Membrane separations have been playing an increasing role in wastewater treatment; this is the most evident in processing whey wastewater with ultrafiltration in the dairy industry, one of many types of membrane separation technologies, and in membrane pervaporation of volatile organic compounds (VOCs) from wastewater. Currently, a new field of membrane separations called nanomembrane technology is said to be the future nano-technology for achieving clean water with this purported type of "smart membranes." It is predicted that nanomembranes are able to separate molecules by differences between molecular weights of a mixture of com-pounds—a feat that current membranes (polymeric or inorganic membranes) are unable to achieve. Whether this prediction of smart membranes can really hold water in the future is anyone's guess; however, the use of membrane-based technologies will undoubtedly increase among the evolving water and wastewater treatment landscapes.
A membrane can be viewed as a discreet (or discriminating) barrier that allows some components of the wastewater feed to pass through the membrane faster than the other components. A membrane provides a third phase, mostly a solid phase that straddles between two fluid phases serving as the origin and the destination of the separation process. The membrane is most likely polymeric, though new inorganic membranes are now emerging at a rapid pace. The principal mechanisms of membrane separations are molecular diffusion in solid and tortuous viscous flows in micro-porous solids.
Membrane technology is an evolving separation technology, and because of its multidisciplinary characters it can be used to perform a large number of separations in food and agricultural wastewater treatment. The membrane processes that are commonly found in processing plants or research laboratories include microfiltration (MF), reverse osmosis (RO), ultrafiltration (UF), nanofiltration (NF), electrodialysis (ED), membrane distillation (MD), and pervaporation (PV). Membrane processes are based upon different separation principles or mechanisms and their applications in food processing range from concentration of food fluids to aromatic flavor recovery. The membrane is at the center of every membrane process. However, membrane separations can be achieved only when a driving force is applied to the underlying membrane process. A schematic diagram of a two-phase conceptual system is shown in Fig. 3.12. No perfect man-made membrane ever existed. This situation will be with us in the foreseeable future until perhaps we fully understand the mechanisms that regulate the mass transfer in the membrane, and we are able to tailor the membrane structures to the need of separation of molecules of interest by using the latest advancement in nanotechnology. In assessing membrane systems, two experimental parameters that determine the overall performance of membrane processes should be the main focus of designers' attention. The first one is selectivity, the other, permeation flux.
The selectivity of a membrane toward a mixture, which characterizes the extent of separation, is customarily expressed by one of two quantities: the retention, R, and the separation factor, a. The R is more suitable for the membrane separation of a dilute binary system and given by Equation 3.19:
Cf where Cf is the solute concentration in the feed stream and Cp the solute in the permeate. The value of R varies between 100% (complete rejection or retention) and 0% (complete permeation). For most mixtures, however, the selectivity factor is more adequate (Equation 3.20):
where Ci and Cj are the concentrations of components i and j in the permeate and in the feed. The value of ajj is greater than 1 if the component i is more readily permeable than component j and if the separation occurs.
The other parameter, permeation flux, takes many forms depending upon the underlying membrane processes. It is normally expressed as the following (Equation 3.21):
' dz where K is the phenomenological coefficient and dg/dz is the driving force expressed as the gradient of g (concentration, temperature, pressure) in the z direction toward the membrane. The phenomenological coefficient K is strongly related to the driving force, module configuration, and operating conditions.
Membrane processes can be classified according to the nature of their driving forces and pore size of the membrane. Although all membrane processes are driven by the electrochemical potential gradient, one particular driving force is usually dominated in a membrane process. Three types of membrane separation processes relevant to the food industry can be considered: those driven by hydrostatic pressure difference, those driven by the partial vapor pressure gradient, and those driven by electrical potential differences. The next sections present brief general descriptions of the membrane processes used or potentially usable in various operations of the food industry.
Membrane separation by hydrostatic pressure difference
Membrane performance of a pressure-driven system is usually described by the flow rates of water (solvent) and solute (permeate). The flow of water (volume flux) through a membrane without considering concentration polarization and fouling (or/and gel layer) is expressed by the following (Equation 3.22)
where Kw is water permeability; AP is the applied pressure; An is the osmotic pressure difference; and a is the reflection coefficient of the membrane toward the solute, which is a measure of degree of solute rejection. The driving force is (AP — An) as dg/dz in Equation 3.21. Because there is no perfect membrane, we may suspect (it can be verified) that some solutes, including those undesirable, also transport across the membrane, though less freely than water (solvent) (Equation 3.23):
where Ks is the solute permeability and Cm and Cp are concentrations on the upstream side of the membrane and on the permeate side, respectively. Note that Ks has a different unit from that of Kw because the driving force in Equation 3.23 is expressed as the solute concentration difference.
Although Equations 3.22 and 3.23 are generally considered as valid phenomenological expressions, the true meanings of Cm and Cp are not what they seem to be. This is because in a pressure-driven membrane process, the retained solutes transported by convective transmembrane flux can accumulate at the membrane surface leading to high concentration of solutes near the membrane, a phenomenon called concentration polarization. This concentration gradient is encompassed in a region designated as the boundary layer (velocity has its own gradient due to the viscous effect at the water-membrane interface and the no-slip condition for a common cross-flow membrane configuration, thus a velocity boundary layer). In a steady state situation, the concentration polarization is the result of solute buildup counterbalanced by the solute flux through the membrane plus the diffusive flux of solute at the membrane surface toward the bulk flow on the upstream side of the membrane. The magnitude of the concentration polarization is expressed by the Equation 3.24 as a result of the solute mass balance based upon the concentration profile of the film model illustrated in Fig. 3.13:
Was this article helpful?