Quality by Design (QbD) in the pharmaceutical industry is a systematic approach to process development. The entire focus of the QbD concept is that quality should be built into a product using scientific knowledge, prior experience, and technological understanding. In the QbD system, the process is designed to consistently meet the product Critical Quality Attributes (CQA). This requires the engineers and chemists to study the effects of reactor hardware and operating parameters on the CQA so that they may identify the Critical Process Parameters (CPP) that dictate the product quality.
Conventionally, the correlation between the CPP and CQA is studied by performing experiments. However, in case of design and scale-up of processes, experiments may not be able to provide the insights required for successful operation. For instance, consider an example of a solid-liquid operation where the particle size distribution of the solids is a critical product attribute. The critical hydrodynamic parameters in such operations are – liquid circulation rate, level of suspension of solids and shear distribution in the liquid. Of these, the liquid circulation rate can be measured in terms of blending time and solid suspension can be visualized. The shear distribution, however, cannot be quantified using experiments.
Shear distribution in the reactor is the function of impeller speed as well as impeller geometry. High impeller speed results in high average shear in the vessel. Also, the radial flow impellers such as the Rushton Turbine impeller typically give higher shear as compared to the axial flow impeller. Although the effect of impeller speed and impeller type can be studied through experimentation what makes the scale-up process complex is the distribution of shear in the reactor. The region near the impeller is subject to high shear and is known as the deformation zone while the region away from the impeller experiences low shear and is known as the relaxation zone. The particle size depends on the shear rate experienced by the particles over a period time as it passes through these zones. As the scale of the vessel changes, the size of the deformation and the relaxation zone changes along with the shear rate history of the particles. This makes predicting the final particles size distribution at a commercial scale a challenge. The level of process understanding that is required to address this challenge is best derived by integrating computational fluid dynamics (CFD) in your scale-up studies.
A 3D CFD analysis allows engineers to visualize the shear rate distribution in their vessel. They can also track the motion of individual particles in the vessel and record the shear rate they experience in the tank. The shear rate history curves thus produced are one of the best indicators of the process performance of a given vessel. For processes, where the shear rate has considerable influence on the hydrodynamic performance of the system, the shear rate history of the particles (or fluid parcels) is an important scale-up criterion. Consider the example shown in the Figure-1 where the objective of the scale-up was to obtain the desired particle size distribution (PSD) in the polymerization reactor. The two plant reactors available for production were completely dissimilar with one having an anchor impeller and the other equipped with three hydrofoil impellers. In order to identify a suitable reactor for scale-up and to determine the appropriate operating conditions, we performed a CFD analysis of both the reactors and compared the shear rate history of the particles with the CFD results of the lab reactor. The comparison (Figure 2) showed that the shear history experienced by the particles in a reactor with 3-hydrofoil impellers is similar to that of the lab reactor and hence was a more suitable option plant scale production.
The Option-2 design was implemented in the plant, multiple trial batches were conducted. Figure-3 shows the particle size distribution obtained.
Thus integrating CFD analysis in the QbD approach can help chemists and engineers better understand the correlation between the process parameters and CQA of the products thereby increasing the chances of a successful scale-up.
CFD Automation of Stirred Tank Vessels