Correlating Die-Filling Performance with Powder Properties

Many pharmaceuticals are produced in premetered doses: as tablets, for example, or as powders for inhalation or for dissolution before intravenous or oral delivery. Filling processes into tablet dies, capsules, and other vessels, are therefore a routine aspect of pharmaceutical manufacture. The goal of these processes is to ensure the consistent and precise extraction of the required dose at an acceptable processing speed.

Filling machines may have complex and varied designs and geometries, depending on the application for which they are used, and on the properties of the process material. Oral capsules, for example, typically have a fill weight in the range of 50 to 500 mg, while dry-powder inhaler (DPI) applications require a much smaller mass, usually between 0.5 and 15 mg, and tend to involve highly cohesive formulations. The optimal filling procedure is different in each case, but well-controlled, uninterrupted powder flow through the selected geometries is a prerequisite for efficient processing. Success is judged by the consistency of the fill weight.

Identifying a suitable filling solution for a given formulation at the outset minimizes operability problems over the long term. Relevant characterization of the formulation (i.e., measuring the properties that predict how it will behave during processing) is the first step. This principle is especially true in the context of recent regulatory initiatives, such as quality by design, which emphasize a thorough understanding of the process-control space. To select the best filling geometry and properly control manufacture, it is necessary to measure the powder in a way that reveals how it will perform as it flows through the filling machine.

This article details a recent experimental study by Freeman Technology and Innovative Technology for Custom Machinery (ITCM) to explore correlations between powder properties and behavior in two filling geometries. The results confirmed that no one-size-fits-all filling solution exists, but pinpointed the properties that can be usefully measured to achieve a compatible powder–plant match in filling applications.

Different geometries for different formulations

A recent study evaluated the performance of four pharmaceutical excipients (i.e., S1, S2, S3, and S4) for two filling configurations using a commercially available dosing and filling machine. Figure 1 shows configurations A and B. In both configurations, powder flows down through a channel to a pocket on a rotating wheel. As the wheel rotates, powder is transferred from the pocket to a waiting receptacle, and the wheel rotates back to the filling position.

Figure 1: Filling geometry configuration A (left) and filling geometry configuration B (right).

The volume of the pocket into which powder flows is the same (~410 mm³ ) in both geometries. On the other hand, with configuration A, the feed channel is narrower and more closely matched with the diameter of the pocket opening. In contrast, the feed channel in configuration B is much wider, encircling the entrance to the pocket.

In both systems, the extraction of a consistent volume relies on the complete filling of the pocket every time the wheel rotates, and on the efficient transfer of that volume to the waiting receptacle. Consistent volume translates directly into consistent mass, which is the goal of the process, as long as bulk density remains the same. Process performance for each of the excipients was therefore described using a process-capability index, C_pk, derived from the weight variance of the doses produced (1).

Measuring powder properties

To fully characterize the properties of each of the four formulations, personnel subjected them to multifunctional testing using a powder rheometer. During the past decade, powder rheometers have evolved to offer multiple measuring techniques in a single instrument, including shear, bulk, dynamic, and axial methods. Personnel measured shear, bulk, and dynamic properties to quantify properties, such as cohesiveness, flow function, aeration ratio, basic flow energy, and permeability.

Cohesion and flow function are parameters derived from shear-test data. With this methodology, one powder face is sheared against another to ascertain the strength of particle interactions. Shear testing is a well-established, widely recognized technique, but has the limitation that powders are measured in a consolidated state. Information about how the powders will behave when flowing freely or when aerated is derived theoretically rather than measured directly. High flow functions typically are associated with powders with low cohesion that are expected to flow freely, whereas low flow functions assume high cohesion.

Dynamic testing is a modern technique that involves measuring the forces acting on a blade as it rotates along a defined path through the powder sample. These measurements are used to determine the energy required to induce flow. The powder is measured in motion, hence the term "dynamic." This technique enables the testing of consolidated, conditioned, aerated, and fluidized powders. Aeration ratio, for example, is a parameter that indicates the extent to which flow energy changes when the sample is aerated. Generally speaking, the flow energy of cohesive materials changes little when the sample is aerated. The attractive forces between particles resist particle separation, and the air channels through the powder bed, having relatively little effect. In contrast, with less cohesive materials, the air flows between the particles, substantially facilitating movement within the bed and dramatically reducing flow energy. Aeration is therefore a complementary way of investigating cohesiveness in a way that arguably reflects in-process behavior more closely than shear testing alone. Reference 2 provides a more complete description of all of the powder testing techniques used in this study.

Correlating process performance with powder properties

Figures 2 and 3 show correlations between filling performance and four powder properties for configurations A and B. Data are shown for all four formulations for configuration A, and for S1, S2, and S4 for configuration B. S3 was not intended to be processed on configuration B during production, and data for this combination are therefore unavailable.

Figure 2: Correlations between filling-performance Cpk in configuration A and various powder properties.

The results indicated that for configuration A, less cohesive materials were more compatible than cohesive materials. High basic flowability energy, aeration ratio, and flow function, along with low cohesion, were correlated to improved processing performance. All of these features indicated low cohesiveness. The findings suggested that for this configuration, the powder's ability to flow freely into the feed channel and pocket determined process performance. The small feed diameter of the configuration made this design more susceptible to the powder arching and to the discontinuous flow that can occur with cohesive materials, which could have a dramatic effect on processing performance.

Figure 3: Correlations between filling performance Cpk in configuration B and various powder properties.

The results for configuration B showed different behavior. For this configuration, cohesive formulations (i.e., those with low basic flowability energy, aeration ratio, and flow function) performed better. This suggested that with the wider feed channel, the ability of the powder to flow freely under gravity was less important. Further research would be required to fully understand the mechanisms that gave rise to the observed behavior, but it is worth reiterating that the filling process involves more than simply filling the pocket. Successful transfer of the extracted volume as the wheel rotates is also essential. The wide feed channel may have encouraged successful pocket filling for all formulations, regardless of cohesiveness, but as the wheel rotates, less cohesive materials may have been lost more easily, thus compromising dose weight. The types of powders that were optimal for configuration A clearly were suboptimal for configuration B, and vice versa.

Conclusion

To succeed in matching filling-machine geometry to the demands of a specific formulation, it is critical to gain a good understanding of the relationships between powder characteristics and process performance. The results presented in this article suggest that the optimal powder property set depends on machine geometry. Although free-flowing formulations are preferable for certain geometries, as might be expected, cohesive formulations perform better for other geometries.

The data indicated that certain types of geometry were better suited to cohesive materials, thus highlighting the importance of studies such as these for personnel who develop filling solutions. Equally important, the data suggested that new, free-flowing formulations will not necessarily perform better on an existing unit than previous products. Appropriate powder testing holds the key to achieving a compatible powder–geometry match that delivers efficient operation over the long term.

Tim Freeman is director of operations at Freeman Technology, Boulters Farm Centre, Castlemorton Common, Welland, Worcestershire, WR13 6LE, UK, tel. +44 0 1684 310860, fax +44 0 1684 310236, info@freemantech.co.uk.

References

1. S. Kotz and N.L. Johnson, Process Capability Indices (Chapman and Hall, London, 1993).

2. R. Freeman, Powder Technol. 174 (1–2), 25–33, (2007).