Homogeneous Versus Inhomogeneous Methods

The second drawback will be spatial inhomogeneities. In general, most simulations are concerned with bulk properties of a fluid. The presence of a surface, however, introduces some anisotropic behavior to the fluid. For exam- 

Other kinds of inhomogeneity can also arise with this approach. For example, in experiments the container is placed in contact with a heat bath that regulates the temperature of the system. Shear motion tends to generate viscous heating, which is then conducted through the container wall and out to the heat bath. Hence, fluid particles closer to the wall will be colder than those farther away. If we limit ourselves to atomistic length scales, a simple hydrodynamic theory predicts that the maximum local temperature depends quadratically on the distance from the wall. The combination of these and other inhomogene- 

The effect of surfaces on fluid density as a function of distance from the 

To address this, Liem, Brown, and Clarke ° simulated in excess of 40,000 particles interacting via a Weeks-Chandler-Andersen (WCA) potential. While the x and z directions were treated normally, the y direction was divided into three regions two atomistic walls separated by a fluid region. The walls consisted of three hexagonally close-packed layers of particles. The wall atoms interacted with the fluid particles and with each other through the same WCA potential used for the fluid-fluid interactions. Additionally, each wall particle felt a harmonic potential centered at its triangular lattice site. This setup allowed heat transfer from the fluid to the wall while allowing the wall to remain crystalline. The momenta of the wall particles were rescaled to keep the total 

Hopefully, we have hinted that there are interesting nonequilibrium systems that can be explored using homogeneous, field-driven simulations. Because these approaches are often extensions of equilibrium methods, it is natural to first present the equilibrium foundation, as is done in the next section. From there, we will be able to develop the theoretical basis for NEMD simulations and practical guidelines for implementing them. Once the tools are in place, we will discuss applications and the kinds of question that can be tackled by NEMD methods. 

---

A plot of tan 8 versus temperature gives a strong indication of blend homogeneity when a single T is detected or inhomogeneity when two T s are detected. An example of this method in studying blends of ethylene-vinyl acetate copolymers (45 wt.- % Ac) with chlorinated polyethelene (52 wt.- % Cl) at a constant frequency of 11 Hz is shown inFig, 10. Similar results obtained for blends ofchlorinated polyethylene (44 wt.-%Cl) with chlorinated polyethylene (62 wt.- % Cl) are shown in Fig. 11. 

---