An improved cosmological bound on the thermal axion mass

TL;DR

The paper tightens cosmological bounds on thermal hadronic axions by incorporating a hot dark matter component from massive neutrinos and leveraging updated cosmological datasets. It models axion decoupling in the hadronic framework, connects $m_a$ to the decoupling temperature via $f_a$ and $g_{\star S}(T_D)$, and constrains the joint hot DM content with eight-parameter MCMC analyses. The main findings are $m_a<0.42$ eV (95% c.l.) without neutrino-mass priors and $m_a<0.35$ eV (95% c.l.) with full data, accompanied by $\sum m_{\nu}<0.20$ eV (or $<0.17$ eV with Lyman-\alpha), indicating only a few percent of CDM in hot DM. The results highlight an anti-correlation between $m_a$ and $\sum m_{\nu}$ and emphasize the potential for future cosmological and laboratory data to refine these limits.

Abstract

Relic thermal axions could play the role of an extra hot dark matter component in cosmological structure formation theories. By combining the most recent observational data we improve previous cosmological bounds on the axion mass m_a in the so-called hadronic axion window. We obtain a limit on the axion mass m_a < 0.42eV at the 95% c.l. (m_a < 0.72eV at the 99% c.l.). A novel aspect of the analysis presented here is the inclusion of massive neutrinos and how they may affect the bound on the axion mass. If neutrino masses belong to an inverted hierarchy scheme, for example, the above constraint is improved to m_a < 0.38eV at the 95% c.l. (m_a < 0.67eV at the 99% c.l.). Future data from experiments as CAST will provide a direct test of the cosmological bound.

Figures (2)

• Figure 1: Likelihood contour plot in the $\sum m_{\nu}-m_a$ plane showing the 68% and 95% c.l. from the conservative dataset (left panel) and from the complete dataset (right panel). Note different axes.
• Figure 2: $95\%$ CL limits on the axion mass obtained in the conservative and full analysis (shaded regions), assuming three possible values of the sum of the neutrino masses in the $m_a$-$g_{a \gamma \gamma}$ plane. From right to left the region represent the exclusion limits assuming a prior $\sum m_\nu>0$, $\sum m_\nu>0.05$ eV (N.H.) and $\sum m_\nu>0.1$ eV (I.H). As a comparison, we show the recent results from the CAST experiment (blue contour) as well as the theoretical KSVZ parameter region (within the green lines), following Fig. 8 from Ref. cast, and the CAST prospects (blue dashed line) castfuture.