Crystallization is an extremely important process with extensive industrial applications including, but not limited to, the manufacture of electronics, explosives, fine chemicals, and pharmaceuticals. As such, controlling both crystal shape and crystal structure is vital for the production of high-quality products with desirable properties. However, the processes that govern crystallization, crystal growth, and crystal nucleation are not well understood at present. This is due in part to the limitations of experimental techniques in studying such processes because of the small number of molecules, often tens or hundreds, involved. Furthermore, experimental strategies for identifying and analyzing crystal structures (which may have serious implications in terms of intellectual property rights) and controlling crystal shape are not always successful in yielding the optimal product and often can be costly and time consuming.

Melt crystallization is an important separation, purification, and concentration technique used in the chemical, pharmaceutical, and food industries. Crystallization from melt is a very powerful separation process for the purification of organic compounds up to very high purities of 99.99 percent. Therefore, the objectives of melt crystallization (i.e., purity, separation, or concentration) are quite often different from crystallization from solution (i.e., purity and defined crystal size distribution). Good background information about the theory of melt crystallization can be found, for example, in Arkenbout (1995), Atwood (1972), Jansens and van Rosmalen (1994), Matsuoka (1991), Matz (1969), Molinari (1967), Mullin (2001), Özoğuz (1992), Rittner and Steiner (1985), Sloan and McGhie (1988), Toyokura and Hirasawa (2001), Ulrich and Bierwirth (1995), Ulrich and Kallies (1994), Ulrich and Nordhoff (2006), Ulrich and Stelzer (2011), Verdoes et al. (1997), and Wintermantel and Wellinghoff (2001). In the following sections, the basics and design examples of plants for melt crystallization will be given.

Protein crystals are of interest for several fields of science and technology. Their formation underlies several human pathological conditions. An example is the crystallization of hemoglobin C and the polymerization of hemoglobin S that cause, respectively, the CC and sickle cell diseases (Charache et al. 1967; Hirsch et al. 1985; Eaton and Hofrichter 1990; Vekilov 2007). The formation of crystals and other protein condensed phases of the so-called crystallines in the eye lens underlies the pathology of cataract formation (Berland et al. 1992; Asherie et al. 2001). A unique example of benign protein crystallization in humans and other mammals is the formation of rhombohedral crystals of insulin in the islets of Langerhans in the pancreas. The suggested function of crystal formation is to protect the insulin from the proteases present in the islets of Langerhans and to increase the degree of conversion of the soluble proinsulin (Dodson and Steiner 1998).

Crystallization from solutions is a complex process completed in several stages. The first stage is the formation of supersaturated solution because the spontaneous appearance of a new phase can occur only when a system is in a nonequilibrium condition. In the next stage, molecules dissolved in solution begin to aggregate to relieve the supersaturation and move the system toward equilibrium. The molecular aggregation process eventually leads to the formation of nuclei that can act as centers of crystallization. A nucleus can be defined as the minimum amount of a new phase capable of independent existence (Khamskii 1969). The nature of nuclei (i.e., whether they are amorphous particles or tiny crystals) is still unknown. The birth of these small nuclei in an initially metastable phase is called nucleation, which is a major mechanism of first-order phase transition. Kashchiev and van Rosmalen (2003) describe nucleation as the process of fluctuational appearance of nanoscopically small clusters of the new crystalline phase, which can grow spontaneously to macroscopic sizes. The growth stage, which immediately follows nucleation, is governed by the diffusion of particles, called growth units, to the surface of the existing nuclei and their incorporation into the structure of the crystal lattice (Khamskii 1969). This stage continues until all the solute in excess of saturation is consumed for the development of mature crystals. The initial stages of crystallization, which can be defined as the period between the achievement of supersaturation and the formation of nuclei, plays a decisive role in determining properties of the resulting solid phase, such as purity, crystal structure, and particle size. Thus higher levels of control over crystallization cannot be achieved without understanding the fundamentals of nucleation.

Mixing determines the environment in which crystals nucleate and grow and is therefore intrinsic to industrial crystallization. Individual nucleating and growing crystals respond directly to their microenvironment and not in a simple way to the macroenvironment, often thought of as the bulk or average environment. Because the growing crystal removes solute from solution and the dissolving crystal releases it, the solute concentration and therefore the supersaturation is in general different at the crystal surface than in the bulk. Crystals grow when the microenvironment is supersaturated, stop when it is just saturated, and dissolve when it is undersaturated. In most cases, impurities are rejected by growing crystals; therefore, each growing crystal face creates a zone of locally higher impurity concentration immediately adjacent to it. The growth rate and amount of impurity taken up by the growing crystal are functions of the impurity concentration where growth is occurring – at the crystal face itself. Mixing is the family of processes that links this local microenvironment to the macroscopic scale of the crystallizer by affecting the mass transfer between crystal and the larger environment and the dynamics of crystal suspension flow in the crystallizer. Mixing, therefore, to a large extent creates the crystal microenvironments. Furthermore, it determines the homogeneity of the macroenvironment, both temporally and spatially. Inhomogeneity in the macroenvironment affects the microenvironments around crystals, causing temporal variations as the crystals circulate from one zone to another inside the crystallizer. This is particularly important because local values of key variables such as supersaturation and solids concentration are often much more important in crystallization than the bulk or global averages of these quantities, as discussed below.

There are many components in foods that crystallize, either partially or completely (Hartel 2001). Most important are sugars (i.e., sucrose, lactose, glucose, and fructose), ice, lipids and starches, although crystallization of salts, sugar alcohols, organic acids, proteins, and emulsifiers may be important in certain applications. Crystallization in the food industry differs to some extent from that in other fields in that, for the most part, the crystals form an integral part of the food. Although separation of crystals is important in certain food applications, crystalline structures within the food itself often define the characteristics of that product.

Crystallization is a separation and purification technique employed to produce a wide variety of materials. Crystallization may be defined as a phase change in which a crystalline product is obtained from a solution. A solution is a mixture of two or more species that form a homogeneous single phase. Solutions are normally thought of in terms of liquids, but solutions may include solids and even gases. Typically, the term solution has come to mean a liquid solution consisting of a solvent, which is a liquid, and a solute, which is a solid, at the conditions of interest. The term melt is used to describe a material that is solid at normal conditions and is heated until it becomes a molten liquid. Melts may be pure materials, such as molten silicon used for wafers in semiconductors, or they may be mixtures of materials. In that sense, a homogeneous melt with more than one component is also a solution, but it is normally referred to as a melt. A solution can also be gaseous; an example of this is a solution of a solid in a supercritical fluid.