Laboratory Scale Development

Besides quality of informahon, quanhty of data shll is a decisive factor in laboratory-scale development. 

After development of a new process scheme at laboratory scale, constmction and operation of pilot-plant faciUties to confirm scale-up information often require two or three years. An additional two to three years is commonly required for final design, fabrication of special equipment, and constmction of the plant. Thus, projections of raw material costs and availabiUty five to ten years into the future become important in adopting any new process significantly different from the current technology. 

The development of the novel Davy-McKee combined mixer—settler (CMS) has been described (121). It consists of a single vessel (Fig. 13d) in which three 2ones coexist under operating conditions. A detailed description of units used for uranium recovery has been reported (122), and the units have also been studied at the laboratory scale (123). AppHcation of the Davy combined mixer electrostatically assisted settler (CMAS) to copper stripping from an organic solvent extraction solution has been reported (124). 

Fluorine was first produced commercially ca 50 years after its discovery. In the intervening period, fluorine chemistry was restricted to the development of various types of electrolytic cells on a laboratory scale. In World War 11, the demand for uranium hexafluoride [7783-81-5] UF, in the United States and United Kingdom, and chlorine trifluoride [7790-91 -2J, CIF, in Germany, led to the development of commercial fluorine-generating cells. The main use of fluorine in the 1990s is in the production of UF for the nuclear power industry (see Nuclearreactors). However, its use in the preparation of some specialty products and in the surface treatment of polymers is growing. 

Tetrafluoroethylene was first synthesized in 1933 from tetrafluoromethane, CF, in an electric arc furnace (11). Since then, a number of routes have been developed (12—18). Depolymerization of PTFE by heating at ca 600°C is probably the preferred method for obtaining small amounts of 97% pure monomer on a laboratory scale (19,20). Depolymerization products contain highly toxic perfluoroisobutylene and should be handled with care. 

Some additives have the ability to lower the pour point without lowering the cloud point. A number of laboratory scale flow tests have been developed to provide a better prediction of cold temperature operability. They include the cold filter plugging point (CFPP), used primarily in Europe, and the low temperature flow test (LTFT), used primarily in the United States. Both tests measure flow through filter materials under controlled conditions of temperature, pressure, etc, and are better predictors of cold temperature performance than either cloud or pour point for addithed fuels. 

A number of special processes have been developed for difficult separations, such as the separation of the stable isotopes of uranium and those of other elements (see Nuclear reactors Uraniumand uranium compounds). Two of these processes, gaseous diffusion and gas centrifugation, are used by several nations on a multibillion doUar scale to separate partially the uranium isotopes and to produce a much more valuable fuel for nuclear power reactors. Because separation in these special processes depends upon the different rates of diffusion of the components, the processes are often referred to collectively as diffusion separation methods. There is also a thermal diffusion process used on a modest scale for the separation of heflum-group gases (qv) and on a laboratory scale for the separation of various other materials. Thermal diffusion is not discussed herein. 

Direct-Liquefaction Kinetics All direct-liquefac tion processes consist of three basic steps (1) coal slurrying in a vehicle solvent, (2) coal dissolution under high pressure and temperature, and (3) transfer of hydrogen to the dissolved coal. However, the specific reac tion pathways and associated kinetics are not known in detail. Overall reaction schemes and semiempirical relationships have been generated by the individual process developers, but apphcations are process specific and limited to the range of the specific data bases. More extensive research into liquefaction kinetics has been conducted on the laboratory scale, and these results are discussed below. 

In recent years researchers at West Virginia University have developed coal-derived pitches on a laboratory scale in quantities sufficient to make 1 kg samples of calcined coke for fashioning graphite test specimens. The pitches were derived by uhlizmg solvent extraction with N-methyl pyrrohdone (NMP). This solvent is able to isolate coal-based pitches m high yield and with low mineral matter content [13]. It is this work that will form the basis of the discussion for the later part of this chapter. 

The importance of chemical syntheses of a-amino acids on industrial scale is limited by the fact that the standard procedure always yields the racemic mixture (except for the achiral glycine H2N-CH2-COOH and the corresponding amino acid from symmetrical ketones R-CO-R). A subsequent separation of the enantiomers then is a major cost factor. Various methods for the asymmetric synthesis of a-amino acids on laboratory scale have been developed, and among these are asymmetric Strecker syntheses as well. ... 

The ultimate goal of process development is to achieve feasibility where it is possible to produce amino adds on a large scale at a production cost per kg of amino add comparable to, or cheaper than, the processes currently used by other companies. If we presume that the technical performance (fermentation and recovery) are sorted out on a laboratory scale and scaling up looks promising, then it is time to find out whether it is possible to operate economically on a large scale. 

The main role of pilot plant is in the scale-up of polymer formulations from laboratory to full scale production and the development of new processes and techniques, including trials of new equipment. The laboratory is normally where the chemistry of new products and processes is investigated and established. When scale-up is contemplated, the use of commercial quality materials will normally be investigated, test procedures established and certain processing tolerances examined. An experienced chemist can frequently learn much on the laboratory scale that will indicate likely scale-up behaviour, but it is always prudent to then go through the pilot stage before embarking on full scale production. 

Felcht reports that the testing of industrial-scale processes can be performed with low expenditure by using micro reactors, since this should result in a faster time to market of the development [137]. He also sees uses for micro reactors at the laboratory scale as a means of high-throughput screening and model examinations such as fast determination of reaction kinetics. 

Finally, yield improvements were also reported for industrial process developments. For the Merck Grignard process, a yield of 95% was obtained by a micro mixer-based process, while the industrial batch process (6 m stirred vessel) had only a 72% yield (5 h, at -20 °C) [11]. The laboratory-scale batch process (0.5 1 flask ... 

The main driver was to develop a laboratory-scale micro-channel process and transfer it to the pilot-scale, aiming at industrial fine-chemical production [48, 108]. This included fast mixing, efficient heat transfer in context with a fast exothermic reaction, prevention offouling and scale-/numbering-up considerations. By this means, an industrial semi-batch process was transferred to continuous processing. 

The solution of these problems is based on a simple idea the developed laboratory-scale process is used for manufacturing of a chemical product by parallelization of many small units. Although promising great advantages over scale-up, this procedure, denoted numbering-up , is not trivial by far. It cannot be carried out in a simple way due to the tremendous technological effort necessary a chemical plant with hundreds or even thousands of small-scaled vessels, stirrers, heaters, pumps. 

Figure 5.13 Solvent demands during the development of a process for the synthesis route in Scheme 5.7. For percentage changes see Table 5.2 (footnote b recycling considered). Due to the introduction of a second solvent in the pilot and operation scale solvent demand is higher than in the laboratory scale (laboratory scale 6.1 kgkg versus pilot scale 10.6 kgkg and operation scale 7.9 kgkg, see also Figure 5.1 0).

---