Table of Contents

Figure 1: Overview of Crystallization Process [1]

What is Crystallization?

Crystallization is a process by which a solid is formed, where the atoms or molecules are arranged into a repeating well-defined three-dimensional crystal lattice in order to minimize their energy state [1]. The process can be completed by freezing, precepting from a solid, or deposition from a gas. Crystallization can be caused by a physical change, such as temperature and air pressure, or a chemical change, such as acidity [2]. During crystallization, the atoms or molecules bind-together with well-defined angles that form a crystal with smooth surfaces and facets on the macroscopic level [1]. Crystallization can be a slow process, between two and seven days for a good quality crystal that is able to diffract well, compared to precipitation because of their regular microscopic structure [3]. Although crystallization occurs in nature, it has a wide range of industrial application. Some of those applications are purifications of solid compounds and separation of solids from a solution. It is used to produce essential materials, such as sodium chloride, sodium and aluminum sulfates, and sucrose.

Crystallization Process

The driving force behind the formation and growth of crystals is supersaturation [4]. Supersaturation is caused when the concentration of the solute exceeds the equilibrium (saturation) solubility concentration. Thermodynamically, the driving force is the change in Gibbs' free energy. However this is difficult to determine, instead the concentration difference is commonly used in practical correlation [4]. The crystallization process occurs through a two-step stage, nucleation and crystal growth, accompanied by multiple intermediate states. The structures and transition dynamics of the intermediate states are too difficult to observe, however macroscopic molecular dynamics simulations revealed the two-step process [5].

Nucleation

The first step of crystallization is nucleation. During the nucleation process, a small number of ions, atoms, or molecules are arranged in cluster in a pattern of a crystalline solid known as a nucleus. Total nucleation is the sum of two categories - primary and secondary. These clusters must be stable under current experimental conditions to grow into the "critical cluster size" or they will re-dissolve [6]. Nucleation is a debated topic in the academic and industrial community because of its mystery and concealment. The most popular and oldest is classical nucleation theory, which states that the nucleation process is the result of competition between free energy and total surface energy [7].

Primary Nucleation

Primary nucleation is the formation of a crystal in absence of any other crystal or with a crystal that will not affect or influence the process [8]. Classical theories of primary nucleation are a build-up of lattice-structured bodies which may or may not achieve thermodynamic stability caused by sequences of bimolecular collisions and interactions in as supersaturated fluid [9]. At low supersaturation levels, primary nucleation isn't significant; at higher supersaturation levels, primary nucleation rates increase drastically. There are two types of primary nucleation, the homogeneous nucleation and heterogeneous nucleation. In homogeneous nucleation, the nuclei are formed in a perfectly clean solution where there are no foreign particles such as impure molecules, dust particles, or ions. The nucleation is also not affected by particles of any substance or the walls of equipment used in the crystallization process, also known as crystallizer. Normally, the nucleation process occurs randomly and spontaneously. Homogeneous nucleation is rare because the nucleation occurs without any favorable nucleation sites and the enthalpy is higher to start the nucleation because of the absence of solid particles. For homogeneous nucleation, after a certain amount of time the nucleation starts, the crystals grow and become bigger at the same time new crystals start to nucleate. In heterogeneous nucleation, the nuclei are formed in a solution that contains foreign particles such as impure molecules, dust particles, or ions. Heterogeneous nucleation occurs more often than homogeneous nucleation, therefore, in industrial crystallization, most primary nucleation is almost certainly heterogeneous. Normally, this process occurs at favorable sites such as surfaces of containers, impurities, and phase boundaries. At these sites, the effectiveness of the surface energy is lower, thus it can ease the nucleation process and reduce the free energy barrier. However, the exact mechanism of heterogeneous nucleation is not fully understood, it may start with the absorption of the crystallizing species on the surface of solid particles, therefore creating crystalline bodies that are larger than the critical nucleus size, which then grow into macro-crystals [9]. In heterogeneous nucleation, after a certain time when the nucleation process starts, the crystal starts to grow and become bigger and no nucleation occurs [8].

Figure 2: Effect of supersaturation on the rates of homogeneous and heterogeneous nucleation [9]	Figure 3: Homogeneous Nucleation [8]	Figure 4: Heterogeneous Nucleation [8]

Secondary Nucleation

Secondary nucleation occurs only when crystals of the species under consideration are already present [9]. Since this is usually the case in industrial crystallizers, secondary nucleation has a large impact on virtually all industrial crystallization processes. Contacting with previous crystals is the most common and effective method in nucleation. The advantages of secondary nucleation are: the nucleation occurs at low supersaturation since the present crystals reduce the energy required which causes the growth rate to be optimal for good quality, lower energy is needed where the crystal strike avoids breaking the existing crystal to form new crystal, lower kinetic order and rate-proportional to supersaturation which allows easy control without unstable operation, and quantitative fundamentals and practices have already been isolated and are being incorporated within industry [8]. The crystallization process always enters an uncontrollable state due to the stochastic and spontaneous nature of the primary nucleation process. However, secondary nucleation can bring the crystallization process into a more easily controlled state, which is important when controlling crystal size distribution, polymorphism, and chirality of the crystal products [7].

Crystal Growth

Crystal growth takes place in two stages, a diffusional stage in which solute is moved through the bulk fluid through the solution boundary layer adjacent to the crystal structure, and a deposition step in which absorbed solute ions or molecules at the crystal surface are deposited and integrated into the crystal lattice. It is this point of the crystallization process that defines the crystal structure as more ions, atoms, or molecules are deposited as the crystal grows [9]. Changes in the environment during the growth environment has profound effects on growth, such as temperature, supersaturation pH, and impurities. These differences in individual face growth rates give rise to habit changes in crystals. The solution velocity past the fixed crystal is an important growth-determining parameter, sometimes responsible for the size-dependent growth effect in agitated and fluidised-bed crystallisers. Large crystals have higher settling velocities than small crystals and, if their growth is diffusion-controlled, they tend to grow faster. Salts that include the alums such as nickel ammonium sulphate and potassium sulphate have solution velocity growth rates. Although salts such as ammonium sulphate and ammonium or potassium dihydrogen phosphate aren't affected by solution velocity [9]. For some systems, incorporation of solute into crystals is easy and growth is finite by diffusion to the surface through bulk solution or boundary layer. In other systems, surface integration of solute is rate controlling and growth on a flat crystal face is difficult, as growth mainly occurs on kinked or stepped edges. In extreme cases, growth is strongly dependent on dislocations in the crystal. Since the rate of primary nucleation is not significant in low supersaturation levels but the rate drastically increases in higher supersaturation levels; this leads to the concept of a metastable zone in which growth dominates, and a liable zone in which primary nucleation dominates. Usually metastable zone widths range between 1-2°C to 30-40°C; inorganic compounds generally have lower widths than organic ones [4]. Crystal growth kinetics often depend on crystal size because the size depends on the surface deposition kinetics. Different crystals of the same size also have different growth rates because of the differences in surface structure or perfection. Thus small crystals of many substances grow much more slowly than larger crystals, and some don't grow at all. Since the rate of growth complexly depends on temperature, supersaturation, size, habit, and system turbulence, there is not a simple and exact way of expressing the rate of crystal growth. Under carefully defined and maintained conditions, growth can be expressed as an overall mass deposition rate, an overall linear growth rate, or as a mean linear velocity [9].

Impurities

The perfect separation with optimal productivity, yield, and purity is very difficult to achieve. In crystallization, unwanted impurities are common in contaminating a crystallization product. This can be detrimental in the pharmaceutical industry if residual impurity compounds are greater than a threshold concentration value, as it could have unwanted biological effects. Awareness of the mechanism in which the impurity incorporates is key to understanding how to form crystals of high purity. Lattices can also have many defects and these form sites for rapid or dominant crystal growth [4]. The quality of a crystallization product is characterized by four principal attributes: chemical purity, polymorphic form, particle size, and crystal morphology [11]. Impurities can considerably increase or decrease the solubility of the solute. They can also have profound effects on other characteristics, such as crystal growth kinetics, morphology, and nucleation. Impurities are usually more effective at low supersaturation because of the favorable competition of impurity species with low concentration of solute particles for the occupation of same growth sites available. This result is a threshold supersaturation, below which no growth takes place [10]. There are five principal methods of crystal contamination during crystallization, which are based on location of the impurity: agglomeration, surface deposition, inclusions, cocrystal formation, and solid solution formation [11]. Depending on the solubility, there are different techniques to remove the impurities from the crystallized matrix. Most of the impurities can often be removed by dissolving the crystals in a small amount of fresh hot solvent and cooling the solution to produce a fresh batch of purer crystal. The solubility of the impurities must be greater that that of the main product. Re-crystallization may also have to be repeated multiple times for crystals of the required purity [9].

*Figure 5: Various Methods of Low-Level Impurity Incorporation During Crystallization [4]***

Figure 6: Procedural Sequence to Remove a Soluble Impurity [12]	Figure 7: Procedural Sequence to Remove an Insoluble Impurity [12]

Selection of Solvents

Before the crystallization process is started, choosing a solvent is needed. At the growth temperature, an ideal solvent should ensure sufficient solubility and a good temperature gradient of solubility in it. In addition, it should have low vapour pressure, low toxicity, low flammability, low-viscosity, and shouldn't corrode the growth apparatus. A rule of thumb for the selection of a solvent is have chemical similarity between the solvent and the compound to be grown since chemical similarity also defines solute solubility in the solvent. For instance, if a non-polar organic compound is used then crystals of nonpolar organic compounds can be grown easily. To ensure good crystal growth, it is important to use a solvent in which the substance is moderately soluble, medium solution supersaturation, and non-turbulent conditions. Additives can also be added to change the properties of solution to make crystal growth more favourable and leads to changes in crystal growth habits. The most common solvent used in low-temperature growth is water because more than 90% of the crystals produced by low-temperature solution methods are soluble in water. Water is also not toxic and is less volatile than organic solvents, which are generally volatile, toxic, and flammable. Moreover water is readily available in its pure state, inexpensive, and because its boiling point is higher than that of most organic solvents that are used, it provides a reasonably wide range for the selection of the growth temperature. A disadvantage of water is that it isn't a reversible solvent for some materials. It hydrolyzes some materials and introduces water of crystallization to other compounds, which may be desired in the anhydrous form or in the hydrated form of particular composition [10]. Water of crystallization is water molecules that are present inside crystals. Generally, a solvent in which the compound has a solubility of 10%-60% at a given temperature is economically suitable for crystal growth as very low and very high solute solubility provide low growth rates [10].

Crystallization Techniques

There are many steps to the many methods of crystallization. For instance, high pressure crystallization, freeze crystallization, recrystallization, crystallization from solution, crystallization from melts, fractional crystallization, crystallization from vapours, and prilling and granulation. This section will only focus on simple techniques. Which technique is chosen is mostly personal preference, however, some techniques are more suitable for some compounds.

Slow Evaporation

This is the simplest technique to grow crystals. It is only suitable for compounds that are air and moisture stable at room temperature. With a nearly saturated solution in a suitable solvent, transfer at least a couple of millimetres of the compound into a clean container, for the best results a large surface container with a cover. Do not cover the container tightly as it will hinder the evaporation process of the solvent. Aluminum foil with punched holes is proven to work very well. Set the container aside and do not disturb the container. A vial placed at an angle in a beaker can also be used. Due to the shape and narrowness of the vial, the crystals will form on the sides of the vial and will be easier to remove from the vial without damaging them [7]. Slow evaporation is easy compared to other techniques, however it requires a lot of material, can lead to too much nucleation, produces small crystals, and isn't suitable for air-sensitive compounds [3].

Slow Cooling

This is also a very simple, but successful technique that involves cooling of a saturated solution. Most substances are less soluble at lower temperatures than at higher temperatures. As the temperature decreases, the solvent's ability to dissolve the solute decreases and excess solute precipitates out. If the rate of cooling is slow enough, crystals should be produced. Heat the solvent just below its boiling point before dissolving the compound in it. Slowly cooling the solution to room temperature generally works for tested cases, however then placing the solution in a fridge or freezer produces good results. The solution must be cooled in a stepwise manner – hot → room temperature → cool – and do not place the hot solution directly into the fridge or freezer [13]. Suitable solvents are those in which the solute displays high and low solubility at high and low temperature, respectively. Fig. 8 is a variation of the slow cooling method using a Dewar flask and a water bath, and will take several days or weeks for crystals to form. This method is suitable if the solvent has a boiling point between 30-90°C. A saturated solution is heated to just below the solvent’s boiling point and transferred to a stoppered tube. The tube is transferred to a Dewar flask filled with water 2–3°C below the solvent. The Dewar flask is left in a safe and undisturbed area until the solvent has cooled and crystals have formed [13]

Vapour Diffusion

The vapour diffusion technique is a common technique to crystallize proteins. It involves the diffusion of the vapour of a volatile solvent, into the solvent containing the sample to be crystallized, in which the compound is not soluble. The vapour of the volatile solvent diffuses into the sample, decreasing the overall solubility of the product and forcing it out of the solution. The outer solvent diffuses into the inner slowly, precipitating the product, which forms the crystals. It is used only with milligram quantities of the sample. For air-stable compounds, one narrow is placed into one wide vial. The solution with the compound is added to the inner vile and the solvent in the inner vile should be less volatile than the the outer vial. Some examples of the inner vile solvent are tetrahydrofuran, benzene, methanol, or toluene. Some example of the outer vile solvent are pentane, diethyl ether, or hexane. For air and/or moisture sensitive compounds, two Schlenk flasks are connected by an obstructed glass tube with quick-fit joints at each end. The solution with the compound is prepared with standard air-sensitive methods: dry and degassed solvents. Removing and adding the flasks with inert gas before use, and always having a back pressure of inert gas when removing the stopper. The more volatile solvent is also prepared with standard air-sensitive methods, in which the compound is not soluble, is placed in the second flask and diffuses into the first flask [14].

Liquid/Liquid Diffusion or Layering

This technique involves the slow diffusion of one solvent into another. A solvent in which the compound is moderately soluble is used as the first solvent and placed at the bottom. The second solvent which is less dense and insoluble forms the top layer. Solvent two gradually diffuses into solvent one, creating a mixture with a lower solubility than that of pure solvent one. As solvent two increasingly diffuse into solvent one, the solubility further decreases causing the product to precipitate, which might form crystals at the liquid/liquid boundary. This technique is difficult as it requires some skill and a steady hand to add the second solvent on top of the first solvent slowly and carefully so that the liquid/liquid boundary is clear and distinct. This will ensure diffusion of solvent two will be slow, which in turn decreases the solubility slowly, allowing large and good crystals to grow. Narrow Schlenk flasks or NMR tubes are preferably used as it is easier to build up layers when the surface area of the boundary is small. This technique is suitable for milligram-scale crystallizations [14].

Figure 8: Slow Cooling [13]	Figure 9: Liquid/Liquid Diffusion or Layering [14]

Figure 10: Vapour Diffusion, Air-Stable [13]	Figure 11: Vapour Diffusion, Air-Sensitive [13]

Industry

Crystallization is one of the oldest unit operators. It occurs in nature and has a wide range of industrial applications. It is used to produced sodium chloride, sodium and aluminum sulfates, and sucrose which all have production rates in excess of 10⁸ tonne/year [9]. Crystallization is also vital in the freeze concentration of fruit juices, the desalination of seawater, the production of materials for the electronic industry, the recovery of materials such as metal sals from electroplating processes, and in biotechnological operations such as processing of proteins. Since crystallization can be done closer to ambient temperature and its enthalpies of crystallization are generally much lower than enthalpies of vaporization, many organic liquids are purified by crystallization rather than by distillation [9]. As stated above, there are many crystallization methods, there are also many different types of equipment that are used in the crystallization process. A multistage process is commonly used in the industry to achieve the required product purity. Further separation, melting, washing, and refining might be required in the multistage process.

Batch Crystallization

Batch crystallization has many advantages over continuous crystallization such as simplicity of equipment and reduced encrustation on heat-exchanger surfaces. In certain cases, only a batch crystalliser can produce the required crystal form, size distribution, or purity. However, the operating costs can be significantly higher than those of a comparable continuous unit, and there might be problems from the inconsistencies from each batch. One main advantage of using a batch crystallizer is the ability to clean the crystalliser thoroughly at the end of each batch, thus reducing contamination and impurities, which is especially important in the pharmaceutical industry [9]

Continuous Crystallization

Continuous, steady-state operation is not always ideal for the operation of crystallisation processes. They do however have built-in flexibility for control of temperature, supersaturation nucleation, crystal growth, and other conditions that affect size distribution of the product. The product slurry in a continuous crystallizer may have to be transported to a holding tank to allow equilibrium between the crystals and mother liquid to be reached. Also, in continuous crystallisation systems, undesired self-seeding may occur after a certain time, forcing frequent shutdowns and washouts [9].

Figure 12: Forced-Circulation Swenson Crystallizer [9]	Figure 13: Proabd Refiner, batch cooling crystallization for melts [9]	Figure 14: Swenson Draught-Tube-Baffled (DTB) Crystalliser [9]

Figure 15: Desalination of seawater by freezing [9]	Figure 16: Standard-Messo Turbulence Crystallizer [9]	Figure 17: Tsukishima Kikai (TSK), countercurrent cooling crystallization process [9]

References

Contributors:

User	Last Update

Browser not supported