Industrial scale membrane production

Membrane Chemistry Three chemical families dominate the RO-NF membrane industry. Many other products are made on a small scale, and the field continues to attract significant R D resources. But three types command most of the market. 

At present moment, no generally feasible method exists for the large-scale production of optically pure products. Although for the separation of virtually every racemic mixture an analytical method is available (gas chromatography, liquid chromatography or capillary electrophoresis), this is not the case for the separation of racemic mixtures on an industrial scale. The most widely applied method for the separation of racemic mixtures is diastereomeric salt crystallization [1]. However, this usually requires many steps, making the process complicated and inducing considerable losses of valuable product. In order to avoid the problems associated with diastereomeric salt crystallization, membrane-based processes may be considered as a viable alternative. 

In general, high selectivities can be obtained in liquid membrane systems. However, one disadvantage of this technique is that the enantiomer ratio in the permeate decreases rapidly when the feed stream is depleted in one enantiomer. Racemization of the feed would be an approach to tackle this problem or, alternatively, using a system containing the two opposite selectors, so that the feed stream remains virtually racemic [21]. Another potential drawback of supported enantioselective liquid membranes is the application on an industrial scale. Often a complex multistage process is required in order to achieve the desired purity of the product. This leads to a relatively complicated flow scheme and expensive process equipment for large-scale separations. 

The main disadvantage of all these systems is the Hmitation of scale-up. Monoclonal antibodies are produced by multiplying the hollow fiber systems and stirred tank reactors with membrane aeration are known up to 100 liter. Small quantities of product can be produced by these systems but they are not suitable for real industrial scale-up. 

Although considerable research has been conducted with Pd-alloy foils, tubes, and thinner composite membranes, long-term durability and stability need to be further demonstrated, especially in the fuel reforming or WGS operating conditions, for acceptance of this technology in a commercial sector. Furthermore, mass-scale and cost-effective production of industrial-scale Pd-alloy thin-film composite membranes need to be demonstrated to be competitive in the hydrogen production and purification market. 

The use of membranes to separate gases commercially is a relatively new application. Both Du Pont and Union Carbide had ventures in the early 1970s for recovering H2 and He ffom industrial processes, but these projects were never fully commercialized. Recently, however, a combination of improved economics and better technology has resulted in membrane products that signal a new era in the commercial use of membranes for large-scale gas separation. 

Researchers at Degussa AG focused on an alternative means towards commercial application of the Julia-Colonna epoxidation [41]. Successful development was based on design of a continuous process in a chemzyme membrane reactor (CMR reactor). In this the epoxide and unconverted chalcone and oxidation reagent pass through the membrane whereas the polymer-enlarged organocatalyst is retained in the reactor by means of a nanofiltration membrane. The equipment used for this type of continuous epoxidation reaction is shown in Scheme 14.5 [41]. The chemzyme membrane reactor is based on the same continuous process concept as the efficient enzyme membrane reactor, which is already used for enzymatic a-amino acid resolution on an industrial scale at a production level of hundreds of tons per year [42]. 

Screen printing is an excellent method for fabrication of ISEs on an industrial scale. Ag/AgCI electrodes can be screen printed onto a suitable support such as Kapton polyimide. The selective membrane is then printed onto the ISE, often with an intervening layer of a hydrogel such as poly(vinyl alcohol) soaked with NaCl solution to act as the internal reference solution. The reductions of cost associated with the mass-production of these electrodes allow them to be sold as single-use disposable devices. Also their small size relative to conventional ISEs allows them to be assembled into sensor arrays. 

N-McLhylmorpholine-N-oxidc monohydrate, a tertiary, aliphatic amine N-oxide, is able to dissolve cellulose directly, i.e. without chemical derivatization, which is used on an industrial scale as the basis of the Lyocell process [ 1, 2], This technology only requires a comparatively low number of process steps compared for instance to traditional viscose production. Cellulose material - mainly fibers - are directly obtained from the cellulose solution in NMMO no chemical derivatization, such as alkalization and xanthation for rayon fibers, is required [3]. The main advantage of the Lyocell process lies in its environmental compatibility very few process chemicals are applied, and in the idealized case NMMO and water are completely recycled, which is also an important economic factor. Even in industrial production systems NMMO recovery is greater than 99%. Thus, compared with cotton and viscose the Lyocell process pertains a significantly lower specific environmental challenge [4]. Today, Lyocell fibers are produced on an industrial scale, and other cellulosic products, such as films, beads, membranes and filaments, are also currently being developed or are already produced commercially. 

Electromembrane processes such as electrolysis and electrodialysis have experienced a steady growth since they made their first appearance in industrial-scale applications about 50 years ago [1-3], Currently desalination of brackish water and chlorine-alkaline electrolysis are still the dominant applications of these processes. But a number of new applications in the chemical and biochemical industry, in the production of high-quality industrial process water and in the treatment of industrial effluents, have been identified more recently [4]. The development of processes such as continuous electrodeionization and the use of bipolar membranes have further extended the range of application of electromembrane processes far beyond their traditional use in water desalination and chlorine-alkaline production. 

For practical applications in the food industry, where large-volume production is conducted, it is especially important to obtain high disperse-phase flux. Abrahamse et al. [8] reported on the industrial-scale production of culinary cream. In this study they evaluated the required membrane area for different types of membranes an SPG membrane, an a-Al203 membrane and a microsieve filter. The requirements for culinary cream production were a droplet size between 1 and 3 pm and a production volume of 20 m3/h containing 30% disperse phase. They concluded that to produce large quantities of monodisperse emulsions the most suitable was a microsieve with an area requirement of around 1 m2. 

The chemzyme membrane reactor is based on the same continuous process concept as the efficient enzyme membrane reactor, which has already been applied for enzymatic a-amino acid resolution on industrial scale at a production level of hundreds of tons per year (Drauz and Waldmann 2002 Wandrey and I iaschcl 1979 Wandrey et al. 1981 Groger and Drauz 2004). 

Industrial production of perfluorinated ionomers, Nafion membranes, and all perfluorinated membranes is costly due to several factors first, the monomers used are expensive to manufacture, since the synthesis requires a large number of steps and the monomers are dangerous to handle. The precautions for safe handling are considerable and costly. Secondly, the PSEPVE monomer is not used for other applications, which limits the volume of production. The most significant cost driver is the scale of production. Today, the volume of the Nafion market for chlor-aUcali electrolysis (150,000 m year ) and fuel cells (150,000 m year ) is about 300,000 m year resulting in a production capacity of 65,000 kg year. When compared to large-scale production of polymers like Nylon (1.2 x 10 m year ), the perfluorinated ionomer membrane is a specialty polymer produced in small volumes. 

---