Date of Award

Fall 2006

Document Type


Degree Name

Doctor of Philosophy in Chemical Engineering - (Ph.D.)


Chemical Engineering

First Advisor

Piero M. Armenante

Second Advisor

Robert Benedict Barat

Third Advisor

Rajesh N. Dave

Fourth Advisor

Russell V. Plank

Fifth Advisor

Laurent Simon


Dissolution testing is routinely carried out in the pharmaceutical industry to determine dissolution rate of solid dosage forms. The United States Pharmacopoeia (USP) Dissolution Apparatus II is the device most commonly used for this purpose. Despite its widespread use, dissolution testing remains susceptible to significant error and test failures. Limited information is available on the hydrodynamics of this apparatus, although hydrodynamic effects can play a major role on test performance.

Laser-Doppler Velocimetry (LDV) and Computational Fluid Dynamics (CFD) were used here to experimentally map and computationally predict the velocity distribution inside a standard USP Apparatus II under the typical operating conditions mandated by the dissolution test procedure. The flow in the apparatus is strongly dominated by the tangential component of the velocity, but a low recirculation zone exists in the lower part of the hemispherical vessel bottom where the tablet dissolution process takes place. The velocities in this region change significantly over short distances along the vessel bottom, implying that small variations in the location of the tablet on the vessel bottom caused by the randomness of the tablet descent through the liquid result in significantly different velocities and velocity gradients near the tablet.

CFD was also used to study the hydrodynamics when the impeller was placed at four different locations, all within the limits specified by USP. Small changes in impeller location, especially off-center, produced extensive changes in the velocity profiles and shear rates. The blend time to homogenize the liquid content was also obtained for a number of operating conditions using different experimental methods, a CFD-based computational approach, and a semi-theoretical model. Excellent agreement between data and predictions was obtained. The CFD results show that blend time is inversely proportional to the agitation speed, and that blend time is some two orders of magnitude smaller than the time typically required for appreciable tablet dissolution during the typical dissolution test, implying that the contents of this device can be considered to be well mixed during the typical test.

Finally, dissolution tests with prednisone and salicylic acid tablets were conducted, in which the tablets were placed at different locations in the dissolution vessel in order to study the effect of local hydrodynamics on dissolution. The results show that tablet location has a major effect, and that statistically significant differences exist in the dissolution profiles between centrally located tablets and tablets positioned off-center, at it is often the case during testing. The dissolution process was modeled using an approach based on the use of experimentally determined mass transfer coefficients, mass transfer coefficient equations, CFD-predicted velocity profiles, and mass balances. The results can satisfactorily predict the data.

The hydrodynamics of dissolution testing depends strongly on small differences in equipment configurations and operating conditions, which can have a profound effect on the flow field and shear rate experienced by the oral dosage form being tested, and hence the solid-liquid mass transfer and dissolution rate.