The
research capsule filling machine (Labby) has only one
nozzle dosator and can be operated using relatively
small quantities of powder. At the moment of submitting this abstract, nozzle dosators that correspond to capsule size 3 and 5 have been
investigated and the powder used is microcrystalline cellulose (Avicel 102). The scope of the study will be enlarged to
include cohesive as well as free flowing powders and also larger nozzles (size
0 and 1). The main finding using the research capsule filling machine is that
powder net weight in capsules increases as we increase the speed of capsule
filling. It was evident to the experimenter that, although subtle, a different
level of mechanical vibration caused by the driving engine is associated to the
different speeds of operation of the capsule filling machine. The latter causes
different degrees of powder densification or porosity (Woodhead,
1980) in the bowl (as we could also assess), which easily explains the different
net powder weight in capsules. Nozzle dosators
operate on a volumetric basis: the nozzles collect a specified volume of powder
from a rotating bowl and subsequently transfer the powder into an open capsule.
The following correlation is observed: sampling denser powders leads to larger
net weight in capsules.

When
the operations are carried out in an industrial capsule filling machine (Planeta, MG2), the effect of speed on powder net weight was
negligible. In fact, all MG2 production machines have a weight control system
available. This system automatically changes the dimension of the dosing
chamber in the nozzle to keep the target weight whenever it is necessary (i.e. when
powder density inside the rotary bowl changes). The weight control system makes
it possible to automatically adjust the average net weight in the capsules,
maintaining it always as close as possible to the target weight (?theoretical
net weight?). This process is called ?automatic adjustment?.

In
addition to the development of capsule filling operations, the work presented
here is relevant to current good manufacturing practices (cGMP)
in the pharmaceutical industry. Mechanical vibration is an environmental
variable that, just as relative humidity, affects powder properties and the
quality of powder based products. Therefore, just as relative humidity, it
should be carefully assessed and controlled.

Methods

1. Capsule filling: Performed using MG2 equipment (Labby and Planeta).
2. Capsule weight: groups of 100 capsules manufactured with different nozzles: 22 mg (nozzle size 5), 50 mg (nozzle 3), 120 mg (nozzle 3) are weighed.
3. Powder density in the bowl: A graduated glass cylinder is fixed to the driver of the rotating powder bowl (therefore bowl and glass cylinder are exposed to equivalent vibration). The powder densification can be visually observed and measured in the glass cylinder.
4. Vibration of the research capsule filling machine: Assessed using Laser Doppler Vibrometer (LDP).

Results

1. Effect of capsule filling speed (and vibration) on powder density in the bowl

Figure 1:
This figure shows that the powder density (measured in a glass cylinder that is
fixed on the driving wheel of the rotating bowl) increases as the speed of the capsule
filling machine increases.

2. Effect of capsule filling speed (and vibration) on capsule net weight

Figure 2:
This figure shows an increasing trend of powder net weight in capsules as the speed
of the capsule filling machine increases. Data obtained using a nozzle size
number 5.

Figures 3 (left
and right): These figures show an increasing trend of powder net weight in
capsules as the speed of the capsule filling machine increases. Data obtained
using a nozzle size number 3 and two different dosing chamber spaces.

Conclusion

Higher
capsule filling speed of the research capsule filling machine (Labby) is coupled to more intense vibration of the engine,
which leads to a further densification of the powder in the bowl. As a result,
the powder mass collected in the volume of the dosing chamber of the nozzle dosator (and hence powder net weight in capsules) is larger.
Although the effect is less evident (and not statistically significant), there
is a clear trend for low doses (~20 mg). As we increase the volume of powder
collected by the nozzle (~120 mg), the effect of powder bed densification on
capsule net weight becomes more obvious. However, the effect of capsule filling
speed on powder net weight in capsules is not observed in the industrial scale
(Planeta) due to the corrective actions of the weight
control system.

Acknowledgments: 
To the financial and scientific support from MG2 and GSK.

References

Philip John Woodhead. Doctoral Thesis: ?The influence of powder bed
porosity variations on the filling of hard gelatin capsules by a dosator system?. University of Nottingham,
1980.