Philippe Aptel

Impact of coagulation conditions on the in-line coagulation/UF process for drinking water production☆

An in-line coagulation (without settling)/UF process has been studied to improve membrane performance and water quality for surface water treatment. Using coagulation before UF increases permeate quality; the extent of dissolved organic matter removal is controlled by the coagulation step. Efficient coagulation conditions for a coagulation/settling process can be applied for the in-line coagulation/UF process and membrane fouling is reduced. Floc cake resistance is lower than resistance due to the unsettled floc and the uncoagulated organics. For an inside-out hollow-fiber system, the impact of coagulation dose depends on filtration mode: for instance, in cross-flow with a feed-and-bleed configuration, a reduction of coagulant dose induces an increase of the mass transfer resistance even in quasi-stable hydrodynamic operating conditions.

Polysulfone hollow fibers — effect of spinning conditions on ultrafiltration properties

Asymmetric hollow fibers with the skin on the inside of the fiber were prepared from a three-component dope mixture containing polysulfone (18 wt%) polyvinylpyrrolidone (18 wt%) and N,N-dimethylacetamide (64 wt%). A wet—dry spinning technique was employed, using a tube-in-orifice spinneret. Water was used as the inside coagulation agent. The principal variables of the spinning process studied are the average velocity of the polymer solution through the spinneret, the rate of the inside coagulation agent and the air gap between the spinneret and the collecting bath. The results show that the spinning conditions have a pronounced effect on the ultrafiltration characteristics. As extrusion rate was decreased from 8 cm/sec to 1.6 cm/sec, the air gap decreased from 50 to 25 cm, the water flux inside the nascent fiber increased from 4 X 10−2 to 6 X 10−2 cm3/sec, the hydraulic permeability increased from 2 X 10−8 to 6 x 10−8 cm/sec-Pa, while rejection for polyvinylpyrrolidone (Mw ≅ 10,000) decreased from 0.8 to 0.4. An empirical correlation is proposed which represents all results. The experimental findings are discussed in terms of the orientation introduced into the polymer solution as it passes through the spinneret.

Growth of the polarization layer in ultrafiltration with hollow-fibre membranes☆

Abstract A mathematical model is proposed which represents the concentration polarization phenomenon in ultrafiltration. A hollow-fibre membrane geometry has been chosen because it allows well-defined hydrodynamic conditions to be obtained and simplifies the measurement of local membrane permeation rates, and also because its mathematical description, which requires a number of special considerations, has rarely been attempted. Hollow-fibre modules have been built which are divided, on the low-pressure side, into a number of compartments; this allows the measurement of local permeation rates at different positions along the length of the fibre bundle. Experimental values of the local permeation rates are compared with the values predicted by the mathematical model.

Basic aspects of pervaporation

Abstract Pervaporation is a recently introduced membrane technique designed to separate a liquid mixture by partially vaporizing it through a non-porous permselective film. The feed is circulated in contact with the membrane and the permeate is generally evolved from the opposite surface of the barrier by maintaining the downstream half-cell of the device under low pressure (vacuum-pervaporation). Other variant processes may also be used. The performance of a membrane in the separation of a given feed-mixture is characterized by permeation rate and selectivity. Since membrane swelling depends on the composition of the facing liquid mixture both characteristics also vary according to this composition. Pervaporation mechanism is rather complex, due to the multiplicity of gradients involved in this process. The situation is quite straightforward when the feed is a pure liquid. In this case, the determining parameters are reduced to the diffusivity D* of the unique penetrant in the pure polymeric membrane material (M) and to the γ coefficient which characterizes the plasticizing action of this permeant. In the case of a binary A-B feed-mixture, the problem is more complex since, at every point within the swollen film, the system behaves as a ternary M-A-B mixture. Attempts have been made to interpret experimentally measured fluxes and selectivities by considering the γ AB and γ BA plasticizing cross-coefficients or by referring to the FLORY-HUGGINS equation which governs the thermodynamics of a polymer-mixed solvent system. In fact, a steady-state vacuum-pervaporation regime is chiefly determined by the large swelling gradient established within the film. As long as the permeate is rapidly evolved under a low pressure, the downstream layer of the barrier is kept in a nearly dry state and transmembranar diffusion may be considered as the rate-determining step of the transport. On the other hand, if downstream pressure is allowed to increase, a transition is observed when the permeate becomes saturated under the conditions then established in the downstream half-cell. Desorption slows down and therefore turns out to be the rate-limiting step of the process.

Mass transfer improvement by secondary flows: Dean vortices in coiled tubular membranes

A helically wound hollow-fibre (or tubular) membrane module was studied in an oxygenation operation with water flowing in the laminar regime inside the tube. Data are compared with a conventional module where straight hollow-fibre membranes are in parallel alignment. In the former design, the results are analyzed taking into account the formation of secondary flow (Dean vortices) and a new mass transfer correlation is presented. In the latter design, results are consistent with the Leveque equation. It is shown that the presence of vortices gives better performance in terms of oxygen transfer. Improvement factors were in the range of 2 to 4.

Treatment of textile dye effluents using a new photografted nanofiltration membrane

A nanofiltration membrane has been developed by UV-photografting. Sodium p-styrene sulfonate was used for the modification of a polysulfone ultrafiltration membrane. The membrane cut-off was estimated. The grafted membranes have been evaluated for the removal of five different dyes with an aim to reuse water in the process house. The effect of different parameters such as dye class, pH and the presence of salt was evaluated. It is observed that the newly developed membranes show acceptable performance both in terms of flux and rejection. Dye retention was higher than 97% and hydraulic permeability 0.23–0.28 m3.m−2.d−1 at 0.4 MPa. The influence of pH on the performance of membranes in terms of fouling and retention was established and compared to a commercial membrane (Desal 5DK).

From ultrafiltration to nanofiltration hollow fiber membranes: a continuous UV-photografting process☆

A new way to prepare nanofiltration membranes, consisting of in-line external modification of the skin of a polysulfone ultrafiltration hollow fibre skin, is described. The paper presents a continuous process (dip-coating followed by photografting) and the influence of the operating conditions on the membrane characteristics. This study is focused on a simplified model based on the evaluation of the two sources of monomer available for grafting: one is monomer contained in the thin film drawn from the dip bath, second is contained in the pores of the membrane. The results show that two extreme types of modification can be produced, depending on the experimental conditions: a high rate coupled with a short dip contact time leads essentially to an external grafting whereas a low rate coupled with a long dip residence time leads essentially to an “in pore” grafting.

Use of air sparging to improve backwash efficiency in hollow-fiber modules

The use of air in backwash of hollow-fiber modules was investigated experimentally from bench to full scale. Modules operated in a dead-end and outside-in mode: they were fouled by either a bentonite suspension or a raw river water and then backwashed in presence of air. The air was injected into the retentate compartment either in combination with a reversed permeate flux or together with feed water after a brief permeate back flow. Results indicate that the cake layer is instantaneously lifted off by the reversed permeate flux and is concentrated in the free volume of the module. To remove it from the module and recover the feed concentration, this volume has to be rinsed with a volume at least three times as big. The air, by its piston-like action, improves material removal and reduces the volume of concentrated foulant to be flushed. So the backwash time is reduced and its efficiency is improved. An optimum air flow rate can be found that is independent of the water flow rate used to flush the module free-volume.

Dead-end ultrafiltration in hollow fiber modules: Module design and process simulation

Abstract Ultrafiltration in a hollow-fiber module operating with outside-in and dead-end flow at a constant flow rate was simulated using a model that takes into account the longitudinal pressure drops inside the fibers and within the fiber bundle. The model considers both the filtration phase during which the membrane is fouled by the formation of a filter cake and the backwash phase in which it is cleaned, so as to predict the net rate of production of the module during an operating cycle. The results show that there is a combination of packing density and fiber diameter that gives a maximum net flow rate. Furthermore, this model allows the influence of operating conditions and feed properties on the module performance to be estimated. This can be used to determine how operating parameters must be modified when there is a change in the feed properties.

Hollow-fibre membrane module design: comparison of different curved geometries with Dean vortices

Abstract The performance of several designs of curved membrane modules with Dean vortices was compared through experiments using a colloidal bentonite suspension and cellulose acetate hollow-fibre ultrafiltration (UF) membranes. The different module geometries were: straight, helically coiled, twisted and sinusoidal, or meander-shaped. The experiments show a remarkable increase in mass transfer in curved modules as compared to conventional straight ones. Comparisons were made for modules equipped with the same hollow fibres and the same Dean number (De) for a given Reynolds number (Re). At the same Dean number, all the curved geometries gave the same limiting permeate flux. A mass transfer correlation relating limiting UF flux with the mean wall shear stress has been obtained.