Process design scale

Liquid adsorption processes hold a prominent position ia several appHcations for the production of high purity chemicals on a commodity scale. Many of these processes were attractive when they were first iatroduced to the iadustry and continue to iacrease ia value as improvements ia adsorbents, desorbents, and process designs are made. The UOP Parex process alone has seen three generations of adsorbent and four generations of desorbent. Similarly, Hquid adsorption processes can be applied to a much more diverse range of problems than those presented ia Table 3. 

Hydrothermal Synthesis Systems. Of the unit operations depicted in Figure 1, the pressurized sections from reactor inlet to pressure letdown ate key to hydrothermal process design. In consideration of scale-up of a hydrothermal process for high performance materials, several criteria must be considered. First, the mode of operation, which can be either continuous, semicontinuous, or batch, must be determined. Factors to consider ate the operating conditions, the manufacturing demand, the composition of the product mix (single or multiple products), the amount of waste that can be tolerated, and the materials of constmction requirements. Criteria for the selection of hydrothermal reactor design maybe summarized as... 

Process plant design has come a long way from the early 1930s when process designers used the rule-of-thumb that a process faciUty could not be scaled-up more than 10-fold (2). American Oil s Ultracracking unit (Texas City, Texas) for example, was designed from data from a small pilot plant with a scale-up factor of 80,000 (3). 

The influence of transport process in two-phase reaction systems depends on flow conditions, which change with the size of the equipment. This is the reason for the historic observation that performance changes as processes are scaled up and therefore scale-up should be done in several steps, each limited to a small increase in size. This is a slow and expensive method and still does not guarantee optimum design. 

After many years of improvements in technical safety methods and process design, many orgaruzations have found that accident rates, process plant losses and profitability have reached a plateau beyond which further improvements seem impossible to achieve. Another finding is that even in orgarriza-tions with good general safety records, occasional large scale disasters occur which shake public confidence in the chemical process industry. The common... 

The purification of value-added pharmaceuticals in the past required multiple chromatographic steps for batch purification processes. The design and optimization of these processes were often cumbersome and the operations were fundamentally complex. Individual batch processes requires optimization between chromatographic efficiency and enantioselectivity, which results in major economic ramifications. An additional problem was the extremely short time for development of the purification process. Commercial constraints demand that the time interval between non-optimized laboratory bench purification and the first process-scale production for clinical trials are kept to a minimum. Therefore, rapid process design and optimization methods based on computer aided simulation of an SMB process will assist at this stage. 

In summary, the Koji process serves as a small scale, relatively localised and spedalised process designed largely to take advantage of the available carbohydrate source. 

As an example for continuous process design, 2-keto-3-deoxy-D lycero-D-galacto-nonosouate (KDN) (S) has been produced on a 100-g scale from D-mannose and pyruvate using a pilot-scale EMR at a space-time yield of 375 gl d and an overall crystallized yield of 75% (Figure 10.6) [47]. Similarly, L-KDO (6) can be synthesized from L-arabinose [48]. 

In process design, it is frequently discovered that many of the basic data needed to rmderstand a process are lacking. Because most crrrrent mathematical models are not sufficiently accrrrate to permit direct scale-up of the process from laboratory data to full plant size, a pilot plant must be constracted. As models are improved, it may become possible to evaluate design decisions with more confidence, and bypass the pilot plant stage. 

Since electrochemical processes involve coupled complex phenomena, their behavior is complex. Mathematical modeling of such processes improves our scientific understanding of them and provides a basis for design scale-up and optimization. The validity and utility of such large-scale models is expected to improve as physically correct descriptions of elementary processes are used. 

The time scale of deactivation has profound consequences for process design. This is easiest illustrated for continuous processes by plotting the catalytic activity as a function of the time-on-stream. Qualitatively, often the type of behaviour shown in Fig. 3.34 is observed. 

---