Search for a solar-bound axion halo using the Global Network of Optical Magnetometers for Exotic physics searches

Technical Summary: Search for a Solar-Bound Axion Halo using GNOME

Problem and Motivation
The nature of dark matter remains a central open question in astrophysics and particle physics. Ultralight bosons, specifically axions or axion-like particles (ALPs) with masses $m_a \ll 1$ eV, are well-motivated candidates. Recent theoretical work suggests that dense astrophysical bodies, such as the Sun, can gravitationally capture axions. If axions possess quartic self-interactions, these interactions can provide the necessary energy dissipation mechanism to allow axions to thermalize into the ground state of the Sun's gravitational potential well. This process forms a "solar axion halo," a coherent, ground-state configuration of the axion field centered on the Sun.

Unlike the standard galactic dark matter halo, which is virialized and stochastic, a solar axion halo in the ground state ($n=1, \ell=0$) would exhibit a distinct density profile ($\rho \propto e^{-2r/R_\star}$) and a coherent, monochromatic oscillation at the axion Compton frequency $\omega_a = m_a c^2 / \hbar$. The presence of such a halo would imply a significant overdensity of axions near the Earth compared to the average galactic dark matter density, potentially enhancing detection sensitivity. The axion field in this halo couples to atomic spins via gradient interactions (linear or quadratic), generating a pseudo-magnetic field ($B_p$) that oscillates with a specific daily modulation pattern due to the Earth's rotation and orbital motion relative to the Sun.

Methodology
This study utilizes data from the Global Network of Optical Magnetometers for Exotic physics searches (GNOME), specifically the fifth Science Run (SR5), which comprised 69 days of data from 12 geographically distributed, magnetically shielded atomic magnetometers.

1. Signal Model: The analysis assumes the axions are in the ground state of the solar gravitational potential. The resulting axion field $a(r,t)$ generates a pseudo-magnetic field $B_p$ through linear ($H_l \propto S \cdot \nabla a$) and quadratic ($H_q \propto S \cdot \nabla a^2$) gradient couplings to proton spins. The signal consists of two components: a radial component (due to the spatial gradient toward the Sun) and a transverse component (due to the "axion wind" from the Earth's orbital motion). The relative strength of these components depends on the axion mass (frequency).
2. Data Pre-processing: Raw time-series data from each station were filtered to remove power-line noise (50/60 Hz) and slow drifts (high-pass filter at 0.0003 Hz). Data were binned to a 64 Hz sampling rate. Hourly calibration pulses were used to correct for gain drifts.
3. Time-Shifted Cross-Correlation: To distinguish a global signal from local noise, the analysis employs a cross-correlation pipeline. Because the daily modulation of the signal amplitude varies for each station based on its geographic location and orientation, data from different stations are time-shifted ($\Delta t_j$) to align their predicted signal peaks before cross-correlation.
4. Spectral Analysis: Pairwise cross-correlations are computed in 20-minute windows. The resulting signals are converted to Power Spectral Densities (PSDs), normalized, and combined into a weighted network-level spectrum.
5. Candidate Selection and Fitting:
   • Stage 1: Frequencies where the normalized PSD exceeds a threshold ($P_\star = 3.35$) on every day of the run are identified as candidates.
   • Stage 2: For candidate frequencies, the daily modulation pattern of the combined PSD is fitted to the theoretical model derived from the axion halo simulation. A test statistic $Z$ compares the fit quality of the axion model against a constant (noise-only) model.
   • Significance: A candidate is considered significant only if the fitted amplitude is non-zero and the $p$-value of the test statistic $Z$ falls below a global significance threshold ($p_{crit}$) derived from the empirical distribution of $Z$ values.

Key Contributions

• First Dedicated Search: This work presents the first dedicated search for a gravitationally bound solar axion halo using a global network of magnetometers.
• Signal Modeling: The authors developed a comprehensive signal model for the pseudo-magnetic fields generated by both linear and quadratic axion-proton gradient couplings in a ground-state solar halo, explicitly accounting for the daily modulation and the interplay between radial and transverse field components across the frequency spectrum (0.05–20 Hz).
• Analysis Pipeline: The study implements a robust cross-correlation pipeline that utilizes time-shifting to align global signals across a worldwide network, effectively suppressing uncorrelated local noise while preserving the coherence of a solar-bound signal.

Results

• No Detection: The analysis of 69 days of data from 12 GNOME stations revealed no statistically significant candidate signals consistent with a solar axion halo. No observed $p$-values for the test statistic $Z$ fell below the critical threshold for global significance.
• Upper Limits: In the absence of a detection, the study sets 95% confidence-level upper limits on the amplitude of the axion-induced pseudo-magnetic field over the frequency range $\approx 0.05 - 20$ Hz.
• Coupling Constraints: These field limits were translated into constraints on the linear ($1/f_l$) and quadratic ($1/f_q$) axion-proton coupling constants. The constraints were derived for two benchmark halo density scenarios:
  1. The maximum overdensity allowed by planetary ephemerides.
  2. The density predicted by the quartic self-interaction gravitational capture model.
• Quadratic Coupling Sensitivity: For the quadratic coupling case, the derived limits surpass existing astrophysical bounds (e.g., from Supernova 1987A) by more than two orders of magnitude across much of the accessible parameter space. For linear couplings, the limits are competitive with or complementary to existing bounds, depending on the assumed halo density.

Significance and Claims
The paper claims to have established the first experimental constraints on the existence of a solar axion halo formed via gravitational capture and thermalization. The results demonstrate that GNOME is capable of probing exotic spin couplings with a global network approach, specifically targeting the unique daily modulation signature of a solar-bound field.

The authors modestly note that their constraints are conditional on the assumption that the axions are in the ground state and that the formation mechanism (quartic self-interaction) is efficient. They state that for quadratic couplings, their results represent a significant improvement over previous astrophysical limits. The work also highlights that the data could be reanalyzed for other exotic physics targets, such as topological defects or stochastic fields, and that future iterations of GNOME using alkali-noble-gas comagnetometers will offer enhanced sensitivity to both proton and neutron spins.