Kinetic energy fluctuations and specific heat in generalized ensembles
Sergio Davis, Catalina Ruíz, Claudia Loyola, Carlos Femenías, Joaquín Peralta · 2026
Read Kinetic energy fluctuations and specific heat in generalized ensembles by Sergio Davis, Catalina Ruíz, Claudia Loyola, Carlos Femenías, Joaquín Peralta (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-02168.txt.
Introduction
Sergio Davis, Catalina Ruíz, Claudia Loyola, and colleagues extend a classic heat formula in this 2026 physics paper. The classic result links swings in motion energy to how much heat a closed system can hold. It was derived for one narrow case, the sealed system at fixed energy. Many real systems sit in broader settings, though, and need a wider rule.
Their new formula holds for any steady setting and any system size. It ties motion energy swings to total energy swings and to heat capacity. The old result falls out when total energy stops changing. Tests used computer sampling of vibrating springs and exact math for a flat energy case. The rule may help study small systems with odd heat behavior.
Catalog record
| CREATOR | Sergio Davis, Catalina Ruíz, Claudia Loyola, Carlos Femenías, Joaquín Peralta |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26789 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.02168
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Kinetic energy fluctuations and specific heat in generalized ensembles Sergio Davis, Catalina Ruíz, Claudia Loyola, Carlos Femenías, Joaquín Peralta Abstract We derive an exact generalization of the well-known Lebowitz–Percus–Verlet (LPV) formula that relates the kinetic energy fluctuations of an isolated system to its specific heat. Our general formula, obtained by the application of expectation identities, is valid for arbitrary steady–state ensembles and system sizes, expressing the relative variance of the kinetic energy in terms of the variance of total energy and the microcanonical specific heat. The usual microcanonical LPV formula can be readily recovered as a particular case where energy fluctuations vanish. We test the validity of the generalized formula by performing Monte Carlo simulations of a superstatistical system of harmonic oscillators, as well as by exact calculation of energy variances in a uniform–energy ensemble, discussing its relevance to systems exhibiting negative heat capacity and ensemble inequivalence, as encountered in finite nuclei and self–gravitating models. Our results may provide useful in the study of non-equilibrium phase transitions in finite systems. I Introduction Studying fluctuations in physical observables within statistical mechanics is crucial for understanding the distinct characteristics of various statistical ensembles 20 ; 13 . Among these ensembles, the microcanonical ensemble holds a fundamental place 12 ; 10 . This ensemble, where energy, volume, and the number of particles are conserved, allows for a focused examination of isolated systems. By analyzing the fluctuations of physical observables within the microcanonical framework, we can clearly depict the underlying behaviors of these systems. One important observable is the specific heat ( C ), typically measured at constant volume. The specific heat is a crucial quantity that provides insights into a wide range of material properties, particularly because of its unique behavior near phase transitions 22 ; 23 . Beyond the thermodynamic limit, the behavior of finite and isolated systems reveals a richer phenomenology. In particular, negative heat capacities and ensemble inequivalence have been observed in a variety of finite systems, ranging from fragmenting nuclei and atomic clusters to long–range self–gravitating models 3 ; 16 ; 21 ; 15 ; 19 . These anomalies emerge near first–order phase transitions and signal the presence convex intruders in the microcanonical entropy, implying that the canonical and microcanonical ensembles can yield different equilibrium behaviors 3 ; 16 . Partial energy fluctuations have been proposed as a diagnostic of such phase transitions and have been used to reconstruct the microcanonical heat capacity even in small systems 3 . Understanding kinetic energy fluctuations in generalized ensembles may provide a diagnostic tool that could, in principle, be applied to systems known to exhibit negative heat capacities and ensemble inequivalence. A key theoretical link between these anomalous behaviors and equilibrium thermodynamic quantities is provided by the relationship between kinetic energy fluctuations and the specific heat. A landmark result by Lebowitz, Percus, and Verlet (LPV) 14 established that the microcanonical fluctuations of kinetic energy are related to the specific heat. This relationship provides a deeper understanding of how specific heat can describe the fluctuations and behavior within different statistical ensembles. The formula is given by \big<(\delta K)^{2}\big>_{E}=\frac{3N}{2{\beta_{E}}^{2}}\left(1-\frac{3N}{2C}\right), (1) where \big<(\delta K)^{2}\big>_{E} is the variance of kinetic energy, N is the number of particles, \beta_{E}\mathrel{\mathop{:}}=\frac{1}{k_{B}T(E)} (2) is the inverse temperature at energy E , and C is the total specific heat of the system in units of k_{B} . Studies have shown that partial energy fluctuations can be used to reconstruct microcanonical heat capacities even in small systems and near phase transitions. For example, in the context of nuclear multrifragmentation, the relative variance of the kinetic energy, \sigma_{K}^{2}/T^{2} , provides an estimate of the heat capacity; simulations with lattice gas and Lennard–Jones models validate this approach 11 . These results motivate our present generalization of the LPV formula and emphasize the importance of kinetic energy fluctuations in detecting phase–transition signatures. Small and finite systems also exhibit fluctuations of intensive quantities such as temperature, which cannot be neglected. The formalism of superstatistics 1 ; 2 has been advocated as a framework for describing temperature fluctuations in nonequilibrium steady states; however, its interpretation of \beta as a fluctuating inverse temperature faces conceptual difficulties 9 . Recent work has introduced generalized ensembles described by an ensemble function \rho(E;\boldsymbol{\lambda}) and shown that superstatistics describes only a subclass of such states. Particularly, the inverse temperature covariance \mathcal{U}\mathrel{\mathop{:}}=\big<(\delta\beta_{\Omega})^{2}\big>_{\boldsymbol{\lambda}}+\big<{\beta_{\Omega}}^{\prime}\big>_{\boldsymbol{\lambda}}, (3) where \beta_{\Omega}(E)\mathrel{\mathop{:}}=\frac{\partial}{\partial E}\ln\Omega(E) (4) is the microcanonical inverse temperature and \Omega(E) is the density of states, allows to classify nonequilibrium steady states into supercanonical states (including superstatistics), where \mathcal{U}>0 and subcanonical states 8 , with \mathcal{U}<0 . This classification helps to identify whether temperature fluctuations can be encoded by a distribution P(\beta|\boldsymbol{\lambda}) or more general descriptions are needed. In this study, we propose to investigate the fluctuations of kinetic energy and their relationship to the specific heat, aiming to extend our understanding to different ensembles, in this particular case, the generalized ensemble 4 ; 17 ; 18 . From here, we derive a generalization of the LPV formula for an arbitrary steady-state ensemble with parameters \boldsymbol{\lambda} , namely \frac{\big<(\delta K)^{2}\big>_{\boldsymbol{\lambda}}}{\big<K\big>_{\boldsymbol{\lambda}}^{2}}=\frac{2}{3N}\frac{C+1}{C+2}\left[\left(\frac{3N}{2}+1\right)\frac{\big<(\delta E)^{2}\big>_{\boldsymbol{\lambda}}}{\big<E\big>_{\boldsymbol{\lambda}}^{2}}+1-\frac{3N}{2(C+1)}\right], (5) where now \big<(\delta E)^{2}\big>_{\boldsymbol{\lambda}} represents the energy fluctuations within the ensemble parameterized by \boldsymbol{\lambda} , while \big<E\big>_{\boldsymbol{\lambda}} is its mean total energy. Similarly, \big<K\big>_{\boldsymbol{\lambda}} stands for the mean kinetic energy in the ensemble. Importantly, our method for arriving at this result does not rely on the assumptions used in the original LPV formula, and it holds true for any system size N\geq 1 . In the following sections, we will outline the structure of the paper. Section 2 will detail the derivation of the generalized equation 5 , and Section 3 will explain our computational procedure using Monte Carlo simulations to test our findings. We will conclude with some final remarks in Section 4. II Derivation of the main result We will assume a classical system of N particles with Hamiltonian \mathcal{H} given by \mathcal{H}(\bm{r}_{1},\ldots,\bm{r}_{N},\bm{p}_{1},\ldots,\bm{p}_{N})=\sum_{i=1}^{N}\frac{\bm{p}_{i}^{2}}{2m_{i}}+\Phi(\bm{r}_{1},\ldots,\bm{r}_{N}) (6) and constant specific heat C in units of k_{B} , that is, such that \beta_{\Omega}(E)=\frac{C}{E}, (7) with \Omega(E) the density of states, \Omega(E)\mathrel{\mathop{:}}=\int d\bm{r}_{1}\ldots d\bm{r}_{N}d\bm{p}_{1}\ldots d\bm{p}_{N}\delta\left(E-\left[\sum_{i=1}^{N}\frac{\bm{p}_{i}^{2}}{2m_{i}}+\Phi(\bm{r}_{1},\ldots,\bm{r}_{N})\right]\right), (8) such that, in our case, it can be obtained from ( 7 ) as \Omega(E)=\Omega_{0}\,E^{C} . We will assume that the interaction potential \Phi is bounded from below, and without loss of generality, we will set this minimum value to zero so that the energy is always non-negative. From here, we will consider the system in a non-equilibrium, steady-state ensemble described by the microstate distribution: P(\bm{R},\bm{P}|\boldsymbol{\lambda})=\rho(\mathcal{H}(\bm{R},\bm{P});\boldsymbol{\lambda}) (9) where \bm{R}\mathrel{\mathop{:}}=(\bm{r}_{1},\ldots,\bm{r}_{N}) and \bm{P}\mathrel{\mathop{:}}=(\bm{p}_{1},\ldots,\bm{p}_{N}) , and the function \rho(E;\boldsymbol{\lambda}) is the ensemble function . The distribution of total energy is simply P(E|\boldsymbol{\lambda})=\rho(E;\boldsymbol{\lambda})\Omega(E), (10) while the distribution of kinetic energy in this ensemble can be computed from P(K|\boldsymbol{\lambda})=\int d\bm{R}d\bm{P}\;\rho\big(K(\bm{P})+\Phi(\bm{R});\boldsymbol{\lambda}\big)\delta\big(K(\bm{P})-K\big)=\left[\int d\bm{R}\;\rho(K+\Phi(\bm{R});\boldsymbol{\lambda})\right]\;\Omega_{K}(K;N) (11) where \Omega_{K} is the kinetic density of states, which is of the form \Omega_{K}(K;N)=W_{N}\;K^{\frac{3N}{2}-1}, (12) with W_{N} a constant only dependent on N and the particle masses. We can replace the multidimensional integral in square brackets with a one-dimensional integral weighted by the configurational density of states \mathcal{D}(\phi) , \int d\bm{R}\;\rho(K+\Phi(\bm{R});\boldsymbol{\lambda})=\int_{0}^{\infty}d\phi\mathcal{D}(\phi)\;\rho(K+\phi;\boldsymbol{\lambda}), (13) where \mathcal{D}(\phi)\mathrel{\mathop{:}}=\int d\bm{R}\delta(\Phi(\bm{R})-\phi). (14) Replacing ( 12 ) we obtain P(K|\boldsymbol{\lambda})=W_{N}K^{\frac{3N}{2}-1}\int_{0}^{\infty}d\phi\mathcal{D}(\phi)\rho(K+\phi;\boldsymbol{\lambda}). (15) Moreover, the full density of states \Omega(E) can be computed from \mathcal{D}(\phi) as \Omega(E)=W_{N}\,\int_{0}^{\infty}dK\,K^{\frac{3N}{2}-1}\mathcal{D}(E-K). (16) In the case of the microcanonical ensemble, we have