Black-Hole Instabilities and New Phases

An AdS black object can fail as a phase in several inequivalent ways: its thermodynamic Hessian can be indefinite, a linear perturbation can grow, a superradiant mode can be trapped, a charged scalar can condense, or a localization mode can bifurcate toward another geometry. A zero mode identifies an onset; it does not determine the nonlinear endpoint or prove that the new branch dominates.

Required background. AdS Black Branes and Holographic Thermodynamics fixes the neutral phase. Rotating and Charged AdS Black Holes supplies chemical potentials and superradiant thresholds.

Helpful background. Rotating and Charged Horizons, Superradiance, and Chemical Potentials treats horizon physics. Metastability, Spinodals, and Thermal Decay distinguishes local instability from decay.

First application. Locate a charged-scalar zero mode around an AdS black hole and identify the branch leading toward a holographic superconductor.

Four stability questions

For a stationary saddle, ask separately:

1. Thermodynamic: is the Hessian of the correct ensemble potential positive in allowed fluctuations?
2. Euclidean: does the gauge-fixed quadratic action have negative modes on the chosen integration cycle?
3. Lorentzian: do normalizable perturbations obeying physical boundary conditions have $\operatorname{Im}\omega>0$?
4. Global: is there another saddle with lower thermodynamic potential?

Relations among these tests require hypotheses. Translationally invariant black branes motivate correlated-stability results, but compact horizons, gauge constraints, superradiance, and additional conserved charges can break a naive equivalence Reall 2001, §§2–5.

Charged-scalar zero mode

Consider a scalar of mass $m$ and charge $q$ on a charged AdS black hole. At linear order,

$$\left(D_aD^a-m^2\right)\psi=0, \qquad D_a=\nabla_a-iqA_a.$$

At the onset of a static homogeneous instability, solve this equation with regular horizon data and a normalizable, source-free boundary falloff. Only discrete values of temperature or chemical potential satisfy both conditions. The first such zero mode is a candidate bifurcation point to a hairy branch.

Near an extremal horizon, the electric field lowers the effective mass in the AdS$_2$ throat. Violation of the throat’s effective Breitenlohner–Freedman bound is a useful sufficient signal in many models, but the full asymptotic boundary-value problem decides the actual instability. Gubser’s mechanism Gubser 2008 and the explicit holographic-superconductor construction Hartnoll, Herzog, and Horowitz 2008 exemplify this calculation.

To determine the new phase, continue beyond the zero mode, solve the backreacted nonlinear equations, renormalize both saddles in the same ensemble, and compare free energies. The condensate’s existence does not establish thermodynamic dominance.

Superradiance and localization

A rotating or charged horizon can amplify a bosonic mode when

$$0<\omega<m\Omega_H+q\Phi_H.$$

Reflecting AdS boundary conditions can trap the amplified wave and produce exponential growth if a compatible normal mode exists. The endpoint may be a hairy black hole, a time-dependent state, or more complicated behavior; the inequality alone does not select it.

Extended horizons or internal compact directions can develop Gregory–Laflamme-type localization modes. A static threshold mode again marks a branch point. One must distinguish localization on an internal space, boundary spatial modulation, and thermodynamic phase separation because their conserved quantities and order parameters differ.

Adversarial controls

Hessian without spectrum. A positive heat capacity does not exclude a charged scalar or superradiant instability. Solve all relevant fluctuation channels.

Near-horizon test without UV completion. An AdS$_2$ bound can signal an infrared tendency, but a mode that cannot satisfy the asymptotic source condition is not a physical zero mode of the full saddle.

Zero mode as endpoint. Linear theory fixes the onset and eigenfunction only. If backreaction drives a first-order transition or a runaway, the infinitesimal branch does not give the final state.

Evidence ceiling

A converged normalizable zero mode or unstable quasinormal frequency establishes linear instability of a specified saddle under specified boundary conditions. A backreacted branch and lower ensemble potential can establish a competing thermodynamic phase. Neither result alone identifies the nonlinear time-dependent endpoint or proves stability against every channel.

The chapter overview contains the structure diagram and validity and failure diagram. They are embedded there once so that their shared chapter-level context is not repeated on every article.

For the chapter-wide comparison of assumptions, counterevidence, falsifiers, and claim ceilings, see the claim-domain table.

References

• Gubser, Steven S. “Breaking an Abelian Gauge Symmetry Near a Black Hole Horizon.” Physical Review D 78, 065034 (2008). DOI; arXiv:0801.2977.
• Hartnoll, Sean A., Christopher P. Herzog, and Gary T. Horowitz. “Building a Holographic Superconductor.” Physical Review Letters 101, 031601 (2008). DOI; arXiv:0803.3295.
• Reall, Harvey S. “Classical and Thermodynamic Stability of Black Branes.” Physical Review D 64, 044005 (2001). DOI; arXiv:hep-th/0104071.