The Attractor Mechanism

The attractor mechanism explains why a regular extremal black hole can have horizon scalars and entropy fixed by conserved charges even though the same scalars are freely chosen at infinity. The mechanism is a radial dynamical flow, not a claim that all scalar fields everywhere lose their moduli dependence. We derive its effective potential, solve a two-charge example, and identify basin and small-black-hole failures.

Required background. Gauge Theory and String-Theory Preparation supplies electric–magnetic charges and scalar couplings. Gravitational Actions and Boundary Terms supplies the reduced variational principle and the asymptotic definition of mass.

Helpful background. Quantum Gravity supplies the effective-theory regime, Quantum Information helps distinguish protected data from generic degeneracy, and Black-Hole Geometry and Mechanics supplies extremal throats and horizon entropy.

The black-hole potential

Consider static, spherically symmetric, asymptotically flat solutions of four-dimensional ungauged $\mathcal N=2$ supergravity at two derivatives. Write $Z_E$ for the energy-normalized superalgebra charge and $Z_{\rm geo}=G_4Z_E$ for the corresponding length entering the metric. For vector-multiplet scalars $z^i$ and symplectic charge $\Gamma=(p^I,q_I)$, eliminating the electric and magnetic fields gives the positive, length-squared black-hole potential

$$V_{\rm BH}(z,\bar z;\Gamma) =|Z_{\rm geo}|^2+g^{i\bar j}D_iZ_{\rm geo}\,\overline{D_jZ_{\rm geo}},$$

where the conventional special-geometry expression for $Z_E$ is $e^{K/2}(q_IX^I-p^IF_I)$ and $D_iZ_{\rm geo}=(\partial_i+\tfrac12\partial_iK)Z_{\rm geo}$. In a radial coordinate $\tau$ increasing toward the horizon, the one-dimensional equations include

$$\ddot U=e^{2U}V_{\rm BH}, \qquad \ddot z^i+\Gamma^i{}_{jk}\dot z^j\dot z^k =e^{2U}g^{i\bar j}\partial_{\bar j}V_{\rm BH}.$$

The Hamiltonian constraint sets the non-extremality constant to zero. These equations are the second-order setting in which the first-order BPS flow lives. Their inputs are the charge sector, scalar metric, gauge-coupling matrix, asymptotic moduli, and a choice of radial orientation.

Fixed points and entropy

A regular extremal throat has constant scalars at its endpoint, so

$$\partial_iV_{\rm BH}(z_*,\bar z_*;\Gamma)=0.$$

For a BPS critical point, $D_iZ_{\rm geo}(z_*)=0$. The effective potential then reduces to $V_{\rm BH,*}=|Z_{{\rm geo},*}|^2$, and the near-horizon geometry is $\mathrm{AdS}_2\times S^2$ with squared radius $|Z_{{\rm geo},*}|^2$ in the standard normalization. The area and entropy are therefore

$$A_H=4\pi |Z_{{\rm geo},*}|^2, \qquad S_{\rm BH}=\frac{A_H}{4G_4} =\frac{\pi |Z_{{\rm geo},*}|^2}{G_4}.$$

Ferrara and Kallosh state the minimal-central-charge principle and derive $D_iZ=0$ and the charge-controlled area in Ferrara and Kallosh 1996, §1, pp. 2–4, Eqs. (1)–(17). Strominger gives the symplectic electric–magnetic stabilization equations in Strominger 1996, pp. 2–4, Eqs. (4)–(10).

The phrase “horizon values are independent of asymptotic moduli” means that every regular flow in the same basin reaches the same isolated $z_*(\Gamma)$. The ADM mass remains

$$M_{\rm ADM}=|Z_E(z_\infty,\bar z_\infty;\Gamma)| =\frac{|Z_{\rm geo}(z_\infty,\bar z_\infty;\Gamma)|}{G_4},$$

so it generally does depend on $z_\infty$. The entire radial profile also remembers the boundary data.

A two-charge dilaton attractor

Let a real scalar $\phi$ couple to magnetic and electric charges through

$$V_{\rm BH}(\phi)=\frac14\left(p^2e^{-2\phi}+q^2e^{2\phi}\right).$$

Extremizing gives

$$\frac{dV_{\rm BH}}{d\phi} =\frac12\left(-p^2e^{-2\phi}+q^2e^{2\phi}\right)=0, \qquad e^{2\phi_*}=\left|\frac{p}{q}\right|.$$

Substitution yields $V_{\rm BH,*}=|pq|/2$ with this choice of the overall normalization. The Hessian,

$$\left.\frac{d^2V_{\rm BH}}{d\phi^2}\right|_{\phi_*} =2|pq|=4V_{\rm BH,*}\gt0,$$

shows that the fixed point is locally attractive when both charges are nonzero. If $p=0$ or $q=0$, the critical point runs to infinite coupling or zero coupling and the two-derivative area vanishes. That is a small-black-hole limit; higher-derivative physics may create a finite corrected entropy, but the classical derivation no longer controls it.

Basins, flat directions, and non-BPS extrema

An attractor equation can have several critical points for the same charges. Singular divisors in moduli space can separate their basins, and an asymptotic choice outside a basin may generate a singular flow rather than the advertised horizon. A local Hessian check is therefore necessary but not sufficient: one must integrate the flow and verify that the scalar metric, warp factor, and curvature remain regular.

Not every extremum is BPS. Non-BPS extremal attractors satisfy $\partial_iV_{\rm BH}=0$ while some $D_iZ$ remain nonzero. They can still have charge-controlled entropy. Conversely, an extremum can have flat directions. Along them, some horizon scalars remain undetermined, yet $V_{\rm BH,*}$ and the entropy can be constant. Thus “all moduli are fixed” is stronger than the entropy-attractor statement.

Hypermultiplets provide a standard boundary: in ungauged two-derivative $\mathcal N=2$ supergravity they do not enter the central charge and are not generally attracted by this vector-multiplet argument. Gauging, rotation, higher derivatives, multicenter structure, and quantum corrections each modify the equations or the object being extremized.

Independent checks

For a proposed attractor, check the symplectic invariance and dimensions of $V_{\rm BH}$; solve both the stationarity equations and the Hamiltonian constraint; verify a positive finite area and a regular throat; and perturb the asymptotic moduli numerically to confirm convergence to the same endpoint within one basin. A second useful check is to compute the entropy directly from the horizon geometry and compare it with $\pi V_{\rm BH,*}/G_4$.

The attractor mechanism alone does not supply a microscopic count. It predicts a macroscopic charge function to be compared with a protected index or degeneracy, with the distinctions developed on the next page.

Exercises

For the two-charge potential above, show that the arithmetic–geometric mean inequality proves the same attractor value without differentiation, and determine the equality condition.

Solution

Set $x=|p|e^{-\phi}$ and $y=|q|e^\phi$. Both are nonnegative and their product is fixed: $xy=|pq|$. Then $V_{\rm BH}=(x^2+y^2)/4\geq xy/2=|pq|/2$. Equality holds exactly when $x=y$, which gives $|p|e^{-\phi_*}=|q|e^{\phi_*}$ and hence $e^{2\phi_*}=|p/q|$. Because the inequality is global, this also shows that the critical point is the global minimum on the real dilaton line when both charges are nonzero.

References

• Ferrara, Sergio, and Renata Kallosh. “Supersymmetry and Attractors.” Physical Review D 54 (1996): 1514–1524. DOI. Open manuscript.
• Strominger, Andrew. “Macroscopic Entropy of N=2 Extremal Black Holes.” Physics Letters B 383 (1996): 39–43. DOI. Open manuscript.