Dingeman van der Haven ^a, Maria Mikoroni ^b, Umair Zafar ^c, James Elliott ^a, Ioannis Fragkopoulos ^b

^a Department of Materials Science & Metallurgy, University of Cambridge, 27 Charles Babbage Road, Cambridge, CB3 0FS, United Kingdom
^b Oral Drug Product Process Development, Novo Nordisk A/S, Måløv, Denmark
^c Oral Formulation Research, Novo Nordisk A/S, Måløv, Denmark

Purpose. The production of pharmaceutical tablets, particularly with direct compression, is challenging due to defects that can often occur during tablet ejection. Although ejection has been studied using various metrics such as static friction, kinetic friction, and ejection force, extensive comparisons of such metrics are still missing, especially at different ejection speeds and tablet densities. We aim to improve the current understanding of friction during tablet ejection so that such potential defect causes can be identified more swiftly.

Methods. We performed a systematic study on numerous ejection metrics by varying the powder, the tablet density, and the ejection speed. The materials studied were microcrystalline cellulose (MCC), dibasic calcium phosphate dihydrate (DCPD), and partially pre-gelatinised corn starch (STA). Five different densities were considered per powder. Each density was ejected using five different ejection speeds, ranging from 0.3 to 200 mm/s. Machine sampling speed and dwell time were also investigated.

Results. Ejection forces showed distinct trends depending on the powder, with e.g. DCPD showing a strong dependence on density and ejection speed, whereas STA showed a weak dependence on both. Moreover, methods were developed to estimate the static and kinetic friction during tablet ejection using only parameters already available from compaction simulations with an instrumented die. Static and kinetic friction show opposite trends, decreasing vs. increasing, as density increases.

Conclusions. Following the markedly different trends observed for static friction, kinetic friction, and ejection force, we propose that a distinction between these parameters is warranted when investigating ejection defects.