Multivariate analysis of relationships between material properties, process parameters and tablet tensile strength for α-lactose monohydrates

The present work describes an approach to quantify relationships between the material properties of various α-lactose monohydrate grades (αLM), process parameters (punch velocity, lubricant fraction) and the tablet tensile strength (TS). Milled, sieved, agglomerated and spray-dried αLMs were studied. Each material was tableted (11 mm flat punches, constant true volume of 0.2833 cm3) on a compaction simulator at a pressure of 104.4 ± 0.1 MPa. The force–displacement data was analyzed by applying a combination of compression descriptors (derived from Kawakita and Heckel equations, work-related parameters). The relationships were evaluated and quantified by principal component analysis (PCA) and partial least square regression (PLS-1). PCA verified fundamental relationships between different powder and compression properties of studied materials. It was found that the compression descriptors Kawakita ‘1/b’ and WoC were sufficient to distinguish the tested αLM-grades, even in combination with different lubricant fraction or by utilizing different punch velocities; the identified descriptors correlated with TS. These relationships were quantified by PLS-1. Finally, TS were successfully predicted for all αLM with the help of separate optimized PLS-1 models. The present study shows an approach how to extract relevant information about tableting behavior from a limited amount of material.