Characterization of Single Phase Flows in Stirred Tanks via Computer Automated Radioactive Particle Tracking (CARPT)

Single phase flows in stirred tanks have been extensively characterized using different experimental techniques like Hot Wire Anemometry1, Laser Doppler Anemometry2, and Digital Particle Imaging Velocimetry3. None of these techniques however, show much promise for the interrogation of opaque multiphase flows. Hence, little or no information of the local fluid dynamics of multiphase flows in stirred tanks is available. Non-optical techniques like Computer Automated Radioactive Particle Tracking (CARPT) and Computed Tomography (CT) have been successfully applied to probe a variety of multiphase reactors4–5 such as bubble columns6–7, risers8 etc., over a range of dispersed phase holdups. CARPT provides the local fluid dynamic information such as velocities and the turbulence parameters throughout the system that is investigated. CT provides time averaged local dispersed phase holdup profiles in various planes of the entire reactor. In this study, it is proposed to extend these techniques to characterize gas-liquid flows in stirred tank reactors. As a first step, CARPT is implemented in characterization of single phase flows in stirred tanks. CARPT experiments have been performed with water at 150 rpm in a 0.20m cylindrical tank quipped with a six bladed Rushton turbine (0.067m dia) conforming to the standard Holland and Chapman9 configuration. The CARPT technique is shown to capture some of the important flow phenomena observed in such flows, like the two recirculating loops above and below the impeller and the dead zones at the bottom of the tank. Radial pumping numbers determined by CARPT (0.67 near the impeller tip) compare reasonably well with data reported in the literature. Comparison of the complete three dimensional mean velocity profiles from CARPT with similar PIV, LDA and other data reported in the literature reveals that CARPT captures the right order of magnitude of the radial and the tangential velocities. Comparisons of the fluctuating velocity components, like the root mean squared (rms) velocity and the turbulent kinetic energy, suggest that the CARPT experiments were limited by large tracer particle size (dp = 2.3 mm) from sampling the high frequency fluctuations of the fluid. In addition, the three dimensional profiles of the components of the Reynolds stress tensor are measured. The detailed comparisons, even with the large tracer particle, indicate that CARPT measurements capture all the important qualitative features of the flow and quantitatively capture the right order of magnitude of the mean flow parameters. The quantitative comparisons suggest that the current size of the tracer particle restricts it from responding completely to the fluid phase fluctuations. Some Lagrangian measures of the fluid dynamics like the ‘Sojourn’ time distributions (STDs) in different zones of the reactor, Circulation Time Distributions (CTDs), Particle Return Maps to specific planes, Poincarre sections and Hurst exponents are evaluated from the collected CARPT data.