(477g) Coarse-Grained Modeling of Polarizable Nanoparticle Assembly

Polarizable nanoparticles are of special interest in materials science because of their rich phase behavior, and consequently in designing engineered materials with several promising applications. With short inorganic ligands such as metal chalcogenide complexes as capping groups, polarizable nanoparticles in polar solvents can form highly ordered 3D superlattices with safer and excellent optoelectronic properties. This however depends on the delicate interplay between the dispersion and the strength of the electrostatic interactions governed by the design parameters such as temperature, size-charge of the nanoparticle and the dielectric strength of the solvent. To understand and rationally navigate the design space for forming highly ordered lattices, we develop a coarse-grained model by also taking polarizability of the nanoparticle into account using the image method for capturing the many-body polarization effects. Thereby, extending the design space typically considered for nanoparticles with polarizability and in advancing the experimental realization of simulations.

We employ an active learning framework to efficiently sample the free energy landscapes within the high-dimensional particle design space using enhanced sampling molecular dynamics simulations. The trends in the phase boundaries quantitatively predict assembly behavior and reveal that the attraction regime widens with increase in particle size, dielectric constant of solvent and dielectric contrast between nanoparticle and solvent. Similar to experiments, we observed that the structure of self-assembled aggregates become more ordered with decreasing nanoparticle size and increasing dielectric constant of solvent. Having validated our computational techniques, we then employ our approach for high throughput virtual screening and rational nanoparticle design to engineer switchable materials systems composed of polarizable nanoparticles that are capable of triggered assembly and disassembly upon changes in temperature and solvent quality. We demonstrate and refine these computational designs in experimental tests.