Formulation Variables on Pellets Containing Self-Emulsifying Systems

, , ,

Pharmaceutical Technology Europe

Pharmaceutical Technology Europe, Pharmaceutical Technology Europe-06-01-2005, Volume 17, Issue 6

This high quality of pellet roundness is surprising in view of the high extrusion forces required to extrude the wet masses.

Recent work has shown that it is possible to prepare pellets containing a high proportion of a self-emulsifying system as a potential method to convert a liquid into a solid dosage form.1 When a poorly water soluble drug was formulated in such a system, the product was equally bioavailable when administered to dogs.2 The system reported involved the use of a mixture of equal parts polysorbate 80 USNF (United States National Formulary) and mono- and diglycerides USNF.

This mixture forms one of many possibilities for a set of self-emulsifying systems, which can be used for the oral drug delivery of solid compounds.3 So was the system studied a specific combination of self-emulsifying system and microcrystalline cellulose (MCC) or there are other self-emulsifying systems that will function in a similar manner? A further combination of self-emulsifying systems has been compared with the original systems to answer this question. Because the inclusion of lactose in the formulation reduced the ability to incorporate the self-emulsifying system and did not enhance pellet preparation, this ingredient was excluded from the design.

Materials and Methods

Polysorbate 80 USNF (Tween 80) (P) and sorbitan mono-oleate (Span 80) (S) were supplied by Honeywill and Stein (UK). The mixture of mono- and diglycerides USNF (Imwitor 742) (M) was supplied by Condea (Germany) the soyabean oil (SB) was supplied by Croda Chemicals Ltd (UK), while the MCC was supplied by FMC Corp (Avicel PH101; Ireland). The water used throughout was prepared by reverse osmosis.

From these materials, three self-emulsifying systems were prepared with the following compositions:

  • A mixture of equal parts by weight of mono- and diglycerides (M) and polysorbate 80 (P) to give a composition MP.

  • A mixture containing 70% w/w mono- and diglycerides(M), 20% w/w polysorbate 80 (P) and 10% sorbitan mono-oleate (S) to give a composition MPS.

  • A mixture containing 30% w/w soyabean oil (SB), 49% Polysorbate 80 (P) and 21% Sorbitan mono-oleate (S) to give a composition SBPS.

These mixtures were dispersed with 10, 20, 30 and 40% w/w water to provide self-emulsifying systems, the appearance of which is recorded in Table 1. The mixtures were produced by agitating with a magnetic stirrer until an even texture was achieved.

Table 1 Consistency of self-emulsifying systems mixed with water.

All the mixtures were used on the same day that they were prepared.

To prepare pellets, the appropriate quantity of fluid was added to 50 g of MCC in a mortar and thoroughly mixed with a pestle to provide an even consistency. The correct quantity of fluid required was determined by trial and error. In some instances, the quantities of the self-emulsifying system and the water were added separately to the MCC, the water being incorporated prior to the self-emulsifying system.

The wet powder mass from these mixtures was extruded through a 1 mm diameter die, 6 mm in length attached to a 25 mm diameter barrel of a ram extruder. The ram was attached to the crosshead of a universal testing instrument (MX 50; Lloyd Instruments, UK) and extruded at a cross head speed of 200 mm/min. The force required to cause steady state flow was monitored by the load cell, the output of which was fed to a computer and displayed on a chart recorder.4 The extrudate, about 80 g, was spheronized for 10 min on a 120 mm diameter spheronizer (G.B. Caleva Ltd, UK) fitted with a cross hatch plate rotating at 1880 rpm. The resultant pellets were dried at 40 °C in an oven (Gallenkamp Hot Box, UK).

The particle size of the pellets and its distribution was determined using a set of British Standard sieves (a square root 2 progression of sizes), by agitating for 10 min with a sieve shaker (Endecotts Ltd, UK). The mechanical strength of the pellets was determined by a diametral compression test with a strength tester (CT5; Engineering Systems, UK) at a platen speed of 1 mm/min. The values are the mean of 10 pellets taken at random from the relevant size fraction. The shape of the pellets was assessed with an image analyser (Seescan Ltd, UK) with a black and white camera (CCD-4, miniature video camera, Rengo Co. Japan) connected to a zoom lens (18-108/2.5 from Olympus, Germany) by the method described by Podczeck and Newton.5

Table 2 Proportions and properties of pellets produced with mixtures of a self-emulsifying system containing MP and water and MCC.

A minimum of 100 pellets selected at random from the appropriate size fraction was assessed. The disintegration time of the pellets was measured with British Pharmacopoeia (BP) disintegration apparatus (Copley, UK) fitted with 1.0 mm aperture mesh. The tabled values are the mean of six pellets tested individually.

Table 3 Proportions and properties of pellets produced with mixtures of a self-emulsifying system containing MPS and water and MCC.

Results and Discussion

As with the system used previously,


none of the self-emulsifying systems, when added to MCC without the presence of water, produced a wet powder mass that could be extruded. As pressure was applied to the mass with the ram, fluid was expressed from the die and the force increased to excessive levels. As soon as water was present at a 10% level, the mass could be extruded in all the three systems tested. The results for the steady-state extrusion force, particle size and size distribution characteristics, shape, mechanical strength and disintegration, are presented in Tables 2–4.

Table 4 Proportions and properties of pellets produced with mixtures of a self-emulsifying system containing SBPS and water and MCC.

It is evident from these results that it is possible to prepare pellets from all the compositions tested, thereby extending the generality of the previous studies. There are slight differences in the quantity of fluid, which was required to produce pellets by the different self-emulsifying systems, but these were of the same order of magnitude, ranging from 70–80% of the quantity of MCC. In all instances, the pellets tend to be larger than those that would be produced using a totally aqueous system with the same diameter and length die, and for those made at the higher level of fluid, the modal size exceeded 2.0 mm. Generally the modal size was the 1.4–2.0 mm size fraction and it was usually possible to produce pellets in which an excess of 70% was in this size fraction. It was more difficult to produce the smaller size of pellets when the systems contained 90% of the self-emulsifying systems.

In terms of shape, only on two occasions did the value of the shape factor eR fall below 0.55, a value that indicates that the pellets are approximately spherical.5 These two sets of pellets were formed with the 90% level of self-emulsifying systems and the highest quantity of these used. Reducing the quantity of fluid used in the formulations could produce better pellets. Many of the pellet formulations had values of eR greater than 0.60, a value which cannot always be achieved with aqueous systems.

This high quality of pellet roundness is surprising considering the high extrusion forces required to extrude the wet masses. Equivalent aqueous formulations using a die of this length and diameter would not be expected to produce acceptable pellets at forces above 10 kN. Such extrudates would produce pellets that are elongated rather than round. Thus, more than a simple measure of the extrusion force is required to evaluate the consistency of wet powder masses used in extrusion/spheronization.

Another intriguing phenomenon is the ability of the systems to function, both when the water is added directly to the MCC or to the self-emulsifying system. The extrusion force and the properties of the pellets produced for such systems are not different, whatever self-emulsifying system was used (Tables 2–4).

Another property that could be important if a drug was incorporated into the pellets is whether or not they disintegrate. When tested by a BP procedure, pellets from all the formulations and conditions disintegrated within 15 min.

Generally the higher the water level of the formulation, the longer the disintegration time. Those pellets produced from the self-emulsifying systems containing soyabean oil, Tween 80 and Span 80 (SBPS) had the longest disintegration times while those prepared from the mixture of mono- and diglycerides with Tween 80 (MP) had the shortest.

As disintegration times increase with an increase in water used to produce the product so does the mechanical strength for all these systems (Tables 2–4). There is little difference between the strength of pellets prepared from the three different component self-emulsifying systems. In all cases, the pellets prepared with the lowest level of water in the fluid are the softest. These pellets not only have the lowest breaking load they tend to be the most deformable. Hence these could be the most difficult pellets to handle if they are to be filled into hard gelatin capsules.

In general, it is preferable to have the highest level of self-emulsifying component in the pellet to obtain the potential for drug incorporation. The results in Tables 2–4 show a reduction in achievable self-emulsifier system content from 40–30% when the water content is increased from 10–40% in the wet mass. By use of either 20 or 30% water, the levels of self-emulsifying systems can be maintained in the region of 35% within pellets, which disintegrate within 10 min. There is obviously a compromise to be made in terms of water incorporation to maximize emulsifier content, strength and minimize disintegration times.


The preparation of pellets by extrusion/spheronization, which can contain up to 40% of self-emulsifying systems and are of an appropriate pharmaceutical quality has been demonstrated for other self-emulsifying systems. It has been confirmed that water must always be present to enable pellets to be manufactured. A consideration of mechanical properties and self-emulsifying content must be undertaken, but levels in the region of 35% of self-emulsifying content of the pellet weight are achievable, allowing the potential administration of poorly soluble drugs as solid dosage forms.

John Michael Newton is the emeritus professor of pharmaceuticals and Anabela Godinho is a post-graduate research student both at University of London, UK.

Ashley Peter Clarke is a research fellow and Steven William Booth is a senior director both at Merck Sharp & Dohme, UK.


1. J.M. Newton

et al


J. Pharm. Sci.

90, 987–995 (2001).

2. C. Tuleu et al., J. Pharm. Sci. 93, 1495–1502 (2004).

3. C.W Pouton, Int. J. Pharm. 27, 335–348 (1985).

4. P.J. Harrison, J.M. Newton and R.E. Rowe, J. Pharm. Pharmacol. 37, 686–691 (1985).

5. F. Podczeck and J.M Newton, J. Pharm. Pharmaco1. 46, 82–85 (1994).