Membrane Processes

In the chemical process industry one often encounters the problem of separating a mixture in its components. Using membranes one can in principle carry out the majority of these separation processes and may complement or form

Copyright © 2004 by American Scientific Publishers All rights of reproduction in any form reserved.

Encyclopedia of Nanoscience and Nanotechnology Edited by H. S. Nalwa Volume 7: Pages (47-82)

v ^yWfHSrSzf,',' . »MM "•• WESEE&t OHM

Figure 1. Scanning electron microscope (SEM) photographs. Left: a microporous skeleton of a diatomaceous earth fossil. Diatom cells are enclosed by cell walls made largely of silica (SiO2). This glassy shell, or frustule, consists of two tightly fitting halves often resembling a flat, round, or elongated box. The minute perforations allow dissolved gases and nutrients to enter and exit. Right: cross-section of a microporous synthetic polyethersulfon membrane obtained with a phase inversion process. The membrane may have at least one outer skin layer with a mean pore size smaller than that of the inner part. Courtesy of Aqua-marijn Research/Membrane Technology Group, University of Twente.

an alternative for chemical processes like distillation, extraction, fractionation, adsorption, etc. Advantages of membrane filtration are, among others, that there is a low energy consumption, the separation can be carried out continuously, up-scaling is relatively simple, and membrane technology can be used for almost any kind of separation.

The success of a membrane application or process is closely related to the intrinsic properties of the membrane. Interfacial interactions between membrane surface, surrounding environment, and solutes govern membrane

Endoface

Figure 2. SEM photograph of a human blood cell retained on a microsieve membrane. Courtesy of Aquamarijn Research. Right: Drawing of the lipid bilayer (cell membrane) of a human red blood cell, showing the single-pass transmembrane protein glycophorin. Reprinted with permission from [4], S. Chien, Microvascular Res. 44, 243 (1993). © 1993, Academic Press.

Endoface

Figure 2. SEM photograph of a human blood cell retained on a microsieve membrane. Courtesy of Aquamarijn Research. Right: Drawing of the lipid bilayer (cell membrane) of a human red blood cell, showing the single-pass transmembrane protein glycophorin. Reprinted with permission from [4], S. Chien, Microvascular Res. 44, 243 (1993). © 1993, Academic Press.

performance to a great extent. These interactions have considerable impact on transport characteristics, selectivity, fouling propensity, and bio- and hemocompatibility of the membrane, especially in biotechnological and medical applications, where highly adsorptive solutes such as proteins are present and the adsorption and consecutive fouling of the membrane lead to considerable losses in flux, selectivity, and performance [5]. See Tables 1 and 2.

1.2.1. Filtration through Size Exclusion Molecular Sieves In 1756 Cronstedt discovered a selection of natural porous minerals which, when heated, produced steam. Due to this phenomenon he called them zeolites from the Greek for boiling stone (zeo from zein— to boil; lithos—stone). In 1949, Milton working at Union Carbide produced the world's first manmade zeolites, the most important being Linde A and X. Linde A has gone on to become one the most widely used zeolites. Zeolites are highly crystalline alumino-silicate frameworks comprising [SiO4]4- and [AlO4]5- tetrahedral units. T atoms (Si,Al) are joined by oxygen bridges. Introduction of an overall negative surface charge requires counterions (e.g., Na+, K+ and Ca2+).

Synthetically they are made by a crystallization process with use of a template (e.g., tetrapropyl ammonium) that builds a self-assembling skeleton together with the silicon, aluminum, and oxygen atoms. Afterward the template is removed (e.g., by heating, the calcination process) leaving small nanosized holes and tubes along the crystal planes of the zeolite structure. The framework structure may contain linked cages, cavities, or channels, which are of the right size to allow small molecules to enter (i.e., the limiting pore sizes are roughly between 3 and 10 A in diameter). In all, over 130 different framework structures are now known.

Zeolites have the ability to act as catalysts for chemical reactions which take place within the internal cavities. An important class of reactions is that the class is catalyzed by hydrogen-exchanged zeolites, whose framework-bound protons give rise to very high acidity. This is exploited in many organic reactions, including crude oil cracking, isomeriza-tion, and fuel synthesis. Zeolites can also serve as oxidation or reduction catalysts, often after metals have been introduced into the framework. Examples are the use of titanium ZSM-5 in the production of caprolactam, and copper zeolites in NOx decomposition. See Figure 3.

Top layers of zeolites have also been deposited on many microporous support materials (e.g., alumina) to study their gas selective properties or to separate branched monomers (e.g., alkenes) from unbranched ones.

Table 1. Some membrane separation processes arranged according to the mechanism of separation.

Separation mechanism

Membrane separation process

Size exclusion (filtration)

nanofiltration, ultrafiltration,

microfiltration

Solubility/diffusivity

reverse osmosis, gas separation,

pervaporation, liquid membranes

Charge

electrodialysis

Table 2. Filtration processes with their properties and application.

Pressure

Filtration process Pore size Separation capability (bar) Application

Molecular sieving 0.3-1 nm branched molecules versus unbranched 0-3 gas separation with zeolite molecular molecules sieves

Nanofiltration 1-10 nm permeation of low molecular 5-25 purification of sugar from acids and salts weight

(200-20,000 daltons) substances from dyes, water treatment

Ultrafiltration 5-100 nm retention of viruses, bacteria, dissolved 0.5-5 dairy industry, beverage industry, pharmaceutical substances with molecular weight industry, separation of water from crude oil, between 10,000 and 500,000 daltons separation of fruit and vegetable extracts, waste water treatment

Microfiltration 50 nm-5 fxm retention of bacteria, Colloids, 0.5-3 prefiltration in water treatment, dye industry, protozoa (Cryptosporidium Legnonella) 0.5-3 beverage clarification, removal of bacteria, etc.

Nanofiltration Nanofiltration is a relatively young description for filtration processes using membranes with a pore size ranging from 1 to 10 nm. This term has been introduced to indicate a specific domain of membrane technology in between ultrafiltration and reverse osmosis. One of the first researchers using the term "nanofiltration" was Eriksson in 1988 [6]. But already some years earlier a company called FilmTec started to use this term for their NF50 membrane which was supposed to be a very loose reverse osmosis membrane or a very tight ultrafiltration membrane [7]. See Figure 4.

But the history of nanofiltration dates back further in time. In 1970 Cadotte et al. already showed that reverse osmosis membranes, a nanofiltration membrane with slightly smaller pore size, could be made out of polyethyleneimine and toluenediisocyanate, forming a polyurea [8]. This membrane, the NS l00, was different in several aspects from the then-existing reverse osmosis membranes. It was the first reverse osmosis membrane not made by a phase inversion process using cellulose acetate or polyamide plus it showed a high salt retention.

All ceramic and some polymeric nanofiltration membrane are considered to be nanoporous, and most of the polymeric nanofiltration membranes are regarded as "dense," which means that no fixed pores are present in the membrane, but they contain a network structure that can be charged or uncharged.

Figure 3. Left: Paraxylene molecule can diffuse freely in the open channels of silicalite. Right: top view of MFI-type ordered zeolite (molecular sieve) crystals having orthogonal nanopores of 0.55 nm on a silicon substrate. Courtesy of TOCK, TU Delft.

In several transport models for nanofiltration the pore size is used as a parameter to describe the morphology of the membrane [9, 10]. However, this concept of pores may be considered as hypothetical for several polymeric nano-filtration membranes. Bowen and Mukthar [11] mentioned that the determination of an effective pore size by transport models should not mean that those pores really exist in nanofiltration membranes. The hindrance to transport is the same for ions passing through the polymer network of a specific membrane as for ions passing through pores having these effective sizes.

In general nanofiltration membranes are used to separate relatively small organic compounds and (multivalent) ions from a solvent. However, in most of the applications with aqueous solutions, such as waste water, drinking water, and process water, the transport mechanisms are not yet fully understood. In the case of separation of organic solvents new developments of polymeric membranes are continuously showing up, where the solvent stability of these polymeric membranes is frequently studied and improved. These polymeric membranes may suffer from swelling when wrongly chosen.

Typical applications of these membranes are the separation of salts from dye solutions or the separation of acids from sugar solutions for the extraction of the purest products in highly concentrated form.

Ultrafiltration and Microfiltration Ultrafiltration membranes are porous membranes (in the range of 5 to 50 nm) that have a distinct, permanent porous network through j Li/

membrane

Figure 4. Schematic drawing of separation processes that occur in nanofiltration membranes.

which transport occurs [12, 13]. The separation of microorganisms, bacteria, and solids from the liquid is mainly based on size and is pressure driven. The term ultrafiltration has been introduced to discriminate the process from microfiltration with a more narrow retention behavior. The membranes, in most cases polymeric, are either used in dead-end mode or in cross-flow mode. In dead-end mode all liquid is forced to pass the membrane, where in cross-flow mode, a tangential flow across the membrane is used to minimize a fouling layer, thereby keeping the filtration process ongoing [14]. See Figures 5 and 6.

For the retention of larger particles microporous membranes (with pore size from 50 nm to 5 /m) can be used. As an example, a feed containing macromolecules (e.g., proteins) is contacted with a membrane that contains small pores. The particles to be separated (the macromolecules) are withheld from flowing through the pores; the solution itself can flow through the pores. The reason that the particles cannot freely enter the pores is their geometric size, although interaction with the pore may also be of importance. In this case the properties of the membrane are also dependent on the structure of the membrane, together with the intrinsic properties of the membrane material. Separation of larger particles (e.g., micro-organisms) usually proceeds according to the same mechanism.

Was this article helpful?

0 0

Post a comment