Issues in Crystallization Practice

Direct scale-up from an undeveloped laboratory procedure may result in a different product from the expected one. These differences may include smaller mean particle size, wide and/or bimodal particle size distribution (PSD), needles or plates with difficult downstream processing properties, and lack of consistent results. 

Most examples in this book illustrate successful manipulation to achieve particular goals. In some cases, however, the natural growth and/or morphology has not been modified sufficiently by the corrective action to achieve success. Example 9-3 is a case in which satisfactory growth could not be achieved. For difficult situations such as these, downstream problems can be minimized by optimization of process variables (Johnson et al. 1997 Kim et al. 200.5). 

In the absence of control of supersamration, nucleation will usually predominate. Nyvlt et al. (1985, p. 36) notes that the primary requirement in control of the crystallization process is thus control of the number of crystals formed . As discussed below, this control may be 

Crystallization- of Organic Compounds An Industrial Perspective. By H.-H. Tung. E. L. Paul, M, Midler, and J. A. McCauley 

A specific crystallization can be dominated by either nucleation or growth. Which of these does dominate depends on the methods of control of critical variables, the amount of seed, the size distribution and surface qualities of the seed, and the environment in which supersamration is created, as well as the specific nucleation and growth rate properties of the compound. 

---

A highly detailed picture of a reaction mechanism evolves in-situ studies. It is now known that the adsorption of molecules from the gas phase can seriously influence the reactivity of adsorbed species at oxide surfaces[24]. In-situ observation of adsorbed molecules on metal-oxide surfaces is a crucial issue in molecular-scale understanding of catalysis. The transport of adsorbed species often controls the rate of surface reactions. In practice the inherent compositional and structural inhomogeneity of oxide surfaces makes the problem of identifying the essential issues for their catalytic performance extremely difficult. In order to reduce the level of complexity, a common approach is to study model catalysts such as single crystal oxide surfaces and epitaxial oxide flat surfaces. 

The effective removal of residual palladium has become one of the most critical issues in the implementation of homogeneously catalyzed Heck reactions in industrial syntheses, particularly in the production of pharmaceuticals, in which tight specifications (sometimes less than 0.5 ppm Pd) must be met. Only very few scalable, inexpensive removal techniques are known [47], for example, binding of Pd to N-acetylcysteine and removal of the adduct by extraction or crystallization [48], When such practical, inexpensive separation and recycling tools for palladium catalysts have been fully developed and industrially implemented, the Heck reaction will certainly find commercial use to the large extent that would be expected from its enormous synthetic utility. 

In the case of non-Newtonian behavior and especially in the case of viscoplastic behavior, such as that typical for moderately concentrated colloidal dispersions, Poiseuille s law gradually loses its validity. This happens because in the shear force-free central region close to the capillary axis, the structure of the concentrated colloidal system remains intact, so that the viscous shear exists only in the peripheral regions of the capillary. This process causes serious issues in the pumping of cement slurries or crude oil containing crystallizing phases. In laboratory practice, it is beneflcial to conduct such measurements in combination with other measurements that utilize uniform states. 

In addition to the challenges cited above, there are some special issues associated with steroid chemistry that should be noted. The steroidal impurities formed in the process are generally similar in structure to the desired product and, in some cases, co-crystallization with the product is a problem. It is, therefore, critical to limit the formation of steroidal impurities in the reactions. The structural similarity between product and impurities also creates challenges in developing assays for reaction monitoring and purity determination. Furthermore, the poor solubility of these compounds in the solvents typically used in a manufacturing process makes it very difficult to achieve practical volume productivity in process development. 

Monodisperse spherical colloids and most of the applications derived from these materials are still in an early stage of technical development. Many issues still need to be addressed before these materials can reach their potential in industrial applications. For example, the diversity of materials must be greatly expanded to include every major class of functional materials. At the moment, only silica and a few organic polymers (e.g., polystyrene and polymethylmethacrylate) can be prepared as truly monodispersed spherical colloids. These materials, unfortunately, do not exhibit any particularly interesting optical, nonlinear optical or electro-optical functionality. In this regard, it is necessary to develop new methods to either dope currently existing spherical colloids with functional components or to directly deal with the synthesis of other functional materials. Second, formation of complex crystal structures other than closely packed lattices has been met with limited success. As a major limitation to the self-assembly procedures described in this chapter, all of them seem to lack the ability to form 3D lattices with arbitrary structures. Recent demonstrations based on optical trapping method may provide a potential solution to this problem, albeit this approach seems to be too slow to be useful in practice.181-184 Third, the density of defects in the crystalline lattices of spherical colloids must be well-characterized and kept below... 

This takes us back to the issue of surface heterogeneity. Suspended oxides expose a variety of crystal planes between which the various pK s may vary. Hence, in practice, measurements refer to very heterogeneous surfaces. Accounting for heterogeneity can be done along the same lines as introduced in sec. 1.7 with the potential as an additional parameter but the analysis is very laborious. Perhaps the trend is that the electric potential tends to smooth the... 

---