Journal of Multiscale Modelling

Cardiovascular stents are tiny scaffolds that are used in the treatment of heart disease. The recent development of drug-eluting stents has lead to stent implantation in arterial regions that would previously have been considered too complex. Deployment in these tortuous and branched regions results in an increased deformation of the stent. It is thus important to assess whether there is an increased likelihood of stent failure in deployment in such regions. A multiscale approach, incorporating the results of microscale modeling of failure in individual stent struts and macroscale modeling of stent deployment in realistic arterial geometries is considered in this work. Such an approach allows for a more accurate assessment of failure than is obtainable through the macroscale modeling of deployment in idealized arterial geometries alone, as is presented in previous studies. Results give an insight into failure risks for different stent implantation scenarios: stent failure is unlikely in deployment in tortuous vessels, however there may be risks associated with certain bifurcational stenting techniques.

In the THM modeling of multiphase medium, the coupling effects of skeleton, suction, and temperature have been integrated via the concept of state surfaces of void ratio and degree of saturation. Based on proposed formulation, a fully coupled numerical model for the behavior of soil deformation, water flow, air flow, heat flow in unsaturated soil has been developed and integrated in a finite element code θ-Stock by the first author. This program is conceived with this idea that it will be able to analyze the response of a soil in different states of humidity to mechanical, thermal loading, and also damage phenomena. Damage model is dedicated to unsaturated brittle rocks. It mixes phenomenological and micromechanical concepts and is formulated based on the use of independent state variables. The expression of the liquid permeability is modified in order to represent the influence of fracturing on interstitial fluid flows. The final matrix form of established field equations of the proposed model for unsaturated case has been encoded for this particular purpose, in a finite element program which had been developed for dry and saturated soils previously.

We study the n-dimensional wave equation with an elastoplastic nonlinear stress–strain relation. We investigate the case of heterogeneous materials, i.e., x-dependent parameters that are periodic at the scale η>0. We study the limit η→0 and derive the plasticity equations for the homogenized material. We prove the well-posedness for the original and the effective system with a finite-element approximation. The approximate solutions are also used in the homogenization proof which is based on oscillating test functions and an adapted version of the div-curl Lemma.

The paper reports the use of the finite element method to model longitudinal and transverse deformation of representative volume elements (RVE) of ceramic matrix composites subjected to uniaxial loading parallel to fibers. Cohesive elements have been used to model two forms of damage: fracture initiation and propagation both within the matrix, and along the fiber–matrix interface. From the knowledge of the constituent materials behavior, the FE technique has been used to predict the stress–strain behavior and the variation of Poisson's ratio of the RVE due to these two damage forms; but the model does not cater for fiber failure. The RVE predictions have been benchmarked against experimental results for Nicalon-CAS material and good agreement has been obtained. Comparison of the predicted behavior of the single Nicalon-CAS RVE with experimental data for unidirectional tows indicates that the stress–strain curve is predominantly controlled by the Weibull distribution of fiber failure stress, while the degradation of Poisson's ratio is determined by the Weibull distribution of interfacial strength. The same approach has been used for a HITCO C/C material for which transverse deformation behavior is unknown. The results for the HITCO C/C material, as for the Nicalon-CAS, show that the fiber behavior determines the ultimate failure of the RVE, and that interface debonding is the controlling mechanism for the variation of the Poisson ratio with axial strain.

This study focuses on the movement of particles and extracellular fluid in soft tissues and microvessels. It analyzes modeling applications in biological and physiological fluids at a range of different length scales: from between a few tens to several hundred nanometers, on the endothelial glycocalyx and its effects on interactions between blood and the vessel wall; to a few micrometers, on movement of blood cells in capillaries and transcapillary exchange; to a few millimetres and centimetres, on extracellular matrix deformation and interstitial fluid movement in soft tissues. Interactions between blood cells and capillary wall are discussed when the sizes of the two are of the same order of magnitude, with the glycocalyx on the endothelial and red cell membranes being considered. Exchange of fluid, solutes, and gases by microvessels are highlighted when capillaries have counter-current arrangements. This anatomical feature exists in a number of tissues and is the key in the renal medulla on the urinary concentrating mechanism. The paper also addresses an important phenomenon on the transport of macromolecules. Concentration polarization of hyaluronan on the synovial lining of joint cavities is presented to demonstrate how the mechanism works in principle and how model predictions agree to experimental observations quantitatively.

A discrete element method (DEM) has been developed to provide highly accurate and detailed predictions of the Lagrangian particle phase. Especially in this study, DEM has been used together with an Eulerian approach for the fluid phase to look at interphase exchange phenomena in a multiphase-multiscale modeling approach. The drying process inside a fluidized bed coffee bean roaster has been chosen. Herein, heat, mass, and momentum transport are solved on a fluid cell level; heat, mass, and momentum transfer coefficients are solved at a particle scale level; and 1D temperature and moisture content profiles are solved inside each coffee bean on a sub-particle scale level. Therefore, this multiscale approach provides much more information compared to existing coffee bean roaster models. In this work, a detailed description of this method is provided and results on different scale levels have been discussed. Modeling data and experimental results are compared and found to be in good agreement.