Computer simulation study on the self-assembly of unimodal and bimodal polymer-grafted nanoparticles in a polymer melt.

The controllable distribution of nanoparticles (NPs) in polymer nanocomposites (PNCs) is a challenge in materials science. An important method is grafting chains that are chemically identical to the polymer matrix on NPs. By performing comprehensive molecular dynamics simulations, the self-assembly behavior of polymer-grafted NPs in a polymer matrix is investigated in this study. The relationship between the grafted chain length N, grafting density σ and the NPs' self-assembly morphologies is studied. Phase diagrams of the NP self-assembly structures for both unimodal and bimodal grafted NP systems are constructed on a parameter space, where P is the matrix polymer chain length. NP self-assembly structures of strings, connected/sheet and small clusters are identified in different regions. In order to quantitatively characterize the NP self-assembly morphology, we define a morphological measurement parameter which characterizes the distribution of the Voronoi cell volume of the NPs. Using this parameter, we discuss the influences of both long and short grafted chains on the dispersion of bimodal polymer-grafted NPs in a polymer melt. We find that the short grafted chains can not only shield the NP surface from the polymer matrix but also elongate the long grafted chains into the polymer matrix, therefore favoring a better dispersion of NPs. Our results also indicate that the bimodal grafted NPs will not be fully dispersed until the short grafted chains are dense enough to elongate the long grafted chains, hence forming a wetting NP/matrix interface.