We have developed a simulation technique of multiscale Lagrangian fluid dynamics to tackle hierarchical problems relating to historical dependency of polymeric fluid. We investigate flow dynamics of dilute polymeric fluid by using the multiscale simulation approach incorporating Lagrangian particle fluid dynamics technique (the modified smoothed particle hydrodynamics) with stochastic coarse-grained polymer simulators (the dumbbell model). We have confirmed that our approach is nicely in agreement with the macroscopic results obtained by a constitutive equation corresponding to the dumbbell model, and observed microscopic thermal fluctuation appears in macroscopic fluid dynamics as dispersion phenomena.
For the purposes of long-term use of tungsten divertor walls, the formation process of the fuzzy tungsten nanostructure induced by exposure to the helium plasma was studied. In the present paper, the fuzzy nanostructure's formation has been successfully reproduced by the new hybrid simulation method in which the deformation of the tungsten material due to pressure of the helium bubbles was simulated by the molecular dynamics and the diffusion of the helium atoms was simulated by the random walk based on the Monte Carlo method. By the simulation results, the surface height of the fuzzy nanostructure increased only when helium retention was under the steady state. It was proven that the growth of the fuzzy nanostructure was brought about by bursting of the helium bubbles. Moreover, we suggest the following key formation mechanisms of the fuzzy nanostructure:(1) lifting in which the surface lifted up by the helium bubble changes into a convexity, (2) bursting by which the region of the helium bubble changes into a concavity, and (3) the difference of the probability of helium retention by which the helium bubbles tend to appear under the concavity. Consequently, the convex-concave surface structure was enhanced and grew to create the fuzzy nanostructure.
Predicting the flow of an entangled polymer melt is still difficult because of its multiscale characteristics. We have developed a novel multiscale simulation technique to investigate the history-dependent flow behavior of entangled polymer melts. The technique involves using a smoothed particle hydrodynamics simulation that is coupled at each fluid element to microscopic simulators that can accurately account for the dynamics of entangled polymers. The multiscale simulation is used to investigate the flow of an entangled polymer melt around a cylindrical obstacle subject to periodic boundary conditions. It is found that the macroscopic flow behavior is dependent on the history of the microscopic states of the polymers and that this memory causes nonlinear behavior even in the regions where the local Weissenberg number defined using the local strain-rate is less than unity. The spatial distribution of the entanglements Z suggests that, in a region around the obstacle, a slight depletion of the entanglements is observed and that this region broadens along the downstream direction. The totality of the presented results suggests that we have succeeded in describing the entangled polymer melt flow without using any constitutive equation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.