| Summary: | Motionless mixers are pipe inserts that generate high rates of energy dissipation compared to an empty pipe, and more uniformly than in a stirred tank. Four types of motionless mixer were chosen for investigation: the Kenies and HEV mixers (manufactured by Chemineer); the SMV and SMXL mixers (from Sulzer). A range of experimental techniques were used to investigate the performance of the mixers andthe results interpreted by using appropriate correlations and insights from turbulence theory. New data is presented on LOA measurements in two motionless mixer flow fields. The mixer flow patterns, turbulent diffusivities and local turbulent energy dissipation rates are now available for the Kernes andHEV mixers. Correlations ofthe type developed for pipe mixers (whereaxial and radial dispersion can be expressed in terms ofthe square root of the friction factor) were found to be inappropriate for characterising the macromixing data of motionless mixers. However, the prediction of axial dispersion coefficients, based onthe analysis ofRTD measurements presented inthiswork, isnowpossible. All the motionless mixers tested have narrow RTDs in the fully turbulent regime. RTDs differ significantly in the transitional regime. It wasfound that the total power consumption of a motionless mixer is not a good indicator of its likely mixing performance for macro-, meso-Iinertial-convective mass transfer) or micrornixing limited applications. As a consequence, designing or retro-fitting a motionless mixer on the basis ofpressure drop information alone is not recommended. The turbulence generating efficiency (T}) of a motionless mixer is defined as; the ratio of the average turbulent energy dissipation rate ina mixer, to thetotal energy dissipation rate (calcluated from pressure drop measurements). Approximate values of n (determined from the results of micromixing-controlled azo-coupling experiments) are: 80% (HEV mixer); 70%(SMXL mixer); 60% (Kenies mixer validated at two scales); 15% (SMV mixer). For fast chemical processes, the optimum feed position into a motionless mixer is several diameters downstream of the first mixer element, and not upstream of the mixer as often recommended. . Thecharacteristic time-scales of the different mixing mechanisms have been derived. Comparison of the time-scales revealed the rate-limiting (i.e. slowest) mixing step in the azo-coupling experiments carried out in all four mixers. The scale-up/-down criteria for motionless mixers in semi-batch or continuous operation have been derived for macromixing, mesomixing and micromixing limited processes. The scaling rules for macromixing and mesomixing are the sam
|
|---|
