Origin of particle clustering in a simulated polymer nanocomposite and its impact on rheology

• Published in: The Journal of chemical physics 2003-07, Vol.119 (3), p.1777-1788
• Main Authors: Starr, Francis W., Douglas, Jack F., Glotzer, Sharon C.
• Format: Article
• Language: English
• ISSN: 0021-9606; 1089-7690
• DOI: 10.1063/1.1580099

Many nanoparticles have short-range interactions relative to their size, and these interactions tend to be “patchy” since the interatomic spacing is comparable to the nanoparticle size. For a dispersion of such particles, it is not a priori obvious what mechanism will control the clustering of the nanoparticles, and how the clustering will be affected by tuning various control parameters. To gain insight into these questions, we perform molecular dynamics simulations of polyhedral nanoparticles in a dense bead–spring polymer melt under both quiescent and steady shear conditions. We explore the mechanism that controls nanoparticle clustering and find that the crossover from dispersed to clustered states is consistent with the predictions for equilibrium particle association or equilibrium polymerization, and that the crossover does not appear to match the expectations for first-order phase separation typical for binary mixtures in the region of the phase diagram where we can equilibrate the system. At the same time, we cannot rule out the possibility of phase separation at a lower temperature. Utilizing the existing framework for dynamic clustering transitions offers the possibility of more rationally controlling the dispersion and properties of nanocomposite materials. Finally, we examine how nanocomposite rheology depends on the state of equilibrium clustering. We find that the shear viscosity for dispersed configurations is larger than that for clustered configurations, in contrast to expectations based on macroscopic colloidal dispersions. We explain this result by the alteration of the polymer matrix properties in the vicinity of the nanoparticles. We also show that shear tends to disperse clustered nanoparticle configurations in our system, an effect particularly important for processing.