Feed-side concentration polarization

A phenomenon that is particularly important in the design of reverse osmosis units is that of concentration polarization. This occurs on the feed-side (concentrated side) of the reverse osmosis membrane. Because the solute cannot permeate through the membrane, the concentration of the solute in the liquid adjacent to the surface of the membrane is greater than that in the bulk of the fluid. This difference causes mass transfer of solute by diffusion from the membrane surface back to the bulk liquid. The rate of diffusion back into the bulk fluid depends on the mass transfer coefficient for the boundary layer on feed-side. Concentration polarization is the ratio of the solute concentration at the membrane surface to the solute concentration in the bulk stream. Concentration polarization causes the flux of solvent to decrease since the osmotic pressure increases as the boundary layer concentration increases and the overall driving force (AP - An) decreases. 

This effect, usually known as feed-side concentration polarization, may become particularly relevant for solutes with a high sorption affinity towards the membrane, which may lead to its depletion near the membrane interface if external mass-transfer conditions are not sufficiently good to guarantee their fast transport from the bulk feed to the interface [32, 36] (see Figure 11.3). As a consequence of their depletion near the interface the driving force for transport, and the resulting partial fluxes, become lower. 

Feed-side concentration polarization When feed components sorb in the membrane, a local concentration gradient develops in the feed phase adjacent to the membrane upstream face. Owing to this gradient, transport of components from the bulk into the bulk-membrane interface occurs, thus replenishing the components that were absorbed by the membrane. The transport of a solute across the phase adjacent to the membrane can be either convective or diffusive, depending on the solute concentration as well as the fluid dynamic conditions over the membrane surface. 

This phenomenon is denoted feed-side concentration polarization and, in practice, affects mainly the fluxes of compounds of high sorption coefficient, even under turbulent hydrodynamic conditions over the membrane, as their permeability (and hence flux across the membrane) is high. It should at this point be emphasized that contrary to the non-ideal transport phenomena discussed earlier, feed-side concentration polarization is not a membrane-intrinsic phenomenon, but stems from poor design of the upstream flow conditions in practice it may in fact not be overcome owing to module design limitations (Baker et ah, 1997). 

In vapor permeation, feed-side concentration polarization is much less prone to occur than in pervaporation, owing to the high mass-transfer rate of the solute in the vapor feed phase. In fact, this feature is one of the main factors that distinguish the two processes. 

Feed-side and strip-side concentration polarization result in a reduction in the driving force for mass transfer. There is a decrease in water activity at the feed-membrane interface and an increase at the strip-membrane interface. This results in a reduction in the water vapor pressure gradient across the membrane. The feed side and strip side mass transfer co-efficients, Kf and K, respectively, can be expressed in terms of the solute diffusion co-efficient in the boundary layer, D, ... 

In general, concentration-polarization effects are much more noticeable on the feed side of the membrane than on the product side. As discussed earlier, the trans-membrane flux is relatively insensitive to the metal ion concentration in the product solution but varies substantially with feed solution metal ion concentration, particularly with low concentration feed solutions. Concentration polarization effects can be a serious problem when treating these solutions with practical membrane systems. Figure 9.25 shows the effect of concentration polarization with a hollow fiber supported membrane system described in the following section. In this system, a feed solution velocity of greater than 20 cm/sec is required to overcome polarization. 

Fig. 23. Two types of hollow-fiber modules used for gas separation, reverse osmosis, and ultrafiltration applications, (a) Shell-side feed modules are generally used for high pressure appHcations up to - 7 MPa (1000 psig). Fouling on the feed side of the membrane can be a problem with this design, and pretreatment of the feed stream to remove particulates is required, (b) Bore-side feed modules are generally used for medium pressure feed streams up to - 1 MPa (150 psig), where good flow control to minimise fouling and concentration polarization on the feed side of the membrane is desired.

The dead-end setup is by far the easiest apparatus both in construction and use. Reactor and separation unit can be combined and only one pump is needed to pump in the feed. A cross-flow setup, on the other hand, needs a separation unit next to the actual reactor and an additional pump to provide a rapid circulation across the membrane. The major disadvantage of the dead-end filtration is the possibility of concentration polarization, which is defined as an accumulation of retained material on the feed side of the membrane. This effect causes non-optimal membrane performance since losses through membrane defects, which are of course always present, will be amplified by a high surface concentration. In extreme cases concentration polarization can also lead to precipitation of material and membrane fouling. A membrane installed in a cross-flow setup, preferably applied with a turbulent flow, will suffer much less from this... 

The layer of solution immediately adjacent to the membrane surface becomes depleted in the permeating solute on the feed side of the membrane and enriched in this component on the permeate side. Equivalent gradients also form for the other component. This concentration polarization reduces the permeating component s concentration difference across the membrane, thereby lowering its flux and the membrane selectivity. The importance of concentration polarization depends on the membrane separation process. Concentration polarization can significantly affect membrane performance in reverse osmosis, but it is usually well controlled in industrial systems. On the other hand, membrane performance in ultrafiltration, electrodialysis, and some pervaporation processes is seriously affected by concentration polarization. 

Using this model and the assumption that concentration polarization occurs only on the feed side of the membrane, the flux across the combined resistances of the feed side boundary layer and the membrane can be written as... 

Figure 4.1 shows the concentration gradients that form on either side of a dialysis membrane. However, dialysis differs from most membrane processes in that the volume flow across the membrane is usually small. In processes such as reverse osmosis, ultrafiltration, and gas separation, the volume flow through the membrane from the feed to the permeate side is significant. As a result the permeate concentration is typically determined by the ratio of the fluxes of the components that permeate the membrane. In these processes concentration polarization gradients form only on the feed side of the membrane, as shown in Figure 4.3. This simplifies the description of the phenomenon. The few membrane processes in which a fluid is used to sweep the permeate side of the membrane,...

The effect of concentration polarization on specific membrane processes is discussed in the individual application chapters. However, a brief comparison of the magnitude of concentration polarization is given in Table 4.1 for processes involving liquid feed solutions. The key simplifying assumption is that the boundary layer thickness is 20 p.m for all processes. This boundary layer thickness is typical of values calculated for separation of solutions with spiral-wound modules in reverse osmosis, pervaporation, and ultrafiltration. Tubular, plate-and-ffame, and bore-side feed hollow fiber modules, because of their better flow velocities, generally have lower calculated boundary layer thicknesses. Hollow fiber modules with shell-side feed generally have larger calculated boundary layer thicknesses because of their poor fluid flow patterns. 

In the discussion of concentration polarization to this point, the assumption is made that the volume flux through the membrane is large, so the concentration on the permeate side of the membrane is determined by the ratio of the component fluxes. This assumption is almost always true for liquid separation processes, such as ultrafiltration or reverse osmosis, but must be modified in a few gas separation and pervaporation processes. In these processes, a lateral flow of gas is sometimes used to change the composition of the gas on the permeate side of the membrane. Figure 4.14 illustrates a laboratory gas permeation experiment using this effect. As the pressurized feed gas mixture is passed over the membrane surface, certain components permeate the membrane. On the permeate side of the membrane, a lateral flow of helium or other inert gas sweeps the permeate from the membrane surface. In the absence of the sweep gas, the composition of the gas mixture on the permeate side of the membrane is determined by the flow of components from the feed. If a large flow of sweep gas is used, the partial... 

The different solute concentrations on the feed and permeate sides are linked to the volumetric permeate flux in terms of the concentration polarization model, which is based on the stagnant film theory [14] ... 

Permeate flux decreases with an increase in feed concentration. This phenomenon can be attributed to the reduction of the driving force due to decrease of the vapor pressure of the feed solution and exponential increase of viscosity of the feed with increasing concentration. The DCMD flux gradually increases with an increase in temperature difference between feed and cooling water. Lagana et al. [63] reported that the viscosity of apple juice at high concentration induces severe temperature polarization. It may be noted that temperamre polarization is more important than concentration polarization, which is located mainly on the feed side. 

The SGMD is a temperature driven process, and it involves (a) evaporation of water at the hot feed side, (b) transport of water vapor through the pores of hydrophobic membrane, (c) collection of the permeating water vapor into an inert cold sweeping gas, and (d) condensation outside the membrane module. A decrease in driving force has been observed due to polarization effects of both temperature and concentration [80,82]. To calculate both heat and mass transfer through microporous hydrophobic membrane as well as the temperature and concentration polarization layer, the theoretical model suggested by Khayet et al. [58] can be written as... 

Mass transfer in the feed and strip solutions is limited by the extent of concentration polarization. On the feed side of the membrane, concentration polarization refers to an increase in the concentration of solutes at and near the feed-membrane interface because of evaporation of water into the membrane pores (Fig. 1). The resulting solute concentration gradient between the membrane-feed interface, where the concentration is greatest, and the bulk solution induces diffusive transport of rejected solutes back through the concentration polarization boundary layer into the bulk stream. Bulk solution is simultaneously transported to the membrane wall by convection. When equilibrium has been established under a given set of operating conditions (stream flow rate, temperature, fluid dynamics imposed by membrane module design), the rate of back diffusion is equal to the rate at which the solutes are carried to the membrane surface by convective flow. ... 

Thus, the resistance lo mass transfer from ihe bulk feed stream to the bulk dialysate stream can be considered the sum of three terms concentration polarization on the feed and dialysate sides of the membrane and the resistance to diffusion in the membrane itself. For laminar flow devices, these resistaoces... 

The contribution of concentration polarization to overall dialytic mass transfer resistance suggests that dialyzera should be operated with flows in the lurbolenl flow region to minimize boundary-layer formation. While this is typically the case on the dialysate side of the membrane, diaiyzers usually are operated with leminar flow on the fsed side. In hemodialysis, this is partly to avoid undue mechenical stress on the blood cell9 which mey result in their destmction. However, a consideration of fluid dynamics dictates that feed-side flow be leminar in nearly all applications. 

Figure 9.22A illustrates the purely diffusion-controlled process, in which the effects of boundary layers and interfacial reaction rates are negligible. In this case, the concentrations of the complex at the interfaces are the equilibrium concentrations. Figure 9.22B illustrates the partially boundary-layer-controlled case. Here, prior to steady state, the permeant diffuses across the membrane faster in the feed-side boundary layer and accumulation of permeant in the product-side boundary layer. The consequence of this concentration polarization is a reduction in the net concentration gradient across the membrane, and a reduced flux compared with the diffusion-controlled case. The last case is that of partially reaction-rate-controlled flux, illustrated by the concentration profile in Figure 9.22C. Here, either the permeant initially diffuses away from the feed interface faster than it can be replenished by the interfacial reaction, or the dissociation reaction is not fast enough to prevent accumulation of the complex at the product interface. Again, the net result is a decrease in the concentration gradient compared with that in the purely diffusion-controlled case. In all three cases, the flux is proportional to the slope of the concentration profile across the liquid membrane.

Introduction of feed gas on the shell side of the module can reduce concentration polarization. Figure 10 shows three shell side feed configurations [13]. Notice that in each configuration, one tube sheet is open to the bore of the hollow fibers and the opposite end is closed. Concentration... 

Fig. 3.6-11 The principle of concentration polarization (a) on the membrane upstream side in contact with the feed liquid and (b) on the membrane downstream side facing the vacuum. The straight lines indicating the concentration profiles are strongly oversimplified.

Industrial-size plate-and-frame modules, for example, consist of a stack of tightly packed membranes over which the feed solution is recirculated (Mulder, 1997). The membranes are separated by spacers and the permeate is withdrawn by a central permeate pipe (Stiirken, 1994). Pressure losses occur on both the feed and the permeate side of the packed membranes and need to be accounted for in the module design. On the feed side, the fluid dynamic conditions over the membrane may be less uniform than on the laboratory scale, resulting in more pronounced concentration polarization. On the permeate side, the packed configuration of the membranes may lead to considerable pressure losses, rendering the instantaneous removal of solutes from the membrane downstream surface more difficult. Both aspects may cause solute fluxes lower than expected (Chapter 3.2) and a possible... 

It should be noted that Cp is the salt concentration at the membrane surface on the feed side that was used to calculate Ng, while Cfi, is the salt concentration in the bulk of the solution on the feed side of the membrane. Cp and Cf may become different due to a phenomenon called concentration polarization. This occurs due to the rejection of the salt and its accumulation at the membrane surface. The concentration polarization is schematically illustrated in Figure 3.6. 

---