On the Influence of Mixing and Scaling-Up in Semi-Batch Reaction Crystallization

Sammanfattning: Semi-batch crystallization experiments have been performedboth in a loop reactor and in stirred tank reactors.Hydrochloric acid was fed to a stirred solution of sodiumbenzoate, and benzoic acid immediately formed. Benzoic acid isformed in excess of the solubility making the solutionsupersaturated.The loop reactor is U-shaped. In one leg a propeller stirrerwas placed to circulate the solution and in the other a turbinestirrer was placed in front of the feed point to vary the localmixing intensity. The objective was to analyse the relativeimportance of different levels of mixing on the product sizedistribution. The importance of mixing as well as scaling-upeffects on the product size distribution were studied in threestirred tank reactors of volumes 2.5 L, 10 L, and 200 L. Thestirred tank reactors had different geometry and were equippedwith either a marine propeller or a pitched blade turbine.The weight mean size generally increases with increasingtotal feeding time and increasing mixing intensity. The weightmean size increases by locating an extra turbine impeller atthe feed point in the 10 L stirred tank reactor. The turbineimpeller provides the desired feed point mixing intensitywithout raising the mixing intensity of the whole tank.The weight mean size increases with decreasing feed pipediameter in the loop reactor and for low feed rates in the 10 Lstirred tank reactor. The weight mean size increasessignificantly by changing the feed pipe opening from circularto rectangular with a constant cross-sectional area at equalfeed rates. Backmixing is visually observed in the largest feedpipe diameter in the loop reactor, thus, reducing the weightmean size. However, backmixing is not considered to be adominant phenomenon in the present work.Mesomixing time constants have been calculated according tothe turbulent dispersion mechanism and the inertial-convectivemechanism. The time constants for mesomixing are generallylonger than the time constant for micromixing. Thus, the ratioof the mesomixing and the micromixing time constants shows aninfluence of mesomixing as is shown by the experimentalresults. The experimental results are best described by theinertial-convective disintegration mechanism showing that thefeed plume mixing increases with decreasing feed pipe diameterand increased feed point mixing.The weight mean size is not strongly affected by the reactorvolume. However, the mixing conditions in the reactors have astrong influence on the weight mean size. No suggestedscaling-up rule can satisfactorily predict the weight mean sizein the different volumes. No single physical parameter, such asthe local energy dissipation rate, the mean energy dissipationrate or the circulation time, can satisfactorily explain theexperimental results. A new dimensionless mixing parameter, TR,has been defined as the ratio of the total feeding time and themesomixing time constant. The mesomixing time constant isdefined as the shortest dimension of the feed pipe divided bythe resultant bulk velocity passing the feed pipe entrance. Theexperimental results from both the loop reactor and the stirredtank reactors of different volumes can be correlated with TR.The weight mean size increases with increasing TR.Keywords: reaction crystallization, precipitation,benzoic acid, macromixing, mesomixing, micromixing,semi-batch, loop reactor, backmixing, colour experiments,scaling-up.Keywords: reaction crystallization, precipitation,benzoic acid, macromixing, mesomixing, micromixing,semi-batch, loop reactor, backmixing, colour experiments,scaling-up.