Image-based characterization of powder flow to predict the success of pharmaceutical minitablet manufacturing

• Published in: International journal of pharmaceutics Vol. 581; p. 119280
• Main Authors: Blanco, David, Antikainen, Osmo, Räikkönen, Heikki, Mah, Pei Ting, Healy, Anne Marie, Juppo, Anne Mari, Yliruusi, Jouko
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 15.05.2020
• Subjects: Cellulose - analysis; Cellulose - chemistry; Chemistry, Pharmaceutical - methods; Cohesive powders; Forecasting; Image analysis; Microcrystalline cellulose; Powder flow; Powders; Rheology - methods; Small scale; Tableting; Tablets; Powder flow; Image analysis; Microcrystalline cellulose; Cohesive powders; Tableting; Small scale
• ISSN: 0378-5173; 1873-3476
• DOI: 10.1016/j.ijpharm.2020.119280

[Display omitted] Powder flowability plays an important role in die filling during tablet manufacturing. The present study introduces a novel small-scale measuring technique for powder flow. Based on image analysis, the flow was defined depending on the variation of luminous intensity and the movement of powder inside the measurement cuvette. Using quantities around 100 mg it was possible to characterize a wide range of common pharmaceutical powders, especially in distinguishing subtle differences in flow caused by minor changes in samples characteristics. The method was compared with powder rheometry, which is widely used in the pharmaceutical literature, and showed a significant improvement in predicting the success of pharmaceutical minitablet manufacture (d = 5 mm). Tablet weight variation (RSD) was defined as the most efficient way to assess relevant powder flow behaviour in tablet production when using the novel device. The proposed method was distinguished from others by its ability to classify different grades of microcrystalline cellulose in the die-filling process. Subsequently, eight common pharmaceutical powders, both excipients and APIs, were properly ranked as a function of flowability based on their physical properties. The method showed a high repeatability, with a relative standard deviation not more than 10%.