Dissolution testing is routinely used in the pharmaceutical industry to provide critical in vitro drug release information for quality control purposes and especially to assess batch-to-batch consistency of solid oral dosage forms such as tablets. The United States Pharmacopoeia (USP) lists several types of dissolution apparatuses. Of these, the most commonly used for solid dosage forms is the UPS Dissolution Testing Apparatus 2 consisting of an unbaffled, hemispherical-bottomed vessel equipped with a 2-blade radial impeller.
Significant information can be found in the literature of the power consumption of impeller-agitated, baffled systems. However, despite extensive industrial use and a large body of work on Apparatus 2, little or no information can be found in the literature on the power dissipated by this apparatus. Power consumption is a key parameter to predict for the rate of the dissolution process in such a system, especially for the prediction of mass transfer rate between the solid oral dosage form and the dissolution medium, since it has been shown previously that the power dissipated by an impeller rotating in a fluid correlates directly to the mass transfer coefficient experience by suspended particles. Thus, the determination of power dissipation is critical to quantify the mass transfer rate and hence be able to predict eventually the drug release rate from an oral dosage form during dissolution tests. Therefore, the main objective of this work was to experimentally determine and computationally predict the power dissipated by the impeller in the USP Dissolution Testing Apparatus 2 and the associated power number Po, and to relate these values to the mass transfer coefficient experienced by suspended drug particles.
A sensitive torque measurement apparatus was specifically built to measure the small torque applied to the liquid in the dissolution vessel when stirred by the paddle. To do so, the dissolution vessel and its content were floated in a shallow tank, and the force required to prevent rotation of this assembly was experimentally measured, from which torque and the power dissipation could be determined knowing the system geometry and the agitation speed.
A commercial CFD software package (FLUENT, 6.3.26; Gambit, 2.4.6) was used to predict the velocity distribution profiles in the USP 2 vessel, as well as the pressure distribution on the surface of the impeller blades from which the torque applied to the impeller and hence the impeller power dissipation were numerically calculated. Furthermore, the power dissipated by the impeller was additionally predicted by calculating, via CFD, the distribution of the local energy dissipation rate per unit mass in the USP 2 vessel, and then by integrating this variable over the entire liquid mass in the vessel.
The power dissipation was computationally and experimentally determined for a number of agitation speeds and liquid volumes. In general, good agreement was found between the experimental measurements and the computational results obtained with both methods. When Po was plotted against the impeller Reynolds number Re, a similar good agreement was typically found. As in other unbaffled system, Po was found to decrease slightly with Re, and was found to be in the range 0.25-0.38 for the case of the 900 mL volume fill level and in the range 0.20-0.29 for the 500-mL volume case. The values of the power dissipation per unit liquid mass were used to predict the mass transfer coefficient to suspended particles of different size and under different agitation conditions.