Modeling of fouling phenomena in cross-flow ultrafiltration of suspensions containing suspended solids and oil droplets

In cross-flow ultrafiltration permeate flux has three types of flow regimes. Initially it has a transient regime with a decreasing permeate flow rate. Then it attains a steady state, and the flux reaches a plateau. In the third stage the flow rate becomes independent of applied pressure and is called the limiting flux. In this paper a general membrane fouling modeling was developed that accounts for internal clogging, partial and total clogging, cake deposition and cake deposition with retroflux for any type of filtration; and it is validated with the experimental data of cross-flow ultrafiltration of suspensions containing suspended solids and oil droplets. A relationship for the limiting flux was developed using the experimental data. The fouling was then quantified and linked to the operational parameters. The main operating parameter affecting the filtration process was temperature. Indeed, this determined the droplet and particle size distribution, which is found highly dispersed at any temperature leading to standard deviations of the same order of magnitudes as or even larger than the mean diameters. The major limiting processes that controlled the fouling were evidenced and explained by mean droplet size and size distribution. A systematic study of the influence of the different operational parameters on the transient filtration rate is presented for different temperatures.

1. Flux performance depends on the droplet and particle size distribution, which depends strongly on temperature. 2. Transient and steady-state data are fitted to a model of cake deposition with retroflux and model parameters are evaluated. 3. The transient and limiting flux are satisfactorily quantified. 4. For mixed suspensions, it seems necessary to decrease the temperature in order to increase the droplet and particle size. 5. The classical filtration models are derived from using classical relationships established for dead-end filtration; it was found that they are less accurate for cross-flow filtration. 6. For cross-flow filtration modeling, it is necessary to use a cake deposition with a [1] T. Jiang, M. Kennedy, W.G.J. Van der Meer, P.A. Vanrollenghem and J.C. Schippers, The role of blocking and cake filtration in MBR fouling, Desalination, 157 (2002) 335–343. [2] R. Audinos, Les bases théoriques de l’ultrafiltration, Techniques Séparatives par Membranes, PhD Thesis, Toulouse, France, 1988. [3] S.G. Yiantsios and A.J. Karabelas, Colloidal fouling of desalination membranes and fouling indices, IDA World Congress on Water Desalination and Reuse, Manama, Bahrain, 2002. [4] L.M. Gang, Etude de la texture des membranes et de leur interaction avec des suspensions au cours d’une filtration dynamique, Ph D Thesis, Université de Compiègne, France, 1992. [5] D.A. Drew, J.A. Schonberg and G. Belfort, Lateral inertial migration of a small sphere in fast laminar flow through a membrane duct, Chem. Eng. Sci., 46 (1991) 3219–3224. [6] M. Mouihi, Contribution à l’étude expérimentale et a la modélisation du colmatage des membranes en ultrafiltration tangentielle d’une émulsion huile/eau, PhD Thesis, Ec. Cent. Arts Manuf., Châtenay, Malabry, France, 1992. [7] J. Murkes and C.J. Calson, Cross-flow Filtration (Theory and Practice), Wiley, New York, 1988. [8] P. Aimar and V. Sanchez, A novel approach to transfer limiting phenomena during ultrafiltration of macromolecules, Ind. Eng. Chem. Fundam., 25 (1986) 789. [9] D.E. Wiley, C.J.D. Fell and A.G. Fane, Optimization of membrane module design for brackish water desalination, Desalination, 52 (1985) 249. [10] G.K. Anderson, C.B. Saw and M.S. Le, Oil/water separation with surface modified membranes, Environ. Tech. Lett., 8 (1987) 75. [11] M. Assadi and D.A. White, A model for determining the steady state flux of inorganic microfiltration membranes, Chem. Eng. J., 48 (1992) 11–16. [12] M.W. Chudacek and A.G. Fane, The dynamics of