Holzhey, Larsen and Wilczek (Nucl. Phys. B 424 (1994) 443-467) made the geometric (entanglement) entropy of 1+1 dimensional conformal field theory sharp, defined a renormalized entropy of moving-mirror radiation by subtracting the vacuum contribution, found it non-extensive and locally negative, and observed the identity S_ren = -(1/24) ln(C/C_0) relating it to the excess of energy-momentum correlations over the vacuum. The paper itself left open: the effect of zero modes and fermionic field content, the incorporation of energy conservation and back-reaction into the prescribed mirror trajectory, the generality of the renormalized entropy and its correlation formula beyond the moving-mirror setting, and the implications for the black hole information problem. Of these, the infrared zero-mode sector, the entropy production of self-consistent dynamics (fully dynamical mirror motion with quantum back-reaction has recently been realized by Kumar-Reyes-Wintergerst 2024), and the generality of the correlation formula remain open.

Background

In quantum field theory the geometric entropy (also called entanglement entropy) of a spatial subregion is the von Neumann entropy of the reduced density matrix obtained by tracing the state over the subregion's complement; because a field theory has infinitely many degrees of freedom per unit volume, this entropy diverges for sharply localized subregions. It is computed by the replica trick: the trace of the n-th power of the reduced density matrix is expressed as a partition function on an n-sheeted space, analytically continued in n, and differentiated at n = 1. For a two-dimensional conformal field theory (CFT - a massless relativistic quantum field theory invariant under angle-preserving transformations, characterized by a central charge c, a dimensionless count of its degrees of freedom), Holzhey, Larsen and Wilczek (1994) used the replica trick, conformal mapping onto a torus, and modular invariance (invariance of the torus partition function under exchange of its two cycles; the modular parameter is the complex shape parameter of the torus) to obtain the now-standard result that the geometric entropy of an interval of length Sigma with ultraviolet cutoff epsilon is (c/3) ln(Sigma/epsilon). They compared this analytic result with numerical computations of the geometric entropy for free fields by Srednicki (1993), finding approximate agreement of the logarithmic coefficient together with an additional infrared-cutoff-dependent term that they ascribed to the neglected zero modes.

The moving mirror is an idealized, perfectly reflecting pointlike boundary in 1+1 dimensional Minkowski spacetime whose trajectory is prescribed externally; Carlitz and Willey (1987) showed that a mirror on an appropriate eternally accelerating trajectory - one that accelerates away from rest and asymptotes to the speed of light - radiates thermally, and the specific ray-tracing profile f(z) proportional to exp(-z/4M) (z and f(z) are the null coordinates of a light ray before and after reflection; M is a mass parameter), which emulates a collapsing black hole horizon, reproduces Hawking radiation at the Hawking temperature T_H = 1/(8 pi M). Holzhey-Larsen-Wilczek (HLW) computed the geometric entropy of the radiation in such mirror-excited states and defined the renormalized entropy S_ren by subtracting the stationary-mirror (vacuum) contribution, rendering it cutoff-independent. They found S_ren is non-extensive (not additive over independent subsystems, unlike thermodynamic entropy) and can be locally negative - analogous to locally negative Casimir energies - and, in the settings they examined, found what they called an intriguing coincidence: S_ren = -(1/24) ln(C/C_0), where C is the connected two-point correlation function of the energy-momentum tensor in the observed radiation and C_0 its stationary-mirror (vacuum) value, so that S_ren measures correlations in excess of the vacuum. Physically this expresses that the mirror radiation is pseudo-thermal - a reflection of vacuum correlations rather than a genuinely equilibrated thermal ensemble - which bears on the black hole information paradox: the tension between the approximately thermal character of Hawking's semiclassical radiation and the expectation from quantum gravity that evaporation end in a globally pure final state. In the later covariant reformulation of Bianchi and Smerlak (2014), who re-derived the renormalized entropy of the radiation through mutual information (one half of the mutual information between the radiation region and the complement of its broadened version, the broadening serving as a covariant regulator), the entropy of the radiation emitted up to retarded time u takes the compact form S(u) = -(c/12) ln p'(u), where p(u) is the ray-tracing function - the incoming null coordinate as a function of the outgoing null coordinate - and p' its derivative.

HLW were explicit about the limits of their derivation: their torus argument ignored zero modes (field configurations of vanishing frequency, allowed depending on boundary conditions, whose inclusion can change the infrared-sensitive normalization of the partition function and add subleading cutoff dependence); fermions required a separate treatment; the mirror trajectory is prescribed a priori, so the model does not enforce energy conservation or back-reaction (the reaction of the radiation on the mirror's motion, or the analogous gravitational response in a black hole spacetime); and whether the renormalized entropy and its correlation formula have useful meaning outside the moving-mirror problem was left as a question worthy of further investigation. Later companion and follow-up work - Fiola-Preskill-Strominger-Trivedi (1994) on two-dimensional black hole thermodynamics and information loss, Larsen-Wilczek (1995) on fermionic wave functionals, and Calabrese-Cardy (2004), who systematized 1+1 CFT entanglement entropy - built the framework into a standard tool. Early attempts to make the mirror dynamics energy-conserving - the dynamical mirror of Chung-Verlinde (1994), equivalent to two-dimensional dilaton gravity (a two-dimensional gravity theory whose metric couples to an additional scalar field, the dilaton), and the WKB energy-conserving computation of Tuning-Verlinde (1996) - found instabilities and approximation breakdowns, so prescribed trajectories remained the norm in entropy computations; only recently has fully dynamical mirror motion with quantum back-reaction been solved (Kumar-Reyes-Wintergerst 2024), without yet computing its entropy production.

Problem Statement

Determine the domain of validity and the dynamical self-consistency of the renormalized geometric entropy framework established by Holzhey, Larsen and Wilczek (HLW) for the moving-mirror conformal field theory. The framework renders the entropy of mirror radiation finite by subtracting the stationary-mirror (vacuum) contribution, yielding a renormalized entropy SrenS_{\mathrm{ren}} that is non-extensive (not additive over independent subsystems) and can be locally negative, and satisfying, in the free 1+1 conformal settings they examined, the identity Sren=−124ln⁡(C/C0)S_{\mathrm{ren}}=-\frac{1}{24}\ln(C/C_0), where CC is the connected two-point correlation function of the energy-momentum tensor in the observed radiation and C0C_0 its stationary-mirror value. The source poses four linked questions:

(1) Field content and boundary conditions. The fermionic geometric entropy has since been computed (Larsen-Wilczek 1995), and zero-mode-induced divergences of the entanglement entropy have been characterized for free fields (Mallayya et al. 2014; Nenmeli-Shankaranarayanan 2022). What remains open is the infrared sector: exactly how zero modes - field configurations of vanishing frequency, allowed depending on boundary conditions - change the normalization of the replica partition function (a power of the modular parameter of the torus) and the subleading ultraviolet cutoff dependence of the HLW formulas, in particular for the free massless scalar with non-compact target, whose spatially constant zero mode makes the vacuum degenerate and whose infrared behavior remains under discussion.

(2) Dynamical self-consistency. The mirror trajectory is prescribed a priori, so the model violates energy conservation. What happens to entropy production and to the late-time behavior - in particular whether a slowly decelerating, remnant-like endpoint is physically realized - once the mirror dynamics (or the corresponding gravitational back-reaction in the black hole context) is made self-consistent with the energy and momentum carried away by the radiation? A fully dynamical model now exists (Kumar-Reyes-Wintergerst 2024: a massive boundary particle obeying its relativistic equation of motion with the quantum stress tensor of a CFT as back-reaction, for which the back-reaction prevents horizon formation for a class of collapse-mimicking incoming data), but the renormalized entropy production and the remnant question have not been computed in any such self-consistent model.

(3) Generality of the framework. Does the definition of renormalized geometric entropy, and specifically the correlation formula Sren=−124ln⁡(C/C0)S_{\mathrm{ren}}=-\frac{1}{24}\ln(C/C_0), generalize beyond the prescribed-trajectory 1+1 free conformal moving-mirror setting - to interacting (non-free) conformal theories, to non-conformal theories, to higher dimensions - or what are the obstructions? The covariant re-derivations of the renormalized entropy (Bianchi-Smerlak 2014 and successors) recover the trajectory formula S(u)=−c12ln⁡p′(u)S(u)=-\frac{c}{12}\ln p'(u) and an exact flux-entropy identity, but do not address the excess-of-correlations interpretation or the correlation formula itself.

(4) Black hole implications. Can the pseudo-thermal, non-extensive character of the renormalized entropy be reconciled with a globally pure final state within a consistent dynamical (or fully quantum-gravitational) treatment? In its general quantum-gravity form this question is now pursued within the separate island/quantum-extremal-surface program (quantum extremal surfaces - surfaces extremizing the sum of horizon area and bulk matter entropy - which bound the entropy of the radiation); within this problem it enters through what the mirror and two-dimensional dilaton-gravity models imply about entropy production and evaporation endpoints.

The renormalized-entropy formulas and the correlation identity remain to be established (or refuted) in any one of the open extended settings - the zero-mode/infrared sector of the replica partition function, energy-conserving dynamical mirror (or two-dimensional dilaton-gravity) models with back-reaction, or beyond free 1+1 conformal fields - and, in any one of these settings, a precise identification of where and why the framework fails resolves the question equally. Sub-questions (1)-(3) are each independently refutable: resolving any one of them suffices, and sub-question (4) is not separately required, its general quantum-gravity form being pursued within the separate island/quantum-extremal-surface program.

References

  1. C. Holzhey, F. Larsen, F. Wilczek, Geometric and renormalized entropy in conformal field theory, Nucl. Phys. B 424 (1994) 443-467, DOI 10.1016/0550-3213(94)90402-2, arXiv:hep-th/9403108, https://arxiv.org/abs/hep-th/9403108
  2. F. Larsen, F. Wilczek, Geometric Entropy, Wave Functionals, and Fermions, Annals Phys. 243 (1995) 280-298, DOI 10.1006/aphy.1995.1100, arXiv:hep-th/9408089, https://arxiv.org/abs/hep-th/9408089
  3. T.-D. Chung, H. Verlinde, Dynamical Moving Mirrors and Black Holes, Nucl. Phys. B 418 (1994) 305-336, DOI 10.1016/0550-3213(94)90249-6, arXiv:hep-th/9311007, https://arxiv.org/abs/hep-th/9311007
  4. N. Tuning, H. Verlinde, Backreaction on Moving Mirrors and Black Hole Radiation, arXiv:hep-th/9605063 (1996), https://arxiv.org/abs/hep-th/9605063
  5. R. D. Carlitz, R. S. Willey, Reflections on moving mirrors, Phys. Rev. D 36 (1987) 2327-2335, DOI 10.1103/PhysRevD.36.2327, https://doi.org/10.1103/PhysRevD.36.2327
  6. M. Srednicki, Entropy and area, Phys. Rev. Lett. 71 (1993) 666-669, DOI 10.1103/PhysRevLett.71.666, https://doi.org/10.1103/PhysRevLett.71.666
  7. T. M. Fiola, J. Preskill, A. Strominger, S. P. Trivedi, Black hole thermodynamics and information loss in two dimensions, Phys. Rev. D 50 (1994) 3987-4014, DOI 10.1103/PhysRevD.50.3987, arXiv:hep-th/9403137, https://arxiv.org/abs/hep-th/9403137
  8. P. Calabrese, J. Cardy, Entanglement Entropy and Quantum Field Theory, J. Stat. Mech. (2004) P06002, DOI 10.1088/1742-5468/2004/06/P06002, arXiv:hep-th/0405152, https://arxiv.org/abs/hep-th/0405152
  9. E. Bianchi, M. Smerlak, Entanglement entropy and negative energy in two dimensions, Phys. Rev. D 90 (2014) 041904, DOI 10.1103/PhysRevD.90.041904, arXiv:1404.0602, https://arxiv.org/abs/1404.0602
  10. E. Bianchi, T. De Lorenzo, M. Smerlak, Entanglement entropy production in gravitational collapse: covariant regularization and solvable models, JHEP 06 (2015) 180, DOI 10.1007/JHEP06(2015)180, arXiv:1409.0144, https://arxiv.org/abs/1409.0144
  11. M. R. R. Good, Y. C. Ong, Signatures of Energy Flux in Particle Production: A Black Hole Birth Cry and Death Gasp, JHEP 07 (2015) 145, DOI 10.1007/JHEP07(2015)145, arXiv:1506.08072, https://arxiv.org/abs/1506.08072
  12. S. Abdolrahimi, D. N. Page, Hawking Radiation Energy and Entropy from a Bianchi-Smerlak Semiclassical Black Hole, Phys. Rev. D 92 (2015) 083005, DOI 10.1103/PhysRevD.92.083005, arXiv:1506.01018, https://arxiv.org/abs/1506.01018
  13. P. Chen, D.-h. Yeom, Entropy evolution of moving mirrors and the information loss problem, Phys. Rev. D 96 (2017) 025016, DOI 10.1103/PhysRevD.96.025016, arXiv:1704.08613, https://arxiv.org/abs/1704.08613
  14. M. R. R. Good, K. Yelshibekov, Y. C. Ong, On Horizonless Temperature with an Accelerating Mirror, JHEP 03 (2017) 013, DOI 10.1007/JHEP03(2017)013, arXiv:1611.00809, https://arxiv.org/abs/1611.00809
  15. M. R. R. Good, E. V. Linder, Finite Energy but Infinite Entropy Production from Moving Mirrors, Phys. Rev. D 99 (2019) 025009, DOI 10.1103/PhysRevD.99.025009, arXiv:1807.08632, https://arxiv.org/abs/1807.08632
  16. A. Myrzakul, C. Xiong, M. R. R. Good, Relativistic quantum information as radiation reaction: entanglement entropy and self-force of a moving mirror analog to the CGHS black hole, Entropy 23 (2021) 1664, DOI 10.3390/e23121664, arXiv:2101.08139, https://arxiv.org/abs/2101.08139
  17. I. Akal, Y. Kusuki, N. Shiba, T. Takayanagi, Z. Wei, Entanglement entropy in holographic moving mirror and Page curve, Phys. Rev. Lett. 126 (2021) 061604, DOI 10.1103/PhysRevLett.126.061604, arXiv:2011.12005, https://arxiv.org/abs/2011.12005
  18. I. Akal, T. Kawamoto, S.-M. Ruan, T. Takayanagi, Z. Wei, Zoo of holographic moving mirrors, JHEP 08 (2022) 296, DOI 10.1007/JHEP08(2022)296, arXiv:2205.02663, https://arxiv.org/abs/2205.02663
  19. K. Mallayya, R. Tibrewala, S. Shankaranarayanan, T. Padmanabhan, Zero modes and divergence of entanglement entropy, Phys. Rev. D 90 (2014) 044058, DOI 10.1103/PhysRevD.90.044058, arXiv:1404.2079, https://arxiv.org/abs/1404.2079
  20. V. Nenmeli, S. Shankaranarayanan, When and why do zero-modes cause a divergence in the entanglement entropy?, arXiv:2212.07174 (2022), https://arxiv.org/abs/2212.07174
  21. G. Penington, Entanglement Wedge Reconstruction and the Information Paradox, JHEP 09 (2020) 002, DOI 10.1007/JHEP09(2020)002, arXiv:1905.08255, https://arxiv.org/abs/1905.08255
  22. A. Almheiri, N. Engelhardt, D. Marolf, H. Maxfield, The entropy of bulk quantum fields and the entanglement wedge of an evaporating black hole, JHEP 12 (2019) 063, DOI 10.1007/JHEP12(2019)063, arXiv:1905.08762, https://arxiv.org/abs/1905.08762
  23. P. Kumar, I. A. Reyes, J. Wintergerst, Relativistic dynamics of moving mirrors in CFT2_2: quantum backreaction and black holes, Phys. Rev. D 109 (2024) 065010, DOI 10.1103/PhysRevD.109.065010, arXiv:2310.03483, https://arxiv.org/abs/2310.03483
  24. I. A. Reyes, Moving mirrors, Page curves and bulk entropies in AdS2_2, Phys. Rev. Lett. 127 (2021) 051602, DOI 10.1103/PhysRevLett.127.051602, arXiv:2103.01230, https://arxiv.org/abs/2103.01230
  25. P. Biswas, S. Das, A. Dinda, Moving Interfaces and two-dimensional Black Holes, JHEP 05 (2024) 329, DOI 10.1007/JHEP05(2024)329, arXiv:2401.11451, https://arxiv.org/abs/2401.11451