Solar Axions Cannot Explain the XENON1T Excess

We argue that the interpretation in terms of solar axions of the recent XENON1T excess is not tenable when confronted with astrophysical observations of stellar evolution. We discuss the reasons why the emission of a flux of solar axions sufficiently intense to explain the anomalous data would radically alter the distribution of certain type of stars in the color-magnitude diagram in the first place and would also clash with a certain number of other astrophysical observables. Quantitatively, the significance of the discrepancy ranges from $3.3\sigma$ for the rate of period change of pulsating white dwarfs and exceeds $19\sigma$ for the $R$ parameter and for $M_{I,\mathrm{TRGB}}$.

Figure 1

Contours of the axion energy-loss rates per unit mass, in $\mathrm{erg}{\mathrm{g}}^{-1}{\mathrm{s}}^{-1}$, for a pure He plasma and $g_{e13}=4.3$.

Figure 2

XENON1T 90% C.L. fit (blue region). $3\sigma$ exclusion limit from solar data (gray hatched region). $2\sigma$ LUX limit (gray dashed line) and CAST limits for $m_a<20\mathrm{meV}$ and $m_a<0.7\mathrm{eV}$ (green lines). Individual $2\sigma$ limits from $R$ parameter, TRGB, WDLF, WDVs (gray lines), and $2\sigma$ global bound from astrophysics (red region).