Influence of Solvents and Impurities

The choice of the solvent plays an important role in crystallization processes. The solute to be crystallized should be readily soluble in the solvent and it should be 

Sometimes mixed solvents, that is, mixtures of two or more solvents, show more favorable solution properties for a particular solute than just one solvent. Therefore, like temperature, the solvent composition is a common variable in designing a crystallization process. In antisolvent crystallization, the application of an additional solvent (antisolvent) is targeted to the reduction of the solubility of a substance in order to cause its crystallization. 

Impurities can also affect the solubility of a solute of interest. Here, both a solubility enhancement and a solubility decrease occur. When electrolytes are involved, the terms salting-in and salting-out apply. Small impurity contents might be evaluated together with the solvent. In presence of higher impurity contents or in cases where the impurity is readily available in sufficient amounts, it should be considered as a third component in the system. Then, SLE data in the ternary system of the target compound, the impurity, and the solvent/solvent mixture have to be measured and instead of a binary a ternary (solubility) phase diagram applies. The representation and application of ternary SLE will be addressed in Section 3.3.7 on the example of enantiomers. 

KCI in a NaCI-saturated solution that still increases with temperature, NaCI shows retrograde solubility in KCI-saturated solutions (lines 3 and 4). This particular solubility behavior is used in the so-called hot leaching process to separate KCI and NaCI from sylvinite crude salts. It is obvious from the solubility curves that cooling a solution saturated with both salts (point of intersection of lines 3 and 4 at about 100 C) leads to selective crystallization of KCI as target compound. (Reproduced with permission from Ref [8].) 

---

Lastly, the mass transport processes at the crystal-liquid interface play a central role in crystallization. The influence of solvent and impurities on the structure and growth rates of faces is considered in this chapter along with its effect on the incorporation of impurities. The solvent solute-impurities interactions in solution will also be shown to interact in subtle, but important, ways with the interface during the crystallization process. With appropriate thermodynamic analysis it is shown how these interactions ultimately affect crystallization as a purification process. 

Typically, crystal habits predicted based on crystal chemistry alone are best compared with crystals grown from sublimation processes, or to solution-based systems where the solvent/impurity interactions are negligible. In fact, significant deviations from structure-based predictions are often best explained by such solvent and impurity interactions (Davey et al. 1992 Winn et al. 2000). There are significant efforts underway to understand such phenomena at the molecular level and to consequently predict the influences of solvent and impurities on crystal growth rates. This will be further highlighted in subsequent sections. 

Gu, C. Influence of Solvents and Impurities on the Crystallization Process and Properties of Crystallized Products. Ph.D. thesis. Department of Pharmaceutics, University of Minnesota, Minneapolis, 2001, pp. 123-160. 

As the next step in understanding the factors influencing crystal shape, and the effect of solvents and impurities, models... 

Some redox couples of organometallic complexes are used as potential references. In particular, the ferrocenium ion/ferrocene (Fc+/Fc) and bis(biphenyl)chromium(I)/ (0) (BCr+/BCr) couples have been recommended by IUPAC as the potential reference in each individual solvent (Section 6.1.3) [11]. Furthermore, these couples are often used as solvent-independent potential references for comparing the potentials in different solvents [21]. The oxidized and reduced forms of each couple have similar structures and large sizes. Moreover, the positive charge in the oxidized form is surrounded by bulky ligands. Thus, the potentials of these redox couples are expected to be fairly free of the effects of solvents and reactive impurities. However, these couples do have some problems. One problem is that in aqueous solutions Fc+ in water behaves somewhat differently to in other solvents [29] the solubility of BCr+BPhF is insufficient in aqueous solutions, although it increases somewhat at higher temperatures (>45°C) [22]. The other problem is that the potentials of these couples are influenced to some extent by solvent permittivity this was discussed in 8 of Chapter 2. The influence of solvent permittivity can be removed by... 

Solvent and impurity effects must also be considered. Solvent effects are important and may play a key role in inclusions and in affecting the width of the metastable zone, as discussed in Example 11-1. However, variations in impurity composition can have more influence and can dominate the course of crystallization in many ways. 

Particular problems are given special emphasis, because certain authors do not take into account the decisive influence of various factors in the process of polymer oxidation. These factors include the content of impurities, catalysts, photosensitizers, the properties of solvents, and the presence of singlet oxygen and ozone. 

All impurities in the pure solvents have to be ehminated. Degassing of solvents (and sometimes of polymers too) is absolutely neeessaty. Polymers and solvents must keep diy. Sometimes, inhibitors and antioxidants are added to polymers. They may probably influence the position of the equilibrium. The thermal stability of polymers must be obeyed, otherwise, depolymerization or formation of networks by chemical processes can change the sample during the experiment. 

The interfacial mass transfer simulator was made of quartz glass with an inner size of 200 mm in length, 20 mm in width, and 40 mm in height. The liquid was initially quiescent in the simulator with a thickness of 10 mm. Nitrogen gas successively passed through activated carbon, silica gel, and molecular sieve to remove the impurities and water, and then presaturated by the solvent in a tank in order to reduce the influence of solvent evaporation. The hquid was likewise presaturated by nitrogen gas to avoid the gas absorption into the liquid. The liquid concentrations near the gas inlet and outlet positions of the simulator were measured by the gas chromatography. 

The theory underlying the removal of impurities by crystaUisation may be understood from the following considerations. It is assumed that the impurities are present in comparatively small proportion—usually less than 5 per cent, of the whole. Let the pure substance be denoted by A and the impurities by B, and let the proportion of the latter be assumed to be 5 per cent. In most instances the solubilities of A (SJ and of B (/Sb) are different in a particular solvent the influence of each compound upon the solubility of the other will be neglected. Two cases will arise for an3 particular solvent (i) the impurity is more soluble than the compound which is being purified (/Sg > SA and (ii) the impurity is less soluble than the compound Sg < S ). It is evident that in case (i) several recrystallisations will give a pure sample of A, and B will remain in the mother liquors. Case (ii) can be more clearly illustrated by a specific example. Let us assume that the solubility of A and 5 in a given solvent at the temperature of the laboratory (15°) are 10 g. and 3 g. per 100 ml. of solvent respectively. If 50 g. of the crude material (containing 47 5 g. of A and 2-5 g. of B) are dissolved in 100 ml. of the hot solvent and the solution allowed to cool to 15°, the mother liquor will contain 10 g. of A and 2-5 g. (i.e., the whole) of B 37-5 g. of pure crystals of A will be obtained. 

The amount of solvent relative to the amount of total catalyst is usually large, and the amount of solvent relative to the number of active catalyst sites larger still very small amounts of inhibitors or poisons can have, therefore, large adverse influences on the rate of reduction. Solvent purity per se is of little regard in this connection, for gross amounts of innocuous impurities can be present without untoward effect. 

Model formulation. After the objective of modelling has been defined, a preliminary model is derived. At first, independent variables influencing the process performance (temperature, pressure, catalyst physical properties and activity, concentrations, impurities, type of solvent, etc.) must be identified based on the chemists knowledge about reactions involved and theories concerning organic and physical chemistry, mainly kinetics. Dependent variables (yields, selectivities, product properties) are defined. Although statistical models might be better from a physical point of view, in practice, deterministic models describe the vast majority of chemical processes sufficiently well. In principle model equations are derived based on the conservation law ... 

Plastic deformation, unlike elastic deformation, is not accurately predicted from atomic or molecular properties. Rather, plastic deformation is determined by the presence of crystal defects such as dislocations and grain boundaries. While it is not the purpose of this chapter to discuss this in detail, it is important to realize that dislocations and grain boundaries are influenced by things such as the rate of crystallization, particle size, the presence of impurities, and the type of recrystallization solvent used. Processes that influence these can be expected to influence the plastic deformation properties of materials, and hence the processing properties. 

---