Marco Ellero

Viscoelastic flows studied by smoothed particle dynamics

Abstract A viscoelastic numerical scheme based on smoothed particle dynamics is presented. The concept goes a step beyond smoothed particle hydrodynamics (SPH) which is a grid-free Lagrangian method describing the flow by fluid-pseudo-particles. The relevant properties are interpolated directly on the resulting movable grid. In this work, the effect of viscoelasticity is incorporated into the ordinary conservation laws by a differential constitutive equation supplied for the stress tensor. In order to give confidence in the methodology we explicitly consider the non-stationary simple corotational Maxwell model in a channel geometry. Without further developments the scheme is applicable to ‘realistic’ models relevant for three-dimensional (3D) viscoelastic flows in complex geometries.

Incompressible smoothed particle hydrodynamics

We present a smoothed particle hydrodynamic model for incompressible fluids. As opposed to solving a pressure Poisson equation in order to get a divergence-free velocity field, here incompressibility is achieved by requiring as a kinematic constraint that the volume of the fluid particles is constant. We use Lagrangian multipliers to enforce this restriction. These Lagrange multipliers play the role of non-thermodynamic pressures whose actual values are fixed through the kinematic restriction. We use the SHAKE methodology familiar in constrained molecular dynamics as an efficient method for finding the non-thermodynamic pressure satisfying the constraints. The model is tested for several flow configurations.

Consistent scaling of thermal fluctuations in smoothed dissipative particle dynamics

Dissipative particle dynamics (DPD) as a model of fluid particles suffers from the problem that it has no physical scale associated with the particles. Therefore, a DPD simulation requires an ambiguous fine-tuning of the model parameters with the physical parameters. A corrected version of DPD that does not suffer from this problem is smoothed dissipative particle dynamics (SDPD) [P. Espanol and M. Revenga, Phys. Rev. E 67, 026705 (2003)]. SDPD is, in fact, a version of the well-known smoothed particle hydrodynamics method, albeit with the proper inclusion of thermal fluctuations. Here, we show that SDPD produces the proper scaling of the fluctuations as the resolution of the simulation is varied. This is investigated in two problems: the Brownian motion of a spherical colloidal particle and a polymer molecule in suspension.

Multiscale modeling of particle in suspension with smoothed dissipative particle dynamics

We apply smoothed dissipative particle dynamics (SDPD) [Espanol and Revenga, Phys. Rev. E 67, 026705 (2003)] to model solid particles in suspension. SDPD is a thermodynamically consistent version of smoothed particle hydrodynamics (SPH) and can be interpreted as a multiscale particle framework linking the macroscopic SPH to the mesoscopic dissipative particle dynamics (DPD) method. Rigid structures of arbitrary shape embedded in the fluid are modeled by frozen particles on which artificial velocities are assigned in order to satisfy exactly the no-slip boundary condition on the solid-liquid interface. The dynamics of the rigid structures is decoupled from the solvent by solving extra equations for the rigid body translational/angular velocities derived from the total drag/torque exerted by the surrounding liquid. The correct scaling of the SDPD thermal fluctuations with the fluid-particle size allows us to describe the behavior of the particle suspension on spatial scales ranging continuously from the dif...

Self-diffusion coefficient in smoothed dissipative particle dynamics

Smoothed dissipative particle dynamics (SDPD) is a novel coarse grained method for the numerical simulation of complex fluids. It has considerable advantages over more traditional particle-based methods. In this paper we analyze the self-diffusion coefficient D of a SDPD solvent by using the strategy proposed by Groot and Warren [J. Chem. Phys. 107, 4423 (1997)]. An analytical expression for D in terms of the model parameters is developed and verified by numerical simulations.

Exact Pressure Evolution Equation for Incompressible Fluids

An important aspect of computational fluid dynamics is related to the determination of the fluid pressure in isothermal incompressible fluids. In particular this concerns the construction of an exact evolution equation for the fluid pressure which replaces the Poisson equation and yields an algorithm which is a Poisson solver, i.e., it permits to time‐advance exactly the same fluid pressure without solving the Poisson equation. In fact, the incompressible Navier‐Stokes equations represent a mixture of hyperbolic and elliptic pde’s, which are extremely hard to study both analytically and numerically. This amounts to transform an elliptic type fluid equation into a suitable hyperbolic equation, a result which usually is reached only by means of an asymptotic formulation. Besides being a still unsolved mathematical problem, the issue is relevant for at least two reasons: a) the proliferation of numerical algorithms in computational fluid dynamics which reproduce the behavior of incompressible fluids only in ...

Shear Thinning of Noncolloidal Suspensions

Shear thinning-a reduction in suspension viscosity with increasing shear rates-is understood to arise in colloidal systems from a decrease in the relative contribution of entropic forces. The shear-thinning phenomenon has also been often reported in experiments with noncolloidal systems at high volume fractions. However its origin is an open theoretical question and the behavior is difficult to reproduce in numerical simulations where shear thickening is typically observed instead. In this letter we propose a non-Newtonian model of interparticle lubrication forces to explain shear thinning in noncolloidal suspensions. We show that hidden shear-thinning effects of the suspending medium, which occur at shear rates orders of magnitude larger than the range investigated experimentally, lead to significant shear thinning of the overall suspension at much smaller shear rates. At high particle volume fractions the local shear rates experienced by the fluid situated in the narrow gaps between particles are much larger than the averaged shear rate of the whole suspension. This allows the suspending medium to probe its high-shear non-Newtonian regime and it means that the matrix fluid rheology must be considered over a wide range of shear rates.

A multiscale SPH particle model of the near‐wall dynamics of leukocytes in flow

A novel multiscale Lagrangian particle solver based on SPH is developed with the intended application of leukocyte transport in large arteries. In such arteries, the transport of leukocytes and red blood cells can be divided into two distinct regions: the bulk flow and the near-wall region. In the bulk flow, the transport can be modeled on a continuum basis as the transport of passive scalar concentrations. Whereas in the near-wall region, specific particle tracking of the leukocytes is required and lubrication forces need to be separately taken into account. Because of large separation of spatio-temporal scales involved in the problem, simulations of red blood cells and leukocytes are handled separately. In order to take the exchange of leukocytes between the bulk fluid and the near-wall region into account, solutions are communicated through coupling of conserved quantities at the interface between these regions. Because the particle tracking is limited to those leukocytes lying in the near-wall region only, our approach brings considerable speedup to the simulation of leukocyte circulation in a test geometry of a backward-facing step, which encompasses many flow features observed in vivo.

Unique representation of an inverse-kinetic theory for incompressible Newtonian fluids

Fundamental aspects of inverse kinetic theories for the incompressible Navier-Stokes equations [Ellero and Tessarotto, 2004, 2005] include the possibility of defining uniquely the kinetic equation underlying such models and furthermore, the construction of a kinetic theory implying also the energy equation. The latter condition is consistent with the requirement that fluid fields result classical solutions of the fluid equations. These issues appear of potential relevance both from the mathematical viewpoint and for the physical interpretation of the theory. Purpose of this work is to prove that under suitable prescriptions the inverse kinetic theory can be determined to satisfy such requirements.

Continuous inverse kinetic theory for incompressible fluids

A fundamental aspect of CFD for incompressible fluids and magnetofluids is the algorithmic complexity which characterizes both conventional direct simulation methods and kinetic approaches based on microscopic or phenomenological models based on asymptotic kinetic theories. A possible solution to this problem, can be provided by so‐called inverse kinetic theories suitably constructed in such a way to avoid the computational complexity of previous numerical approaches, in particular due to the Poisson equation for the fluid pressure, which typically (for example, for so‐called fractional step due to Kim and Moin), involves an algorithmic complexity of order N2, being N the grid number characterizing the numerical solution. This may explain why, in massive parallel DNS simulations of turbulent flows, the maximum grid number remains substantially limited. Goal of the present Note is propose a solution to this problem, pointing out a new inverse kinetic theory which does not require the solution of Poisson eq...