees of freedom in the early universe

TL;DR

The paper analyzes whether extra relativistic energy density in the early universe can be concealed by a nonzero electron neutrino degeneracy $\xi_e$ without conflicting with Big Bang Nucleosynthesis (BBN) and Wilkinson Microwave Anisotropy Probe (WMAP) data. By connecting $\xi_e$, the effective number of neutrino species $\Delta N_ν$, and the baryon density $\eta_{10}$ through BBN and CMB likelihoods, the authors show that $|\xi_e|$ must be small ($\lesssim 0.1$) in the standard case ($ΔN_ν=0$), but larger $ΔN_ν$ can be accommodated if compensated by $\xi_e$ and other relativistic components. The joint analysis finds a best-fit near $(ΔN_ν,\xi_e) \approx (0.22, 0.057)$, with $ΔN_ν=1$ allowed for $ξ_e\sim 0.1$ and $ΔN_ν=3$ possible at $ξ_e\sim 0.2$; marginalization broadens the 2σ range to roughly $-1.7 \lesssim ΔN_ν \lesssim 4.1$. These results imply that nonstandard relativistic sectors or delayed sterile neutrino thermalization (as motivated by LSND) can be cosmologically consistent, though the required compensation is model-dependent and finely balanced with observational constraints on $\eta$ from CMB data.

Abstract

We quantify the extent to which extra relativistic energy density can be concealed by a neutrino asymmetry without conflicting with the baryon asymmetry measured by the Wilkinson Microwave Anisotropy Probe (WMAP). In the presence of a large electron neutrino asymmetry, slightly more than seven effective neutrinos are allowed by Big Bang Nucleosynthesis (BBN) and WMAP at 2σ. The same electron neutrino degeneracy that reconciles the BBN prediction for the primordial helium abundance with the observationally inferred value also reconciles the LSND neutrino with BBN by suppressing its thermalization prior to BBN.

Figures (5)

• Figure 1: Isoabundance curves for D and $^4$He in the $\xi_{e}$ -- $\eta_{10}$ plane for $N_{\nu}$ = 3. The nearly horizontal curves are for $^4$He (from top to bottom: Y = 0.23, 0.24, 0.25). The nearly vertical curves are for D (from left to right: $y_{\rm D}$$\equiv 10^{5}$(D/H) = 3.0, 2.5, 2.0). The data point with error bars corresponds to $y_{\rm D}$ = $2.6\pm 0.4$ and Y = $0.238\pm 0.005$.
• Figure 2: The $1\sigma$ and $2\sigma$ contours in the $\eta_{10} - \xi_e$ plane for $N_{\nu}$ = 3 from a joint CBR -- BBN fit using WMAP data and the adopted D and $^4$He abundances. The cross marks the best-fit point $(\eta_{10},\xi_e)=(6.16,0.044)$.
• Figure 3: The approximate $\xi_{e}$ -- $\Delta N_\nu\;$ relation corresponding to $y_{\rm D}$ = 2.6 and Y = 0.238. The numbers shown are the corresponding values of $\eta_{10}$ at those points.
• Figure 4: Same as Fig. \ref{['f2']}, except $N_{\nu}$ = 4. The cross marks the best-fit point $(\eta_{10}, \xi_e)=(6.22, 0.096)$.
• Figure 5: The $1\sigma$ and $2\sigma$ contours in the $\Delta N_\nu\;$ -- $\xi_{e}$ plane using WMAP data and BBN with the adopted D and $^4$He abundances. The cross marks the best-fit point $(\Delta N_\nu, \xi_e) = (0.22, 0.057)$.