Synthesis of a biocompatible polymer using siloxane-based surfactants in supercritical carbon dioxide

Poly(N-vinyl-2-pyrrolidone) (PVP) particles were prepared by dispersion polymerization in the presence of 2,2′-azobisisobutyronitrile as the initiator and siloxane-based surfactant in supercritical carbon dioxide (scCO₂). The dispersants used in this study were non-ionic, non-reactive and commercially produced siloxane-based surfactants (Monasil PCA and KF-6017). We investigated the effect of kinds and concentrations of the surfactants, in addition to the reaction temperature and the concentration of the monomer on the particle size and morphology. PVP microspheres were prepared in 0.23–0.74 μm size range with Monasil PCA and 0.71–1.98 μm size range with KF-6017, respectively. The resulting polymer particle of >90% yield was obtained. Particle size slightly increased with the amount of monomer in polymerization with Monasil PCA. In the case of KF-6017 as the surfactant, there was not an obvious variation in particle size with increasing monomer. Particle size of PVP decreased as surfactant concentration increased from 5.0 to 15.0 wt.% basis on concentration of monomer. The narrow particle size distribution (D_n = 0.23 μm and PSD = 1.06) was presented at the high concentration of Monasil PCA (15 wt.% on monomer concentration). As indicated by the reaction temperature and the addition of organic solvent, which affected solubility of monomer, polymer and surfactant in scCO₂, particle size and particle size distribution of PVP varied. PVP particles with Monasil PCA strongly aggregated at 75 °C in contrast to KF-6017 which showed discrete particles at 65 and 70 °C, but particle size distribution was broad. Particle size was slightly reduced with a little amount of hexane, with an inverse relationship of adding hexane reduced the particle size. The amount of the relative residual surfactants on surface of the polymer after extracting with supercritical fluid process (SFE) was measured by using SEM/EDS and EPMA analysis to map out the distribution of silicon element qualitatively. The original polymer particle before the extraction using CO₂ had the high level of silicon element, but the average level of silicon element became low after CO₂ extraction.