ASSESSMENT OF SUGAR POLARISATION USING AN ELECTRICAL CONDUCTIVITY METHOD

A STUDY of the potential for using an electrical conductivity (EC) method for sugar polarisation (pol) assessment was undertaken during 2004. The method involves measuring EC in a 27o brix solution of the sugar being assessed, and using the result to calculate the purity. This in turn is used in conjunction with an infrared moisture analysis to calculate the pol. The results of this study were encouraging and the decision was taken to implement the method routinely for shift sugar analysis. The method has been in use for three seasons and has performed well. Over this period the standard error of prediction (SEP, calculated as the standard deviation of the difference between the results for conventional and EC pol methods) has been 0.093oZ, with 95% of results being within 0.189oZ. The best performance was for the 2006 season when the SEP was 0.077oZ, with 95% of results being within 0.152oZ. This compares favourably with conventional sugar pol analysis repeatability of 0.147oZ. Maintaining a calibration equation that adequately represents the relationship between EC and sugar purity has been found to be integral to achieving good results. The precision of results returned using the EC method has been shown to be related to the precision of the analyst undertaking the analysis used for calibration. As such, having a competent analyst undertake this work is advantageous. The EC sugar pol assessment method offers advantages over the conventional method for routine shift analysis. It is simple, requires little analytical skill, experience or training to undertake, and has a high degree of repeatability. It is also significantly faster to complete than the conventional method. The adoption of this method in conjunction with an EC method for the assessment of pan product true purity has allowed a significant reduction in laboratory manning without a reduction in the analytical output. Both methods also have the advantage of allowing large numbers of samples to be analysed during trials with far less work than conventional methods.