Çay Bitkisinden Ekstraksiyon Yöntemiyle Kafein Eldesinin Optimizasyonu

Abstract

In this thesis study, the rates and yields of caffeine extraction from tea fiber, exposed as a waste in tea factories with no economic value, brewed tea pulp waste and packed tea (black tea) was compared in a batch stirred system and soxhlet extractor. In the first part of studies the effects of solid/liquid ratio, temperature, stirring rate, type of solvent; addditionaly the effect of particle size range for packet tea on the rate and yield of caffeine production were examined in the batch stirred extractor with three types of tea. For soxhlet extraction only the effects of solid/liquid ratio and type of solvent on the caffeine extraction rate and yield were investigated.For all the teas, kinetic constants were determined by using second-order kinetic model in batch stirred system extraction and by using first-order kinetic model in soxhlet extraction. The highest caffeine extraction rates and yields for each type of tea were found at 50:1 liquid/solid ratio, at 60°C, at 140 rpm with water solvent in the batch stirred extraction. In these working conditions the highest caffeine extraction yields were found as 2,66% (26,57 mg caffeine/g dry tea fiber) for tea fiber, 2,35% (23,54 mg caffeine/g dry brewed tea pulp waste) for brewed tea pulp waste and 4,65% (46,47 mg caffeine/g dry packet tea) for packet tea, respectively.In soxhlet extraction studies the maximum caffeine extraction rates and yields for each type of tea were found at 50:1 liquid/solid ratio, with water solvent at its boiling temperature. In these working conditions the highest caffeine extraction yields were found as 2,69 (26,91 mg caffeine/g dry tea fiber) for tea fiber, 2,4 % (24,04 mg caffeine/g dry brewed tea pulp waste) for brewed tea pulp waste and 4,7% (47,02 mg caffeine/g dry packet tea) for packet tea, respectively. In batch stirred extraction the combined effects of stirring rate, temperature, liquid/solid ratio and extraction time on the amount of caffeine extracted from per unit dry weight of tea were analyzed by using Response Surface Methodology. Second-order quadratic polynomial equations giving the amount of caffeine extracted from per unit dry weight of tea for each tea type were found and three dimensional response curves were obtained. The results of proposed models were evaluated statistically by ANOVA test.