Process optimization in pharmaceutical hot-melt extrusion: real-time volatile detection via SIFT-MS combined with multivariate analysis

Establishing robust processing windows for pharmaceutical polymers during hot-melt extrusion (HME) remains challenging, as conventional thermal analyses reveal little about early chemical change. Here, selected-ion-flow-tube-mass-spectrometry (SIFT-MS) combined with principal component analysis (PCA) was used to characterise real-time volatile evolution under both thermogravimetric (TGA) and extrusion conditions. Centroid-distance mapping and PCA loadings revealed distinct transitions, providing a data-driven means of defining the onset of significant chemical change. Across four representative polymers (Soluplus®, Affinisol™15LV, Kollidon® VA64, and Plasdone™ S630 Ultra), each exhibited changes in volatile composition that marked the onset of temperature-driven chemical evolution. Soluplus® and Plasdone™ S630 Ultra remained stable up to ≈190 °C with optimum extrusion ranges of 150-170 °C. Kollidon® VA64 showed earlier volatile emergence near 180 °C, defining a 160-180 °C window, while Affinisol™15LV, the most viscous system, degraded above 190-200 °C, narrowing its range to 170-185 °C. A brief rheological assessment supported these chemically defined limits, confirming that changes in volatile composition coincide with softening behaviour. Overall, SIFT-MS detected subtle, low-level volatile changes that emerge well before conventional thermal indicators, enabling rapid, non-destructive definition of polymer-specific extrusion windows and enhancing process understanding in amorphous solid dispersion manufacture. Through this analysis we were able to provide a narrower processing range than those defined by their respective manufacturers.