Screening of the ultrasonic spray congealing process variables by factorial design.

The aim was to evaluate, using the factorial design, the contribution of the process variables on the mean particle diameter of microparticles prepared by the ultrasonic spray congealing technique. An asymmetric factorial design was used to reduce the number of experiments. Three parameters were studied at different levels: presence or absence of a surfactant (5% w/w of soya lecithin), amplitude of the ultrasound (25, 60 and 100% of the rated power) and kind of the lipid (6 different waxes were employed: stearyl alcohol, stearic acid, Gelucire 50/13, Cutina HR, carnauba wax and microcrystalline wax). The 18 experiments were run randomly in triplicate. In all the conditions, non aggregated and spherical shaped microparticles were obtained. The mean particle diameter was selected as experimental response. The analysis of the results, performed estimating the “weight” of total effects, showed that both the presence of the surfactant in the formulation and the ultrasound amplitude had not a significant (a > 0.05) effect on the mean particle size. On the contrary, the kind of employed lipid is a very important variable for determining the microparticles dimensions: stearic acid and Gelucire 50/13 microparticles were significantly bigger (a <0.01), while Cutina HR ones were significantly smaller (a < 0.05) with respect to stearyl alcohol, carnauba wax and microcrystalline wax microspheres. In conclusions factorial design results show that the kind of the lipid is the most important variable determining the sizes of the spray congealed microspheres; studies are in progress to investigate if this effect could be correlated to the physico-chemical properties (viscosity, density, surface tension) of the molten lipid.