Accelerating Black Holes and C-Metric Families

An accelerating black hole is not accelerated by empty asymptotically flat space alone. In the C-metric family, acceleration is supported by a conical defect interpreted as a cosmic string or strut, by an external field, or by a nontrivial AdS boundary. The metric, coordinate cell, axis periods, and conformal boundary must therefore be specified together before words such as mass, horizon, or regularity have an unambiguous meaning.

C-metric geometry and interpretation

Property	Statement
Object class	Exact solution family
Theory	Four-dimensional GR, Einstein–Maxwell, and $\Lambda$ extensions
Parameters	Metric mass parameter $m$, acceleration $A$, charge, $\Lambda$, and axis periods
Horizons	Black-hole and acceleration horizons; cosmological variants add roots
Acceleration support	String, strut, external field, or AdS boundary conditions
Physical status	Exact geometry; realization depends on defects and boundaries

A useful static, uncharged representative is written

$$ds^2=\frac{1}{A^2(x-y)^2} \left[-F(y)dt^2+\frac{dy^2}{F(y)} +\frac{dx^2}{G(x)}+G(x)d\phi^2\right],$$

with, in one common factorized convention,

$$G(x)=(1-x^2)(1+2mAx), \qquad F(y)=-G(y),$$

Here $m$ is a metric parameter, not yet a conserved mass. For $m>0$, $A>0$, and $0<2mA<1$, the standard static exterior cell is $-1<x<1$ and $-1/(2mA)<y<-1$: $y=-1/(2mA)$ is the black-hole horizon, $y=-1$ the acceleration horizon, and $F,G>0$ between them. This root ordering and its global interpretation are detailed by Griffiths, Krtouš, and Podolský 2006, §§ 2–4. Other sign, coordinate, charge, or cosmological-constant conventions move the roots and rescale $t$ and $\phi$; translating only the symbols can change the physical patch. The factorized form and its domains are developed in Hong and Teo 2003, §§ 2–3, eqs. (3)–(15), while Coordinate Domains and Horizon Structure provides the canonical root analysis.

Axes, strings, and force balance

At a simple axis root $x=x_i$, elementary flatness fixes the azimuthal period

$$\Delta\phi_i=\frac{4\pi}{|G'(x_i)|}.$$

Choosing this period regularizes that axis segment. At a second root $x_j$, the remaining deficit is

$$\delta_j=2\pi-\frac{|G'(x_j)|}{2}\,\Delta\phi_i, \qquad \mu_j=\frac{\delta_j}{8\pi G}.$$

Positive $\mu_j$ describes a string pulling the hole; a negative deficit is a strut exerting compression. In the generic vacuum C-metric one period cannot regularize both axes. Removing the visible defect requires added physics—such as electromagnetic force in an Ernst solution—or a boundary construction that carries the force. Calling the metric “a pair of freely accelerating black holes” without this support omits a distributional part of the spacetime.

The axis choice also enters conserved charges because it changes the angular range. Mass and charge formulas quoted with $\phi\in[0,2\pi)$ cannot be copied unchanged after a deficit is introduced. A physical mass $M$ can be defined only after fixing the angular normalization and timelike generator; for the common azimuthal normalization $\Delta\phi=2\pi/K$, for example, the static AdS conserved mass is proportional to $m/K$ (with the factor of $G$ fixed by the charge convention), rather than being identifiable with $m$ before that normalization is chosen.

Horizons, mechanics, and global data

Simple roots of $F(y)$ are Killing horizons in the chosen static cell. One root is associated with the black hole and another with acceleration; their surface gravities are generally unequal. A Euclidean section is smooth at both only when the corresponding periods agree or when a controlled conical singularity is retained. This restriction is central in pair-creation instantons, not a minor coordinate choice. Hawking and Ross 1995, complete article, printed pp. 3382–3385 give a classic charged-pair application.

The conformal factor vanishes at $x=y$, so conformal infinity can intersect the coordinate cell in a way unlike Schwarzschild infinity. In AdS, the acceleration scale relative to the AdS radius separates “slow” configurations with a single black-hole horizon from “fast” configurations with an additional acceleration horizon, as made explicit in Podolský and Griffiths 2006, §§ 2–3. The normalization of the timelike generator must be fixed at the appropriate boundary before assigning surface gravity or mass.

Thermodynamics with a conical defect includes work terms conjugate to the string tensions. Schematically,

$$dM=T_H\,dS+\Phi\,dQ-\lambda_+\,d\mu_+-\lambda_-\,d\mu_-+\cdots,$$

where the thermodynamic lengths $\lambda_\pm$ and even the definition of $M$ depend on the ensemble and boundary normalization. Appels, Gregory, and Kubizňák 2016, printed p. 131303-3, eqs. (14)–(16) established the consistent fixed-tension thermodynamics, while their Appels, Gregory, and Kubizňák 2017, §§ 2–4 derives the thermodynamic lengths when the tensions vary. The general bookkeeping belongs with Multiple Horizons and Variable Couplings.

Controlled limits

Zero acceleration. The limit $A\to0$ requires an $A$-dependent change from $(x,y)$ to spherical coordinates because the conformal prefactor is singular term by term. With mass, charge, and $\Lambda$ held in the translated physical normalization, it recovers Schwarzschild, Reissner–Nordström, or their $(A)dS$ versions.

Zero mass. Setting $m\to0$ removes the black-hole curvature while retaining an acceleration horizon in accelerated coordinates. A residual deficit can still describe a string in otherwise flat or constant-curvature spacetime. This distinguishes acceleration-horizon thermality from black-hole curvature.

Vanishing deficit. Sending one tension to zero regularizes one axis; making both tensions vanish is not a generic independent limit at nonzero $mA$. External fields or boundary forces must be varied consistently if they are what balances the acceleration.

Limitations

C-metrics are mathematically established exact solutions, but their global interpretation is cell- and boundary-dependent. A regular black-hole horizon does not remove the string/strut source, and a smooth axis choice does not by itself give a globally asymptotically flat isolated system. Pair-creation rates add an ensemble, Euclidean saddle, subtraction, and negative-mode problem. There is no observational evidence for an astrophysical C-metric black hole.

References

• Appels, Michael, Ruth Gregory, and David Kubizňák. “Thermodynamics of Accelerating Black Holes.” Physical Review Letters 117 (2016): 131303. DOI.
• Appels, Michael, Ruth Gregory, and David Kubizňák. “Black Hole Thermodynamics with Conical Defects.” Journal of High Energy Physics 2017(05): 116. DOI.
• Griffiths, Jerry B., Pavel Krtouš, and Jiří Podolský. “Interpreting the C-Metric.” Classical and Quantum Gravity 23 (2006): 6745–6766. DOI.
• Hawking, Stephen W., and Simon F. Ross. “Pair Production of Black Holes on Cosmic Strings.” Physical Review Letters 75 (1995): 3382–3385. DOI.
• Hong, K., and Edward Teo. “A New Form of the C-Metric.” Classical and Quantum Gravity 20 (2003): 3269–3277. DOI.
• Podolský, Jiří, and Jerry B. Griffiths. “Accelerating Kerr–Newman Black Holes in (Anti-)de Sitter Space-Time.” Physical Review D 73 (2006): 044018. DOI.