Work probability distribution of weakly driven process in overdamped dynamics

Pierre Nazé · 2025

Read Work probability distribution of weakly driven process in overdamped dynamics by Pierre Nazé (2025) 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-2504-05836.txt.

Introduction

Pierre Nazé, a solo author, derives work chance patterns for gently driven systems in 2025. Work in a small system varies from run to run, and exact formulas for its spread are rare. The study covers the overdamped regime, where friction dominates motion, and works for any switching time.

A test case uses a Brownian particle in a stiffening spring trap. The work spread has an odd shape: it starts at a floor value, spikes there, and is not bell-shaped. Nazé checks where the linear response view holds by a standard self-consistency test. The first four moments of the spread come out right for small drives.

Catalog record

CREATORPierre Nazé
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE19 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-05836

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Work probability distribution of weakly driven process in overdamped dynamics Pierre Nazé Abstract Analytical work probability distributions for open classical systems are scarce; they can only be calculated in a few examples. In this work, I present a new method to derive such quantities for weakly driven processes in the overdamped regime for any switching time. The white noise Brownian motion in a harmonic linear stiffening trap illustrates the result. The work probability distribution is non-tabulated, with positive, semi-finite support, diverging at the minimal value, and non-Gaussian. An analysis of the range of validity of linear response is made by using the self-consistent criterion of the fluctuation-dissipation relation. The first, second, third, and fourth moments are correctly calculated for small perturbations. I Introduction The statistical description of nonequilibrium processes has garnered significant attention over the past decades, particularly in systems driven away from equilibrium by external forces Jarzynski 1997 ; Crooks 1999 ; Seifert 2008 . Among the quantities of interest in such scenarios, the work performed on a system during a time-dependent process plays a central role Speck and Seifert 2004 ; Schuler et al. 2005 . Although equilibrium statistical mechanics provides a comprehensive framework to describe static thermodynamic quantities, obtaining analytical results for work distributions in nonequilibrium settings remains a challenging endeavor Jarzynski 2006 ; Esposito et al. 2009 . In classical systems, exact work probability distributions are known only for a few specific cases, often requiring severe simplifications such as harmonic potentials and idealized driving protocols Imparato and Peliti 2005 ; Engel 2009 ; Kwon et al. 2013 . The situation becomes even more restrictive for open systems in the presence of thermal noise and dissipation Seifert 2012 . In this context, linear response theory emerges as a powerful approximation tool, especially suitable for weakly driven processes Agarwal 1972 ; Kubo 1966 . By expanding observables to first or second order in a small perturbation parameter, linear response theory offers analytical insights into dynamical behavior without requiring full solutions to stochastic differential equations Speck and Seifert 2009 . In the case of overdamped Brownian particles in harmonic traps, linear response theory allows for explicit analytical predictions of the average work and its variance under small driving amplitudes Nazé and Bonança 2020 ; Nazé et al. 2022 ; Nazé 2023 . These results serve as benchmarks for stochastic thermodynamics, especially when testing the limits of fluctuation theorems and nonequilibrium relations Seifert 2005 . However, obtaining the full probability distribution of work analytically, even within linear response theory, remains a nontrivial task and has not receive any attention in the literature. The focus of this work is to derive the work probability distribution for classical systems weakly driven in the overdamped regime. More specifically, I propose a method to obtain the probability distribution analytically for arbitrary switching times, provided the driving remains within the linear response domain. The formulation leverages the relation between the work and the system initial conditions, transforming the problem into one of variable change in the probability measure. To illustrate the generality and effectiveness of the method, I consider the paradigmatic example of an white noise overdamped Brownian particle, subjected to a harmonic linear stiffness trap. This example is particularly appealing, as it combines analytical tractability with physical relevance Bustamante et al. 2005 ; Schmiedl and Seifert 2007 ; Seifert 2012 . Comparisons with numerically exact simulations confirm the accuracy of the predictions, including a detailed analysis of the first four moments of the distribution. Furthermore, the behavior of the work distribution under varying switching times and driving amplitudes reveals how the system transitions from linear to nonlinear response. In particular, I present a self-consistent method based on the fluctuation-dissipation relation Nazé 2023 \langle\overline{W}_{\tau}\rangle=\Delta F+\frac{\beta}{2}\sigma_{W_{\tau}}^{2}, (1) where no extra calculation of next order is necessary to determine the range of validity of linear response theory. Here, \langle\overline{W}_{\tau}\rangle is the averaged work, \Delta F the difference in Helmholtz’s free energies and \sigma_{W_{\tau}}^{2} the variance of work. This manuscript is organized as follows. First, I introduce the theoretical framework and derive the general expression for the work distribution. Next, I apply the method to the stiffening trap model and obtain the probability distribution function. Then, I perform a thorough comparison between the linear response prediction and exact numerical results, including an evaluation of the range of validity of the method. Finally, I conclude with a discussion of the implications of the results and potential directions for future extensions. II Weakly driven processes I start by defining notations and developing the main concepts to be used in this work. Consider a classical system with a Hamiltonian \mathcal{H}(\mbox{\boldmath$z$}(\mbox{\boldmath$z_{0}$},t)),\lambda(t)) , where \mbox{\boldmath$z$}(\mbox{\boldmath$z_{0}$},t) is a point in the phase space \Gamma evolved from the initial point z_{0} until time t , with \lambda(t) being a time-dependent external parameter. Initially, the system is at equilibrium with a heat bath of temperature \beta\equiv{(k_{B}T)}^{-1} , where k_{B} is Boltzmann’s constant. During a switching time \tau , the external parameter is changed from \lambda_{0} to \lambda_{0}+\delta\lambda , with the system always in contact with the initial heat bath. The configuration of the system is such that the driving is made in the overdamped regime, where the acceleration is negligible. The work performed on the system during this interval of time is \overline{W}_{\tau}(x_{0})\equiv\int_{0}^{\tau}\overline{\partial_{\lambda}\mathcal{H}}(x_{0},t)\dot{\lambda}(t)dt, (2) where \partial_{\lambda} is the partial derivative in respect to \lambda and the superscripted dot the total time derivative. Since the work is a random variable, the generalized force \overline{\partial_{\lambda}\mathcal{H}}(x_{0},t) depends on the initial position x_{0} , which obeys the canonical ensemble \rho_{x_{0}}(x_{0}) . The overline symbol \overline{\cdot} denotes an average over the noise produced by the heat bath. The external parameter can be expressed as \lambda(t)=\lambda_{0}+g(t)\delta\lambda, (3) where to satisfy the initial conditions of the external parameter the protocol g(t) must satisfy the following boundary conditions g(0)=0 , g(\tau)=1 . Linear-response theory aims to express average quantities until the first-order of some perturbation parameter considering how this perturbation affects the observable to be averaged and the probabilistic distribution Kubo et al. 2012 . In our case, I consider that the parameter does not considerably change during the process, |g(t)\delta\lambda/\lambda_{0}|\ll 1 , for all t\in[0,\tau] and \lambda_{0}\neq 0 . Also, the average of the initial conditions does not need to be performed. The work can be approximated up to second-order as \begin{split}\overline{W}_{\tau}(x_{0})=&\,\delta\lambda\overline{\partial_{\lambda}\mathcal{H}}(x_{0})-\frac{\delta\lambda^{2}}{2}\widetilde{\psi}(x_{0})\\ &+\frac{1}{2}\int_{0}^{\tau}\int_{0}^{\tau}\psi(x_{0},t-t^{\prime})\dot{\lambda}(t^{\prime})\dot{\lambda}(t)dt^{\prime}dt,\end{split} (4) where \psi(x_{0},t)=\beta\partial_{\lambda}\mathcal{H}(x_{0},0)\overline{\partial_{\lambda}\mathcal{H}}(x_{0},t)-c(x_{0}) (5) is the (not-averaged) relaxation function, with \widetilde{\psi}_{0}(x_{0})\equiv\psi_{0}(x_{0},0)-\overline{\partial_{\lambda\lambda}^{2}\mathcal{H}}(x_{0},0) Kubo et al. 2012 . The constant c(x_{0}) is chosen to nullify the (averaged) relaxation function for long times. Such an equation holds for finite-time and weakly driven processes. In particular, the averaged work could be split into the difference of Helmholtz’s free energy and irreversible work, given by \Delta F(x_{0})=\delta\lambda\overline{\partial_{\lambda}\mathcal{H}}(x_{0})-\frac{\delta\lambda^{2}}{2}\widetilde{\psi}(x_{0}), (6) W^{\rm irr}_{\tau}(x_{0})=\frac{1}{2}\int_{0}^{\tau}\int_{0}^{\tau}\psi(x_{0},t-t^{\prime})\dot{\lambda}(t^{\prime})\dot{\lambda}(t)dt^{\prime}dt. (7) Observe that both quantities depend on the initial position x_{0} . Our objective is to derive the work probability distribution and to corroborate the analytical prediction with a comparison with the exact result for the white noise overdamped Brownian motion, subject to a harmonic linear stiffening trap. III Work probability distribution To find the work probability distribution, one needs to make a change of variables in the initial position probability distribution, considering the inverted function x_{0}=x_{0}(\overline{W}_{\tau},\tau). (8) In this manner \rho_{x_{0}}(x_{0})dx_{0}=\rho_{x_{0}}(x_{0}(\overline{W}_{\tau},\tau))\left|\frac{dx_{0}}{d\overline{W}_{\tau}}\right|d\overline{W}_{\tau}. (9) Therefore, the

Read the complete text