The research, carried out by Jia-Zhou Liu, Shan-Ping Wu, Shao-Wen Wei, and Yu-Xiao Liu of Lanzhou University, tackles a subtle but fundamental question in the covariant phase space formalism, the modern mathematical framework in which conserved quantities such as energy and entropy arise as surface terms tied to diffeomorphism symmetries, the gauge redundancies of general relativity. Within this framework, the Iyer-Wald construction derives the entropy of a stationary black hole by evaluating a local functional built from the Lagrangian’s response to curvature, integrated over a cross section of the horizon. Crucially, the derivation assumes that all fields involved, including any matter fields that couple directly to curvature, can be smoothly extended onto the so-called regular bifurcation surface, the two-dimensional slice where the horizon generator vanishes. The Lanzhou team asked what happens when that smoothness assumption breaks down, as it does for some vector and tensor fields whose components become singular in horizon-adapted coordinates.

When a matter field cannot be smoothly extended to the bifurcation surface, the standard argument that forces certain horizon contributions to vanish no longer goes through. In particular, two types of terms can survive in the horizon surface charge variation, the quantity from which the entropy is extracted. The authors organize the full entropy entering the thermodynamic first law into a clean three-part decomposition: the entropy equals the standard Wald entropy, plus a correction coming from the non-Wald part of the entropy-sector Noether charge, plus a further integrable remainder coming from the presymplectic potential, the one-form that encodes the symplectic structure of the field theory on its space of solutions. This decomposition, which they label schematically as the entropy being the Wald term plus two additional pieces, provides a direct computational criterion for deciding when the familiar Wald density is sufficient and when the full horizon calculation is unavoidable.

Importantly, the authors distinguish their contribution from the well-known Jacobson-Kang-Myers ambiguity of the covariant phase space, in which one may shift the Lagrangian and the presymplectic potential by exact forms, reshuffling local expressions without changing the physics. Instead of exploiting that freedom, the team fixes a single representative, obtained by directly varying the displayed Lagrangian, and works consistently within it. After separating out ordinary matter and gauge work terms, such as electromagnetic or vector kinetic contributions that carry independent charges or fixed background variations, everything that remains in the entropy sector is subject to the new decomposition. The two corrective pieces are then evaluated explicitly for three representative families of static, spherically symmetric black holes, each illustrating a different physical possibility.

The first example involves gravity coupled to a background Kalb-Ramond antisymmetric two-form field, a nonminimal coupling inspired by string theory and by models of spontaneous Lorentz symmetry breaking. Here the coupling does modify the Wald entropy itself, rescaling it in terms of the tensor field’s vacuum expectation value, so the naive area law is already altered. Yet when the team carried out the full horizon analysis, both of the new correction terms vanished exactly. The Kalb-Ramond background is regular enough at the bifurcation surface that the standard Iyer-Wald reduction applies, and the entropy entering the first law is precisely the Wald entropy. This case demonstrates that nonminimal coupling by itself does not guarantee corrections; it is the regularity of the matter fields at the horizon that matters.

The second and perhaps most striking example is bumblebee gravity, a vector-tensor theory in which a fixed background vector field acquires a vacuum expectation value that spontaneously breaks Lorentz symmetry in the gravitational sector. The team examined two branches of exact black hole solutions, distinguished by whether the bumblebee vacuum expectation value is spacelike or lightlike. On the spacelike branch, the Noether charge correction vanishes, but a nonzero remainder survives from the presymplectic potential, and integrating that remainder on the space of solutions adds exactly half of the coupling-dependent contribution back into the entropy, yielding a first law entropy larger than the Wald value. On the lightlike branch, where the radial vector component diverges at the horizon in the static frame, both corrections are nonzero and, remarkably, they cancel each other exactly, so that the final entropy reduces to the plain area law even though neither the Wald term nor the individual corrections would suggest it. The cancellation is a genuine surprise, showing that pieces of the entropy can hide in different sectors of the phase space and conspire in unexpected ways.

The third class of examples pushes the analysis into higher-curvature territory: extended Gauss-Bonnet gravity formulated in Weyl geometry, where the connection is torsionless but not metric-compatible, with the failure of metric compatibility encoded in a Weyl vector field. In five spacetime dimensions, the team studied two solution branches, one in which the two coupling constants coincide and one in which they differ. Here both corrective mechanisms are active simultaneously: the curvature dependence of the tensor that multiplies the Riemann tensor in the Lagrangian becomes entangled with the Weyl vector, and the non-Wald Noether charge contribution as well as the presymplectic remainder both enter the answer. Assembling the full entropy from the three pieces, the authors verified that the enlarged thermodynamic first law closes properly, with the energy obtained from the asymptotic surface charge and an additional conjugate potential appearing for the solution’s hair parameter, which can be promoted to a genuine thermodynamic variable.

Beyond its technical elegance, the work carries a practical message for the black hole thermodynamics community. In recent years, debates have swirled around the correct entropy assignments for black holes in Horndeski theories, generalized Proca models, and other nonminimally coupled frameworks, with different calculations sometimes producing conflicting answers. The Lanzhou decomposition offers a common language for resolving such disputes: by checking whether the Noether charge correction and the presymplectic remainder vanish, vanish partially, or cancel, one can systematically determine the entropy that the first law demands. The authors also emphasize that their criterion separates regular tensor backgrounds, singular vector backgrounds, and higher-curvature Weyl-vector examples within one unified covariant phase space treatment, something no single earlier prescription accomplished.

The study leaves open a number of tantalizing directions. Rotating black holes, whose horizons involve an additional angular momentum term and a second Killing field, were deliberately excluded to keep the analysis tractable, and extending the decomposition to spinning and genuinely dynamical horizons is a natural next step. The authors also note that their analysis could be combined with the recently developed extended Iyer-Wald formalism and universal thermodynamic-volume constructions, which aim to give geometric meaning to the full suite of black hole thermodynamic variables. If the extra entropy terms identified here survive in broader settings, they could subtly alter phase diagrams, thermodynamic stability analyses, and even microscopic state counts of black holes in the modified gravity theories that increasingly populate the cosmological literature.

For now, the lesson is a humbling one. The Wald entropy, elegant as it is, rests on regularity assumptions that real solutions of realistic theories can violate. When a vector field refuses to behave smoothly at the horizon’s deepest point, finite pieces of entropy can leak out of the local Wald density and reappear elsewhere in the covariant phase space, sometimes enhancing the entropy, sometimes hiding inside a Noether charge, and sometimes canceling themselves away entirely. Black holes, it seems, still guard their secrets about the quantum nature of spacetime, and even our most trusted formulas must be checked against the full architecture of the theory in which they live.

Subject of Research: Corrections to black hole entropy beyond the Wald term in gravity theories with nonminimal matter-curvature couplings

Article Title: Black hole entropy beyond the wald term in nonminimally coupled gravity: a covariant phase space decomposition

Article References: Liu, J.-Z., Wu, S.-P., Wei, S.-W., & Liu, Y.-X. (2026). Black hole entropy beyond the wald term in nonminimally coupled gravity: a covariant phase space decomposition. The European Physical Journal C, 86(9), Article 1026. https://doi.org/10.1140/epjc/s10052-026-16290-y

Image Credits: AI Generated

DOI: 10.1140/epjc/s10052-026-16290-y

Keywords: black hole entropy, Wald entropy, covariant phase space, nonminimal coupling, bumblebee gravity, Kalb-Ramond field, Gauss-Bonnet gravity, Weyl geometry, first law of thermodynamics, Noether charge, Lorentz symmetry breaking, black hole thermodynamics