The Crystallization of High Grade Massecuite in Crystallizers

That is, the mass of sucrose crystallized per unit time is It is shown that equations developed for predicting the mass ~ ~ o ~ o r t i o n a l to the product of the film mass transfer cotransfer coefficient in agitated vessels can be t~ efficient, concentration gradient and crystal surface area. crystallizers. Calculations done with a mathematical model of a A correlation has been developed by Calderbank and Moocrystallizer based on this equation indicate that the main factors Young2 for calculating the film mass transfer coefficient in in high grade crystallization are the residence time, power input agitated vessels which is independent of the geometry of the per unit volume and cooling rate. It is also shown that addition agitation equipment. of molasses to crystallizers adversely affects the crystallization rate. Introduction The advantages of increasing the exhaustion of high grade massecuite by means of crystallizers has been pointed out by various authors. Lamusse et a17 reviewed previous work on this subject in 1962, and since then Grapa5, Parashar9 and Julienne6 have written about practical results obtained in the factory. The potential purity drop in A massecuite is more than thirty points, of which between fifteen and twenty points are obtained in vacuum pans and the difference could be obtained by suitable crystallizer installations. In Table 1 are shown purity drops observed under South African conditions, and as can be seen the purity drop obtained in crystallizers were in general less than three points. The study presented in this paper was undertaken to determine what are the factors that influence high grade crystallizer operation. Estimation of mass transfer coefficients The crystallization process is a mass transfer operation which takes vlace in two s t e~s . The accuracy of equation (2) for sugar crystallizers was verified by comparison with measured values obtained from equation (1). The data were obtained from the crystallizer installations shown in Table 2 and Fig. 1 and 2. It was necessary to take the average values from a number of observations because of variations in massecuite quality during the tests. The weight of sucrose crystallized, S, was calculated from the difference in crystal content between the inlet and outlet, the crystal content being obtained from the relation X = Bm (Pm Pn) / (100 Pn) (3) When molasses was being added to the crystallizer, the percentage added was calculated from M = (Bmi Bmo) @mi Bmol> , (4) and the crystal content at the outlet corrected by the equation 1. Diffusion of the sucrose molecule through the stagnant fluid X*, = Xo. (1 + M) ( 5 ) surrounding the crystal. The crystal surface area at the outlet was obtained from a size 2. Incorporation of the molecule into the crystal lattice. analysis of the sugar produced, and the following equation Silinlo has pointed out that for high viscosity impure sugar derived from those suggested by Gilett4. solutions the fust step is rate controlling, and that the second step can be neglected so that crystallization can be represented 3,986.X0 a. = (6) by the following equation: Lo The crystal size at the crystallizer inlet was calculated from TABLE 1 Observed purity drops in A-massecuites