Study of a low-dose capsule filling process by dynamic and static tests for advanced process understanding

Experimental and Model-Based Study of the Vibrations in the Load Cell Response of Automatic Weight Fillers

The paper presents a study of the vibrations in the load cell response of automatic weight fillers for fluids, due to the dynamics of the system. The aim is to characterize vibratory phenomena through both experimental and model-based analysis, in order to identify the main causes and identify compensation strategies. Two test campaigns were conducted, on a test bench and on a sixteen stations machine, with the simultaneous acquisition of acceleration signals and load cell signals. A detailed sensitivity analysis based on experimental data, as many system parameters vary, has been developed. For the system modelling, a one Degree of Freedom (1 DoF) model, with lumped parameters and time-variant mass, including fluidic forces, was considered and numerically implemented. Genetic algorithms were used for the identification problems in the model-based analysis. The model allowed a deeper understanding of the phenomena that occur, showing promising results for the vibration prediction in a compensation process.

Study of the capsule-filling dosator process via calibrated DEM simulations

Capsule filling is frequently accomplished via the dosator process. Controlling the main quality attributes, i.e., the fill weight and the fill weight variability, requires excellent process understanding. For the investigation of critical process parameters of low-dose capsule filling, DEM simulations were used. Two contact models (Hertz model and Luding model) were calibrated and validated to represent the lactose powder Lactohale 100. Both models gave good results, yet the Luding model resulted in better predictions. Since the dosator process is a volume-based approach, the dosator geometry is an important factor. We provide evidence that not only the volume, but also the proportions of the dosator affect the capsule fill weight. Slim nozzle chambers cause lower mass due to increased wall friction impact. The process is not volumetric, as a fill weight to volume correlation does not exist. Furthermore, the fill weight increases with higher powder beds and smaller gaps between the nozzle tip and the container bottom due to higher densification of the powder. Fill weight variations due to bed inhomogeneity were investigated by varying the bulk density of the initial powder bed.

Predicting capsule fill weight from in-situ powder density measurements using terahertz reflection technology

The manufacturing of the majority of solid oral dosage forms is based on the densification of powder. A good understanding of the powder behavior is therefore essential to assure high quality drug products. This is particularly relevant for the capsule filling process, where the powder bulk density plays an important role in controlling the fill weight and weight variability of the final product. In this study we present a novel approach to quantitatively measure bulk density variations in a rotating container by means of terahertz reflection technology. The terahertz reflection probe was used to measure the powder density using an experimental setup that mimics a lab-scale capsule filling machine including a static sampling tool. Three different grades of α-lactose monohydrate excipients specially designed for inhalation application were systematically investigated at five compression stages. Relative densities predicted from terahertz reflection measurements were correlated to off-line weight measurements of the collected filled capsules. The predictions and the measured weights of the powder in the capsules were in excellent agreement, where the relative density measurements of Lactohale 200 showed the strongest correlation with the respective fill weight (R2=0.995). We also studied how the density uniformity of the powder bed was impacted by the dosing process and the subsequent filling of the holes (with excipient powder), which were introduced in the powder bed after the dosing step. Even though the holes seemed to be filled with new powder (by visual inspection), the relative density in these specific segments were found to clearly differ from the undisturbed powder bed state prior to dosing. The results demonstrate that it is feasible to analyze powder density variations in a rotating container by means of terahertz reflection measurements and to predict the fill weight of collected capsules.