This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

Are regular black holes from pure gravity classified within the same thermodynamical topology?

Sheng-Wei Wang, Shan-Ping Wu, Shao-Wen Wei 111Corresponding author. E-mail: [email protected] 1Lanzhou Center for Theoretical Physics, Key Laboratory of Theoretical Physics of Gansu Province, and Key Laboratory of Quantum Theory and Applications of MoE, Lanzhou University, Lanzhou, Gansu 730000, China,
2Institute of Theoretical Physics, Research Center of Gravitation, and School of Physical Science and Technology, Lanzhou University, Lanzhou 730000, People’s Republic of China
Abstract

Regular black holes, which avoid the essential center singularities, can be constructed through various methods, including nonlinear electrodynamics and quantum corrections. Recently, it was shown that via an infinite tower of higher-curvature corrections, one can obtain different regular black hole solutions in any spacetime dimension D5D\geq 5. Utilizing the concept of thermodynamical topology, we examine these black holes as topological thermodynamic defects, classifying them into distinct topological categories based on their generalized free energy. We find that the Hawking temperature of the black hole has at least one zero point at the small horizon radius limit. Under this fact, the regular black holes generated through the purely gravitational theories exhibit universal thermodynamical behaviors, strongly suggesting they belong to the same topological class. We presents a comprehensive analysis of these properties, providing a clearer understanding of the fundamental nature of regular black holes and their classification within the framework of thermodynamical topology.

Regular black hole, thermodynamical topology

I Introduction

Black hole thermodynamics is an advanced field of study that investigates the intricate relationship between thermodynamic principles and black hole characteristics. This research interests emerged from the groundbreaking recognition that black holes exhibit entropy and temperature, parameters conventionally linked to thermodynamic systems Hawking:1975ParticleCreation ; Hawking:1976BHTherm ; Bardeen:1976FourLaws ; Gibbons:1976ActionIntegral . Despite significant advancements in exploring black holes, it seems that we still need further means to study them more thoroughly and clearly. To enhance our understanding and characterization of different black hole systems, the concept of thermodynamical topology has been effectively introduced Wei:2022dzw . This method views black hole solutions as topological thermodynamic defects, analyzing their topological numbers based on the asymptotic behavior of the generalized free energy, and classifies nearly all black hole systems into three distinct categories: +1, 0, and -1. This groundbreaking work has offered new insights into the fundamental nature of black holes and gravity. The topological approach introduced in Ref. Wei:2022dzw has gained significant popularity due to its wide applicability and efficiency of study. So far, this method has been applied to studies of rotating AdS black holes Wu:2023sue ; Wu:2022whe and Lovelock AdS black holes in AdS spacetime Bai:2022klw ; Liu:2022aqt . In addition, many other black holes have been studied, such as quantum BTZ black holes, Born-Infeld black holes, dyonic AdS black holes, and many more Wu:2024txe ; Chen:2023elp ; Chen:2024sow ; Fang:2022rsb ; Chatzifotis:2023ioc ; Wei:2023bgp ; Du:2023wwg ; Chen:2023ddv ; Rizwan:2023ivp ; Hazarika:2024dex .

The singularity problem of black holes, as a general prediction of general relativity, signifies the limitations of this theory HawkingLargeScale ; Senovilla:1998oua ; Penrose:1964wq . In order to address this issue, many fruitful approaches have emerged. Black holes that lack essential singularities but possess only coordinate singularities are termed regular black holes Lan:2023cvz . The study of regular black holes can be traced back to the pioneering work of Sakharov and Gliner, who proposed that essential singularities could be circumvented by substituting the vacuum with a vacuum-like medium characterized by a de Sitter metric Sakharov:1966aja ; Gliner:1966 .

The early model of regular black hole was developed by Bardeen by making the mass of the Schwarzschild black hole as a position-dependent function Bardeen:1968 , which is now referred to as the Bardeen black hole. Ayon-Beato and Garcia provided the first explanation of the Bardeen black hole Ayon-Beato:2000mjt that they are described by the Lagrangian with nonlinear electrodynamics source. The solution to the Einstein equations, coupled with the energy-momentum tensor associated with the magnetic field strength, corresponds to a self-gravitating magnetic monopole charge Ayon-Beato:2000mjt . Later, many other regular black hole solutions were obtained by making use of the nonlinear electrodynamics Ayon-Beato:1998hmi ; Bronnikov:2000vy ; Bronnikov:2000yz ; Ayon-Beato:2004ywd ; Dymnikova:2004zc ; Berej:2006cc ; Balart:2014jia ; Fan:2016rih ; Bronnikov:2017sgg ; Junior:2023ixh .

Recently, Bueno, Cano, and Hennigar introduced a novel method that presents the first instances of regular black hole solutions within purely gravitational theories, requiring no specific fine-tuning or constraints among the relevant parameters Bueno:2024dgm . Their work unveiled innovative families of regular black holes in spacetime dimensions of larger than five. These black holes introduce modifications to the conventional Schwarzschild black hole and belong to the class of quasi-topological gravities, a well-established category of metric theories of gravity with higher-curvature terms. Notably, these modifications are advantageous for investigating black hole solutions as their equations of motion simplify to second order when applied to static and spherically symmetric metrics Bueno:2024dgm ; Oliva:2010eb ; Myers:2010ru ; Dehghani:2011vu ; Ahmed:2017jod ; Cisterna:2017umf . Thus far, Ref. Bueno:2024dgm has inspired a series of remarkable studies utilizing its proposed approach Konoplya:2024hfg ; DiFilippo:2024mwm ; Konoplya:2024kih ; Ma:2024olw ; Bueno:2024qhh ; Zhang:2024ljd ; Estrada:2024moz ; Bueno:2024eig ; Bueno:2024zsx .

We carefully study the thermodynamics of regular black holes constructed using the method of Bueno, Cano, and Hennigar Bueno:2024dgm and found that they exhibit certain universal properties. We conjecture that regular black holes constructed by this method share some common thermodynamic features, classifying them as the same type of black hole. Is there a simple and intuitive way to illustrate this? Fortunately, through the analysis of the general case, we conclude that these black holes can reach zero temperature. By focusing on cases where the Hawking temperature has only one single zero point, our analysis in asymptotic limits shows that thermodynamic topology not only effectively describes their thermodynamic properties but also confirms, from a topological perspective, that they belong to the same category. Moreover, this category is actually W0+W^{0+} topological classification given in Ref. Wei24 , through which one can find that the the innermost and outermost black hole states are local thermodynamical stable and unstable. In particular, at low temperature limit, there could be the unstable large and stable small black holes appeared in pair. These result provide a clearer and more comprehensive understanding of the thermodynamic properties of these regular black holes, serving as the motivation for this paper.

The paper is organized as follows. In Sec. II, we review the construction of regular black holes using pure gravity methods. In Sec. III, we analyze their thermodynamical topology and draw our main universal conclusions that these regular black holes belong to one topological class. Two examples to validate these results, including specific details about the thermodynamical topology are given in Sec. IV. Finally, we summarize and discuss our results. The appendix provide some mathematical details.

II Regular Black Holes from Pure Gravity

Quasi-topological gravity is especially effective for analyzing black hole solutions, and they exist across all curvature orders in dimensions D5D\geq 5 Oliva:2010eb ; Myers:2010ru ; Dehghani:2011vu ; Ahmed:2017jod ; Cisterna:2017umf . This family of theories is extensive enough to serve as a foundational set of polynomial densities for expanding general relativity within the framework of effective field theory. As a result, they offer a robust framework for exploring the influence of higher-curvature corrections near black hole singularities. In this section, we would like to give a brief review for such quasi-topological gravity.

In quasi-topological gravity theory, the action is

IQT=116πGdDx|g|[R+n=2nmaxαn𝒵n],I_{\mathrm{QT}}=\frac{1}{16\pi G}\int\mathrm{d}^{D}x\sqrt{|g|}\left[R+\sum_{n=2}^{n_{\mathrm{max}}}\alpha_{n}\mathcal{Z}_{n}\right], (1)

where αn\alpha_{n} are arbitrary coupling constants with dimensions of length2(n1)\mathrm{length}^{2(n-1)}, and 𝒵n\mathcal{Z}_{n} are the quasi-topological densities Oliva:2010eb ; Myers:2010ru ; Dehghani:2011vu ; Ahmed:2017jod ; Cisterna:2017umf . Here nmaxn_{\mathrm{max}} is an integer greater than or equal to 2, and under certain conditions, it can approach positive infinity. Then, let us consider the line element for the DD-dimensional (D5D\geq 5) static and spherically symmetric black hole

ds2=N(r)2f(r)dt2+dr2f(r)+r2dΩD22.ds^{2}=-N(r)^{2}f(r)dt^{2}+\frac{dr^{2}}{f(r)}+r^{2}d\Omega_{D-2}^{2}. (2)

To simplify the procedure, one can introduce Bueno:2024dgm

ψ(r)1f(r)r2,\psi(r)\equiv\frac{1-f(r)}{r^{2}}, (3)

and series function

h(ψ)ψ+n=2nmaxαnψn.h(\psi)\equiv\psi+\sum_{n=2}^{n_{\max}}\alpha_{n}\psi^{n}. (4)

Then, the equations of motion derived from the action (1) can be expressed as

dNdr=0,ddr(rD1h(ψ(r)))=0,\frac{dN}{dr}=0,\quad\frac{d}{dr}\left(r^{D-1}h(\psi(r))\right)=0, (5)

and its solution reads Bueno:2024dgm

N(r)=1,h(ψ(r))=h(1f(r)r2)=μrD1,N(r)=1,\quad h\left(\psi(r)\right)=h\left(\frac{1-f(r)}{r^{2}}\right)=\frac{\mu}{r^{D-1}}, (6)

where μ\mu is an integration constant which is proportional to the ADM mass of the solution. Note that if α2,α3,,αn\alpha_{2},\alpha_{3},...,\alpha_{n} are given, then the metric function f(r)f(r) can be obtained by solving Eq. (6).

Expanding this solution around r=0r=0, one has

f(r)=1(mαnmax)1/nmaxr2(D1)/nmax+.f(r)=1-\left(\frac{m}{\alpha_{n_{\mathrm{max}}}}\right)^{1/n_{\mathrm{max}}}r^{2-(D-1)/n_{\mathrm{max}}}+\cdots. (7)

If nmaxn_{\mathrm{max}} is finite, a curvature singularity occurs at r=0r=0. However, as nmaxn_{\mathrm{max}}\rightarrow\infty, the metric near r=0r=0 approaches that of a de Sitter universe with a cosmological length scale determined by m/αnmaxm/\alpha_{n_{\mathrm{max}}}, which resolves the singularity. To ensure that the black holes with regular interiors exist as solutions to the full theory, the following conditions are sufficient,

αn0n,limn(αn)1n=C>0.\alpha_{n}\geq 0\mathrm{~{}}\forall\;n,\quad\lim_{n\to\infty}(\alpha_{n})^{\frac{1}{n}}=C>0. (8)

Here, the condition that αn\alpha_{n} is positive ensures the monotonicity of h(ψ)h(\psi) for ψ>0\psi>0, while the limit limn(αn)1n=C>0\lim_{n\to\infty}(\alpha_{n})^{\frac{1}{n}}=C>0 determines the radius of convergence of the series (4) as ψ0=1/C\psi_{0}=1/C. At ψ=ψ0\psi=\psi_{0}, the function h(ψ)h(\psi) diverges. These conditions imply that hh is a smooth, bijective mapping from (0,1/C)(0,1/C) to (0,)(0,\infty), ensuring the existence of an inverse function for h(ψ)h(\psi). To clearly illustrate the divergence of h(ψ)h(\psi), we show its behavior under various conditions in Fig. 1. Our calculations reveal that the divergence radius for the four cases is ψ=1/C=1/α\psi=1/C=1/\alpha. These properties are crucial for the following analysis and will significantly influence the thermodynamic characteristics of the black holes under consideration, which will be discussed in detail subsequently.

Refer to caption
Figure 1: h(ψ)ψh(\psi)-\psi diagram of the four regular black holes given in Ref. Bueno:2024dgm ; Konoplya:2024kih . The definition of the function Ł\L can be found in Eq. (42). The lines in four different colors represent the plots of h(ψ)h(\psi) corresponding to four different h(ψ)h(\psi) configurations. Here, α=1\alpha=1 is chosen. It is easy to prove that the radius of convergence is the same in all four cases.

III Thermodynamic topology

Topological methods have emerged as a powerful tool for studying black hole thermodynamics, offering direct insights into their fundamental properties and enabling classification based on the topological invariants Wei:2022dzw . This approach provides a novel perspective on black hole physics, allowing us to understand their behavior from a topological viewpoint. Quasitopological gravity, an important class of gravitational theories discussed earlier, plays a key role in constructing various regular black hole solutions. Our goal is to further classify these solutions by using the framework of thermodynamic topology.

For black hole, the geometry near its event horizon is of critical importance. To analyze the horizon condition, one needs to examine f(r)=0f(r)=0. From Eq. (3), we find that

ψ+ψ(r+)=1r+2.\psi_{+}\equiv\psi(r_{+})=\frac{1}{r_{+}^{2}}. (9)

Here, r+r_{+} denotes the largest root of the equation f(r)=0f(r)=0, which corresponds to the radius of the event horizon. The term ψ+\psi_{+} indicates the value of ψ(r)\psi(r) at r=r+r=r_{+}. Simultaneously, Eq. (6) gives

h(ψ+)=μr+d1.h(\psi_{+})=\frac{\mu}{r_{+}^{d-1}}. (10)

Since hh is closely related to the black hole metric, analyzing the behavior of hh is crucial for further studying the thermodynamic properties of black holes. Firstly, as discussed earlier, hh is a smooth, monotonically increasing function within its domain (0,1/C)(0,1/C). Secondly, when ψ+0\psi_{+}\rightarrow 0, we have r+r_{+}\rightarrow\infty and h(ψ+)0h(\psi_{+})\rightarrow 0; when ψ+1/C\psi_{+}\rightarrow 1/C, r+Cr_{+}\rightarrow\sqrt{C} and h(ψ+)h(\psi_{+})\rightarrow\infty, as listed in Table 1.

      ψ+\psi_{+}       r+r_{+}       h(ψ+)h(\psi_{+})
ψ+0\psi_{+}\rightarrow 0 r+r_{+}\rightarrow\infty h(ψ+)0h(\psi_{+})\rightarrow 0
ψ+1/C\psi_{+}\rightarrow 1/C r+Cr_{+}\rightarrow\sqrt{C} h(ψ+)h(\psi_{+})\rightarrow\infty
Table 1: Boundary behaviors of h(ψ+)h(\psi_{+}).

Thus, r+r_{+} should belong to the range (C,)(\sqrt{C},\infty). The value of the lower bound C\sqrt{C} arises from the analytic domain of function hh itself, as shown in Fig. 1. Another point to note is that C\sqrt{C} obtained here is not necessarily the lower bound of r+r_{+}, which is because that Eq. (9) only indicates that f(r+)=0f(r_{+})=0, while not implies r+r_{+} is the largest root.

Drawing upon the conclusions from Ref. Bueno:2024dgm , one derives the thermodynamic quantities of the regular black holes as follows:

M=(D2)ΩD2r+D116πGh(ψ+),M=\frac{(D-2)\Omega_{D-2}r_{+}^{D-1}}{16\pi G}h(\psi_{+}), (11)
T=14πr+[(D1)r+2h(ψ+)h(ψ+)2],\begin{aligned} T=\frac{1}{4\pi r_{+}}\left[\frac{(D-1)r_{+}^{2}h(\psi_{+})}{h^{\prime}(\psi_{+})}-2\right]\end{aligned}, (12)
S=(D2)ΩD28Gh(ψ+)ψ+D/2dψ+,S=-\frac{(D-2)\Omega_{D-2}}{8G}\int\frac{h^{\prime}(\psi_{+})}{\psi_{+}^{D/2}}\mathrm{d}\psi_{+}, (13)

where MM, TT, SS are the ADM mass, Hawking temperature, and Wald entropy Wald:1993nt , respectively. The volume of the DD-dimensional unit sphere reads

ΩD=2πD+12Γ(D+12).\Omega_{D}=\frac{2\pi^{\frac{D+1}{2}}}{\Gamma(\frac{D+1}{2})}. (14)

Here, we continue to analyze the general expression of the Hawking temperature. Substituting Eqs. (4) and (9) into Eq. (12), one will have

T=14πr+[(D1)(1+n=2nmaxαnψ+n1)1+n=2nmaxnαnψ+n12].T=\frac{1}{4\pi r_{+}}\left[\frac{(D-1)\left(1+\sum_{n=2}^{n_{\max}}\alpha_{n}\psi_{+}^{n-1}\right)}{1+\sum_{n=2}^{n_{\max}}n\alpha_{n}\psi_{+}^{n-1}}-2\right]. (15)

Here, we note that as r+r_{+}\to\infty and ψ+0\psi_{+}\to 0, the two series n=2nmaxαnψ+n1\sum_{n=2}^{n_{\max}}\alpha_{n}\psi_{+}^{n-1} and n=2nmaxnαnψ+n1\sum_{n=2}^{n_{\max}}n\alpha_{n}\psi_{+}^{n-1} in Eq. (15) tend to zero. Consequently, we obtain

limr+T(r+)14πr+(D3).\lim_{r_{+}\to\infty}T(r_{+})\to\frac{1}{4\pi r_{+}}(D-3). (16)

Clearly, since D5D\geq 5, the temperature TT will approach 0+0^{+} for large horizon radius. Conversely, as r+Cr_{+}\rightarrow\sqrt{C}, ψ+\psi_{+} reaches its radius of convergence. Based on the analysis results in Appendix A, we conclude that when r+r_{+} is sufficiently close to C\sqrt{C}, the temperature TT can become negative. Thus, on the interval (C,)(\sqrt{C},\infty), T(r+)T(r_{+}) will have at least one zero and must have an odd number of zeros if the number of zero points is finite. Considering the complexity of multiple horizon scenarios, we mainly focus on the simplest case where T(r+)T(r_{+}) has a single zero in the following discussion. It is worth noting that when T(r+)<0T(r_{+})<0, r+r_{+} will represent the inner horizon rather than the outer horizon, losing its original interpretation. To prevent this, r+r_{+} gains a new lower bound r+minr_{+\mathrm{min}}, given by T(r+)=0T(r_{+})=0. This new lower bound is greater than the previously mentioned lower bound C\sqrt{C}, establishing r+minr_{+\mathrm{min}} as the true lower bound of r+r_{+}.

III.1 Topological approach

To describe the thermodynamical topology of black holes, we consider the generalized free energy of a regular black hole given by

F=MSτ,F=M-\frac{S}{\tau}, (17)

where τ\tau represents an extra variable, which can be regarded as the inverse temperature of the cavity surrounding the black hole. When τ=1/T\tau=1/T, the free energy becomes on-shell. We then examine the vector ϕ\phi, defined as

ϕ=(Fr+,cotΘcscΘ),\phi=\begin{pmatrix}\frac{\partial F}{\partial r_{+}},-\cot\Theta\csc\Theta\end{pmatrix}, (18)

where r+(0,)r_{+}\in\left(0,\infty\right) and Θ(0,π)\Theta\in\left(0,\pi\right). The component ϕΘ\phi^{\Theta} diverges at Θ=0\Theta=0 and Θ=π\Theta=\pi, implying that the vector ϕ\phi points outward.

Following Duan’s theory of ϕ\phi-mapping topological currents YSD ; YSD1 , the topological current can be constructed as

jμ=12πεμνρεabνnaρnb,μ,ν,ρ=0,1,2,j^{\mu}=\frac{1}{2\pi}\varepsilon^{\mu\nu\rho}\varepsilon_{ab}\partial_{\nu}n^{a}\partial_{\rho}n^{b},\quad\mu,\nu,\rho=0,1,2, (19)

where ν=/xν\partial_{\nu}=\partial/\partial x^{\nu} and xν=(τ,r+,Θ)x^{\nu}=(\tau,r_{+},\Theta). To normalize ϕ\phi, we use nr=ϕrϕn^{r}=\frac{\phi^{r}}{\|\phi\|} and nΘ=ϕθϕn^{\Theta}=\frac{\phi^{\theta}}{\|\phi\|}. The topological current jμj^{\mu} is conserved, satisfying

μjμ=0.\partial_{\mu}j^{\mu}=0. (20)

Using the Jacobi tensor ϵabJμ(ϕ/x)=ϵμνρνϕaρϕb\epsilon^{ab}J^{\mu}(\phi/x)=\epsilon^{\mu\nu\rho}\partial_{\nu}\phi^{a}\partial_{\rho}\phi^{b} and the two-dimensional Laplacian Green’s function Δϕalnϕ=2πδ2(ϕ)\Delta_{\phi^{a}}\ln\|\phi\|=2\pi\delta^{2}(\phi), the current jμj^{\mu} can be expressed as

jμ=δ2(ϕ)Jμ(ϕx).j^{\mu}=\delta^{2}(\phi)J^{\mu}\left(\frac{\phi}{x}\right). (21)

Since jμj^{\mu} is nonzero only at ϕa(xi)=0\phi^{a}(x^{i})=0, one can denote its ii-th solution as x=zi\vec{x}=\vec{z}_{i}. The density of the topological current is given by

j0=i=1Nβiηiδ2(xzi).j^{0}=\sum_{i=1}^{N}\beta_{i}\eta_{i}\delta^{2}(\vec{x}-\vec{z}_{i}). (22)

The positive Hopf index, denoted as βi\beta_{i}, quantifies the number of loops that ϕα\phi^{\alpha} traces in the vector ϕ\phi space as xμx^{\mu} circles the zero point ziz_{i}. Meanwhile, the Brouwer degree, denoted as ηi\eta_{i}, is given by the expression ηi=sign(J0(ϕ/x)zi)=±1\eta_{i}=\mathrm{sign}(J^{0}(\phi/x)_{z_{i}})=\pm 1. Within a certain parameter region Σ\Sigma, the associated topological number can be determined by

W=Σj0d2x=i=1Nβiηi=i=1Nwi,W=\int_{\Sigma}j^{0}d^{2}x=\sum_{i=1}^{N}\beta_{i}\eta_{i}=\sum_{i=1}^{N}w_{i}, (23)

where wiw_{i} represents the winding number associated with the ii-th zero point of ϕ\phi contained within Σ\Sigma. If loops Σ\partial\Sigma and Σ\partial\Sigma^{\prime} enclose the same zero point of ϕ\phi, we will have the same winding number. If there are no zero points in the enclosed region, W=0W=0. When Σ\Sigma covers part or all of the parameter space, it reveals the local or global topological number, respectively.

III.2 Thermodynamic number

To study the thermodynamical topology, we will further analyze the asymptotic behavior of the thermodynamic quantities. To obtain Eq. (18), we need to differentiate Eq. (17) with respect to r+r_{+}, resulting in Wei24

Fr+=Mr+1τSr+=Sr+(MS1τ)=Sr+(T1τ).\frac{\partial F}{\partial r_{+}}=\frac{\partial M}{\partial r_{+}}-\frac{1}{\tau}\frac{\partial S}{\partial r_{+}}=\frac{\partial S}{\partial r_{+}}\left(\frac{\partial M}{\partial S}-\frac{1}{\tau}\right)=\frac{\partial S}{\partial r_{+}}\left(T-\frac{1}{\tau}\right). (24)

Here, we have used the first law of black hole thermodynamics. Further considering the most general expressions (11), (12), and (13), we shall obtain the general conclusions.

By substituting Eq. (9) into Eq. (13), we obtain

S=(D2)ΩD28G2h(ψ+)r+3ψ+D/2dr+.S=\frac{(D-2)\Omega_{D-2}}{8G}\int\frac{2h^{\prime}(\psi_{+})r_{+}^{-3}}{\psi_{+}^{D/2}}\mathrm{d}r_{+}. (25)

Here, the prime symbol denotes the first derivative of function hh with respect to ψ\psi, which is given by

h(ψ)=1+n=2nmaxnαnψn1.h^{\prime}(\psi)=1+\sum_{n=2}^{n_{\max}}n\alpha_{n}\psi^{n-1}. (26)

Thus, we can derive

Sr+=(D2)ΩD28G2h(ψ+)r+3ψ+D/2.\frac{\partial S}{\partial r_{+}}=\frac{(D-2)\Omega_{D-2}}{8G}\frac{2h^{\prime}(\psi_{+})r_{+}^{-3}}{\psi_{+}^{D/2}}. (27)

Clearly, it follows that when r+>0r_{+}>0, it is always true that

Sr+>0.\frac{\partial S}{\partial r_{+}}>0. (28)

Note that in Eq. (24), τ\tau is positive. Considering the conclusions we reached earlier, we can derive that when r+r+minr_{+}\rightarrow r_{+min}, T0T\rightarrow 0, and T1/τ<0T-1/\tau<0; when r+r_{+}\rightarrow\infty, T0T\rightarrow 0, and T1/τ<0T-1/\tau<0. Therefore, by combining the signs of two terms Sr+\frac{\partial S}{\partial r_{+}} and (T1τ)\left(T-\frac{1}{\tau}\right) in Eq (24), we can conclude that when r+r_{+} takes r+minr_{+min} or tends to infinity, Fr+\frac{\partial F}{\partial r_{+}} is less than zero, as listed in Table 2.

      r+r_{+}       TT       T1/τT-1/\tau       Fr+\frac{\partial F}{\partial r_{+}}
r+r+minr_{+}\rightarrow r_{+min} T0T\rightarrow 0^{-} T1/τ<0T-1/\tau<0 Fr+<0\frac{\partial F}{\partial r_{+}}<0
r+r_{+}\rightarrow\infty T0+T\rightarrow 0^{+} T1/τ<0T-1/\tau<0 Fr+<0\frac{\partial F}{\partial r_{+}}<0
Table 2: Boundary behavior of Fr+\frac{\partial F}{\partial r_{+}}.

This implies that in the Θr+\Theta-r_{+} diagram, as r+r_{+} approaches the lower bound and infinity, the direction of the vector ϕ\phi remains consistent (leftwards). At the same time, we will note that when Θ\Theta approaches zero, the direction of the arrows in the Θr+\Theta-r_{+} diagram remains fixed (downwards); and when Θ\Theta approaches π\pi, the direction of the arrows in the Θr+\Theta-r_{+} diagram also remains fixed (upwards). The behaviors of the above vectors at the boundaries have been illustrated in Fig. 2. Employing with above results, we can derive some universal properties on a global scale. If we consider a sufficiently large loop that encompasses all the zeros of Fr+\frac{\partial F}{\partial r_{+}} in the Θr+\Theta-r_{+} diagram, the global topological number can be obtained as zero

W=0.W=0. (29)
Refer to caption
Figure 2: Θr+\Theta-r_{+} schematic diagram of the regular black holes. According to our discussion of the general case, the Φr+\Phi-r_{+} diagram of such black holes exhibits certain common features. The vector direction on the boundaries is fixed, thus they have the same winding number. This preserves the technical terminology in a way that should be suitable for your context in black hole physics.

Since all the discussions derived thus far are general, the black holes obtained under the aforementioned considerations can be classified as the same type with a topological number W=0W=0. Combining with the vector direction on the boundaries, we find their topological class actually is W0+W^{0+} suggested in Ref. Wei24 . Meanwhile, we can also derive the result that the small and large black hole states are stable and unstable from the general property of W0+W^{0+}. This results imply that the black hole solutions constructed from the pure gravity approach outlined above belong to the same topological class, which indicates that certain inherent topological properties are embedded in the method of constructing such regular black holes. Specifically, the geometric and topological constraints of these construction techniques ensure that universal features are preserved across various gravitational backgrounds.

A key requirement in our analysis is the condition that the temperature T(r+)T(r_{+}) has a single zero point. This constraint directly influences the topological classification and physical interpretation of the black hole states. In black hole thermodynamics, topological numbers W=1W=1 and W=1W=-1 correspond to stable and unstable black hole solutions, respectively Wei:2022dzw . A global topological number W=0W=0 thus indicates a balance between stable and unstable black hole states. This balance arises because the positive topological numbers of stable states are exactly canceled by the negative topological numbers of unstable states, leading to a net topological number of zero.

This equilibrium is not coincidental but is deeply rooted in the inherent symmetry and geometric characteristics of the gravitational field. It reveals an underlying topological structure, wherein perturbations in the gravitational field within a multidimensional spacetime background generate equal numbers of stable and unstable solutions, thereby maintaining a global topological balance. This phenomenon provides a novel perspective on the construction and evolution of regular black holes, highlighting their significance within the framework of generalized gravity theories.

In the subsequent discussion, we will provide specific examples to verify our perspective, exploring the topological structure and thermodynamic properties of these regular black hole solutions.

IV Two regular black hole solutions

In the previous section, we consider the thermodynamical topology for the general case and find that the total topological number vanishes. In this section, we will take two specific black hole solutions as specific examples to verify our above results.

IV.1 Regular black hole with αn=nαn1\alpha_{n}=n\alpha^{n-1}

Taking αn=nαn1\alpha_{n}=n\alpha^{n-1} and substituting it into Eq. (4), one will find

h(ψ)=ψ(1αψ)2.h(\psi)=\frac{\psi}{(1-\alpha\psi)^{2}}. (30)

Then, the metric function reads

f(r)=12mr2rD1+2αm+r2(D1)+4αmrD1.f(r)=1-\frac{2mr^{2}}{r^{D-1}+2\alpha m+\sqrt{r^{2(D-1)}+4\alpha mr^{D-1}}}. (31)

Through Eqs. (11), (12), and (13), we obtain the following thermodynamic quantities

M\displaystyle M =\displaystyle= (D2)πD32r+D+18(r+2α)2Γ(D12),\displaystyle\frac{(D-2)\pi^{\frac{D-3}{2}}r_{+}^{D+1}}{8(r_{+}^{2}-\alpha)^{2}\Gamma(\frac{D-1}{2})}, (32)
T\displaystyle T =\displaystyle= 3r+2Dr+2+α+Dα4πr+3+4παr+,\displaystyle-\frac{3r_{+}^{2}-Dr_{+}^{2}+\alpha+D\alpha}{4\pi r_{+}^{3}+4\pi\alpha r_{+}}, (33)
S\displaystyle S =\displaystyle= πD12r+D((D2)r+4+D(r+2α)22F1[2,1D2;2D2;αr+2])4(r+3αr+)2Γ(D12).\displaystyle\frac{\pi^{\frac{D-1}{2}}r_{+}^{D}\left(-(D-2)r_{+}^{4}+D(r_{+}^{2}-\alpha)^{2}\ _{2}F_{1}[2,1-\frac{D}{2};2-\frac{D}{2};\frac{\alpha}{r_{+}^{2}}]\right)}{4(r_{+}^{3}-\alpha r_{+})^{2}\Gamma(\frac{D-1}{2})}. (34)

It is worth recalling that, using Eq. (8), we determine that the lower bound for r+r_{+} is α\sqrt{\alpha}, which is constrained by the convergence radius of h(ψ+)h(\psi_{+}). Furthermore, from Eq. (33), it is straightforward to see that this regular black hole becomes extremal when r+=D+1αD3r_{+}=\frac{\sqrt{D+1}\sqrt{\alpha}}{\sqrt{D-3}}. By comparing these two lower bounds, it is clear that for D5D\geq 5, this expression provides the true lower bound for r+r_{+}, avoiding the cases of negative temperature. Thus, the horizon radius r+r_{+} lies in the range (D+1αD3,)\left(\frac{\sqrt{D+1}\sqrt{\alpha}}{\sqrt{D-3}},\infty\right). In Fig. 3(a) and Fig. 4(a), we present our results in the r+τr_{+}-\tau graphs, specifically considering α=0.1\alpha=0.1 for D=5D=5 and D=7D=7, respectively. The lower bounds of r+r_{+} are described by black dashed lines. As r+r_{+} approaches this bound, τ\tau diverges to infinity. When r+r_{+} increases from its lower bound, τ\tau decreases to a turning point, corresponding to the condition Tr+=0\frac{\partial T}{\partial r_{+}}=0, marked by black dots in the figures. Beyond this point, as r+r_{+} continues to increase, τ\tau gradually increases again, eventually diverging to infinity as r+r_{+}\rightarrow\infty. This behavior is consistent with the properties we have previously analyzed.

Based on Eqs. (32), (34), and (17), we can derive the expression for the free energy

F=πD32r+D[(D2)r+3(2πr++τ)2Dπ(r+2α)22F1[2,1D2;2D2;αr+2]]8(r+3αr+)2τΓ(D12),F=\frac{\pi^{\frac{D-3}{2}}r_{+}^{D}\left[(D-2)r_{+}^{3}(2\pi r_{+}+\tau)-2D\pi(r_{+}^{2}-\alpha)^{2}\,\ _{2}F_{1}[2,1-\frac{D}{2};2-\frac{D}{2};\frac{\alpha}{r_{+}^{2}}]\right]}{8(r_{+}^{3}-\alpha r_{+})^{2}\tau\Gamma\left(\frac{D-1}{2}\right)}, (35)

and the components of the vector ϕ\phi (18) can be calculated as

ϕr+\displaystyle\phi^{r_{+}} =[(2+D)πD32r+D[4πr+(r+2+α)(3+D)r+2τ+(1+D)ατ]]8(r+2α)3τΓ(12(1+D)),\displaystyle=-\frac{\left[(-2+D)\pi^{\frac{D-3}{2}}r_{+}^{D}\left[4\pi r_{+}\left(r_{+}^{2}+\alpha\right)-(-3+D)r_{+}^{2}\tau+(1+D)\alpha\tau\right]\right]}{8\left(r_{+}^{2}-\alpha\right)^{3}\tau\Gamma\left(\frac{1}{2}(-1+D)\right)}, (36)
ϕΘ\displaystyle\phi^{\Theta} =cotΘcscΘ.\displaystyle=-\cot\Theta\csc\Theta. (37)

Using the conclusions from Eq. (36) and Eq. (37), we can locate and represent the zeros of Fr+\frac{\partial F}{\partial r_{+}}, employing the vector ϕ\phi field diagram for characterization. We select values of τ\tau on the right of the turning point in Fig. 3(a) (r+τr_{+}-\tau graph), ensuring that for a given τ\tau, the r+τr_{+}-\tau plot has two intersections, thereby guaranteeing two zeros in the Θr+\Theta-r_{+} plot. The two zeros, labeled ZP1aZP^{a}_{1} and ZP2aZP^{a}_{2} from left to right, are marked by black dots in Fig. 3(b). Here r0r_{0} is an arbitrary length scale set by the size of a cavity surrounding the black hole. To determine the winding number of the zero point, we need to construct a closed loop around it and count the changes in the direction of the vector. For this purpose, we parametrize the closed loop CaC^{a} using the angle ϑ\vartheta, as follows:

{r+/r0=acosϑ+rc,Θ=bsinϑ+π2.\displaystyle\left\{\begin{aligned} r_{+}/r_{0}&=a\cos\vartheta+r_{c},\\ \Theta&=b\sin\vartheta+\frac{\pi}{2}.\end{aligned}\right. (38)

As ϑ\vartheta varies from 0 to 2π2\pi, one traces a counterclockwise path around the zero point along this closed loop. To track the change in the direction of the vector, it is useful to define the deflection angle Wei:2020rbh

Ω(ϑ)=Cϵabnainbdxi.\Omega(\vartheta)=\oint_{C}\epsilon_{ab}n^{a}\partial_{i}n^{b}dx^{i}. (39)

Then the winding number shall be

W=Ω(2π)/2π.W=\Omega(2\pi)/2\pi. (40)

Using the method described above, by examining the deflection angle along the green circle C1aC^{a}_{1} and the blue circle C2aC^{a}_{2} shown in Fig. 3(c), we determine that the winding number for ZP1aZP^{a}_{1} is W1=1W_{1}=1, indicating a thermodynamical stable black hole state, while for ZP2aZP^{a}_{2}, it is W2=1W_{2}=-1, indicating an unstable black hole state. These support that such topological class is W0+W^{0+} rather than W0W^{0-}. The black circle CaC^{a} represented by the black line surrounds the entire global structure shown in Fig. 3(c), leads to the winding number W=0W=0. It is noteworthy that if we choose a τ\tau to the left of the turning point, where no intersection points exist in the r+τr_{+}-\tau graph, there will be no zeros in the Θr+\Theta-r_{+} graph. Certainly, by summing the topological numbers of all the local zeros, we can also obtain the total topological number, which is W=1+1=0W=-1+1=0. Using the same method, we can obtain the results for D=7D=7 and α=0.1\alpha=0.1, with similar results shown in Figs. 4(a), 4(b), and 4(c).

Certainly, we can discuss the cases corresponding to different values of α\alpha in various dimensions, all of which show consistent results, though we will not present all of them here. It is quite natural to observe this, as our previous discussions were based on general conditions, and the values of DD and α\alpha do not affect our conclusions. In other words, when α\alpha and DD vary independently, they do not affect the asymptotic behavior of τ\tau or Fr+\frac{\partial F}{\partial r_{+}}. Regardless of how α\alpha and DD vary, this regular black hole consistently exhibits a topological number of W=0W=0, indicating that black holes in different dimensions exhibit the same topological behavior. Therefore, this result supports our earlier conclusions.

Refer to caption
(a)
Refer to caption
(b)
Refer to caption
(c)
Figure 3: Schematic diagrams characterizing the thermodynamic topology of the regular black hole from pure gravity with αn=nαn1\alpha_{n}=n\alpha^{n-1} for the case of D=5D=5 and α=0.1\alpha=0.1. (a) The r+τr_{+}-\tau diagram. The left side of the inflection point (marked with black dots) indicates no black hole state, and the right side indicates the presence of two black hole states. The black dashed line represents the lower bound of r+r_{+}; values less than this will result in T<0T<0. (b) Θr+\Theta-r_{+} diagram. The red arrows depict the unit vector field nn on a section of the Θr+\Theta-r_{+} plane. Here r0r_{0} is an arbitrary length scale set by the size of a cavity surrounding the black hole. The zero points (ZP1aZP^{a}_{1} and ZP2aZP^{a}_{2}) are indicated by black dots. The green contour C1aC^{a}_{1} and the blue contour C2aC^{a}_{2} are two closed loops surrounding the zero points. And the black contour CaC^{a} representing the closed loop surrounding the entire global structure. (c) The deflection angle Ω\Omega as a function of ϑ\vartheta for contours C1aC^{a}_{1} (green), C2aC^{a}_{2} (blue) and CaC^{a} (black).
Refer to caption
(a)
Refer to caption
(b)
Refer to caption
(c)
Figure 4: Schematic diagrams characterizing the thermodynamic topology of the regular black hole from pure gravity with αn=nαn1\alpha_{n}=n\alpha^{n-1} for the case of D=7D=7 and α=0.1\alpha=0.1. (a) The r+τr_{+}-\tau diagram. The left side of the inflection point (marked with black dots) indicates no black hole state, and the right side indicates the presence of two black hole states. The black dashed line represents the lower bound of r+r_{+}; values less than this will result in T<0T<0. (b) Θr+\Theta-r_{+} diagram. The red arrows depict the unit vector field nn on a section of the Θr+\Theta-r_{+} plane. Here r0r_{0} is an arbitrary length scale set by the size of a cavity surrounding the black hole. The zero points (ZP1bZP^{b}_{1} and ZP2bZP^{b}_{2}) are indicated by black dots. The green contour C1bC^{b}_{1} and the blue contour C2bC^{b}_{2} are two closed loops surrounding the zero points. And the black contour CbC^{b} representing the closed loop surrounding the entire global structure. (c) The deflection angle Ω\Omega as a function of ϑ\vartheta for contours C1bC^{b}_{1} (green), C2bC^{b}_{2} (blue) and CbC^{b} (black).

IV.2 Dymnikova black hole from corrections

The Dymnikova black hole can be interpreted as a solution to the Einstein equations (in four or higher dimensions) with an energy-momentum tensor describing matter that exhibits exponential decay Dymnikova:1992ux ; Paul:2023pqn . Konoplya and Zhidenko derived this solution using the pure gravity approach in Ref. Konoplya:2024kih . Although the expression of h(ψ)h(\psi) differs from that in the previous subsection, detailed analysis yields similar results.

The function h(ψ)h(\psi) for this Dymnikova black hole reads Konoplya:2024kih

h(ψ)=ψ1+αψŁ(e1/αψαψ),α>0,h(\psi)=\frac{\psi}{1+\alpha\psi\L\left(-\frac{e^{-1/\alpha\psi}}{\alpha\psi}\right)},\quad\alpha>0, (41)

where Ł(x)\L(x) is the principal Lambert function, which fulfills the equation

Ł(x)eŁ(x)=x.\L(x)e^{\L(x)}=x. (42)

Then the metric function is

f(r)=1μrD3(1erD1αμ).f(r)=1-\frac{\mu}{r^{D-3}}\left(1-e^{-\cfrac{r^{D-1}}{\alpha\mu}}\right). (43)

We can naturally obtain the corresponding thermodynamic quantities by Eqs. (11), (12), and (13),

M\displaystyle M =\displaystyle= (D2)πD32r+D18Γ(D12)(r+2+αA),\displaystyle\frac{(D-2)\pi^{\frac{D-3}{2}}r_{+}^{D-1}}{8\Gamma(\frac{D-1}{2})\left(r_{+}^{2}+\alpha A\right)}, (44)
T\displaystyle T =\displaystyle= (D3)+(D1)A4πr+,\displaystyle\frac{\left(D-3\right)+(D-1)A}{4\pi r_{+}}, (45)
S\displaystyle S =\displaystyle= (D2)πD124Γ(D12)2r+D1(1+A)(r2+αA)dr+,\displaystyle\frac{\left(D-2\right)\pi^{\frac{D-1}{2}}}{4\Gamma(\frac{D-1}{2})}\int\frac{2r_{+}^{D-1}}{\left(1+A\right)\left(r^{2}+\alpha A\right)}\mathrm{d}r_{+}, (46)

where

A=Ł(exp(r+2α)r+2α).A=\L\left(-\frac{exp\left(-{\frac{r_{+}^{2}}{\alpha}}\right)r_{+}^{2}}{\alpha}\right).

Similar to the previous section, it is straightforward to verify that the lower bound for r+r_{+}, determined by the convergence radius of the function h(ψ)h(\psi), is α\sqrt{\alpha}. To avoid negative temperatures, we obtain r+minr_{+\text{min}}, and the event horizon must lie above this value. While providing a general expression is challenging, we illustrate the r+minr_{+\text{min}} for different cases using dashed black lines in Fig. 5(a) and Fig. 6(a). By substituting specific values, it is easy to verify that r+minr_{+\text{min}} is always greater than α\sqrt{\alpha}. Thus, the lower bound derived from the zero-temperature condition is indeed the true lower bound for r+r_{+}.

Based on Eqs. (44), (46), and (17), we can derive the expression for the free energy

F=(2+D)π3+D28Γ(1+D2)[4πτr+D1(1+A)(r+2+αA)dr++r+D1r+2+αA],F=\frac{(-2+D)\pi^{\frac{-3+D}{2}}}{8\Gamma\left(\frac{-1+D}{2}\right)}\left[\ \frac{-4\pi}{\tau}\int{\frac{r_{+}^{D-1}}{\left(1+A\right)\left(r_{+}^{2}+\alpha A\right)}\,\mathrm{d}r_{+}}+\frac{r_{+}^{D-1}}{r_{+}^{2}+\alpha A}\right], (47)

and the components of the vector ϕ\phi can be calculated as

ϕr+\displaystyle\phi^{r_{+}} =(2+D)π3+D2r+2+D[4πr++(3+D)τ+(1+D)τA]8τΓ(1+D2)(1+A)(r+2+αA),\displaystyle=\frac{(-2+D)\pi^{\frac{-3+D}{2}}r_{+}^{-2+D}\left[-4\pi r_{+}+(-3+D)\tau+(-1+D)\tau\,A\right]}{8\tau\Gamma\left(\frac{-1+D}{2}\right)\left(1+A\right)\left(r_{+}^{2}+\alpha A\right)}, (48)
ϕΘ\displaystyle\phi^{\Theta} =cotΘcscΘ.\displaystyle=-\cot\Theta\csc\Theta. (49)

Using the same approach as in the previous subsection, we can obtain the corresponding r+τr_{+}-\tau diagram, the vector field ϕ\phi, and the deflection angle shown in Fig. 5 and Fig. 6. The cases for α=0.1\alpha=0.1, D=5D=5, and D=7D=7 are considered separately. Similarly, we can examine cases for different values of α\alpha across various dimensions, all of which yield consistent results, although we will not present every case here. The variations in the two parameters, α\alpha and DD, do not affect the asymptotic behavior of τ\tau or Fr+\frac{\partial F}{\partial r_{+}}, ultimately leading to a topological number W=0W=0, which further confirms our analysis.

Refer to caption
(a)
Refer to caption
(b)
Refer to caption
(c)
Figure 5: Schematic diagrams characterizing the thermodynamic topology of the of the Dymnikova black hole for the case of D=5D=5 and α=0.1\alpha=0.1. (a) The r+τr_{+}-\tau diagram. The left side of the inflection point(marked with black dots) indicates no black hole state, and the right side indicates the presence of two black hole states. The black dashed line represents the lower bound of r+r_{+}; values less than this will result in T<0T<0. (b) Θr+\Theta-r_{+} diagram. The red arrows depict the unit vector field nn on a section of the Θr+\Theta-r_{+} plane. Here r0r_{0} is an arbitrary length scale set by the size of a cavity surrounding the black hole. The zero points (ZP1cZP^{c}_{1} and ZP2cZP^{c}_{2}) are indicated by black dots. The green contour C1cC^{c}_{1} and the blue contour C2cC^{c}_{2} are two closed loops surrounding the zero points. And the black contour CcC^{c} representing the closed loop surrounding the entire global structure. (c) The deflection angle Ω\Omega as a function of ϑ\vartheta for contours C1cC_{1}^{c} (green), C2cC_{2}^{c} (blue) and CcC^{c} (black).
Refer to caption
(a)
Refer to caption
(b)
Refer to caption
(c)
Figure 6: Schematic diagrams characterizing the thermodynamic topology of the of the Dymnikova black hole for the case of D=7D=7 and α=0.1\alpha=0.1. (a) The r+τr_{+}-\tau diagram. The left side of the inflection point(marked with black dots) indicates no black hole state, and the right side indicates the presence of two black hole states. The black dashed line represents the lower bound of r+r_{+}; values less than this will result in T<0T<0. (b) Θr+\Theta-r_{+} diagram. The red arrows depict the unit vector field nn on a section of the Θr+\Theta-r_{+} plane. Here r0r_{0} is an arbitrary length scale set by the size of a cavity surrounding the black hole. The zero points (ZP1dZP^{d}_{1} and ZP2dZP^{d}_{2}) are indicated by black dots. The green contour C1dC^{d}_{1} and the blue contour C2dC^{d}_{2} are two closed loops surrounding the zero points. And the black contour CdC^{d} representing the closed loop surrounding the entire global structure. (c) The deflection angle Ω\Omega as a function of ϑ\vartheta for contours C1dC^{d}_{1} (green), C2dC^{d}_{2} (blue) and CdC^{d} (black).

V Discussion and Conclusion

In this paper, we analyzed the thermodynamical topology of regular black holes arising from pure gravity from a general perspective. We found that regular black holes constructed using the method from Ref. Bueno:2024dgm exhibit certain similar thermodynamic properties. We also observed that these black holes, with at most two horizons, have similar topological properties and the same topological number, W=0W=0, as revealed by the asymptotic behavior of the constructed vector. In particular, they belong to W0+W^{0+} topological classification given in Ref. Wei24 , through which one can find that the the innermost and outermost black hole states are local thermodynamical stable and unstable. Given the general nature of our discussion, this result suggests that the thermodynamic topological property is intrinsically linked to the construction method of these regular black holes.

Specifically, we consider two representative examples of regular black hole solutions: one characterized by αn=nαn1\alpha_{n}=n\alpha^{n-1} and the Dymnikova black holes. We demonstrated the zero points of the free energy by constructing a vector field ϕ=(Fr+,cotΘcscΘ)\phi=\begin{pmatrix}\frac{\partial F}{\partial r_{+}},-\cot\Theta\csc\Theta\end{pmatrix}. Through detailed calculations, we observed that for small τ\tau, there are no black hole states, leading to a vanishing topological number. However, for large τ\tau, two zero points of the constructed vector field emerge, corresponding to the regular black hole states. We then enclosed these zeros with a closed loop and counted the changes in the vector’s direction to determine the winding number associated with each zero point. Consequently, we obtained the overall global winding number, which yields a topological number of W=0W=0. This result further supports our conclusion that the regular black holes (with at most two horizons in pure gravity) belong to the same topological class W0+W^{0+}. Moreover, the small and large black hole states are stable and unstable indicated from their winding numbers, which strongly supports the thermodynamical properties of W0+W^{0+}. Although we only provide a few specific examples, this does not affect the generality of our conclusion, as the choice of specific parameters does not alter the results of our previous discussion on the asymptotic behavior of τ\tau and Fr+\frac{\partial F}{\partial r_{+}}.

On the other hand, our discussion has certain limitations. We only considered the case where the Hawking temperature has a single zero point and did not address more complex scenarios involving multiple horizons. If αn\alpha_{n} is chosen as an appropriate function, such cases may arise, requiring more detailed study. Nevertheless, for the regular black hole solutions constructed by this method that have been identified so far, our findings remain applicable. Furthermore, the reasons behind the topological consistency of these black holes, and whether there are deeper underlying factors in their construction methods, merit further reflection and discussion.

Appendix A Behavior of Hawking temperature at the lower bound r+=Cr_{+}=\sqrt{C}

We provide a simple proof for the conclusion in Sec. III about Eq. (12) or (15) in the main text, namely, that when r+r_{+} is sufficiently close to C\sqrt{C}, the temperature TT can become negative. Before proceeding, we first present a claim: for smooth function f:(a,b)f:(a,b)\rightarrow\mathbb{R} satisfying the condition,

limxbf(x)=+,\lim_{x\rightarrow b^{-}}f(x)=+\infty, (50)

it is always possible to find a sequence Ξ={ξn,n+|a<ξn<b,ξnξn+1,n+}\Xi=\left\{\xi_{n},n\in\mathbb{N}^{+}\middle|a<\xi_{n}<b,\xi_{n}\leq\xi_{n+1},\forall n\in\mathbb{N}^{+}\right\} satisfying limnξn=b\lim_{n\rightarrow\infty}\xi_{n}=b such that limn1/f(ξn)=0\lim_{n\rightarrow\infty}1/f^{\prime}(\xi_{n})=0.

We presents a simple proof for this claim. Since the left-sided limit of f(x)f(x) at x=bx=b is infinity, f(x)f(x) can exceed any number if xx is sufficiently close to bb. Therefore, we can always find such a sequence {xn,n+|a<xn<xn+1<b,f(xn)<f(xn+1),n+;limnxn=b}\left\{x_{n},n\in\mathbb{N}^{+}\middle|a<x_{n}<x_{n+1}<b,f\left(x_{n}\right)<f\left(x_{n+1}\right),\forall n\in\mathbb{N}^{+};\lim_{n\rightarrow\infty}x_{n}=b\right\}. Now we construct the sequence Ξ\Xi by finding the value of xx in the closed interval [x1,xn][x_{1},x_{n}] that maximizes f(x)f^{\prime}(x), and let one of these values be ξn1\xi_{n-1}. This implies the following inequality holds,

f(xn)f(x1)+f(ξn1)(xnx1)<f(x1)+f(ξn1)(bx1),n2.f(x_{n})\leq f(x_{1})+f^{\prime}(\xi_{n-1})(x_{n}-x_{1})<f(x_{1})+f^{\prime}(\xi_{n-1})(b-x_{1}),\ \forall n\geq 2. (51)

It is evident that the constructed ξn\xi_{n} satisfies a<ξnξn+1<b,n+a<\xi_{n}\leq\xi_{n+1}<b,\forall n\in\mathbb{N}^{+}. This indicates that the sequence Ξ\Xi has a limit, denoted as ξ\xi_{\infty}. Furthermore, it can be shown that f(ξn)f(ξn+1),n+f^{\prime}(\xi_{n})\leq f^{\prime}(\xi_{n+1}),\forall n\in\mathbb{N}^{+}. Therefore, if f(ξ)f^{\prime}(\xi_{\infty}) is finite, we have

f(xn)<f(x1)+f(ξn1)(bx1)f(x1)+f(ξ)(bx1),n2.f(x_{n})<f(x_{1})+f^{\prime}(\xi_{n-1})(b-x_{1})\leq f(x_{1})+f^{\prime}(\xi_{\infty})(b-x_{1}),\ \forall n\geq 2. (52)

This contradicts limxf(xn)=\lim_{x\rightarrow\infty}f(x_{n})=\infty. Hence, limxf(xn)=f(ξ)=\lim_{x\rightarrow\infty}f^{\prime}(x_{n})=f^{\prime}(\xi_{\infty})=\infty. Since f(x)f^{\prime}(x) is finite for any x(a,b)x\in(a,b), we have ξ=b\xi_{\infty}=b. Thus, the proof of this claim is complete.

In Sec. III, h(ψ+)h(\psi_{+}) is a smooth function on the interval (0,1/C)(0,1/C), and limx(1/C)h(ψ+)=+\lim_{x\rightarrow(1/C)^{-}}h(\psi_{+})=+\infty. This implies that ln(h(ψ+))\ln(h(\psi_{+})) also exhibits the same property. Our claim asserts that there exists a sequence Ξ={ξn,n+|0<ξn<1/C,ξnξn+1,n+}\Xi=\left\{\xi_{n},n\in\mathbb{N}^{+}\middle|0<\xi_{n}<1/C,\xi_{n}\leq\xi_{n+1},\forall n\in\mathbb{N}^{+}\right\} satisfying limnξn=1/C\lim_{n\rightarrow\infty}\xi_{n}=1/C, such that

limnh(ξn)h(ξn)=0.\lim_{n\rightarrow\infty}\frac{h(\xi_{n})}{h^{\prime}(\xi_{n})}=0. (53)

For the temperature,

limnT(ξn)=12πC.\lim_{n\rightarrow\infty}T(\xi_{n})=-\frac{1}{2\pi\sqrt{C}}. (54)

It is noteworthy that Eq. (53) implies that if limψ+(1/C)h(ψ+)h(ψ+)\lim_{\psi_{+}\rightarrow(1/C)^{-}}\frac{h(\psi_{+})}{h^{\prime}(\psi_{+})} exists, then its result must be zero, which strengthens all the conclusions. Nevertheless, we can ascertain that near the lower bound r+=Cr_{+}=\sqrt{C}, there exists an r+r_{+} for which the temperature is negative. This, combined with the discussion in Sec. III, indicates that the temperature must have at least one zero point.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Grants No. 12475055, No. 12075103, and No. 12247101).

References

  • (1) S. W. Hawking, Particle creation by black holes, Commun. Math. Phys. 43, 199 (1975), Erratum: Commun. Math. Phys. 46, 206 (1976).
  • (2) S. W. Hawking, Black holes and thermodynamics, Phys. Rev. D 13, 191 (1976).
  • (3) J. M. Bardeen, B. Carter, S. W. Hawking, The Four laws of black hole mechanics, Commun. Math. Phys. 31, 161 (1973).
  • (4) G. W. Gibbons and S. W. Hawking, Action integrals and partition functions in quantum gravity, Phys. Rev. D 15, 2752 (1977).
  • (5) S. W. Wei, Y. X. Liu and R. B. Mann, Black Hole Solutions as Topological Thermodynamic Defects, Phys. Rev. Lett. 129, 191101 (2022), [arXiv:2208.01932 [gr-qc]].
  • (6) D. Wu and S. Q. Wu, Topological classes of thermodynamics of rotating AdS black holes, Phys. Rev. D 107, 084002 (2023), [arXiv:2301.03002 [hep-th]].
  • (7) D. Wu, Topological classes of rotating black holes, Phys. Rev. D 107, 024024 (2023), [arXiv:2211.15151 [gr-qc]].
  • (8) N. C. Bai, L. Li and J. Tao, Topology of black hole thermodynamics in Lovelock gravity, Phys. Rev. D 107, 064015 (2023), [arXiv:2208.10177 [gr-qc]].
  • (9) C. Liu and J. Wang, Topological natures of the Gauss-Bonnet black hole in AdS space, Phys. Rev. D 107, 064023 (2023), [arXiv:2211.05524 [gr-qc]].
  • (10) S. P. Wu and S. W. Wei, Thermodynamical topology of quantum BTZ black hole, Phys. Rev. D 110, 024054 (2024), [arXiv:2403.14167 [gr-qc]].
  • (11) Z. Q. Chen and S. W. Wei, Thermodynamics, Ruppeiner geometry, and topology of Born-Infeld black hole in asymptotic flat spacetime, Nucl. Phys. B 996, 116369 (2023).
  • (12) Z. Q. Chen and S. W. Wei, Thermodynamical topology with multiple defect curves for dyonic AdS black holes, [arXiv:2405.07525 [hep-th]].
  • (13) C. Fang, J. Jiang and M. Zhang, Revisiting thermodynamic topologies of black holes, JHEP 01, 102 (2023), [arXiv:2211.15534 [gr-qc]].
  • (14) N. Chatzifotis, P. Dorlis, N. E. Mavromatos and E. Papantonopoulos, Thermal stability of hairy black holes, Phys. Rev. D 107, 084053 (2023), [arXiv:2302.03980 [gr-qc]].
  • (15) S. W. Wei, Y. P. Zhang, Y. X. Liu and R. B. Mann, Static spheres around spherically symmetric black hole spacetime, Phys. Rev. Res. 5, 043050 (2023), [arXiv:2303.06814 [gr-qc]].
  • (16) Y. Du and X. Zhang, Topological classes of black holes in de-Sitter spacetime, Eur. Phys. J. C 83, 927 (2023), [arXiv:2303.13105 [gr-qc]].
  • (17) D. Chen, Y. He and J. Tao, Topological classes of higher-dimensional black holes in massive gravity, Eur. Phys. J. C 83, 872 (2023), [arXiv:2306.13286 [gr-qc]].
  • (18) M. Rizwan and K. Jusufi, Topological classes of thermodynamics of black holes in perfect fluid dark matter background, Eur. Phys. J. C 83, 944 (2023), [arXiv:2310.15182 [gr-qc]].
  • (19) B. Hazarika and P. Phukon, Topology of restricted phase space thermodynamics in Kerr-Sen-Ads black holes, [arXiv:2405.02328 [hep-th]].
  • (20) S. W. Hawking, G. F. R. Ellis, The Large Scale Structure of Space-Time, Cambridge Monographson Mathematical Physics. Cambridge UniversityPress, 2, 2023.
  • (21) J. M. M. Senovilla, Singularity Theorems and Their Consequences, Gen. Rel. Grav. 30, 701 (1998), [arXiv:1801.04912 [gr-qc]].
  • (22) R. Penrose, Gravitational collapse and space-time singularities, Phys. Rev. Lett. 14, 57 (1965).
  • (23) C. Lan, H. Yang, Y. Guo, and Y. G. Miao, Regular Black Holes: A Short Topic Review, Int. J. Theor. Phys. 62, 202 (2023), [arXiv:2303.11696 [gr-qc]].
  • (24) A. D. Sakharov, Nachal’naia stadija rasshirenija Vselennoj i vozniknovenije neodnorodnosti raspredelenija veshchestva, Sov. Phys. JETP 22, 241 (1966).
  • (25) E. B. Gliner, Algebraic properties of the energy-momentum tensor and vacuum-like states of matter, Sov. Phys. JETP 22, 378 (1966).
  • (26) J. M. Bardeen. Non-singular general-relativistic gravitational collapse, In Proceedings of the International Conference GR5, Tbilisi, USSR, 1968.
  • (27) E. Ayon-Beato and A. Garcia, The Bardeen model as a nonlinear magnetic monopole, Phys. Lett. B 493, 149 (2000), [arXiv:gr-qc/0009077 [gr-qc]].
  • (28) E. Ayon-Beato and A. Garcia, Regular black hole in general relativity coupled to nonlinear electrodynamics, Phys. Rev. Lett. 80, 5056 (1998), [arXiv:gr-qc/9911046 [gr-qc]].
  • (29) K. A. Bronnikov, Regular magnetic black holes and monopoles from nonlinear electrodynamics, Phys. Rev. D 63, 044005 (2001), [arXiv:gr-qc/0006014 [gr-qc]].
  • (30) K. A. Bronnikov, Comment on ‘Regular black hole in general relativity coupled to nonlinear electrodynamics’, Phys. Rev. Lett. 85, 4641 (2000).
  • (31) E. Ayon-Beato and A. Garcia, Four parametric regular black hole solution, Gen. Rel. Grav. 37, 635 (2005), [arXiv:hep-th/0403229 [hep-th]].
  • (32) I. Dymnikova, Regular electrically charged structures in nonlinear electrodynamics coupled to general relativity, Class. Quant. Grav. 21, 4417 (2004), [arXiv:gr-qc/0407072 [gr-qc]].
  • (33) W. Berej, J. Matyjasek, D. Tryniecki, and M. Woronowicz, Regular black holes in quadratic gravity, Gen. Rel. Grav. 38, 885 (2006), [arXiv:hep-th/0606185 [hep-th]].
  • (34) L. Balart and E. C. Vagenas, Regular black hole metrics and the weak energy condition, Phys. Lett. B 730, 14 (2014), [arXiv:1401.2136 [gr-qc]].
  • (35) Z. Y. Fan, Critical phenomena of regular black holes in anti-de Sitter space-time, Eur. Phys. J. C 77, 266 (2017), [arXiv:1609.04489 [hep-th]].
  • (36) K. A. Bronnikov, Nonlinear electrodynamics, regular black holes and wormholes, Int. J. Mod. Phys. D 27, 1841005 (2018), [arXiv:1711.00087 [gr-qc]].
  • (37) J. T. S. S. Junior, F. S. N. Lobo and M. E. Rodrigues, (Regular) Black holes in conformal Killing gravity coupled to nonlinear electrodynamics and scalar fields, Class. Quant. Grav. 41, 055012 (2024), [arXiv:2310.19508 [gr-qc]].
  • (38) P. Bueno, P. A. Cano, and R. A. Hennigar, Regular Black Holes From Pure Gravity, [arXiv:2403.04827 [gr-qc]].
  • (39) J. Oliva and S. Ray, A new cubic theory of gravity in five dimensions: Black hole, Birkhoff’s theorem and C-function, Class. Quant. Grav. 27, 225002 (2010), [arXiv:1003.4773 [gr-qc]].
  • (40) R. C. Myers and B. Robinson, Black Holes in Quasi-topological Gravity, JHEP 08, 067 (2010), [arXiv:1003.5357 [gr-qc]].
  • (41) M. H. Dehghani, A. Bazrafshan, R. B. Mann, M. R. Mehdizadeh, M. Ghanaatian and M. H. Vahidinia, Black Holes in Quartic Quasitopological Gravity, Phys. Rev. D 85, 104009 (2012), [arXiv:1109.4708 [hep-th]].
  • (42) J. Ahmed, R. A. Hennigar, R. B. Mann and M. Mir, Quintessential Quartic Quasi-topological Quartet, JHEP 05, 134 (2017), [arXiv:1703.11007 [hep-th]].
  • (43) A. Cisterna, L. Guajardo, M. Hassaine and J. Oliva, Quintic quasi-topological gravity, JHEP 04, 066 (2017), [arXiv:1702.04676 [hep-th]].
  • (44) R. A. Konoplya and A. Zhidenko, Infinite tower of higher-curvature corrections: Quasinormal modes and late-time behavior of D-dimensional regular black holes, Phys. Rev. D 109, 104005 (2024), [arXiv:2403.07848 [gr-qc]].
  • (45) F. Di Filippo, I. Kolář and D. Kubiznak, Inner-extremal regular black holes from pure gravity, [arXiv:2404.07058 [gr-qc]].
  • (46) R. A. Konoplya and A. Zhidenko, Dymnikova black hole from an infinite tower of higher-curvature corrections, [arXiv:2404.09063 [gr-qc]].
  • (47) T. X. Ma and Y. Q. Wang, Frozen boson stars in an infinite tower of higher-derivative gravity, [arXiv:2406.08813 [gr-qc]].
  • (48) P. Bueno, P. A. Cano, R. A. Hennigar and M. D. Li, Kasner eons in Lovelock black holes, [arXiv:2409.00648 [hep-th]].
  • (49) X. Y. Zhang, L. Zhao and Y. Q. Wang, Bardeen-Dirac Stars in AdS Spacetime, [arXiv:2409.14402 [gr-qc]].
  • (50) M. Estrada, A new representation of vacuum Lovelock solutions in d=2N+1d=2N+1 dimensions: Black holes with an integrable singularity and regular black holes, [arXiv:2411.01253 [gr-qc]].
  • (51) P. Bueno, P. A. Cano, R. A. Hennigar and Á. J. Murcia, Dynamical Formation of Regular Black Holes, [arXiv:2412.02742 [gr-qc]].
  • (52) P. Bueno, P. A. Cano, R. A. Hennigar and Á. J. Murcia, Regular black holes from thin-shell collapse, [arXiv:2412.02740 [gr-qc]].
  • (53) S.-W. Wei, Y.-X. Liu, R. B. Mann, Universal topological classifications of black hole thermodynamics, Phys. Rev. D 110, L081501 (2024), [arXiv:2409.09333 [gr-qc]].
  • (54) R. M. Wald, Black hole entropy is the Noether charge, Phys. Rev. D 48, R3427-R3431 (1993), [arXiv:gr-qc/9307038 [gr-qc]].
  • (55) Y.-S. Duan and M.-L. Ge, TSU (2) gauge theory and electrodynamics with N magnetic monopoles, Sci. Sin 9, 1072 (1979).
  • (56) Y.-S. Duan, The structure of the topological current, SLAC-PUB-3301/84 (1984).
  • (57) S.-W. Wei, Topological Charge and Black Hole Photon Spheres, Phys. Rev. D 102, 064039 (2020), [arXiv:2006.02112 [gr-qc]].
  • (58) I. Dymnikova, Vacuum nonsingular black hole, Gen. Rel. Grav. 24, 235 (1992).
  • (59) B. C. Paul, Dymnikova black hole in higher dimensions, Eur. Phys. J. Plus 138, 633 (2023).