Nanoprecipitation of polymers in a bad solvent

Nanoparticles (NPs) composed of polymers are of great interest since they can add a functionality in many applications, ranging from food and pharma to plastics and electronics. A key factor that determines the functionality of NPs is their size. Nanoprecipitation [1] is a commonly used technique to prepare NPs. We have performed a combined experimental and theoretical study on the size of NPs formed by precipitation of polymers into a bad solvent in the presence of a stabilizing surfactant. We propose an extension of the theoretical framework put forward by Lannibois et al. [10] for nanoparticle formation via precipitation. Our theory is based upon a kinetic model for diffusion limited coalescence (DLC) in which the relevant transport and diffusion mechanisms are quantified. We find that the macroscopic mixing time and the diffusivity of the polymer and surfactant are the main parameters determining the final particle size. The theoretical result for the final size can be condensed into a single analytical expression. At given polymer concentration and mixing time, it follows that the smallest particles can be obtained in the excess of surfactant. This situation corresponds well to the experimentally used conditions. The mixing efficiency is predicted to have a profound influence on the final particle diameter: faster mixing results in smaller particles. The final particle size in the slow mixing regime, which is the typical situation in experiments, turns out to be independent of the molar mass of the polymer and scales as a power 1/3 with the initial polymer concentration. An increase of the surfactant molar mass is predicted to lead to larger particles, because of longer mixing time and lower surfactant mobility. We have performed systematic experimental investigations on nanoparticles formation using various systems but focused on polycaprolactone (PCL) polymers in acetone precipitated in aqueous solutions containing polyvinyl alcohol as surfactant. The PCL molar mass and concentration and mixing time were varied. We show that both our experimental results as well as literature data are in good agreement with our theoretical DLC predictions. This work therefore provides a solid framework for tailoring nanoparticles with a desired size