14A-15

Y. H. CHOI¹, H. D. Chung¹, E. J. Ko¹, and J. H. Lee². (1) Dept. of Food Science Technology, Kyungpook National University, 1370 Sankuk-Dong, Puk-Gu, Taegu, 702-701, South Korea, (2) Dept. of Food Science and Engineering, Taegu University, 15 Naeriri, Zinyangmyun, Kyungsan, 712-714, South Korea

Pectin substances, which are contained in vegetables and fruits, are the main pollutant materials during the ultrafiltration process. Thus, the solute concentration on membrane surface and mass-transfer coefficient which are affected on the concentration polarization were measured for the establishment of prediction model on permeate flux. Two models were applied in this study.

Our objective was to measure permeate flux and solute concentration on membrane surface in the ultrafiltration process on the pectin solution. Then, the osmotic (OS) model and the boundary layer (BL) resistance model were applied to the prediction of permeate flux.

The pectin solution (MW=240,000) at the concentration of 0.1, 0.3, and 0.5%, and the polysulfone membrane (MWCO=100,000) were used in this study. The solute concentration and permeate flux of model solutions were measured at the pressure of 50, 100, and 150 kPa, and at the flow rate of 110, 130, and 150 mL/min. Each measurement was conducted every 10 min for 60 min in batch-type system.

Our results showed that the two models were well applied in both theoretical equations and experimental data for the concentration polarization to pectin solutions. The permeate flux was consistent with the predicted value in BL and OS system within relative error of 1%. Mass transfer coefficient (k) in concentration polarization layer was measured by Sherwood number.

The models used in this study could be applied to the prediction of permeate flux for the design of an ultrafiltration system and the correct decision of membrane area. These results also suggest that those prediction models could be used to design an ultrafiltration process for other juice clarifications.