An Explanation for the Slopes of Stellar Cusps in Galaxy Spheroids

The stellar surface mass density profiles at the centers of typical ~L* and lower-mass spheroids exhibit power law ‘cusps’ with $\Sigma \propto R^(-n)$, where 0.5<n<1 for radii ~1-100 pc. Observations and theory support models in which these cusps are formed by dissipative gas inflows and nuclear starbursts in gas-rich mergers. At these comparatively large radii, stellar relaxation is unlikely to account for or strongly modify the cuspy stellar profiles. We argue that the power-law surface density profiles observed are a natural consequence of the gravitational instabilities that dominate angular momentum transport in the gravitational potential of a central massive black hole. The dominant mode at these radii is an m=1 lopsided/eccentric disk instability, in which stars torquing the gas can drive rapid inflow and accretion. Such a mode first generically appears at large radii and propagates inwards by exciting eccentricities at smaller and smaller radii, where M*(<R)«M_BH. When the stellar surface density profile is comparatively shallow with n<1/2, the modes cannot efficiently propagate to R=0 and so gas piles up and star formation steepens the profile. But if the profile is steeper than n=1, the inwards propagation of eccentricity is strongly damped, suppressing inflow and bringing n down again. Together these results produce an equilibrium slope of 1/2 < n < 1 in the potential of the central black hole. These physical arguments are supported by nonlinear numerical simulations of gas inflow in galactic nuclei. Together, these results naturally explain the observed stellar density profiles of ‘cusp’ elliptical galaxies.

💡 Research Summary

The paper addresses a long‑standing puzzle in galactic astronomy: why the central stellar surface‑mass density of typical L* and lower‑mass spheroids follows a shallow power‑law cusp, Σ(R) ∝ R⁻ⁿ, with an exponent 0.5 < n < 1 over radii of roughly 1–100 pc. Earlier work has linked these cusps to dissipative gas inflows and nuclear starbursts triggered during gas‑rich mergers, but the mechanisms that maintain such a specific slope in the regime where two‑body relaxation is ineffective have remained unclear.

The authors propose that the dominant agent shaping the cusp is the m = 1 lopsided (eccentric) disk instability that operates in the gravitational potential of the central massive black hole (BH). In the region where the BH mass M_BH far exceeds the enclosed stellar mass M_*(<R), circular orbits are no longer the lowest‑energy configuration; instead, a coherent eccentric mode can grow. Gas and stars in the disk exchange angular momentum through gravitational torques: the eccentric stellar component torques the gas, removing its angular momentum and driving rapid inflow toward smaller radii, while the gas exerts a back‑reaction that excites eccentricities in the stellar orbits. This “eccentricity wave” propagates inward, and its ability to reach ever smaller radii depends sensitively on the local stellar surface density profile.

A simple analytic argument shows three regimes. If the surface density is very shallow (n < ½), the m = 1 mode cannot propagate efficiently to the centre; gas stalls at a characteristic radius, builds up, and forms stars, steepening the profile. If the profile is too steep (n > 1), the strong central potential damps the eccentric mode quickly, suppressing inflow and causing the profile to flatten. Between these limits the system reaches a self‑regulating equilibrium: the mode propagates far enough to sustain inflow, but not so far as to be completely damped, yielding a quasi‑steady slope ½ < n < 1.

To test this picture, the authors run high‑resolution three‑dimensional hydrodynamic simulations that include self‑gravity, star formation, and feedback, as well as a live stellar component that can develop the eccentric instability. Starting from gas‑rich nuclear disks, the simulations quickly develop a dominant m = 1 pattern. The pattern drives gas inward at rates of several solar masses per year, and the resulting star formation produces a stellar cusp with n ≈ 0.6–0.8, precisely within the observed range. Varying the initial gas fraction, BH mass, and disk thickness changes the transient behaviour but the final cusp slope remains robust, confirming the analytic prediction of an attractor solution.

The study therefore unifies several observational facts: (1) the ubiquity of shallow cusps in low‑mass ellipticals, (2) the lack of significant two‑body relaxation at the relevant radii, and (3) the correlation between cusp strength and evidence for past gas‑rich mergers. By identifying the m = 1 eccentric disk as the key angular‑momentum transport mechanism in the BH‑dominated potential, the authors provide a natural explanation for why the cusp slope settles in the narrow interval 0.5 < n < 1.

Beyond explaining the stellar density profile, the work has broader implications. The same eccentric mode can also drive gas onto the BH itself, potentially linking cusp formation to episodes of active galactic nucleus (AGN) activity. Moreover, the self‑regulating nature of the process suggests that nuclear cusps are a generic outcome of any gas‑rich inflow into a BH‑dominated nucleus, regardless of the detailed merger history. This insight opens new avenues for connecting the small‑scale physics of black‑hole feeding with the larger‑scale structural evolution of early‑type galaxies.