Melts crystallization

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%. Hence, the objectives of melt crystallization (purity, separation, or concentration) are quite often different compared to crystallization from solution (purity and defined crystal size distribution). 

In the following chapters, the basics concerning melt crystallization will be introduced. Examples of plants for melt crystallization are provided in Chapter 16. 

Purification of a chemical species by solidification from a liquid mixture can be referred to as either solution crystallization or crystallization from the melt. The distinction between these two ( rations is somewhat subtle. The term meltaysialHzation is defined as the separation of components of a mixture without addition of solvent, but this definition is somewhat restrictive. In solution crystallization a diluent solvent is added to the mixture the solution is then directly or indirectly cooled and/or solvent is evaporated to effect crystallization. The solid phase is normally formed and maintained somewhat below its pure-component 

FIGURE 11J-12 Case study of sensitivity of product critical size to fines tap design. 

High or ultiahigh product purity is obtained with many of the melt purification processes. Table 

4- 1 compares the product quality and product form that is produced from several of these operations. Zone refining can produce very pure material when operated in a batch mode however, other techniques also provide high purity and b oine attractive if continuous high-capacity processing is desired. 

Zone melting relies on the distribution of solute between the liquid and solid phases to eflect a separation. In this case, however, one or more. liquid zones are passed through the ingot. This extremely versatile technique, which was invented by W. G. Pfann, has been used to purify hundreds of materials. Zone melting in its simplest form is illustrated in Fig. 11.4-1. A molten zone can be passed through an ingot 

Finns Cut Size h Fines Flow Index R (min) Fines Flow ( Mi) Fines Trap Ares (ft2) Weight Fraction Ciystals Lf 

For optimum process control and good economics, seed growth must be initiated without a large degree of supercooling.Theories describe the relationship between all the quantities and the work of seed formation [Volmer 1931]. One of these quantities is the interfacial tension between suface and melt. Thus seed formation in melt crystallization can be improved by optimizing the material combination cooling surface/ 

In suspension crystallization the melt is cooled to below the stauration temperature, and crystals grow under adiabatic conditions. The degree of supersaturation is the driving force. Specialist knowledge is required to obtain crystals with a certain purity, structure, and particle size distribution. The residual mdt, which contains the impurities, must be mechanically separated from the crystals. 

In film crystallization the crystals grow on a cooled wall. Therefore, the crystals are colder than the melt (nonadiabatic process), and the driving force is the temperature gradient. Thus, rates of crystal growth that are 10-100 times higher than in suspension crystallization are attainable. 

Liquid feed 1 to 30 % w/w solute operates below the freezing temperature of the solvent. 

Do not exceed the Kjp of the solute. For juices, the solute concentration increases from about 10% to 50% when the freezing temperature is -10 °C. 

Related topic see solidify liquids to solids flakers, belts, Section 9.12. 

Liquid concentration 60 to 90% w/w. Operates below the freezing temperature to solidify the target solute. Capacity usually 3 kg/s. Use for temperature-sensitive materials, when vp 1-4, when azeotropes form or for solid product. A freeze test shows 50% reduction in impurities. 

Design based on heat transfer. Limited by eutectic formation. Multistaging possible but usually 10 stages. 

---

Transformations in the Solid State. From a practical standpoint, the most important soHd-state transformation of PB involves the irreversible conversion of its metastable form II developed during melt crystallization into the stable form I. This transformation is affected by the polymer molecular weight and tacticity as well as by temperature, pressure, mechanical stress, and the presence of impurities and additives (38,39). At room temperature, half-times of the transformation range between 4 and 45 h with an average half-time of 22—25 h (39). The process can be significantly accelerated by annealing articles made of PB at temperatures below 90°C, by ultrasonic or y-ray irradiation, and by utilizing various additives. Conversion of... 

Melt Crystallization. Crystallization (qv) from a melt is inherently more attractive than distillation because the heat of fusion is much lower than that of evaporation. It also benefits from lower operating temperature. In addition, organic crystals are virtually insoluble in each other so that a pure product is possible in a one-stage operation. 

Separation of a chemical species from a mixture of similar compounds can also be achieved by melt crystallization, which is, for example, an important means of separatingpara- s.yXen.e (p-xylene) from the ortho and meta isomers. -Xylene is crystallized at the top of a vertical column and crystals are moved downward countercurrentiy to Hquid. The Hquid flowing upward is generated by adding heat to melt the crystals at the bottom of the column a portion of the melt is removed as product and the remainder flows up the column to contact the downward-flowing crystals. Effluent mother Hquor, consisting almost entirely of the ortho and meta isomers of xylene, is removed from the top of the column. 

Melt Crystallization. The use of a solvent can be avoided in some systems. In such cases, the system operates with heat as a separating agent, as do several processes involving crystallization from solution, but formation of crystalline material is from a melt of the crystallizing species rather than a solution. 

Melt crystallization is carried out either with a suspension of crystals or an advanciag front (layer) of soHds, although a more complete categorization of melt crystallization is available (71). FoUowiag is a brief review of processes ia which melt crystallization is used a more complete review, including a worked out case study for system design, is available (69). 

In advancing-front or layer melt crystallizations, mother Hquor flows over a cooled surface on which material is crystallized. The advancing front of crystals grows in the direction from the cooled surface into the mother Hquor. A variety of techniques can be used to take advantage of this type of Operation. 

Of these five methods all but pressure-swing distillation can also be used to separate low volatiUty mixtures and all but reactive distillation are discussed herein. It is also possible to combine distillation and other separation techniques such as Hquid—Hquid extraction (see Extraction, liquid-liquid), adsorption (qv), melt crystallization (qv), or pervaporation to complete the separation of azeotropic mixtures. 

Another major type of equipment among the family of crystallizers is the flaker. The flaker is a piece of equipment that is used for the production by chilling of flakes, chips, large crumbs, or crystals from a hot, concentrated solution or melted crystals. 

Vill. EFFECT OF MOLECULAR WEIGHT ON THE MELTING-CRYSTALLIZATION OF PE UNDER HIGH PRESSURE... 

Figure 16-47. Emission spectra of a melt-crystallized thin (300 nm) film of Oocl-OPV5. Excitation energy (a) I pJ, (b) 10 pj, (c) 50 pJ, (el) 70 pJ excitation beam diameter x 1 mm.

A further confirmation that mirrorlcss lasing is restricted to single domains comes from an experiment in which an Oocl-OPV5 film has been crystallized from the isotropic melt phase (above 204 "C). Melt crystallization resulted in the formation of large domains with dimensions up to several millimeters (see Fig. 16-29 C). Tlie normalized emission spectra for different excitation energies are shown in Figure 16-47. The excitation spot diameter was 1 mm in these ex-... 

During geological time, a number of separating and sorting processes—melting, crystallization, solution, precipitation—have concentrated various elements in local deposits. In these, the elements tend to be grouped together in rather stable compounds. These are called minerals. Many of the minerals have compositions similar... 

Another interesting example is the melt crystallization of s-PS. For the case of rapid cooling from the melt, the hexagonal a form is obtained [7-9], while for low cooling rates or for isothermal crystallizations, the crystalline form which is... 

It is known the case of i-PP, for which the copolymerization with small amounts of ethylene tends to stabilize the y form [84] for instance, by melt crystallization of a copolymer with 6% by mol of ethylene more than 80% of the crystalline phase is in the y form [85], It is also known that the obtainment of the y form by melt crystallization, is also favored for samples of low molecular mass [86, 87] and for stereoblock fractions [88]. This seems to suggest that, whenever the preferential crystallization of the y-form is observed, there is the concomitant occurrence of a reduction in the polymer of the length of the chain stretches with polypropylene head to tail constitution and isotactic configuration. 

Particularly relevant is the case of some copolymers of PVDF. Already small amounts (5-20% by mol) of a fluorolefinic comonomer (vinyl fluoride (VF) [89-90], trifluoroethylene [91-93], tetrafluorethylene [94, 95]) can force the polymers to a melt crystallization in the piezoelectric P form. (We recall that the homopolymer crystallizes in the non-piezoelectric a form, by melt crystallization). 

For the case of melt crystallization, the blending with PPO favors the obtainment of the P form. On the basis of several experimental results, it has been suggested that this behavior would be due to the interactions between PS and PPO chains in the melt, which would produce a more rapid disappearence of a memory (that is of possible nuclei) of the a form in the melt [104]. 

Recovery by melt crystallization also is limited by the eutectic composition. 

The approach used in these studies follows idezus from bifurcation theory. We consider the structure of solution families with a single evolving parameter with all others held fixed. The lateral size of the element of the melt/crystal interface appears 2LS one of these parameters and, in this context, the evolution of interfacial patterns are addressed for specific sizes of this element. Our approach is to examine families of cell shapes with increasing growth rate with respect to the form of the cells and to nonlinear interactions between adjacent shape families which may affect pattern formation. 

In this description the temperature field has been taken to be linear in the coordinate y and to be independent of the shape of the melt/crystal interface. This is a good assumption for systems with equal thermal conductivities in melt and crystal and negligible convective heat transport and latent heat release. Extensions of the model that include determination of the temperature field are discussed in the original analysis of Mullins and Sekerka (17) and in other papers (18,19). 

Equation (2) is identified as a second-order, nonlinear differential equation once, the curvature is expressed in terms of a shape function of the melt/crystal interface. The mean curvature for the Monge representation y = h(x,t) is... 

The mapping (7) introduces the unknown interface shape explicitly into the equation set and fixes the boundary shapes. The shape function h(x,t) is viewed as an auxiliary function determined by an added condition at the melt/crystal interface. The Gibbs-Thomson condition is distinguished as this condition. This approach is similar to methods used for liquid/fluid interface problems that include interfacial tension (30) and preserves the inherent accuracy of the finite element approximation to the field equation (27)... 

---