I. G. Jolliffe

Theoretical considerations of the filling of pharmaceutical hard gelatin capsules

Abstract The filling of pharmaceutical hard gelatin capsules by the dosator nozzle system requires that an accurate amount of powder is picked up and retained within the cylindrical nozzle, during a transfer process. Retention may be assisted by the application of compressive stress which should be minimal, so that subsequent ejection can be achieved with minimum effort. Application of theories derived to predict arching conditions in mass flow hoppers has enabled some of the principal factors affecting powder retention within the nozzle to be established. This approach shows that the angle of powder—wall friction is important in determining the stress distribution of powder within the nozzle. There is an optimal value of this angle for which the compressive force required to ensure arch formation is a minimum. Consideration of the powder—wall interaction could improve capsule filling ability.

The effect of dosator nozzle wall texture on capsule filling with the mG2 simulator

The effect on capsule filling, using an mG2 simulator, of the surface texture of the bore of the dosator nozzle has been investigated for size fractions of lactose. Since the angle of powder‐wall friction between the powder and the nozzle cannot be readily measured insitu, this was determined using flat plates with similar surface textures fitted into a Jenike shear cell apparatus. The problems of reproducing surface textures on both types of surface are discussed and a lapping process employed as the most suitable method. Angles of wall friction for the nozzle surfaces were extrapolated from the values obtained from flat plates of similar roughness. Capsule filling experiments, using the mG2 simulator, showed the three resurfaced nozzles produced more uniform fill weights with smaller measured compression and ejection stresses than an untreated nozzle surface. One lapped nozzle surface produced a slightly greater improvement than the others. This supports the concept of an optimum angle of wall friction for powder retention (and hence uniformity of fill) with a minimum of applied compression stress.

Practical implications of theoretical consideration of capsule filling by the dosator nozzle system

Eight lactose size fractions with mean particle sizes ranging from 15.6 to 155.2 μm were characterized by their failure properties using a Jenike shear cell. The effective angle of internal friction was found to be constant for all size fractions, with a mean value of 36.2°. Jenike flow factors could only be obtained for the two most cohesive size fractions presumably due to limitations of the shear cell. Angles of wall friction, ϕ, were determined for all size fractions on face ground and turned stainless steel surfaces. These decreased with increasing particle size up to around 40 μm, above which they became effectively constant for both surfaces. The rougher turned plate gave consistently higher values of ϕ for each particle size. Simple retention experiments with a dosator nozzle and a range of powder bed bulk densities showed good retention was possible only up to a particle size of around 40 μm. Retention was difficult or impossible above this size. Values of ϕ were applied to equations derived in the theoretical approach described previously (Jolliffe et al 1980). This showed that the strength required within a powder to ensure arching increases with increasing particle size up to around 40 μm. Above this size, this strength requirement becomes constant. This is related to the powder retention observations. Finally, the failure data was used to calculate the minimum compressive stresses required to ensure powder retention within the dosator nozzle, by employing the equations described by Jollife et al (1980). This suggested that, as powders became more free flowing, a larger compressive stress is necessary and that the angle of wall friction should be lower to ensure stress is transmitted to the arching zone.

The design and use of an instrumented mG2 capsule filling machine simulator.

The problems of instrumenting a continuous motion dosator nozzle capsule filling machine (mG2 type) are discussed and the construction of an mG2 simulator is described. A standard filling turret is employed with mechanical modification so that the dosator does not rotate. This allows instrumentation in the form of a strain gauged dosator piston, to measure compression and ejection stresses during filling, as well as distance transducers, to measure the corresponding piston and dosator movements. An experimental method is described for using this machine to study the filling of lactose powders.

An investigation of the relationship between particle size and compression during capsule filling with an instrumented mG2 simulator

An instrumented mG2 capsule filling machine simulator has been employed to study the effects of the amount of compression (compression ratio) on the capsule fill weight uniformity and measured compression and ejection stresses. Four size fractions of lactose were studied (mean particle sizes 15.6, 17.8, 37.5 and 155.2 μm). The range of compression over which satisfactory filling could be achieved was large for fine, cohesive powders but decreased with increasing particle size. The lower limit of filling ability was the ability to retain the powder and the amount of compression needed to achieve retention increased with increasing particle size. The upper limit on compression, was the compaction of the powder which prevented the piston acting to cause retention. Large particle sizes were able to undergo only a small change in volume before compaction occurred whilst fine, cohesive powders were considerably more compressible and hence could be filled satisfactorily at higher compression settings.

Capsule filling studies using an mG2 production machine

A production mG2 G36 machine has been employed to study the effects of compression on the capsule filling properties of four particle size fractions of lactose having a range of flow properties. The effect of the surface texture of the dosator nozzle bore on capsule filling is also investigated. Fine, cohesive powders gave uniform fill weights over a whole range of compression settings but as increasingly free‐flowing powders were used, this range diminishes. For both types of powder, the upper limit on compression is set by compaction of powder which produces poor fill weights; coarse, free‐flowing powders, which are less compressible, compact at lower compressions. Free‐flowing powders, in particular, also require a minimum compression to be retained. Resurfaced nozzles produced improved capsule filling. One nozzle surface produced slightly more uniform fill weights and was unaffected by powder coating of the nozzle suggesting that an optimum surface texture exists for capsule filling. The results are similar to those obtained using the mG2 simulator and hence validate the latters use in studying production capsule filling.

Extension of theoretical considerations of the filling of pharmaceutical hard gelatin capsules to the design of dosator nozzles

Abstract A previous paper presented theoretically how the compressive stress, applied to a powder in a capsule-filling dosator nozzle to ensure retention, can be minimised. A theoretical method of reducing these stresses still further is proposed. This involves using a nozzle with two areas of different values of angle of wall friction φ. The main part of the nozzle should have a low value of φ, but the wall near the nozzle outlet, a high value of φ. This ensures that effective stress transmission to the nozzle outlet occurs with minimal frictional losses at the wall and that powder arching (and hence retention) is promoted by the rougher wall surface at the nozzle outlet. This in turn reduces the amount of stress required for arching to occur.

POWDER RETENTION WITHIN A CAPSULE DOSATOR NOZZLE

Capsule filling systems which rely on a dosator nozzle transferring powder from a feed bed to a capsule shell, depend on the formation and retention of a powder plug within the nozzle. This plug must then be ejected completely with minimal force. Powder retention requires the formation of a stable arch at the nozzle outlet. The conditions which govern the formation of stable powder arches in parallel sided containers have been discussed by Walker (1966), who also presented equations to calculate the conditions necessary to form such arches. These equations require a knowledge of the bulk density 1, the angle of powder/wall friction &, and the effective angle of internal friction 6 of the powder. According to Walkers theory, the existance Of a free surface at the arch, implies that the strength of the powder under these conditions is the unconfined yield strength fc, whose value can be calculated from the expression: