Area Entropy and the First Law

The coefficient in black-hole area entropy is fixed by matching two independently normalized statements: the mechanical first law and the Hawking temperature. In Einstein gravity they imply $S_{\mathrm{BH}}=A/(4G)$ up to a state-independent additive constant. Because the same horizon generator defines energy, surface gravity, and temperature, rescaling that generator cancels out of the inference. Higher-curvature and quantum terms modify which entropy enters, not this normalization logic.

Required background. Quantum Information supplies the distinction between entropy as a state functional and an additive thermodynamic state function.

Helpful background. Quantum Fields in Curved Spacetime supplies the Hawking temperature; Einstein Equations supplies the stationary solution family; Black-Hole Geometry and Mechanics supplies the mechanical first law; and Semiclassical Gravity supplies the regime in which temperature and backreaction are compared.

Mechanical area coefficient

For nearby stationary Einstein–Maxwell black holes, with all charges and the horizon generator normalized relative to the same boundary, the first law is

$$\delta M =\frac{\kappa}{8\pi G}\,\delta A +\Omega_{\mathrm H}\delta J +\Phi_{\mathrm H}\delta Q.$$

This is initially a relation among classical solutions. Its area coefficient contains $G$ because it follows from the Einstein–Hilbert action and its Hamiltonian boundary charge. By itself the equation makes $\kappa/(8\pi G)$ look analogous to a temperature but does not identify a physical thermal scale.

Quantum field theory on the stationary background supplies that independent scale,

$$T_{\mathrm H}=\frac{\kappa}{2\pi}.$$

Comparing the two equations with $\delta M=T\,\delta S+\Omega\,\delta J+\Phi\,\delta Q$ gives

$$\delta S_{\mathrm{BH}} =\frac{\delta A}{4G}.$$

Integration on a connected family of stationary solutions yields

$$S_{\mathrm{BH}}=\frac{A}{4G}+S_0.$$

The first law cannot determine the constant $S_0$. A reference state, statistical interpretation, or Euclidean normalization is needed if that constant is consequential.

The word “connected” matters. A differential first law compares nearby solutions on one branch; it cannot compare additive constants assigned to disconnected topologies or superselection sectors. Nor may one infer the entropy of an exactly extremal saddle merely by integrating through a singular change of Euclidean topology. A smooth nonextremal limit can be physically useful, but its assumptions should be stated rather than hidden in the integration constant.

The derivation also identifies which Newton constant appears. It is the coupling multiplying the renormalized Einstein term at the scale where the semiclassical action is defined. Laboratory measurements fix renormalized low-energy couplings, not bare regulator-dependent parameters. If matter loops shift that coupling, the geometric entropy and exterior entropy must be renormalized together before the first law is interpreted.

Why generator normalization cancels

Let the future horizon generator change by $\chi^\mu\mapsto a\chi^\mu$ with $a\gt0$. The Hamiltonian energy variation, surface gravity, angular velocity relative to the rescaled time, electric potential, and temperature all scale by $a$. The area and charge variations do not. Dividing the area coefficient by the temperature gives

$$\frac{\kappa/(8\pi G)}{\kappa/(2\pi)}=\frac1{4G},$$

which is independent of $a$. This is why one must not combine a surface gravity normalized at a cavity wall with a mass and temperature normalized at infinity.

Wald presents the classical mechanics and Hawking-temperature matching with their hypotheses Wald 2001, §§2–4, PDF. The derivation assumes a stationary nonextremal family and first-order variations satisfying the linearized constraints. It does not assert that an arbitrary dynamical horizon cross-section has thermodynamic entropy determined solely by its instantaneous area.

Schwarzschild check

For Schwarzschild,

$$A=4\pi(2GM)^2=16\pi G^2M^2, \qquad S_{\mathrm{BH}}=4\pi GM^2.$$

Consequently

$$dS_{\mathrm{BH}}=8\pi GM\,dM, \qquad T_{\mathrm H}dS_{\mathrm{BH}} =\frac{1}{8\pi GM}(8\pi GM\,dM)=dM.$$

The dimensions agree: $A/G$ is dimensionless in four dimensions when $c=\hbar=1$. The entropy grows quadratically with $M$, while the temperature falls as $M^{-1}$, reproducing the negative canonical heat capacity without altering the positive entropy.

Beyond the area term

The quarter-area formula is the leading entropy of Einstein gravity. For a general diffeomorphism-invariant local Lagrangian, the stationary first law contains the Noether-charge or Wald entropy rather than simply $A/(4G)$. Iyer and Wald derive the covariant charge construction and its ambiguities Iyer and Wald 1994, §§3–6, PDF. Matter-loop renormalization shifts $G$ and higher-curvature couplings at the same time that exterior entanglement entropy diverges.

Accordingly, three quantities should remain distinct:

• bare area entropy uses a regulator-dependent bare $G$;
• renormalized gravitational entropy includes the local terms dictated by the renormalized action; and
• generalized entropy adds the renormalized exterior state entropy.

Fine-grained radiation entropy is a state/region quantity and need not equal the stationary thermodynamic entropy. The area-law derivation fixes a thermodynamic coefficient; it does not by itself establish information recovery.

An independent sign check comes from the physical-process version of the first law. Positive Killing energy crossing a future horizon focuses its generators and increases the late stationary area, so both $\delta M$ and $\delta A$ are positive for $\kappa\gt0$. With positive Hawking temperature, the inferred $\delta S_{\mathrm{BH}}$ has the same sign. Reversing the horizon orientation without changing the charge convention would spoil this agreement.

Exercise

Starting from $A=16\pi G^2M^2$ and $T_{\mathrm H}=1/(8\pi GM)$, derive $S_{\mathrm{BH}}$ by integrating $dS=dM/T_{\mathrm H}$. What part of the answer remains undetermined?

Solution

The differential relation is $dS=8\pi GM\,dM$. Integration gives $S=4\pi GM^2+S_0=A/(4G)+S_0$. The first law fixes the derivative and hence the quarter-area coefficient, but it cannot fix the additive constant $S_0$. Choosing that constant requires an additional reference or microscopic normalization and does not affect ordinary first-law variations.

References

• Iyer, V., and R. M. Wald (1994), “Some Properties of Noether Charge and a Proposal for Dynamical Black Hole Entropy,” Physical Review D 50, 846–864. Open PDF.
• Wald, R. M. (2001), “The Thermodynamics of Black Holes,” Living Reviews in Relativity 4, 6. Open PDF.