Displacement model including localization

The displacement flows can be miscible (brine after polymer solution, C02 after oil, steam after water) or immiscible (water after oil). In the former case, it is the mixing process itself which has to be understood and modeled steam recovery requires the thermal transport problem to be accurately modeled. In the latter case, the two fluid phases coexist within the porous medium their relative proportions are determined not only by flow and mixing processes, but equally by interfacial and surface tensions between the three phases (matrix material included). Local (capillary) variations in pressure between the two fluid phases become important. The overall flow field is determined by large-scale pressure gradients. 

Vibrational states can be described in terms of the normal mode (NM) [50, 51] or the local mode (LM) [37, 52, 53] model. In the former, vibrations in polyatomic molecules are treated as infinitesimal displacements of the nuclei in a harmonic potential, a picture that naturally includes the coupling among the bonds in a molecule. The general formula for the energies of the vibrational levels in a polyatomic molecule is given by [54]... 

To obtain particles which occupy a certain volume in space, contact laws could be defined between the nodes. In the system, linear or non-linear interactions could be considered. The meshless approach of the DEM allows to model problems including large deformations with geometrical and physical non-linearities, large displacements even with free motion and localisation phenomena like damage and fracture. Through the direct control over the applied interactions, the local state of the system is always accessible in a simulation. 

An important simplification results if we can consider the bonding between atoms to be a local phenomenon. In this event, we would need to consider only the immediate neighbours of the adsorbate or defect atoms, and we arrive at the cluster models circled in Fig. 1. Of course, some properties of the system will depend on its extended nature. Others, including the variation in total energy with small displacements of atoms, should be described satisfactorily by a cluster calculation. In such cases, the problem has been reduced to one of molecular dimensions, so that the methods of molecular physics or theoretical chemistry could be used. For many systems of interest to the solid-state physicist, where a typical problem might be the chemisorption of a carbon monoxide molecule on the surface of a ferromagnetic metal surface such as nickel, the methods discussed in much of the rest of the present volume are inappropriate. It is necessary to seek alternatives, and this chapter is concerned with one of them, the density functional (DF) formalism. While the motivation of the solid-state physicist is perhaps different from that of the chemist, the above discussion shows that some of the goals are very similar. Indeed, it is my view that the density functional formalism, which owes much of its development and most of its applications to solid-state physicists, can make a useful contribution to theoretical chemistry. 

However, these state variables are not explicitly utilized in the classical theory, which directly models the phenomenon at the local level, in terms of vectors helds that are the electric field E, the electric displacement D, and the polarization vector P, related by the classical formulas, valid in linear materials (including free space) as no operator is used for the system constitutive property. 

Overall, other than underestimation of compressive strains, the MVLEM proves to be an effective modeling approach for the flexural response prediction of slender RC walls, as the model provides good predictions of the experimentally observed global and local responses, including wall lateral load capacity and lateral stiffness at varying drift levels, yield point, cyclic properties of the load-displacement response, rotations (average over the region of inelastic deformations), position of the neutral axis and tensile strains. 

Figure 7 summarizes the data for two key dynamic observables the mean-square monomer displacement gi,mid(0 and i,end(0 and the stress relaxation fimction. Hgures 7(a) and 7(b) show the data as a limction of time in units where f= l,feBT= 1 and unit length is b in Rouse and semiflexible models, Tq in freely jointed and freely rotating models, and a in the LJ+FENE-based models. Since these units are pretty arbitrary and the local details are very different (including presence of inertia), the results are scattered and it is not easy to see similarities and differences between the models. 

---