Observational bounds on Dark Matter Admixed Neutron Stars from Gravitational Wave Data

TL;DR

This paper addresses whether a fraction of dark matter may reside inside neutron stars and how this admixture would imprint on gravitational-wave signals from NS-BH mergers. It develops a two-fluid Tolman–Oppenheimer–Volkoff framework for bosonic DM with quartic self-interaction, and conducts a Bayesian analysis that directly samples DM parameters $(m_\nChi, \lambda_\nChi, \epsilon^c_{\nBM}, \epsilon^c_{\nDM}, F_\nChi)$, computing stellar structure and tidal deformability $\Lambda$ via the two-fluid solutions. By reanalyzing GW events GW230529, GW200115, GW200105, and GW190814 with realistic baryonic EoSs (SLy4, APR4, MPA1), the study obtains upper bounds on the DM fraction and particle mass that depend on the specific system and EoS; notably, GW200105 yields $F_\nChi < 6.1\%$ (SLy4) with $m_\nChi > 435$ MeV, while GW190814 favors a DM-halo interpretation under soft EoSs but can shift toward a DM-core scenario with a stiff EoS. The findings demonstrate the feasibility of GW data to constrain DM in compact objects and motivate higher-SNR observations to refine these constraints and disentangle DM configurations.

Abstract

Recent gravitational-wave (GW) observations offer a unique opportunity to probe the fundamental nature of compact objects. A growing body of research has focused on exploring the role of dark matter (DM) through the concept of DM-admixed neutron stars (NSs), where the presence of DM can significantly alter key physical properties of NSs, such as their mass, radius, and tidal deformability, ultimately affecting the predicted GW waveform emitted during binary coalescences. In this work, we present a novel observational test that, for the first time, places constraints on the influence of DM inside NSs using real GW data. By reanalyzing signals from events such as GW230529, GW200115, and GW200105, we derive new upper bounds on the DM fraction, $F_χ$, and particle mass, $m_χ$, under the assumption that DM is described by a scalar field with a self-interaction potential. We find that the upper bound on $F_χ$ depends on the specific binary system under analysis, indicating that different DM configurations can be consistent with observations in different ways. In particular, the event GW190814 may be compatible with a DM halo configuration. In contrast, the other events analyzed (GW230529, GW200105 and GW200115) are consistent with DM forming a core inside the NS, yielding strong upper bounds on $F_χ$. The corresponding values for the mass scale $m_χ$ are also discussed in the text. This work offers a new approach to probing DM in the context of compact NS objects through GW observations.

Figures (8)

• Figure 1: Mass–radius relations for DM-admixed configurations corresponding to cases (i) and (ii) discussed in Section \ref{['theory']}. Left panel: scalar field parameters set to $m_\chi = 400~\mathrm{MeV}$ and $\lambda_\chi = \pi$. Right panel: scalar field parameters set to $m_\chi = 120~\mathrm{MeV}$ and $\lambda_\chi = \pi$.
• Figure 2: Simulated polarization signals during the inspiral phase are shown for two scenarios: a neutron star containing dark matter (DM curve) and one composed solely of baryonic matter (BM curve). The waveforms were computed using the IMRPhenomNSBH model, assuming fixed values of $F_\chi = 50\%$, $m_\chi \approx 220~\mathrm{MeV}$ and $\lambda_\chi = \pi$. All remaining intrinsic and extrinsic parameters were set according to those inferred for the GW230529 event.
• Figure 3: Flowchart illustrating the statistical strategy employed in this work to integrate the two-fluid TOV formalism with the EoS parameters, enabling the generation of waveforms required for a robust analysis of gravitational wave event signals. For a parameter $\theta$, we write its posterior as $p(\theta|d)$, where $d$ is the data. This approach ensures that the relevant physical parameters associated with the dark matter sector are properly incorporated into the sampling procedure.
• Figure 4: Left panel: Marginalized posterior distributions and 68% and 95% CL contours for source chirp mass and DM parameters $m_\chi$, $\lambda_\chi$ and $F_\chi$ for merger event GW200115 Abbott_2021_NSBH considering the BM EoSs SLy4 Chabanat:1997un and APR4 Akmal:1998cfRight panel: Same as in left panel, but for event GW200105 Abbott_2021_NSBH
• Figure 5: Same as Fig. \ref{['fig:GW20']}, but for merger event GW230529 Abac_2024.