Predicting the compressibility and compactibility profiles of pharmaceutical active ingredients for design of...

Digital design of multi-component pharmaceutical tablets based on the properties of individual constituents plays a critical role in the rational design and optimisation of pharmaceutical formulations. Most active pharmaceutical ingredients (APIs) used in tablet formulations are crystalline materials with diverse mechanical properties, including elastic, plastic, brittle, or combined deformation characteristics. These properties can lead to manufacturing challenges such as capping and sticking during the tableting process, or the formation of fragile tablets, making the direct compaction and testing of pure API tablets difficult or even impossible. In this study, we present an approach to predict the compressibility and compactibility profiles of APIs that cannot be directly compacted into tablets without the support of excipients. The method is based on the assumptions of additive volume fractions and the geometric mean mixing rule applied to the compactibility models of the individual components. API compressibility and compactibility models were derived from the analysis of "out-of-die" compaction data obtained from binary powder mixtures containing 50% API and 50% microcrystalline cellulose, compressed at different compaction pressures. Five APIs with diverse mechanical properties, i.e., aspirin, carbamazepine, metronidazole, paracetamol, and theophylline, were investigated. The proposed approach successfully predicted tablet solid fractions and tensile strengths for both binary (API and filler) and ternary (API, filler and disintegrant) mixtures of the APIs. The predicted tablet solid fractions were within ± 5% of the measured values, while tensile strength predictions showed errors typically ranging from ± 20% to ± 50%, depending on the API, formulation, and compaction pressure. Overall, the approach provides a practical digital design tool for the formulation of multi-component pharmaceutical tablets based on constituent material properties.