Diffusion-entropy scaling across dimensions

Nayana Venkatareddy, Mohd Moid, Prabal K. Maiti, Biman Bagchi · 2025

Read Diffusion-entropy scaling across dimensions by Nayana Venkatareddy, Mohd Moid, Prabal K. Maiti, Biman Bagchi (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-2505-11036.txt.

Introduction

Nayana Venkatareddy, Mohd Moid, Prabal Maiti, and Biman Bagchi report a 2025 simulation study linking how fast particles spread in a liquid to the liquid's entropy. Such a link would let workers read entropy from spread measures and guess transport traits from heat data. The team computed both values on their own for two model liquids.

Runs covered flat and bulk forms of a simple liquid and of water, plus water held in a thin slit. The spread ratio between two states tracked the entropy gap in a near-perfect curve. Water broke the trend across dimensions, a gap the team ties to jump-style motion.

Catalog record

CREATORNayana Venkatareddy, Mohd Moid, Prabal K. Maiti, Biman Bagchi
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-11036

Browse the full library · About the Tower of Babel

Full text

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

Diffusion-entropy scaling across dimensions Nayana Venkatareddy, Mohd Moid, Prabal K. Maiti, Biman Bagchi Abstract A quantitative relationship between the diffusion coefficient D D of a tagged particle in a liquid and the entropy S S of that liquid has long been sought, as it would allow entropy to be inferred directly from diffusion measurements and transport properties to be predicted from thermodynamic information. Here, we employ extensive computer simulations to independently compute both D D and S S for Lennard-Jones (LJ) liquids and for water across a wide range of thermodynamic state points. Our study covers two and three dimensions for both systems, and additionally explores one-dimensional confinement for water. We find that the ratio of diffusion coefficients between two states follows an almost perfect exponential dependence on their entropy difference . For LJ liquids, the exponential prefactor exhibits a pronounced dependence on dimensionality d d , consistent in trend but quantitatively distinct from theoretical predictions. In contrast, water shows a strikingly weak dimensionality (d) dependence , deviating from theory, which we attribute to the dominant role of jump diffusion. Remarkably, the exponential diffusion–entropy relationship persists even when translational and rotational contributions to entropy are separated and considered individually. This robustness suggests that entropy provides a unifying measure governing particle mobility in liquids, largely independent of microscopic mechanisms or dimensional constraints. Preprint: AIP/123-QED I Introduction Einstein’s 1905 work [ 1 ] on Brownian motion laid the groundwork for viewing diffusion as a stochastic process driven by thermal fluctuations. This seminal contribution not only provided a microscopic foundation for Fick’s law but also established diffusion as a bridge between statistical mechanics and observable macroscopic transport. It helped frame the broader connection between thermodynamics and kinetics, a longstanding pursuit originating from the formulation of thermodynamic transition state theory [ 2 ] , which relates the rate of a reaction to the height of free energy barriers. Building on these ideas, Smoluchowski [ 3 ] , Kramers [ 4 ] , and Eyring [ 5 ] pioneered theoretical descriptions of barrier crossing in condensed phases, in which viscosity and diffusivity explicitly enter as determinants of kinetic rates. In these frameworks, the rate of relaxation or reaction is predicted to scale with the diffusion coefficient, thereby tying dynamical processes to transport properties. Despite these advances, a direct and quantitative relation between diffusion and entropy remains far from obvious. Entropy, as a thermodynamic state function, quantifies the accessible phase space of the system, while diffusion reflects the dynamical rate of exploring this phase space. Over the years, several studies have attempted to bridge this conceptual divide [ 6 , 7 ] . A landmark development came from Kauzmann’s [ 8 ] discussion of the “entropy crisis” in supercooled liquids, which highlighted the central role of entropy in controlling relaxation and glassy dynamics. This motivated the search for direct entropy–dynamics relations. The most influential step in this direction was the Adam–Gibbs theory [ 9 ] , which proposed that structural relaxation requires cooperative rearrangements involving a minimum number of particles, with this number governed by the configurational entropy S_{c} . Their argument yields the well-known Adam–Gibbs relation for the temperature dependence of the relaxation time \tau(T) , given by \tau(T)=\tau_{0}e^{C/TS_{c}} , where \tau_{0} and C are material-specific constants. This relation has been widely applied to supercooled liquids near the glass transition temperature, where the entropy crisis drives both the dramatic slowing down of relaxation and the concomitant decrease in diffusion. A different relation between diffusion and entropy was proposed by Rosenfeld [ 6 ] who observed numerically that a scaled (by microscopic parameters) diffusion coefficient D^{*} in liquids could be fitted rather well by a relation of the form D^{*}=ae^{bS_{ex}} , where a and b are fitting parameters, while the excess entropy S_{ex}=S-S_{id} is the difference between the entropy of liquid under given condition and that of the ideal gas state of the same liquid. This scaling relation, known as Rosenfeld scaling, has been extensively used and tested [ 10 , 11 , 12 , 13 , 14 ] , and found to be semi-quantitatively reliable. Together with Dzugutov’s universal scaling law [ 7 ] , these scaling relations further reinforce the idea that entropy can serve as a unifying descriptor of transport and relaxation, connecting equilibrium thermodynamics with dynamical behavior across a wide range of liquids. While many studies have examined both the diffusion-entropy scaling [ 15 , 16 , 17 , 18 ] and the influence of dimensionality on diffusion [ 19 ] , the specific interest in the dimensional dependence of diffusion-entropy scaling is a recent development [ 20 , 21 ] . Notably, the theoretical works of Sorkin et al. [ 20 ] and Liao et al. [ 21 ] both demonstrate an exponential dependence of the diffusion coefficient D on entropy S , with the prefactors 2/d and 1/d respectively. Sorkin et al. [ 20 ] employed a stochastic approach, deriving a generalized D-S relation by expressing entropy in terms of the Shannon entropy of the steady-state configuration distribution, which itself depends on the diffusion coefficient. In contrast, Liao et al. [ 21 ] derived an effective Hamiltonian for a Brownian particle in a heat bath under the weak-coupling limit, replacing the full microscopic Hamiltonian. This allowed for the analytical computation of the partition function, from which the D-S scaling relation was obtained. Importantly, Sorkin et al. [ 20 ] also established a two-state diffusion–entropy relation, enabling the entropy difference between two distinct states to be expressed in terms of their respective diffusion coefficients. Such relations are particularly valuable, as they provide a practical route to infer entropy changes from diffusion measurements. However, the validity of these theoretical relations remains to be rigorously tested through experiments and computer simulations. In this study, we examine the dimensional dependence of the relationship between diffusion coefficients and entropies of two distinct thermodynamic states of a system. In Section II.1 , we briefly review recent theoretical works [ 20 , 21 ] that have explored how dimensionality influences diffusion–entropy scaling. We also present a theoretical derivation of the two-state diffusion–entropy relation for the specific case of a particle undergoing Brownian motion in d dimensions. Our calculation supports a dimensional prefactor of 2/d consistent with earlier findings. In Section III , we turn to molecular dynamics (MD) simulations to test the validity of this relation. Specifically, we assess the validity of this relationship for two distinct systems: Lennard-Jones (LJ) fluid (an atomic fluid) and water (a molecular fluid). Two-phase thermodynamic (2PT) [ 22 , 23 , 24 ] method is used to calculate the entropy of the given systems. Details of the simulation methodology are presented in section III.1 . Our results, presented in section III.2 , show that the ratio of diffusion coefficients of two distinct thermodynamic states follows an exponential dependence on their entropy difference, regardless of the system considered. In 2 d and 3 d LJ systems, we see a clear dimensional dependence where the prefactor \alpha/d before the entropy difference \Delta S (eqn 15 ) has an intermediate value between 1/d and 2/d . As water is a molecular fluid, the validity of the above relation has been studied while considering both

Read the complete text