Amorphous-Amorphous phase separation (AAPS) is an important phenomenon that can impede the performance of amorphous solid dispersions (ASDs). The purpose of this study was to develop a sensitive approach relying on dielectric spectroscopy (DS) to characterize AAPS in ASDs. This includes detecting AAPS, determining the size of the active ingredient (AI) discrete domains in the phase-separated systems, and accessing the molecular mobility in each phase. Using a model system consisting of the insecticide imidacloprid (IMI) and the polymer polystyrene (PS), the dielectric results were further confirmed by confocal fluorescence microscopy (CFM). The detection of AAPS by DS was accomplished by identifying the decoupled structural (α-)dynamics of the AI and the polymer phase. The α-relaxation times corresponding to each phase correlated reasonably well with those of the pure components, implying nearly complete macroscopic phase separation. Congruent with the DS results, the occurrence of the AAPS was detected by means of CFM, making use of the autofluorescent property of IMI. Oscillatory shear rheology and differential scanning calorimetry (DSC) detected the glass transition of the polymer phase but not that of the AI phase. Furthermore, the otherwise undesired effects of interfacial and electrode polarization, which can appear in DS, were exploited to determine the effective domain size of the discrete AI phase in this work. Here, stereological analysis of CFM images probing the mean diameter of the phase-separated IMI domains directly stayed in reasonably good agreement with the DS-based estimates. The size of phase-separated microclusters showed little variation with AI loading, implying that the ASDs have presumably undergone AAPS upon manufacturing. DSC provided further support to the immiscibility of IMI and PS, as no discernible melting point depression of the corresponding physical mixtures was detected. Moreover, no signatures of strong attractive AI-polymer interactions could be detected by mid-infrared spectroscopy within this ASD system. Finally, dielectric cold crystallization experiments of the pure AI and the 60 wt % dispersion revealed comparable crystallization onset times, hinting at a poor inhibition of the AI crystallization within the ASD. These observations are in harmony with the occurrence of AAPS. In conclusion, our multifaceted experimental approach opens new venues for rationalizing the mechanisms and kinetics of phase separation in amorphous solid dispersions.
Keywords: amorphous phase separation; confocal fluorescence microscopy; dielectric spectroscopy; domain size; molecular interactions; relaxation; solid dispersion.