The fractionation of dextran using ethanol

Abstract

Reviews of molecular weight fractionation based on solubility difference and of gel permeation chromatography theories have been made. Fractional precipitation with ethanol has to be performec at least twice to obtain clinical dextran, which is a poly­glucose, from an aqueous solution of dextran having a broad molecular weight distribution. In the first stage the high molecular weight dextrans precipitate out from the solution as a syrup. In the second stage the lower molecular weight dextrans precipitate out from the remaining supernatant solution, when the ethanol concentration is increased. For the economic optimisation of dextran fractions, a mathematical model has been proposed based on the Boltzmann equation which predicts the weight percentage dextrans in each of the two stages of fractionation, the Boltzmann equation constants C, E and the volume ratios D,F for the two-phase separation. The aims of the project were to test this mathematical model on the laboratory-scale ethanol fractionation of dextran and also to use it to predict actual plant fractionations. In the laboratory-scale ethanol fractionation, the comparison of results on the first siage between the model predictions and experimental values are in very good agree­ment. On the second stage there is an offset present between the two comparable sets of results over the entire experi­mental range of values. The model predicts values that are approximately 10 Wt% higher than the experimental values. A similar pattern to that in the laboratory was found to exist between the two sets of results obtained for the plant fractionations of dextran. The precipitation of dextran molecules on an industrial­scale was also studied and it was found that the current settling times were inadequate. It is shown that a. company producing 100 batches per annum could increase its cash flow by £200,000 per annum by using the model to predict plant fractionations.