Using regression analysis to optimise conditions for water-soluble polysaccharide extraction from sunflower

Abstract

Abstract Common sunflower roots, often a by-product of the agriculture industry, comprise important raw materials for food and pharmaceutical uses. The purpose of this study was to detect the optimal parameters of ultrasonic extraction of water-soluble polysaccharides from roots of common sunflowers. We varied the grinding of raw materials, the temperature of the extractant, the multiplicity and duration of extraction, the ratio of raw materials to extractant, and the ultrasonic frequency in the bath "Grad 40-35". The data obtained during the experiments made it possible to develop a linear multivariate regression model with sufficient accuracy for forecasting and analysis. The adequacy of the model was confirmed using Pearson's χ2-criterion. For determination of the strength of the general relationship between the predicted and observed data values, the correlation coefficient was used. Findings reveal that the most important conditions for the extraction of annual sunflower roots to obtain the maximum yield were: extraction temperature equal to 80 °C; extraction multiplicity equal to 3; ultrasound frequency equal to 35 kHz; grinding of raw materials into particles measuring between 0.5 mm to 2.0 mm, extraction time of 15 to 20 minutes and the ratio of raw materials to extractant ranging from 1 g per 10 ml to 1 g per 15 ml.