Files in this item
|(no description provided)|
|Title:||Process Engineering Characteristics of the Hollow Fiber Ultrafiltration of Skimmilk|
|Department / Program:||Food Science|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Subject(s):||Agriculture, Food Science and Technology
|Abstract:||Performance and mass-transfer characteristics of the ultrafiltration of skimmilk was studied using a pilot-scale hollow fiber membrane unit. The independent variables were transmembrane pressure, flow rate, temperature and feed concentration and the major measured variable was permeate flux (J). Physical properties such as density, viscosity and protein content of the retentate were experimentally determined while diffusivity was estimated from literature values. The dependence of physical properties of the ultrafiltered skimmilk on temperature and concentration were usually nonlinear, largely due to hydration effect and the interactions between protein molecules. The effect of processing factors on ultrafiltration flux were closely related to the degree of concentration polarization. No single property could fully account for the variation of processing performance at various process conditions.
The pressure-independent, mass-transfer limited region was modelled on the popular "gel-polarization" film theory. Assuming a linear dependence of diffusivity on feed concentration in the boundary layer, the following form of the model was derived: J = Ks ln (Cg/Cb) + (PHI) Ks (Cg - Cb), Where Ks is mass-transfer coefficient, Cg is protein concentration at the membrane surface, Cb is bulk solution protein concentration and (phi) is a factor describing the effect of protein concentration on diffusivity. The mass-transfer coefficient was obtained from the dimensionless correlation: Sh = A Re('(alpha)) Sc('(beta)). The parameters in these models were determined by nonlinear model fitting method of Marquardt. The final values of the parameters were: A = 0.087, (alpha) = 0.636, (beta) = 0.333, (phi) = 0.1571, Cg = 20.41.
The second model developed in this study was based on the resistance concept. The two major resistances were assumed to be due to (1) the membrane plus any adsorbed layer (Rem) and (2) the combined resistance due to the "gel" and boundary layers, which was assumed to be a function of pressure. The final form of the model wa J = (DELTA)P(,T)/(Rem + K (DELTA)P(,T)), where K was found to be a function of flow rate and feed concentration, but not significantly affected by temperature. Rem was a constant. This resistance model was found to fit the data quite well over most of the experimental values.
Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1983.
|Date Available in IDEALS:||2014-12-15|
This item appears in the following Collection(s)
Dissertations and Theses - Food Science and Human Nutrition
Graduate Dissertations and Theses at Illinois
Graduate Theses and Dissertations at Illinois