Recent Advances in Engineering Aspects of Pharmaceutical Crystallization - Part 1
This article summarises recent developments from an engineering perspective in the key aspects of the crystallization process.
Crystallization, the most important separation and purification process in the pharmaceutical industry has contributed to significant improvements in building efficient manufacturing practices for production of APIs. Past few decades have seen continuous rise in research and development activities in both coming up with novel approaches for deeper understanding of the process as well combining experimental and modelling methods for more robust and unified approaches in the monitoring, control as well as design and scale-up of industrial crystallization processes. This article summarizes recent developments from an engineering perspective in the key aspects of crystallization process.
The article is in two parts, the first part, published here, covers polymorphism , co-crystallisation, kinetics, nucleation, etc. The second part of the article will be published in the forthcoming issue.
The requirements of therapeutic market place, such as consistent high quality and bioavailability of Active Pharmaceutical Ingredients (APIs) demand efficiency of crystallization process in order to meet demandsupply balance, batch-to-batch consistency and robustness in properties in all phases of drug development. As 70% of all solid products and 90 % of APIs involve a crystallization step (Gao et al., 2017), there are many technology and economic drivers for control and monitoring of crystallization processes. Almost 70 percent of new APIs being pursued are poorly soluble in water (Kawabata et al., 2011). The dissolution and disintegration rates and consequently the bioavailability in the body, particularly for low-solubility APIs, depend strongly on physical properties such as the size and shape of the crystals. The crystal size and shape distribution determines the interfacial surface area and therefore affects the dissolution rate. Different crystal structures of the same molecule, in other words ‘polymorphs’ can exhibit different solubilities. While one polymorph dissolves in the digestive system, others might not, which would hamper the therapeutic effect of a drug. All of these physiochemical properties, i.e., crystal sizes, shapes, and solid forms, can and must be controlled via the final API crystallization step (Variankaval et al., 2008). Furthermore, the crystal size distribution and shape affect the powder’s flow ability, segregation phenomena and downstream operations like filtration, blending, granulation, capsule filling, tabletting, and therefore the entire drug development / manufacturing process. In this context, the field of “crystal engineering” is increasingly receiving attention and the crystallization process plays a key role in defining the physiochemical properties of solid APIs and their final dosage forms. Added to these, strict regulatory requirements related to the variation of quality lead to high economic penalty of producing off-specification product (USD 1-2 million / batch). These call for control of crystal properties followed by efficient downstream operations (filtration, drying) and product effectiveness (tablet stability, bio-availability); quality by design, fast scale-up and product consistency. In this perspective, stand-alone as well as unified approaches presented beneath may be adopted to meet the constraints and requirements of the crystallization of APIs.
2.Thermodynamic Aspects in Crystallization 2.1 Solubility
In solution crystallization, the equilibrium between solid solute and dissolved solute in specific solvent, termed as solubility is directly impacted by the temperature and entropy of fusion, equivalent to the difference of chemical potentials between the solid and liquid of a particular compound. In addition, the solvent composition of mixed solvents, as well as other minor components or impurities can affect the activity coefficient.
In the solubility curve shown in Figure 1, if the curve is steep, i.e. the substance exhibits a strong temperature dependence of solubility (e.g. many salts and organic substances), then a cooling crystallization might be suitable. But if the metastable zone is wide (e.g. sucrose solutions), addition of seed crystal might be necessary. This can be desirable, particularly if a uniformly sized product is required. If on the other hand, the equilibrium line is relatively flat (e.g. for aqueous common salt solutions), then an evaporative process might become necessary. If the yield from either of the processes is low, then a second solvent can be added to reduce the effectiveness of the first and thus decrease the residual solution concentration (sometimes called `drowning-out’ or `watering-out’). If the solute occurs as a consequence of chemical reaction or addition of a common ion, and is relatively insoluble, then precipitation or `fast crystallization’ occurs. Typically, the driving force for crystallization, supersaturation is generated by anti-solvent addition or cooling. In batch crystallization, a crystalline product with uniform size
and shape is desirable, as the crystal size distribution (CSD) determines the efficiency of various downstream processes. Control of supersaturation, by adjusting the cooling or addition of anti-solvent will determine the final crystal properties.
Solubility measurement and its prediction have received a lot of attention in recent years. Predictive and correlative models, for example the NRTL-SAC model (Chen and Song., 2004) which is considered to be the benchmark, have been developed and applied to solve practical industrial applications. For the design of a crystallization process, initially the conditions are chosen such that all of the API or intermediate solid is dissolved. At the end, the conditions are chosen so that the majority of the batch is crystallized. Though it is desirable that impurities such as reaction byproducts or other undesired species remain soluble in the mother liquor at this point, much less effort is spent in the industry to quantify the solubility of impurities in solvents; one key reason being the amount of impurities isolated being insufficient for the solubility measurement. However, the diversity of impurities within one compound cannot be neglected. Instead, simple rules based upon other information are applied in guiding the selection of solvents, for example the retention times of impurities from the liquid chromatogram or the chemical structures of impurities if available. It should be noted that along with the solubility difference between the desired compound and impurities, other factors such as solid solution, absorption of impurity on the crystal surface also affect the rejection efficiency as well (Tung et al., 2009).
2.2 Polymorphism, Crystal Habit
The impact of polymorphism (same molecular species, with different crystal structures) goes beyond the level of crystallization as it affects the downstream formulation and drug bioavailability. The search for possible polymorphs, theoretically / empirically, or in a combination of both, is a subject undergoing extensive research and development efforts. Until recently, this search has been mostly empirical. No particular empirical methods, such as cooling, antisolvent, evaporation, or salt formation, have been demonstrated to be superior to other empirical methods.
The focus of the search for polymorphs could vary at different phases of drug development (Tung, 2013). Further, it is not unheard of that a new crystal form of a drug candidate appears in the late phase of drug development or after the drug is approved and distributed in the market (Chemburkar, 2000). After the issue with Ritonavir in 1998 (18 months after the new commercial product Ritonavir was launched, a new stable polymorph (form II) was identified in supplies of the drug, which greatly reduced Ritonavir’s solubility compared with the original crystal form, leading to an oral bioavailability problem) served as a warning to pharmacists and crystal engineers, polymorphism became increasingly important in both fundamental research and intellectual property rights. In addition to its effect on drug safety, polymorphism is an important factor in the testing of generic drugs, a huge expansion of which has occurred following the expiration of many patents of original drugs (Gao et al., 2017).
Polymorphic crystals are known to have different crystal structures. They have different X-ray diffraction patterns and Raman spectra. In addition, they should have different physical attributes, for example morphology, solid density, heat capacity, heat of melting, melting point or decomposition temperature, solubility, etc. The correlation between crystal form and morphology using the microscope could offer potential advantages over other established techniques, like Raman or powder X-ray diffraction methods, such as simplicity of measurement, a significant reduction of equipment and maintenance cost, and a better sensitivity in detecting crystals of undesired forms within the matrix of crystals of desired form.
Crystal habit (appearance of crystals), though does not reflect the internal structure of crystals, has potential impact on downstream filtration, drying and formulation. From the shape of a crystal, it is possible to infer the surface area and relative growth rates of different crystal surfaces. For needle-like crystals, the surface for crystal growth is primarily on the two tips. The surface on the needle has a much slower growth rate. For plate-like crystals, the surface for crystal growth is at the edges. The surface on the plate has a much slower growth rate. For rod-like or cube-like crystals, all crystal surfaces grow at comparable rates. Inferring the relative growth rates on different crystal surfaces has practical impact on the crystallization process development. In general, for needle-like crystals, more seed is needed since it has less surface area for crystal growth, despite the overall surface area per unit mass of crystals being high. On the other hand, for rod-like crystals, less seed is needed even though the overall surface area per unit mass of crystals is low. Various factors impacting the morphology can be crystal structure, solvents, additives, impurities, and supersaturation or desupersaturation rates during crystal growth and dissolution periods, etc. Additionally, it is shown that multiple heat/cool cycles coupling with wet mill-
ing at each cool cycle or without wet milling (Lovett et al., 2011) can be employed to modify the crystal morphology as well. This approach has been practiced over the years in the industry in modifying the crystal morphology, without modifying the solvents or crystal forms or charging additives during the crystallization due to chemical purification or bioavailability requirements. At present, industry has built a significant amount of know-how of this approach (Tung, 2013).
Crystal shape and polymorph influence solubility, dissolution rate (which influence bioavailability), compressibility (crucial for tabletting), and stability. The crystal enantiomorph is important in the manufacture of chiral materials, which has become a $100 billion industry in recent years. Table 1 indicates the effect on drug product for each specific solid state property.
Moreover, crystal size distribution, and shape have a major impact on the design of the manufacturing process since small crystals are difficult to separate from solution, and needle-like crystals or plate-like crystals can be difficult to filter and dry. These effects have been recognized as the major batch-to-batch variation issues leading to inconsistency of the final tablet properties (Shekunov and York, 2000). Hence, it is necessary to control the quality of the crystals during crystallization process to critically satisfy the performance criteria of the drug product. These solid state properties are thus termed as critical quality attributes (i.e. purity, crystal form, particle size, specific surface area, crystal morphology and habit etc.) that directly or indirectly determine the quality of the API or drug product.
Co-crystallization is a method to improve drug quality. Pharmaceutical co-crystals are multicomponent molecular systems that are typically formed through the hydrogen bonding of a co-former molecule with the API (Powell et al., 2015). The reactants are solids at ambient conditions. Excipients, amino acids, biomolecules, vitamins, minerals, and other APIs can be chosen as co-crystal formers (CCF). Salts and co-crystals are multicomponent crystals and a continuum exists linking cocrystals and salts based on the extent of proton transfer between the components. In recent
years, in order to produce poorly soluble compounds, solid forms such as co-crystals and metastable polymorphs are being developed and processes for their production are being developed and scaled up as well (Chen et al., 2011). After the first commercial product’ Entresto (Novartis)’, in 2015, there has been a surge of interest in the development of pharmaceutical cocrystals. Also, intellectual property (IP) opportunities have led to serious efforts by both innovator and generic companies created to explore, develop and patent unique crystal forms.
An effective way to anticipate the existence of multiple forms of crystals, including solvates and hydrates, is to pre-invest in experimental crystal form screening, laboratory automation and analytical methods. A summary of various recent techniques that used different systematic approaches for screening new polymorphs has been reported (Gao et al., 2017). These include development of organized solvent database with property descriptors for polymorph screening and high-throughput crystallization platforms such as CrystalMax (TransForm Pharmaceuticals, Inc.) and Crystal16™ (Avantium Technologies, Inc.) to screen the polymorphs of a given API with high efficiency. The formation and screening, transformation of a solvate carries high costs and more time which remains a challenge in solvate drug development. However, development of new forms poses challenges regarding the large-scale synthesis and stability of these drugs in the presence of excipients which need immediate attention.
2.4 Solvent Selection and Design
Solvent plays an important role in the crystallization process and its design or selection could sometimes determine the success or failure of the operation, especially to meet the demands of producing pharmaceutical products of high purity, consistent quality and high yield. The choice of the solvent is dictated by many parameters such as solvation power of the solvent, the slope of the solubility curve versus temperature, boiling point, safety and toxicity, cost, ability to participate in forming hydrogen bonding as an acceptor or a donor, and viscosity. Moreover, the use of different solvents and processing conditions during crystallization also alters the crystal habit and morphology of the purified drug affecting the product characteristics such as dissolution profile, bioavailability, flow ability, and the ease with which the crystals are compressed into tablets. Winn et al. (2000) presented a review on modeling of crystal shapes of organic materials grown from solution. It has been observed that crystals grown from a mixture of solvents have different characteristics than the crystal grown from pure solvent, and this effect is significant if the solute has very different solubility in each solvent. Frank et al. (1999) reviewed strategies for solvent selection methods used to estimate the solubility of organic solids in a wide variety of solvents for various types of crystallization processes such as cooling crystallization and drowning out crystallization, where the task of solvent selection is carried out from a list of good solvents from a database with solubility being the only criterion for selection. This approach may miss some better solvents or may require too extensive experimentation to arrive at the desired solvent. Therefore, computer aided molecular design (CAMD) methods for solvent design have been attempted as advancement in this direction. Harini et al. (2013) have provided a comprehensive review on the methodologies available for the prediction of product (solvent) properties from the molecular structure and property estimation techniques, along with the various optimization approaches and computational schemes to solve the solvent design problem for crystallization and solvent extraction in pharmaceutical industry. The properties of interest for crystallization process in a pharmaceutical industry include melting point, boiling point and dielectric constant, solubility (Hildebrand solubility parameter, Hansen solubility parameter), viscosity, heat of fusion, critical temperature, enthalpy of vaporization, flash point, polarity of organic solvents, hydrogen bonding propensity or interaction parameter, the octanol−water partition coefficient, the donor/acceptor numbers, the solvatochromic parameters, and other environmentalrelated parameters. It has been observed that hydrogen bonding tendency can affect crystal morphology or habit. Ibuprofen crystals crystallized from solvents with high hydrogen-bonding ability were plate-like, with a low aspect ratio and large size. On the other hand, ibuprofen crystallized from solvents with low hydrogen bonding ability were needle-like crystals with high aspect ratio (Karunanithi et al., 2007). It has been found and verified that ibuprofen crystals formed from 2-ethoxy ethyl acetate as the solvent are significantly larger and have a low aspect ratio, when compared to crystals formed from the solvent n-hexane, by the combined approach of experiments, database search, and CAMD.
As supersaturation is the driving force for crystal nucleation and growth, ultimately dictating final crystal size distribution, the relationship is defined by a
well known set of equations outlined by Nyvlt (1968). Figure 2 relates supersaturation to nucleation, growth and crystal size. The value of the growth order is typically between 1 and 2, while the value of nucleation order is typically between 5 and 10. At low supersaturation, crystals can grow faster than they nucleate, resulting in larger crystal size distribution. However, at higher supersaturation, crystal nucleation dominates crystal growth, ultimately resulting in smaller crystals. Modern techniques such as Attenuated total reflection (ATR) - Fourier transform infrared (FTIR) spectroscopy allow solubility traces to be developed quickly and easily, and the prevailing level of supersaturation to be monitored continuously throughout a crystallization experiment. The key contributors influencing the kinetics of crystallization (in other words, the size distribution) are Nucleation / Growth kinetics (governed by supersaturation) and Seed.
3.1 Nucleation and Growth, Crystal Size Distribution
Supersaturation affects both crystal growth and nucleation rates, which in turn impact the crystal size distribution. A higher level of nucleation leads to smaller crystals and vice versa. Also, a high degree of nucleation rate over crystal growth rate due to a high degree of supersaturation can lead to poorer rejection of impurities. Given these considerations, control of supersaturation, coupled with the utilization of proper seed, to maximize crystal growth and minimize nucleation is generally preferred. It should be pointed out that, if a higher degree of nucleation is created by a higher level of mixing intensity instead of supersaturation alone, it does not necessarily affect the product purity. Common nucleation situations from solution may be seen in Figure 3.
Crystal growth, on the other hand, is the addition of more solute molecules to the nucleation site or crystal lattice to evolution of macroscopic crystal form of defined size and shape. In other words, crystal size distribution and morphologies produced are a result of the relative rates of reaction of nucleation and crystal growth. Crystal growth is considered to be a reverse dissolution process and certain diffusion theories consider that matter is deposited continuously on a crystal face at a rate proportional to the difference of concentration between the surface and the bulk solution.
To gain good control of supersaturation, some quantitative measures of crystal growth and nucleation rate constants are needed. Due to the diversified nature of API (or intermediate) crystals, nucleation and crystal growth rates can vary drastically over several orders of magnitude. At the current time, it is not possible to quantitatively predict the crystallization behavior on the basis of theoretical models. To address this limitation, a model-based experimental design (MBED) methodology for crystallization was reported (Davey et al., 2002). However, limitations reside in the validity of the population balance equation primarily used for predicting the crystal size distribution. Several solution methods for solving the Population Balance Equations, such as Methods of Moments, Method of Classes, Orthogonal collocation methods etc., reported have been fairly in practice over the past two decades in understanding the crystal size distributions (Cameron et al., 2005). Several approaches to understand crystal size distribution available are summarized in Table 3 (Also shown are the different techniques / methods available for crystallization studies). An interesting interplay between thermodynamic, kinetic and molecular recognition phenomena that governs crystallization is reported, which acts as a good aid in inter-linking these key aspects towards monitoring and control as far as application of crystallization phenomenon in the
industry is concerned. The same is shown in Table 2. 3.2 Seed
Seeding a solution with product crystals is a very well-established method to induce crystallization. Seeding impacts and interferes with all aspects of crystallization including crystal size, size distribution, crystal form, nucleation & growth rates, yield and product purity. Seeding is important for systems that are very difficult to nucleate or tend to induce liquid-phase separation, because seeding reduces nucleation time that may be otherwise too long from an economic perspective. Addition of small seeds to a supersaturated solution can increase the nucleation rate and also can selectively nucleate a desired crystal form (especially a polymorph), provided the product seed crystals are available. Seeding with a desired crystal form can also reduce the risk of formation of undesired crystal forms. A detailed summary on seeding approaches for polymorph control has been presented by Linas and Goodman (2008).
In engineering perspective, the purpose of seeding a supersaturated solution is to provide starting surface area for crystal growth and avoid or reduce nucleation as much as possible. Seed loading (Weight % of seeds), seed size, time of seed addition and addition point of seed are the critical quantitative parameters for a seed- ing policy to attain crystal quality attributes.
Seeding and Metastable zone: Most of the API’s are produced through batch cooling crystallization as they are very much temperature sensitive. For seeding, it is necessary to understand the width of metastable zone (MSZ) and a range of operating conditions over which a solution can be supersaturated. The point in the MSZ where the seed is added (seeding temperature) also has a big impact on the relative rates of nucleation and growth. Seeding close to solubility curve (low supersaturation) results in slow nucleation and less number of fine particles. Whereas, close to MSZ (high supersaturation) results in high nucleation rates and thus, more number of fine particles. Likewise, Seeding close to MSZ results in less crystal growth, whereas close to solubility curve produces more number of large particles. Seeding temperature can be manipulated to have a nucleation or growth dominated process that can in turn influence the product size or size distribution.
Seed loading and seed size: For a growth dominated crystallization process, where nucleation is minimized, final particle size can be predicted or controlled based on amount of seeds, seed size and amount of solid grown on the seed crystals. There has been a lot of discussion in literature about seed loading. Kubota et al. (2001) defined seed loading that suppresses nucleation
effectively as ‘critical seed loading’ as a function of seed size to achieve unimodal distribution of final product crystals. There have been evidences about the effect of concentration of solute and temperature of seeding on the yield of crystallization. A seeding strategy with a combination of seeding technique and manipulating the profile of supersaturation-generating variables increases the batch consistency. This has been demonstrated recently through rigorous simulations using multiobjective optimization approach (Hemalatha & Rani, 2017). In their work, optimal seeding policy in terms of seed mass and seed size has been determined for producing optimal crystal size and distribution of the final product. Seed properties have been included as optimized variables along with operating profile in the formulation of optimization problem. Various combinations of seed mass and size provide options for choice of an operating trajectory (in this case temperature) to attain desired product properties with objectives being maximization of mean crystal size, minimization of coefficient of variation and minimization of nucleated crystals. Also it has been proven that addition of seeds decreases the supersaturation and minimizes nucleation. Several studies report seed loading and size through modelling and optimization (Chung et al., 2001; Choong and Smith, 2004; and Sarkar et al., 2005).
Seeding techniques- External seeding: There are different approaches for generation of seed crystals, either dry or in slurry (wet) form. For the dry approach, a batch of solid API (or intermediate) is dry-milled. A portion of the dry milled API is retained as a fresh seed for the next batch while the bulk is sent for downstream processing. Based on the product particle specification, different equipment are used for dry milling (pin mill and jet mill). As the dry milling operations consume less energy for process development, they are used in the early phase of process development. However, there are many disadvantages associated with it. Some crystals may get dissolved or get struck to the surface of the mill which may interact with the metal surface. Some particles may lose their crystallinity (Tung, 2013). In turn, addition of dry seeds may induce unwanted solvent entrapment and agglomeration, during crystallization process as the dry seeds might form aggregates during storage. During dry milling, the surface of the solids might get ruptured and can form various active and non-uniform edges which form non-uniform growth sites for the product crystal that causes product crystals to grow irregularly. To avoid this ‘Healing’ process for the seed crystals can be done where the seed crystals are kept in a saturated slurry for many hours where the sharp edges get dissolved due to Oswald ripening. Another simple procedure employed often is ‘Isothermal hold’ where the seed crystals are added to the supersaturated solution and kept at initial temperature for some time where dissolution takes place and crystal growth dislocation can be avoided.
Adding the seed in slurry form (either in anti-solvent or mother liquor) may help the seeds disperse very well to their actual crystal size. Hence, wet milling is a preferred operation for later phase in drug development. Wet seeding offers advantages of controlled seed crystal form, tunable seed size and amount, and robustness of scale up. Wet seed can be generated through wet milling the API in a slurry that is saturated with the crystallizing solute. Based on the size requirement of the seeds, different wet milling options are available. Media mill is used for generating submicrometer particles, ultrasound device or rotor/stator homogenizer for producing micrometer particles.
Seeding techniques- Internal seeding: Some of the concerns in external seeding can be addressed by creating the seed particles internally in the process which is known as insitu seed generation approach. In this approach, an API is dissolved in a solvent mixture and fresh solid are generated under conditions of supersaturation with controlled mixing. Rapid precipitation using impinging jet device to generate fine particles also falls under this category.
Advanced seeding techniques have been investigated as possible alternatives to these traditional approaches with advances in particle measurement techniques like Focussed beam reflectance measurement (FBRM). An in situ particle seeding approach, which consists of an automated closed loop feedback control technique using FBRM for unseeded cooling crystallizations, to produce seeds during the process for ensuring consistent and repeatable crystal product quality was reported (Chew et al., 2007). Improvement in the final product particle dis¬tribution and consistency is observed for Glycine-water and Paracetamol-water systems when this feedback control approach is used compared to conventional cooling crystallization approach. For this approach, prediction of model parameters is necessary which may not be possible or may be a tedious job for some API’s. In such cases, ‘Direct nucleation control’ (DNC) seems promising as it is a model free control approach for insitu seed generation (Abu Bakar et al., 2009). DNC is a modification of the feedback control approach. In this approach, the seed
count is measured online through FBRM in the control region and the nucleation and dissolution is controlled through a feedback control strategy. The advantage of this technique is that it does not use pre-determined heating or cooling profiles. The temperature profiles are generated automatically and continuously during the crystallization process in response to the number of particles generated by the nucleation events. No prior knowledge of the model, process kinetics or the metastable zone width is necessary as this approach will automatically determine the optimal operating profile by continuously detecting the metastable zone limit in real time using a feedback control strategy. DNC has applications in fines removal also apart from continuous seed generation as this approach reduces effects of breakage and produces uniform crystals.
These in situ approaches obviate general testing needed for storage and stability of the seeds. In addition, it is very flexible over other seed generation methods as seed is generated on demand which can avoid excess seed amount. But, control of crystal forms during seed generation is another concern. The metastable forms, amorphous solids, or oil droplets can form initially. If the stable crystal form is desired for the seed, an additional aging with intensive mixing to facilitate the conversion of metastable crystal forms to the stable crystal form is a good approach. Addition of dry/wet seed of the stable crystal form prior to the generation of in situ seed is another simple option.
Crystallization is one of the key process steps for manufacturing a large number of products in fine and specialty chemicals sector (including API). Product purity and other desired product attributes (like size distribution, habit, morphology and polymorphs) are often crucial. The article will continue into the second part to be published in the forthcoming issue. References:
1. Abu Bakar, M. R.; Nagy, Z. K.; Saleemi, A. N.; Rielly, C. D., 2009, Cryst. Growth Des., 9(3), 1378-1384.
2. Cameron, I. T.; Wang, F. Y.; Immanuel, C. D., Stepanek, F., Chem. Eng. Sci., 2005, 60(14) , 3723-3750.
3. Chen, C.; Song,Y., Ind. Eng. Chem. Res., 2004, 43, 8354−8362.
4. Chen, J.; Sarma, B.; Evans, J. M.; Myerson, A. S, Cryst. Growth. & Des, 2011, 11(4), 887-895.
5. Chew, J. W.; Chow, P. S.; Tan, R. B. , Cryst. Growth Des., 2007, 7(8), 1416-1422.
6. Choong, K. L.; Smith, R., Chem. Eng. Sci., 2004, 59 (2), 313−327.
7. Chung, S. H.; Ma, D. L.; Braatz, R. D., Can. J. Chem. Eng., 1999, 77 (3), 590−596.
8. Davey, R. J.; Blagden, N.; Righini, S.; Alison, H.; Ferrari, E. S. Phys. Chem. B 2002, 106, 1954−1959.
9. Doki, N.; Seki, H.; Takano, K.; Asatani, H.; Yokota, M.; Kubota, N, Cryst. Growth Des., 2004, 4(5), 949-953.
10. Frank, M.; Höland, W.; Rheinberger, V.; J. non-crystalline solids, 1999, 253 (1-3), 170-177.
11. Gao, Z.; Rohani, S.; Gong, J.; Wang, J., Engineering, 2017, 3(3), 343-353.
12. Harini, M.; Adhikari, J.; Rani, K. Y. , Ind. & Engg. Chem. Res., 2013, 52(21), 6869-6893.
13. Hemalatha, K. , Rani, K.Y. , Ind. Eng. Chem. Res., 2017, 56, 6012–6021.
14. Karunanithi, A. T.; Acquah, C.; Achenie, L. E.; Sithambaram, S.; Suib, S. L.; Gani, R., Chem. Eng. Sci., 2007, 62(12), 3276-3281.
15. Kawabata, Y.; Wada, K.; Nakatani, M.; Yamada, S.; Onoue, S., Int. J. Pharm., 2011, 420(1),1–10.
16. Llinas, A; Goodman J. M., Drug Discov Today, 2008, 13 (5–6), 198–210.
17. Lovett, M. A.; Muratore, M.; Doherty, M. AIChE J. 2011, 58, 1465−1474.
18. Nývlt, J. , J. Cryst. Growth,, 1968 , 3, 377-383.
19. Powell, K.A.; Bartolini, G.; Wittering, K. E.; Saleemi, A.N; Wilson, C.C; Rielly, C. D, Cryst. Growth Des., 2015, 15(10), 4821–36.
20. Rodriguez-Spong B.; Price, C.P.; Jayasankar A., Adv Drug Deliv Rev, 2004, 56, 241-274.
21. Sarkar, D.; Rohani, S.; Jutan, A. Chem. Eng. Sci., 2006, 61 (16), 5282−5295.
22. Shekunov, B.; York, P. J. Cryst. Growth, 2000, 211, 122-36.
23. Tung, H. H., Org. Proc. Res. & Dev., 2013, 17(3) 445-454.
24. Tung, H.-H.; Paul, E.; Midler, M.; McCauley, J.; Wiley: Hoboken, NJ, 2009.
25. Variankaval, N.; Cote A. S.; AIChE J , 2008, 54 (7), 16821688.
26. Winn, D.; Doherty, M. F., AIChE J, 2000, 46( 7), 13481367.
Figure 1: Solubility curve (Adapted from Jones, 2002)
Table 1: Solid-state properties defined by crystallization process and their relationship with specific characteristics of drug substances and drug products (Adapted from Shekunov., & York., 2000)
Figure 2: Relationship between supersaturation, nucleation, growth and crystal size (adapted from www.mt.com)
Figure 3: Common nucleation situations from solution (Adapted from Crowder et al., 2003)
Table 2: Interplay between thermodynamic, kinetic and molecular recognition phenomena in crystallization (Adapted from Rodriguez-Spong et al., 2004)