[论文解读] Self-interacting dark matter solves the final parsec problem of supermassive black hole mergers

Gonzalo Alonso-Álvarez, James M. Cline+1 more|arXiv (Cornell University)|Jan 25, 2024

Pulsars and Gravitational Waves Research被引用 5

一句话总结

本文认为，围绕合并的超大质量黑洞的自相互作用暗物质尖峰提供动力摩擦，可以解决最终 parsec 问题并软化引力波谱，与 PTA 数据一致。

ABSTRACT

Evidence for a stochastic gravitational wave (GW) background, plausibly originating from the merger of supermassive black holes (SMBHs), is accumulating with observations from pulsar timing arrays. An outstanding question is how inspiraling SMBHs get past the "final parsec" of separation, where they have a tendency to stall before GW emission alone can make the binary coalesce. We argue that dynamical friction from the dark matter (DM) spike surrounding the black holes is sufficient to resolve this puzzle, if the DM has a self-interaction cross section of order $\,$cm$^2$/g. The same effect leads to a softening of the GW spectrum at low frequencies as suggested by the current data. For collisionless cold DM, the friction deposits so much energy that the spike is disrupted and cannot bridge the final parsec, while for self-interacting DM, the isothermal core of the halo can act as a reservoir for the energy liberated from the SMBH orbits. A realistic velocity dependence, such as generated by the exchange of a massive mediator like a dark photon, is favored to give a good fit to the GW spectrum while providing a large enough core. A similar velocity dependence has been advocated for solving the small-scale structure problems of cold DM.

研究动机与目标

• Motivate the final parsec problem in SMBH binaries and its tension with GW-only hardening.
• Propose self-interacting dark matter (SIDM) with velocity-dependent cross sections as a mechanism to sustain dynamical friction.
• Model DM density profiles around SMBHs including SIDM-induced isothermal cores and spikes.
• Quantify the impact of SIDM on the SMBH merger timescale and GW spectrum.
• Test SIDM scenarios against PTA data to assess fit quality and parameter ranges.

提出的方法

• Model DM halos around SMBHs with NFW outer profiles and spikes for CDM and SIDM cases.
• In SIDM, replace the inner region with an isothermal core and determine core size from scattering rates and age.
• Compute dynamical-friction powered hardening by equating P_df and P_gw and track R(t) and the corresponding t_df.
• Derive the GW energy spectrum dE_gw/df_s incorporating both DF and GW losses.
• Relate the SMBH merger population to galaxy merger rates and calibrate the model normalization to PTA data.

实验结果

研究问题

• RQ1 Can SIDM with velocity-dependent cross sections provide sufficient dynamical friction from the DM spike to shrink SMBH orbits from ~10 pc to the GW-dominated regime within a Gyr?
• RQ2 Which SIDM parameter ranges (sigma/m, velocity dependence, mediator scenario) best reconcile final parsec solving capability with PTA-observed GW spectrum shapes?
• RQ3 How does DM back-reaction affect spike survival, and can an isothermal SIDM core replenish the spike to sustain hardening?
• RQ4 Does the inclusion of SIDM produced softening of the GW spectrum improve fits to NANOGrav, PPTA, and EPTA data compared to CDM-only models?
• RQ5 What are the implications of SIDM for small-scale structure constraints while explaining PTA signals?

主要发现

• SIDM with cross sections around 0.5–5 cm^2/g can yield sub-Gyr inspiral times for SMBH binaries by preserving the DM spike through rethermalization.
• Collisionless CDM spikes are disrupted by the energy deposited by the SMBH binary and cannot bridge the final parsec.
• A massive mediator (velocity-dependent cross section) can create an isothermal core plus a spike, providing both a reservoir for energy and effective DF.
• The model with a massive mediator and appropriate velocity dependence (v_t ~ 300–1500 km/s) and sigma_0/m ~ 0.5–5 cm^2/g provides good fits to PTA spectra, reducing the required merger normalization psi_0.
• The resulting GW spectrum is softened at low frequencies due to energy loss from dynamical friction, consistent with hints in current PTA data.
• Best-fit chi-squared values are reported as chi^2_min ≲ 15 for SIDM scenarios, indicating improved fits relative to some fiducial PTA analyses.