Prediction ultrafiltration

Microfiltration and ultrafiltration have recently been introduced for the removal of particles down to any desired size. Their capital cost is relatively high. Experience with them is limited, and a short trial with a small-scale pilot element is advisable. Prediction of full-scale performance from such trials is normally quite reliable. 

Osmotic pressure models can be developed from a very fundamental basis. For example, it is becoming possible to predict the rate of ultrafiltration of proteins starting from a knowledge of the sequence and three-dimensional structure of the molecule 19). 

Bowen, W. R. and Williams, P. M. Chem. Eng. Sci. 56 (2001) 3083. Prediction of the rate of cross-flow ultrafiltration of colloids with concentration-dependent diffusion coefficient and viscosity-theory and experiment. 

Livage, J. and M. Henry. 1985. A predictive model for inorganic polymerization reactions. In Ultrafiltration Processing of Advanced Ceramics, eds. J. D. Mackenzie and D. R. Ulrich, pp. 183-95. John Wiley Sons, New York. 

Also included are sections on how to analyze mechanisms that affect flux feature models for prediction of micro- and ultrafiltration flux that help you minimize flux decline. Descriptions of cross-flow membrane filtration and common operating configurations clarify tf e influence of important operating parameters on system performance. Parameters irdlucnc irxj solute retention properties during ultrafiltration arc identified and discussed or treated in detail. 

The development of these equations has been reported elsewhere (12), and it has also been shown using ultrafiltration techniques that the composition of the monomer is well predicted by the equation (12)... 

Because of the broad differences between ultrafiltration equipment, the performance of one device cannot be used to predict the performance of another. Comparisons can only be made on an economic basis and only when the performance of each is known,... 

The theory of permeation through microporous membranes in ultrafiltration and microfiltration is much less developed and it is difficult to see a clear path forward. Permeation through these membranes is affected by a variety of hard-to-compute effects and is also very much a function of membrane structure and composition. Measurements of permeation through ideal uniform-pore-diameter membranes made by the nucleation track method are in good agreement with theory. Unfortunately, industrially useful membranes have nonuniform tortuous pores and are often anisotropic as well. Current theories cannot predict the permeation properties of these membranes. 

A practically useful predictive method must provide quantitative process prediction from accessible physical property data. Such a method should be physically realistic and require a minimum number of assumptions. A method which is firmly based on the physics of the separation is likely to have the widest applicability. It is also an advantage if such a method does not involve mathematics which is tedious, complicated or difficult to follow. For the pressure driven processes of microfiltration, ultrafiltration and nanofiltration, such methods must be based on the microhydrodynamics and interfacial events occurring at the membrane surface and inside the membrane. This immediately points to the requirement for understanding the colloid science of such processes. Any such method must account properly for the electrostatic, dispersion, hydration and entropic interactions occurring between the solutes being separated and between such solutes and the membrane. 

The aim of the present chapter is to provide an overview of recent work on the development of quantitative predictive methods for membrane processes at the Centre for Complex Fluids Processing, University of Wales Swansea. The main aim of such work is the development of predictive methods that require no adjustable parameters—that is methods that are truely ab initio. Three theoretical aspects will be illustrated the prediction of rates of ultrafiltration, the prediction of rejection in microfiltration and ultrafiltration, and the prediction of the rejection in nanofiltration. These examples show the importance of bridging the gap between physical chemistry and process engineering in the development of new process technologies. Finally, the use of AFM in the quantification of the adhesion of... 

There have been many models, both simple and sophisticated, that describe the operating patterns of ultrafiltration processes [4]. Most of these models describe how the rate of ultrafiltration is controlled by the properties of a region of very high solute concentration, a filter cake or concentration polarised layer, close to the membrane surface. Relatively few of these models have a genuinely predictive capability. Remarkably, only a very few [5-7] of these models consider the most important feature of the solutes being separated by ultrafiltration—that they fall in the colloidal size range. For colloidal materials, the properties of the filter cake or concentration polarised layer will be controlled by the interparticle interactions in such a region. The important interactions which need to be taken into account are [8] ... 

Having provided accurate descriptions of solute-solute interactions, there are two approaches to the formulation of a predictive method for ultrafiltration ... 

Equations (22-86) and (22-89) are the turbulent- and laminar-flow flux equations for the pressure-independent portion of the ultrafiltration operating curve. They assume complete retention of solute. Appropriate values of diffusivity and kinematic viscosity are rarely known, so an a priori solution of the equations isn t usually possible. Interpolation, extrapolation, even prediction of an operating curve may be done from limited data. For turbulent flow over an unfouled membrane of a solution containing no particulates, the exponent on Q is usually 0.8. Fouhng reduces the exponent and particulates can increase the exponent to a value as high as 2. These equations also apply to some cases of reverse osmosis and microfiltration. In the former, the constancy of C aji may not be assumed, and in the latter, D is usually enhanced very significantly by the action of materials not in true solution. 

A modular pilot size plant involving coagulation/flocculation, centrifugation, ultrafiltration, and sorption processes was designed and constmeted by Benito et al. [13] for the treatment of oily wastewaters. Empirical equations developed by Shaalan [65] predict the impact of water contaminants on flux decline. These formulae enable decision-making concerning a suitable water pretreatment scheme and also selection of the most appropriate cleaning cycle. 

G. E. Wetterau, M. M. Clark, and C. Anselme, A dynamic model for predicting fouling effects during the ultrafiltration of a groundwater. Journal of Membrane Science 109, 185-204 (1996). 

The formation of ceramic membranes for microfiltration, ultrafiltration or nanofiltration by association of various granular layers is now a common procedure [10]. Each layer is characterized by its thickness, h, its porosity, 8, and its mean pore diameter, dp. These parameters are controlled by the particle size, d, and the synthesis method. Each layer induces a resistance which may be predicted through the classical Carman-Kozeny model ... 

Figure 7. Predicted concentration profiles during ultrafiltration at optimal conditions, T = 23 C, pH = 9.0, F/V = 0.088 min-i...

Perhaps because much attention has centered on reverse osmosis membranes, the fine pores present in their skins were observed prior to the discovery of the functionally larger pores of ultrafiltration (UF) membranes. Recently, pores of v30 A have been observed by Zeman (35) in the skins of UF membranes. Their density, uniformity and diameters leave no doubt that these are actually the pores which function during UF. Our ability to actually "see" the intermicellar defect pores (the population of larger size pores) in the skins of RO membranes extends to the 10 X range. Therefore, it is reasonable to expect that at some point we shall be able to extend this ability to the population of smaller sized pores, whose existence is predicted by Sourlrajan s pore theory (36). 

---