Continuum granular flow model with restitution–derived viscoelastic damping Journal: arXiv

Bodhinanda Chandra, Sachith Dunatunga, Ken Kamrin · 2026

Read Continuum granular flow model with restitution–derived viscoelastic damping Journal: arXiv by Bodhinanda Chandra, Sachith Dunatunga, Ken Kamrin (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-2604-21448.txt.

Introduction

Bodhinanda Chandra, Sachith Dunatunga, and Ken Kamrin build one continuum model for rate-dependent granular flows across regimes. Two rate effects, micro-inertia and viscoelastic damping, sit inside one continuum picture, with a direct link from the restitution coefficient to continuum viscosity via wave attenuation. Viscous damping steers wave motion and collisions without changing the plastic flow rule. Built with the material point method, the model captures wave motion, diffusion, and rate-dependent behavior across steady, transient, shaking, and impact-driven flows.

Catalog record

CREATORBodhinanda Chandra, Sachith Dunatunga, Ken Kamrin
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE114528
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.21448

Browse the full library · About the Tower of Babel

Full text

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

Continuum granular flow model with restitution–derived viscoelastic damping Journal: arXiv Bodhinanda Chandra, Sachith Dunatunga, Ken Kamrin Abstract This work presents a unified viscoelastic–viscoplastic continuum framework for modeling rate-dependent granular flows across regimes. The formulation incorporates two distinct rate-dependent mechanisms, namely micro-inertia and viscoelastic dissipation, within a single continuum description. A central contribution is an explicit link between the coefficient of restitution and a continuum viscosity, derived from an analysis of wave attenuation in granular assemblies, thereby establishing a direct connection between particle-scale collision physics and macroscopic damping. This relation is introduced while retaining inertia-dependent plastic flow governed by the classical μ ⁡ ( I ) \mu(I) rheology. The constitutive model is constructed by meticulously partitioning elastic and viscous responses within the model and corresponding stress-update routine, such that viscous dissipation governs wave propagation and collisional processes without altering the plastic flow rule. The framework is implemented within the material point method to simulate transient processes involving large deformations, material separation, and subsequent reconsolidation. A range of numerical examples, including steady, transient, vibrational, and impact-driven flows, demonstrates that the model captures wave propagation, diffusion, and rate-dependent granular behavior within a unified continuum setting. 1 Introduction Dry granular flow exhibits rate-dependent behavior due to non-vanishing microscopic length and time scales. In this study, we focus on modeling grain-scale collisions and the associated energy dissipation. The presence of granular collisions gives rise to two distinct rate-dependent behaviors: (i) micro-inertia effects in dense granular flows subjected to plastic shear, and (ii) the role of restitution in bulk elastic response. Micro-inertia arises from grain-scale inertia associated with relative particle motion and velocity fluctuations, which become important when the time scale associated with macroscopic deformation is comparable to microscopic rearrangement times, a condition commonly characterized by the inertial number I 38 , 30 . In densely flowing granular states, this mechanism modifies the effective resistance to flow and gives rise to rate-dependent apparent friction, which is commonly described by inertial rheologies such as the \mu(I) model 51 . By contrast, restitution governs the dissipation of kinetic energy during elastic particle-to-particle collisions. Its effects manifest differently across regimes, contributing to energy loss through collisions in gas-like states and to the attenuation of stress waves and oscillations in solid-like states. It is known that the coefficient of restitution e influences the \mu(I) relation. This is particularly apparent in frictionless grains, but its effect diminishes for grains with friction 38 , 30 . In practice, this dependence can be absorbed into the calibration of \mu(I) parameters. While micro-inertia has been widely incorporated into continuum descriptions, both within hydrodynamic models, e.g. 85 , 83 , and elasto-viscoplastic solid formulations, e.g. 53 , 31 , mechanisms for incorporating elastic restitution into the remaining bulk phenomenology, namely, the damping of elastic wave propagation and the physically consistent dissipation during volumetric collisions, have not yet been systematically addressed in continuum mechanics frameworks. The challenge of incorporating restitution into a continuum model in this sense is twofold: (i) how to relate the coefficient of restitution, which is a microscopic granular property, to a well-defined continuum parameter, for example through the concept of viscosity , and (ii) how to introduce its dissipative effects in a manner that does not artificially add extra damping into the continuum plastic flow behavior . The empirical kinematic coefficient of restitution, e , was originally proposed by Newton in 1687 70 , 27 . As defined by Newton, the coefficient of restitution, e , can be computed for a particle-to-particle collision as: \displaystyle e=\frac{-\tilde{v}(t_{c})}{\tilde{v}(0)}\,, (1) where \tilde{v} is the relative velocity in the direction of collision and t_{c} denotes the contact time (or release time), i.e. the time at which the contacting particles separate. For a normal collision in which no external energy is supplied to the system simultaneously, thermodynamic constraints require that e\in[0,1] , where e=1 corresponds to a fully elastic collision and e=0 corresponds to a fully plastic collision. The value of e depends on the colliding material types as well as the contact geometry, which is influenced by particle shape and angle of contact 45 , 92 . Experimental observations, e.g. 15 , 57 , 40 , further indicate that e depends on the impact velocity, with larger values typically observed at lower \tilde{v} and decreasing values as \tilde{v} increases. Restitution has been incorporated in the discrete element method (DEM) 29 through normal contact force models to introduce dissipative particle interactions. The normal contact force in DEM is often expressed as: \displaystyle F_{n}=k_{n}\delta_{n}+c_{n}\dot{\delta}_{n}\,,\qquad F_{n}\geq 0\,, (2) where k_{n} is the contact spring stiffness and c_{n} is the normal damping factor. Here, \delta_{n} and \dot{\delta}_{n} denote the spring compression distance and compression rate, which in DEM are often quantified through particle interpenetration. For example, when the Kelvin–Voigt model 48 is adopted and the contact duration is approximated as a half-period single-degree-of-freedom (SDOF) vibration, a closed-form relation between c_{n} and e can be obtained as: \displaystyle c_{n}=\frac{2|\ln{e}|}{\sqrt{|\ln{e}|^{2}+\pi^{2}}}\sqrt{k_{n}\tilde{m}}\,, (3) where \tilde{m} is the effective mass, which for two colliding particles i and j is given by \tilde{m}=m_{i}m_{j}/(m_{i}+m_{j}) 78 . Different choices of spring–dashpot contact models lead to different expressions for c_{n} 20 , 49 . For example, for the Maxwell model, which is the dual (or reciprocal) formulation of the Kelvin–Voigt model, the corresponding normal damping coefficient is given by: \displaystyle c_{n}=\frac{\sqrt{|\ln{e}|^{2}+\pi^{2}}}{2|\ln{e}|}\sqrt{k_{n}\tilde{m}}\,. (4) The above expressions are derived by solving a damped SDOF oscillator, with the contact time t_{c} assumed to correspond to the half-period of the displacement oscillation, i.e. at t=t_{c} , the displacement satisfies \delta_{n}(t_{c})=0 . This assumption, however, is inaccurate for two contacting grains, since evaluating the relative velocity at this instant occurs after the normal contact force has already changed sign in the oscillator solution. This violates the unilateral contact condition, where contact forces can only be compressive and cannot sustain tension (cf. Eq. ( 2 ) 2 ). Moreover, in the case of high damping, the contact time may not admit a real solution, meaning that particles stick together after a head-on collision, leading to the dissipative capture condition 77 . As highlighted by 77 , a more appropriate way to compute the relation between the damping coefficient c_{n} and e is to define t_{c} as the instant at which the contact force vanishes as particles separate from one another. This corresponds to the half-period of the velocity vibration, occurring when the releasing velocity is maximum or, correspondingly, when the acceleration decreases to zero, i.e. \ddot{\delta}_{n}(t_{c})=0 at t=t_{c} . Moreover, depending on the considered elasticity model, the relation between the damping factor and restitution may differ from that of the classical linear spring–dashpot model. For Hertzian contact model, c_{n} becomes a function of \delta_{n}^{1/2} , since k_{n}\sim\delta_{n}^{1/2} 89 . In this case, the damping factor also depends on the viscous material parameters of the contacting grain, \vartheta_{g} and \eta_{g} (bulk and shear viscosities, with subscript g denoting grain properties), analogous to the dependence of k_{n} on the elastic moduli of the grains, e.g. E_{g} and \nu_{g} 19 . For further derivations and details of nonlinear spring–dashpot contact models, the reader is referred to Chapter 3 in 18 . Beyond the normal contact response, the tangential interaction between grains plays an equally important role in governing

Read the complete text