The impact of crystal habit on the pharmaceutical properties of active pharmaceutical ingredients

Abstract

Crystal habit modification potentially improves the pharmaceutical and biopharmaceutical properties related to active pharmaceutical ingredients (APIs) and is an important aspect of crystal engineering. It has been demonstrated that filterability, compaction properties, flow behavior, and dissolution performance of APIs are dependent on the crystal habit of the compound, which ultimately depends upon the combination of factors including but not limited to the nature of solvent, additives, supersaturation, the environment provided during crystallization, etc. Crystal habit modification may be considered as an economically viable approach to mitigate pharmaceutical manufacturing challenges. This tutorial provides a step-by-step approach to crystal habit modification using different crystallization methods. Also, the factors that affect crystallization experiments at each stage of the different methods are explained in detail. The effects of supersaturation, solute–solvent interactions, additives, rate of crystallization, etc., are described, with examples. The impact of crystal habits on pharmaceutical properties such as filtration, punch sticking, compressibility, dissolution rate, etc., is also explained in detail with case studies. Finally, the analytical methods that are useful in the characterization of crystalline materials (morphological, physicochemical, rheological, and surface characterization) are explained, along with the experimental procedures involved.

---

Niranj H. Ram

Niranj H. Ram is an Integrated Dual Degree student in the PSSRL, IIT(BHU), Varanasi, India.

Madhukiran R. Dhondale

Madhukiran R. Dhondale is a PMRF (Prime Minister Research Fellow) in the PSSRL, IIT(BHU), Varanasi, India.

Maan Singh

Maan Singh is a Ph.D. research scholar in the PSSRL, IIT(BHU), Varanasi, India.

Brahmeshwar Mishra

Prof. Brahmeshwar Mishra is a Professor of Pharmaceutics at IIT(BHU), Varanasi, India. He has more than 40 years of research and teaching experience. Prof. Mishra has published more than 250 research & reviews articles and has successfully guided more than thirty Ph.D. students.

Ashish K. Agrawal

Dr. Ashish K. Agrawal is an Associate Professor at IIT(BHU), Varanasi, India. Currently, he leads a research group focusing on pharmaceutical nanotechnology and controlled drug delivery by using nanotechnology-based approaches.

Anne Marie Healy

Prof. Anne Marie Healy is a Professor of Pharmaceutics and Pharmaceutical Technology at Trinity College Dublin, Ireland. She has more than 140 international publications and graduated 26 Ph.D. students and mentored 29 postdoctoral researchers. She is an editor of the International Journal of Pharmaceutics.

Dr. Dinesh Kumar

Dr. Dinesh Kumar is an Assistant Professor at IIT (BHU), Varanasi, India. He leads the Pharmaceutical Solid-State Research Laboratory (PSSRL), where the research is primarily focused on crystal engineering of APIs and solid-state characterization to improve API performance.

---

Key learning points

• Understanding API crystal habit: gain insights about crystal habit, about different types of habits, changes in habit during different stages of crystallization and its critical role in determining API performance.

• Importance of crystal habit modification: learn the importance of crystal habit modification for APIs and the strategic approaches for improving the drug's processability and performance.

• Factors affecting crystal habit growth: know the intrinsic and extrinsic factors such as temperature, agitation, and understand how changes in these factors leads to changes in API crystal habit and decide the final habit of the crystal.

• Impact on pharmaceutical properties: what is the magnitude of effect of crystal habit on properties of the API like dissolution, flowability, filtration.

• Post-modification characterization techniques: understand the different techniques to characterize morphology, to analyse dissolution rates, powder flow, surface characteristics and compaction behaviour of the API after crystal habit modification.

Introduction

Recrystallization from solutions using solvents is the most employed crystallization method in the pharmaceutical and chemical industry. However, during the process, the crystallizing compound may assume different shapes depending upon factors such as solute–solvent interactions, solution properties, crystallization parameters, and the environmental conditions under which the crystallization is being carried out. The geometrical shape of the crystal or its outer appearance is commonly termed as the “Crystal Habit”.¹ One of the important parameters for defining crystal habits is the aspect ratio which is defined as the ratio of the length of the crystal to the width of the same crystal. The crystal habit of a compound is dependent upon the crystal facets and their relative growth rates during crystallization from solution. Accordingly, crystal habits like rod-shaped, plate-like, needle-like, acicular, and many more with different aspect ratios are formed during crystallization.² In solid-state pharmaceutics, crystal habit is of extensive importance since it has a crucial impact in determining the physicochemical properties of the active pharmaceutical ingredients (APIs) such as filterability, flowability, compaction, dissolution rate, and so on.³ Further, sufficient care must be taken to design the crystallization process such that only the habit is modified without any changes in the polymorphic form of the API.

Depending upon the solute–solvent interactions, rate of crystallization, supersaturation etc., the crystals may exhibit rapid or stunted growth in a particular direction. For instance, the rapid growth of prismatic type crystals can lead to its elongation thereby forming needle-like crystals. On the other hand, the stunted growth gives rise to flat-shaped crystals such as plates and tabular crystals.⁴ It is evident that the unit cell of crystals also influences its habit/shape. For instance, block-shaped crystals are formed from cubic unit cell type and unit cells with two or single short axis tend to form needle/plate-shaped crystals, respectively.⁵ Further, the unit cell parameters such as the edges (a, b, & c), and angles between them (α, β, & γ) area useful parameters to differentiate between changes in polymorphism or crystal habit. Any changes in the unit cell parameters during habit modification experiments is undesired and must be controlled.

The commonly encountered crystal habits in the pharmaceutical manufacturing industry, along with their pros and cons in the context of downstream processability (post crystallization), are provided in Table 1 and are pictorially represented in Fig. 1.

Table 1 Crystal habits and their advantages and disadvantages in terms of downstream processing

Crystal habit	Aspect ratio	Advantages	Disadvantages
Acicular/needle-like^6,87	>10 (Since length is much greater than width and thickness)	• Good compressibility profile	• Prone to aggregation and breakage
			• Poor powder flow
			• Causes considerable filterability issues
			• Difficult to disintegrate compacts
Rods^7,57	3–10 (Not as long as needles but still length > width and thickness)	• Good compressibility profile	• Poor powder flow
			• Prone to breakage
			• Causes filterability issues
Spherical^8–10,94,99	∼1 (Dimensions are almost equal in all directions)	• Best powder flow properties	• Lowest surface area per volume of crystals
		• Does not cause issues during filtration	• Moderate compressibility profile
Blocks^11,12	Between 1–1.5 (Dimensions are almost similar)	• Better physical stability and plasticity than plates	• Intermediate powder flowability
Plates^12,13	2–3 (Length is greater than width, but thickness is smaller than width)	• Exhibit good compaction properties and physical stability	• Poor powder flow properties
Fibrous^14,53	>10	• Good compressibility profile	• They tend to agglomerate and cause breakage
			• Poor powder flow properties

---

Fig. 1 Scanning electron microscopy images of some commonly encountered crystal habits A) needles, B) plates, C) tabular, D) rods, E) rhombohedral/cubic, F) aggregated blades, G) dendritic crystals. Reprinted (adapted) with permission from ref. 15–19. Copyright American Chemical Society.

The description of the crystal habits are as follows:

• Acicular: slender, long and pointed crystals. e.g. needle-shaped crystals.

• Cubic: crystals with similar length, width, and thickness.

• Plate: flat crystals with similar length and width.

• Tabular: crystals thicker than a plate but with similar length and width.

• Blade: long, thin flat crystals.

• Dendritic: tree-like pattern formed by tiny crystallites.

• Rod: cylindrical crystal elongated along one-axis.

• Prismatic: hexagonal crystals having width and thickness great than acicular crystals but with smaller length. The facets are parallel to the growth axis.

These downstream processability aspects (filterability, flowability, compaction, etc.) are of great importance in the case of oral solid dosage form manufacturing, especially for tablets. Tablet manufacturing involves granulation, drying, compaction, etc., which requires the API under consideration to possess optimum stability and compaction properties to enable tableting. The compaction properties of the API are defined by compressibility and compactability, and together they determine the ease with which tableting can be undertaken. Powder compressibility is the measure of how readily the material undergoes a change in volume when compressed and is characterized by both tablet porosity and compression stress, while compactability relates the tablet tensile strength to tablet porosity.²⁰ Similarly, during API synthesis and purification, filtration is an important step in isolating crystals from the mother liquor and the filtration efficiency depends on the size and morphology of the crystals. The relationship between filterability and the crystal habit has been extensively studied, and it is generally the case that needle-like crystals have poor filterability.²¹ Powders possessing poor flowability often pose issues with respect to weight variation and content uniformity during tablet compression and capsule filling. The flowability of crystals with habits closer to spherical or rhombohedral tend to be higher, while that of needle-like crystals tends to be very poor.^22,23

Similarly, dissolution rate is an important aspect during tablet development. The dissolution rate of an API and thus, of the tablet is significantly influenced by the changes in the crystal habit, due to anisotropy of the crystal facets. Modification of crystal habit to increase the surface area of a particular functional group bearing facet is also a feasible solution to improve dissolution rate of APIs.^24–26

Various researchers have reported the advantages of different crystal habit modification techniques to improve the processability or dissolution performance of APIs. Through thorough knowledge of the factors affecting the crystallization process, crystallization scientists can control and alter the crystal habits of compounds. A common practice to modify crystal habits involves changing the solvent system (choosing different polarity solvents), adding crystal growth inhibitors (additives), changing the supersaturation extent, or using ultrasound during crystallization. In addition, changing the method of crystallization (anti-solvent crystallization, cooling crystallization, and evaporative crystallization) can also result in modification of the crystal habit.

The present tutorial outlines the procedures that can be followed along with the process parameters that can affect the crystal habit during crystallization. Also, characterization methods, along with their associated procedures and requirements, are summarized and tabulated.

Crystal habit modification

Crystal habit has an immense impact on the processability and manufacturing process of crystalline pharmaceutical solids. At times, active ingredients may not have the required physical characteristics, and the existing properties must be modified to improve the quality of the product. Precise control over the crystal growth allows researchers to crystallize materials with tailored structures and properties.²⁷ Several factors will determine the resultant habit of the crystals. The factors are summarized in Fig. 2 and explained in detail in Table 2. Fig. 3 depicts the experimental trial-and-error based experimental protocol for modifying the habit of pharmaceutical substances. Alternatively, molecular dynamics studies may be employed to screen solvents and additives for modification of crystal habits. The simulation-based software uses various models such as the Bravais–Friedel–Donnay–Harker (BFDH) model, attachment energy and modified attachment energy models, etc. to compute the habit of crystals grown under different environments (due to change of solvents, presence of additives, etc.). These simulation-based studies can be conducted using software packages such as Mercury by Cambridge Crystallographic Data Centre (CCDC), BIOVIA Materials Studio, etc. Such simulation-based predictions of crystal habit are out of the scope of the current tutorial. However, the interested readers are directed towards the works of Clercq et al.,²⁸ Gu et al.,²⁹ Kumar et al.³⁰ for in-depth understanding of the in-silico crystal habit prediction methods. The theoretical aspects of the molecular modelling is explained in the chapter by Chadwick et al.³¹

Fig. 2 Factors impacting the crystal habit of compounds.

Table 2 Factors influencing crystal habits during different steps of crystallization

Steps	Factors that influence the shape/habit of a crystal
Preparation of saturated solution of API in a suitable solvent^32,33,40	• The degree of supersaturation affects the crystal habit of APIs. Nucleation kinetics can be controlled by altering the super-saturation levels. Different habits are formed at different levels of supersaturation ○ High degree of supersaturation: Leads to the formation of smaller crystals of a smaller size ○ Low degree of supersaturation: leads to the formation of a small number of large size crystals
	• Choice of solvent: depending upon the chemistry of the solute, polarity, and hydrogen bonding capacity of the solvent, there is a differential interaction between the solvent and different crystal facets. This difference in the interaction at the solute–solvent interface leads to changes in the habit of a crystal. Hence, one can expect a different crystal habit by selecting different polarity solvents for a compound
	• Use of seed crystals: using seed crystals prevents uncontrolled nucleation, enabling us to obtain crystals of desired habit and size (similar to that of seed crystals used)
Crystallization by:	• Additives: surfactants, impurities, and polymers at low concentration can selectively adsorb onto a particular crystal facet or modify the interfacial energy, thereby altering its growth rate and resulting in a different habit
Cooling	• Control of cooling/antisolvent addition/evaporation: control of the solvent's cooling/antisolvent addition rate/rate of evaporation can affect the crystal growth kinetics and lead to changes in the habit of crystals
Anti-solvent addition	• Ultrasound: the application of high-energy ultrasound can induce stress, influence the shape of primary crystals, and modify the crystal habit
Evaporation	• Degree of agitation: it is impossible to change the habit per se, but small size and uniform size distribution of crystals can be obtained by controlling the agitation rate
Additive mediated^34–37,40	• Impinging jet crystallization: similar to the effects of agitation, the use of the jet impinging crystallization method can produce crystals of smaller size and uniform size distribution
Filtration/separation of crystals^87,89	• Attrition during filtration (such as when using a glass filter dryer) must be avoided, as it leads to crystal breakage leading to changes in the aspect ratio of crystals after filtration process (e.g., breakage of needle-shaped crystals into smaller fragments)
	• Use of pressure during filtration can cause the formation of a cake and breakage of crystals under its own weight
Washing^38	• During washing with anti-solvent care must be taken to ensure the residual solvent does not contain dissolved API, which otherwise changes the shape of crystals (often forms irregular crystals)
	• Washing is a critical step in additive-mediated crystallization, wherein the removal of additives (polymers, surfactants, etc.) is carried out. So, one must ensure that the solvent selected can dissolve the additive/impurity while the API is not soluble in the washing solvent. Also, it must be miscible with the solvent used and must be sufficiently volatile to ensure its easy removal (during drying) after washing
	• Sometimes, using a mixture of solvents is preferable to avoid API precipitation during the solvent-washing phase
Drying^39,87	• The drying step ensures that the crystallized API is free of any residual solvent that otherwise might cause harm to patients. Drying must be carried out in conditions that ensure the product stability and should not cause degradation of the API under consideration
	• Drying at higher temperatures can induce solid-state transformations in APIs
	• The rate of drying has an impact on particle size distribution of the crystals. Low drying rate for a sample with high capillary force strength can promote solid bridge formation leading to lumping

---

Fig. 3 Procedure for crystal habit modification by different crystallization methods.

Factors influencing the crystal habit

As mentioned in Table 2, various factors play an important role in determining the crystal habit of a compound during the crystallization process. Understanding and achieving precise control over these factors can result in the formation of desired habits and improved characteristics. In this section, the factors influencing crystal habits are briefly discussed. However, for detailed information, the readers are directed towards the review work by Pu and Hadinoto.⁴⁰

Nature of solvent. The solvent's role is paramount in deciding the crystal growth kinetics of different crystal facets. The API should be soluble in the selected crystallizing solvent to attain the required super-saturation levels. Modification of the strength of solute-crystallizing solvent interactions can affect the crystal habit. The dynamics of crystal growth may change due to these interactions, and crystal growth may be enhanced or inhibited.⁴¹ The growth rate of crystal facets with stronger interactions with the solvent will be slower than the other crystal facets.⁴² However, in some instances, the solvent molecule gets entrapped in the lattice of the crystallizing compound, resulting in the formation of solvates (pseudo polymorphs). Solvates tend to exhibit completely different physicochemical properties than those of the non-solvated crystal form. Hence, initial screening must be carried out to rule out polymorphic changes in selected solvents.

The polarity of solvents⁴³ and hydrogen bonding between solute and solvent⁴⁴ can also impact the shape of the crystals formed. The molecular composition at each facet of a crystal is different. For example, oxygen and nitrogen are polar (hydrophilic) in nature, while carbon and halogens are non-polar (hydrophobic). Polar solvents such as water, ethanol, methanol, acetonitrile, etc. interact more with the polar surfaces, while non-polar solvents such as dichloromethane, chloroform, n-hexane, etc. interact with the non-polar surfaces. Hence, different crystal habits are formed if different polarity solvents are used to crystallize a particular compound. However, the solubility of the compound in the selected solvent must be ensured prior to crystallization. The use of a mixture of solvents (in definite volumetric ratios) will also yield crystals with modified habits. For example, lovastatin gives prismatic-type crystals in hexane and methyl-cyclohexane (non-polar solvents), while in the water-acetone mixture (polar solvent system), it forms needle-shape crystals.^45,46

Degree of supersaturation. Super-saturation is the driving force for crystallization and has an influence on the nucleation and crystal growth steps. The super-saturation level is the ratio of solute dissolved in the solution to its equilibrium solubility. It is possible to exceed the saturation solubility by heating the solvent to a higher temperature. At this stage, the solubility of compounds in a selected solvent at higher temperatures is higher than at lower temperatures. For example, a compound's solubility in water is 100 mg mL⁻¹ at 25 °C, but at 75 °C, 500 mg can be dissolved in 1 mL of water. A solution containing API in a higher amount than its saturation solubility is termed a supersaturated solution. The super-saturation level can be controlled by varying the solute concentration, solution temperature, and ratio of solvent to anti-solvent.

High supersaturation levels lead to the formation of a higher number of nuclei in a solution. Hence, the driving force (i.e., super-saturation) is consumed in the formation of the nuclei, and less solute is available for the crystal growth phase. As a result, a large number of smaller crystals are formed at a high degree of supersaturation. In contrast, at lower super-saturation levels, relatively fewer nuclei are formed, and the crystals grow to a larger size. Also, a high degree of supersaturation leads to the formation of isotropic crystals due to higher crystallization rates and less facet discrimination. For example, the aspect ratio of benzoic acid crystals at a super-saturation level of 1.029 was ∼20.5, while at a super-saturation level of 2.941, the aspect ratio decreased to 4.9, with the shape-changing from needle to elongated rectangular plates at low and high supersaturation levels, respectively.⁴⁷ Furthermore, the number of kink sites (sites where the growth takes place in a crystal) is also influenced by the super-saturation level.⁴⁸

Rate of cooling/solvent evaporation/antisolvent addition. Crystallization can be primarily carried out by three methods, viz., cooling method, solvent evaporation method, and antisolvent method. The rate of change of these parameters (i.e., cooling, evaporation, and antisolvent addition) can be controlled to obtain different crystal habits. For example, during cooling crystallization, rapid cooling/evaporation/antisolvent addition can form long crystals with a high aspect ratio (needles/rods). This is observed because the elongated crystals lose heat more effectively, whereas controlled cooling/evaporation/antisolvent addition can result in the formation of lower aspect ratio crystals (plates, orthorhombic). As the rate of cooling/evaporation/antisolvent addition decreases, the crystal growth rate also decreases, forming symmetric crystals. Hence, it is advised to carry out crystallization slowly under controlled conditions to avoid the formation of needle/rod-shaped crystals.⁴⁰

More precise control can be exercised by combining two or three crystallization methods. Crystallization can be carried out using both cooling and antisolvent addition methods. A combination of different methods can also yield different crystal habits compared to a single method.⁴⁹ Also, during the cooling crystallization method, a few instances have been reported wherein temperature cycling was performed, resulting in the formation and dissolution of crystals to modify the growth facets. During the re-dissolution of crystals, there is a chance for the formation of new growth facets on the crystal, thereby modifying the final crystal shape.⁵⁰

Presence of external stimuli (agitation/ultrasonication) during crystallization. Agitation of the solution is undertaken to create a mechanical scouring action. In simpler words, it is the mixing of the solution to displace the particles rapidly. When a supersaturated solution is agitated, the particles collide amongst themselves vigorously, which results in the formation of nuclei that further develop into crystals. Studies have found that agitation influences the metastable zone (region of the phase diagram where no nucleation occurs) and alters the metastable zone width (MSZW). It was discovered that raising the agitation rate for some aqueous salts caused a rise in the nucleation temperature. Initial increase in the stirring speed led to a reduction in the nucleation temperature. Further increase in the stirring speed improved nucleation temperature by causing an increase in the nucleation temperature. This mechanism is quite intricate and involves enhanced mass transfer. The relationship between stirring speed and MSZW was further investigated by looking at the primary nucleation kinetics of the aqueous solution. It was found that the nucleation rate constant was directly proportional to stirrer speed but the nucleation rate remained independent of agitation rate, i.e., the nucleation order was dependent only on the number of molecules required to form a critical nucleus.⁵⁰

Generally, high agitation results in the formation of smaller, finer crystals, and moderate to low levels result in bigger crystals (depicted in Fig. 4). However, if the stirring is not rapid enough, it may lead to the accumulation of particles. Particle agglomeration caused the typical shape of the facets of paracetamol crystals to disappear at moderate stirring speeds, and well-defined crystals formed when the agitation level was raised.⁵¹ Aspirin crystals also showed evidence of agglomeration at lower agitation and the formation of single crystals with increasing stirring speed.⁵² This phenomenon can change when agitation is performed along with drying. Kougoulos et al. reported the transformation of fibrous crystals into irregular granules and fine particles after removing the mother liquor completely followed by agitated drying.⁵³

Fig. 4 Effect of agitation speed on crystal habit and size distribution of crystals.

On the other hand, the application of high energy ultrasound waves to the sample during crystallization can result in the breakage of the crystals along a particular axis and cause a change in the habit of the crystals. Also, due to the application of ultrasound energy, mass transfer is enhanced, thereby modulating the crystal facet growth rates. Hence, it is possible to modify the crystal habit of APIs using sonocrystallization. Cavitation and acoustic streaming phenomena are observed during the sonocrystallization. The phenomenon of inducing fluid flow by acoustic waves is called acoustic streaming. This enhances nucleation by increasing agitation and nucleation sites. The compression and rarefaction cycles during the application of ultrasound lead to bubbles forming and collapsing, a phenomenon known as cavitation (acoustic cavitation). The cavitation process results in the breakage of crystals, leading to the generation of additional growth facets and nucleation sites.⁵⁴ A cavitating bubble symmetrically collapses, thus generating a very high local liquid velocity (>100 m s⁻¹). As a result, surface pitting and localized erosion takes place due to this “jet impinging” process. The collapse of the cavitation bubble also generates shockwaves releasing a large amount of energy into the solution thereby inducing crystal breakage.⁵⁵ Cavitation induces nucleation at much lower supersaturation levels by increasing the intermolecular collisions during crystallization. Due to the cavitation, the nucleation induction time is also reduced.⁵⁶ Therefore, through sonocrystallization, it is possible to control the nucleation process precisely. Control of the nucleation process directly correlates to the control over the crystal size distribution of the powder API sample. The application of ultrasound energy during the crystallization process generally increases the nucleation rate, thereby leading to a higher number of nuclei in solution. Higher nucleation rates usually result in the formation of small-size crystals, and the uniform super-saturation provided due to the application of ultrasound energy results in a uniform and narrow size distribution of crystals. Therefore, through sonocrystallization not only can one modify the crystal habit of APIs but also obtain crystals of narrow crystal size distribution with control over the crystal size. A few research works have also reported the possibility of avoiding API agglomeration by using a sonocrystallization technique. During sonocrystallization, the process parameters such as the run time, ultrasound on/off cycles, ultrasound power rate, super-saturation, and rate of cooling/evaporation/antisolvent addition must be optimized so as to obtain desired results.^57,58

Presence of additives/impurities. Any extra substance present in the solution other than the API is called an impurity, and if added deliberately, it is called an additive, which may be a small organic molecule or even a surfactant/polymer. As shown in Fig. 5, the introduction of additives extensively influences the crystallization process since they control the nucleation and crystal growth.⁵⁹ The additives control the crystallization process by altering the solubility, metastable zone width, nucleation induction time, crystal facet growth rates, prematurely terminating crystal growth rate, etc.⁶⁰ Additives depending upon their binding affinity can selectively inhibit the growth of particular facets in a crystal. Sometimes, a combination of additives can be used to alter the habit of a crystallizing compound. When small molecules (molecular weight <500 Da) are used as additives, there is a high probability that the additive can get incorporated into the crystal lattice of the API. Additives that are structurally similar to the crystallizing compound are known to have the most potent crystal growth inhibitory effect. Such additives can be designed and synthesized for use as crystal growth modifiers. On the other hand, long-chain molecules (including polymers and surfactants) adsorb onto the facets of the crystallizing compound and can be removed post-crystallization through washing steps.⁶¹

Fig. 5 An overview of the impact of the presence of additives/impurities on the crystallization process during different steps.

Habit modification using polymers. As a result of the flexibility of polymers, polymer adsorption properties differ from those of other systems. Different polymers such as hydroxypropyl methyl cellulose (HPMC), methyl cellulose (MC), polyvinyl pyrrolidone (PVP), polyethylene glycol (PEG), carboxymethyl cellulose (CMC), etc., are used in the literature to modify the crystal habit of APIs.⁶² Like the small molecule additives, polymers can influence the nucleation induction and selectively adsorb onto the facets of the growing crystal, thereby controlling its growth kinetics.⁶³ The effect of polymers on crystal habit depends upon the interaction between polymers and the crystal facets. The binding energy of this interaction determines the crystal growth rate. Since the crystal facet with the slowest growth rate is morphologically important, the polymers that can inhibit crystal growth extensively might play a crucial role in habit modification. Table 3 summarizes the examples reported in the literature wherein polymers were used to modify the crystal habit of various APIs.

Table 3 Summary of the polymers utilized for crystal habit modification

API	Resulting crystal habit	Polymer used	Changes seen
Nifedipine	Columnar and plate-like	HPMC (0–0.6% w/v)	Decrease in dissolution rate^64
Metformin HCl	• Needle	• PVP	Improvement in flow property was speculated. Changes not studied^65
	• Thick rods	• PEG
	• Prismatic	• HPMC
Paracetamol	• Subrounded-subangular polyhedral	• PEG-6000 & Brij 58 PVA	Improved flow, compressibility, and dissolution rate^66
	Elongated	N-Hydroxyphenyl methacrylamide	Changes were not studied^67
	• Triangular-winged	• Agar	Rectangular blocks exhibited faster dissolution rates, while triangular-winged shape particles exhibited better tabletability profiles^68
	• Rectangular blocks	• Gelatin
	• Rods	• HPMC, PVP
Naproxen	Needle-like	PVP and HPMC	Changes were not studied^69
Erythromycin A dihydrate	Irregular to acicular and plate-like	HPMC	Better tableting properties^70
	Plates	HPC	Enhanced compaction^27
Mebendazole	Plates and rods	SLS and PVP	Improved dissolution^71
Nitrofurantoin	Dendritic clusters	Poly(N-isopropyl acrylamide)	Increased solubility^14
		Ethyl(hydroxy ethyl cellulose) and HPMC	Changes were not studied^72
Ropivacaine	Block-like	PVP	Uniform crystal size distribution^73
Succinic acid	Plates, blocks and needle-shaped	Pluronic P123	Changes were not studied^74
Fenofibrate	Irregular plates and elongated plates/rods	HPMC and SDS	Changes were not studied^75
Tolazamide	Plates	PEG-b-PLA	Improved flow^76
Triacylglycerols	Spherulites, needle-like	Poly(stearyl methacrylate) homopolymer	Changes were not studied^77
Ethyl vanillin	Block-like	PVP	Better compaction^78
p-HMBA (hexamethyl benzoic acid)	Block-like	PVP	Changes were not studied^79
Indomethacin	• Acicular or needle-shaped	• Eudragit E100	Enhanced dissolution^80
	• Columnar-shaped	• PVP K90
L-Alanine	Rods and blocks	Glycine-based oligopeptides	Changes were not studied^81
o-Amino benzoic acid	Irregular needles and prismatic shape	HPMC	Changes were not studied^82
Salbutamol sulphate	Block-like	PVP K25	Better flow^83

---

Impact of particle size reduction on crystal habit

Particle size reduction has a significant role in pharmaceutical manufacturing since the control over particle size distribution (PSD) improves formulation characteristics. Milling involves impact, attrition, and shear forces for the mechanical communition of API particles to reduce their size. Based on the use of solvents, milling can be classified as either wet or dry milling. The choice of milling method also influences the outcome. During dry milling, breakage usually occurs along crystallographic planes, which changes the aspect ratio and habit of crystals. This happens due to vigorous collisions between particles and the mill surface. This fracture naturally results in changes in the external geometric shape of the crystals, for instance, needle shaped crystals might become plate-like structures. The shear stress that occurs when particles slide against each other or against milling surfaces can also change the shape of crystal faces. This gradual surface removal of material usually observed in techniques like ball milling. Solvent-mediated effects might cause habit modifications during wet milling. Usually, needle-like and platy crystals are more prone to breakage. Crystals with anisotropic growth and strong intermolecular bond strength like the cubes are likely to undergo lesser surface changes than others. A possible explanation is that the applied mechanical stress is more evenly distributed in such cases and uneven fracture of crystals is reduced. Such crystals with weakly formed layers might also suffer more loss from milling shear.⁸⁴

Effect of crystal habit on pharmaceutical properties

In pharmaceutical manufacturing and development, the correlation between API properties and crystal habit must be explored and studied to optimize the formulation process. Numerous methods have been developed to control crystal habit to achieve the desired pharmaceutical properties.^85,86 Here, we have reviewed the impact of crystal habits on manufacturing (filtration, flowability, compaction and punch sticking) and formulation performance characteristics (dissolution rate and bioavailability).

Impact of crystal habit on filtration

Filtration refers to the process in which the solid particulate matter is removed from the fluid portion using a porous membrane (filter). During API manufacturing in the pharmaceutical industry, filtration is frequently employed for crystal product isolation post crystallization. Various types of filtration methods are employed, such as surface, crossflow, ultra-filtration, cake filtration, etc. In the manufacturing industry, the filtration process must be uninterrupted so that there are no delays or compromises in product quality. The crystals are collected on top of the filter medium during filtration. This continues as the crystals stack up on top of one another to create the filter cake, which can be referred to as a cluster of particles. The crystals' size and shape determine the cake's filtration performance. Comprehending particle shape is essential because many APIs have a needle-like morphology, which is prone to breakage and impedes manufacturing by filter clogging and filter integrity breach (physical breakage/tear in filter membrane), as shown in Fig. 6. Breaking occurs by two mechanisms, namely, attrition and fracture. During fracture, particles break into two main fragments along with smaller fragments. During attrition, smaller particles are chipped off from the edges of the larger particles.⁸⁷

Fig. 6 Impact of crystal habit on filtration process post crystallization.

The habit of the crystals can significantly impact the filtration process. Needles are prone to attrition, and under stress, they break down into finer particles, filling up the particle bed's voids and reducing filtrate flow.⁸⁸ However, optimization of the crystallization process, which includes processes like continuous filtration coupled with continuous product removal, can solve filterability-related problems to a significant level. For instance, Acevedo et al. designed a robust continuous crystallization process for cooling crystallization of acetaminophen and anti-solvent crystallization of benzoic acid using the mixed suspension mixed product removal crystallizer with continuous filtration system.⁸⁹

Impact of crystal habit on flow properties

The flowability of pharmaceutical powders is critical in determining their processability, particularly during the solid dosage form manufacturing including tablets and capsules. Poor powder flowability poses significant manufacturing problems such as content uniformity and weight variation issues, obstruction of manufacturing, etc., leading to defects in final products or batch failures. The flow properties of the crystals depend on several factors, such as particle size distribution, crystal habit, particle–particle interaction, particle–machinery interaction etc.

Crystals with a high aspect ratio in particular are found to have poor flow characteristics and often high cohesiveness. In this regard, needles are poorly flowable due to the high aspect ratio (surface energy also contributes to poor powder flow).⁹⁰ Podczeck and Mia studied the powder flowability of eight different powders including, acetaminophen, acetyl salicylic acid, microcrystalline cellulose, etc. They found out that particles with needle-like habits, with higher aspect ratios, showed higher angles of internal friction which explains their poor powder flowability profiles.⁹¹ Modifying needles to allow for much more free-flowing habits is important for facilitating downstream API processing.^92,93 Spherical particles, on the other hand, display superior flow properties, as seen in the case of cycloserine, metformin hydrochloride, ethambutol dihydrochloride, etc.^94–98 This could be due to the low surface area of the spheres (which reduces the impact of cohesive forces) and uniform packing. Also, spheres with smooth surfaces experience low friction and tend to flow very easily. A comparative study was carried out to assess the processability of different crystal habits of 5-aminosalicylic acid (5-ASA). The needle-shaped crystals of 5-ASA had poor powder flowability, while the plate-shaped crystals exhibited moderate flowability, and spheroids and long hexagons were freely flowing.⁹⁹ Thus, habit modification can serve as an essential tool to improve the powder flow properties of APIs.

Impact of crystal habit on compaction

Compressibility, tabletability, and compactability collectively define the compaction properties of pharmaceutical powders. These factors are critical in designing and developing solid dosage forms, like tablets, in filling hard-shell gelatin capsules, and in handling powder formulations.¹⁰⁰

Compressibility is the ability of a pharmaceutical formulation to deform under pressure. Tabletability reflects the ability of a powder to form a tablet of adequate strength at different compaction pressures. Compactability represents the ability of a material to be manufactured into tablets of the required strength under the effect of densification. The three parameters, viz., compressibility, tabletability, and compactability are interrelated as shown in Fig. 7A. For convenience, the three plots can be merged and represented in a three-dimensional format which is shown in Fig. 7B.¹⁰¹

Fig. 7 A) Tablet compaction triangle indicating the compressibility, tabletability, and compactability profiles, B) three-dimensional compression profile.

Crystal habit modification can significantly improve the compression and compaction profiles of the API.^102,103 Tabular and platy crystals have high compressibility due to their layered structure and packing density (example, celecoxib and dabigatran etexilate).^104,105 However, it is not always the case that other crystal habits show poor compaction behavior. For instance, polyhedral and platy paracetamol and ibuprofen crystals showed excellent tabletability.^106,107 Various other examples involving glimepiride,¹⁰⁸ dapagliflozin propanediol monohydrate,⁵⁷ erythromycin A dihydrate,⁷⁰etc., have shown a positive impact of crystal habit modification on the compaction properties of the crystals. It is crucial to understand the mechanical properties of the API to enhance its compaction.¹⁰⁹ The powder compaction properties can be analyzed through Heckel or Kawakita plots, and the reliable method to be employed is determined based on various factors like nature of material, pressure applied and so on.¹¹⁰

Impact of crystal habit on punch sticking

Adhesion of the powder material to the tooling surfaces is one of the primary issues faced during tablet punching, and the phenomenon is called punch sticking. Even though extensive research is being done to find effective solutions, to date no single root cause of sticking has been identified. Various factors like crystal habit,^111,112 punch-particle interactions, API-excipient interactions, geometry of the punch,¹¹³ particle size, electrostatic properties of the API^114,115 are under scrutiny and are depicted in Fig. 8.

Fig. 8 Parameters affecting punch sticking propensity during tableting.

The physicochemical characteristic of a drug largely varies due to the anisotropic surface chemistry of crystalline substances. Since the interactions between API, punch, and excipients depend on the API's surface properties, crystal habit modification is proposed to considerably impact punch sticking. For instance, the crystal habit of ibuprofen was discovered to be the primary factor in the drug's propensity for sticking, and the platy crystals had exhibited higher level of sticking than needles/lath-shaped crystals.¹¹⁶ The molecular aspects of crystalline compounds can be considered to explain the punch-sticking tendency. In case of mefenamic acid, the exposure of polar functional groups from the crystal facets increases the interaction of the particles with the punch surface and hence increases sticking.¹¹² The abundancy in hydrogen bond donors and acceptors and other surface phenomena in ibuprofen form I that enhance particle–particle interactions can also be attributed to sticking.¹¹⁷ Plenty of work has been done on punch sticking, emphasizing that punch sticking is less dependent on particle size and more on crystal habit modification, among other factors, because of the changes in the interparticle bonding areas and surface energy.¹¹⁸ Studies indicate that using lubricants can significantly reduce the tendency to stick for pharmaceutical powders. The addition of magnesium stearate, the most commonly used lubricant, to tablet formulations modifies surface properties like roughness, and the sticking tendency decreases significantly with an increase in lubricant concentration.¹¹⁹ Lubricants reduce friction between the punch face and powder particles. This is called boundary lubrication, and is commonly employed in the pharmaceutical industry.¹²⁰

Impact of crystal habit on dissolution rate

Dissolution is a process in which the solid substances (particles/crystals) are converted into a molecular state in a solution form.¹²¹ The dissolution rate of a compound depends upon various factors such as its solid state (crystalline or amorphous), polymorphic form, particle size, API-solvent interactions, etc.¹²² In addition to the mentioned factors, crystal habit can also play a vital role in determining the dissolution rate of APIs.

Research in this area has made it evident that crystal habit modification can change the dissolution profile of an API. For instance, numerous studies have shown that crystal habit modification of poorly soluble drugs such as tolfenamic acid, terbinafine hydrochloride, etc. can result in an improved dissolution rate.^123,124 Crystals with the highest surface area per volume, for instance, needles and plates, tend to have high dissolution rates.^125–127 However, other factors, along with crystal habit, might be responsible for changes in the dissolution profile, and we cannot solely attribute it to the crystal habit of the API.¹²⁸ When the dissolution rate of a drug is enhanced, its bioavailability can be increased, since a greater fraction of a drug can be absorbed.¹²⁹ The surface anisotropy in crystals contributes to the increased dissolution rate after habit modification. The surface anisotropy in crystals is a result of differences in the atomic arrangements, bonding orientations, and surface energies on each crystal face. As a result, the type of functional groups exposed on each facet of the crystal is modified during the crystal habit experiments.¹³⁰ Through meticulous use of crystal growth inhibitors, the growth rate of facets expressing hydrophobic functional groups can be limited to increase the surface area of facets possessing hydrophilic functional groups, potentially improving the interaction of facets with water molecules during dissolution. This results in an increase in the polar energies of the compound, thereby improving its dissolution rate.^131,132

Other research works that have employed crystal habit modification to improve flowability, compaction, and dissolution rate have been summarized in Table 4. The readers are directed towards the individual research works reported in the literature to obtain thorough understanding of the mechanisms of pharmaceutical property improvement after habit modification.

Table 4 Examples of crystal habit modifications resulting in improved pharmaceutical properties

API	Crystal habit	Property improved
Metformin HCl	Equidimensional^133	Flowability
	Hollow, spherical^134
Mesalazine	Planks^135
Celecoxib	Lath-shaped^136	Flowability, compaction, dissolution
	Acicular^15
	Plates^15
Carbamazepine	Block-shaped^137
Paracetamol	Prismatic^11	Compaction
BMS-561390 (investigational anti-HIV drug)	Irregular^138
Paracetamol	Irregular tabular^139
Carbamazepine	Needle-like^140	Dissolution, compaction
Phenytoin	Long thin plates^141	Dissolution rate
Aspirin	Plates/prisms^142
Meloxicam	Prismatic^143
Tolbutamide	Circular, platy^128
Simvastatin	Rods^144
Piroxicam	Cubical^145
Carbamazepine	Needle^146
Ticagrelor	Plates^147
Dipyridamole	Needle-like^148	Bioavailability
Trimethoprim	Rods^149	Stability

---

Impact of crystal habit on stability

Stability of pharmaceutical products refers to the ability of a drug product to retain its physical, chemical, microbiological, and therapeutic properties until consumption under specific storage conditions. Considering the example of a suspension formulation of trimethoprim (TRM), which is expected to show a degree of physical stability with respect to its sedimentation volume. However, it was found that the habit of TRM influenced the sedimentation rate, with the order sedimentation volume of trimethoprim suspension being thin rods > thick rods > cubes > polyhedra.¹⁵⁰ Hygroscopicity can also cause serious stability issues due to absorption of moisture from environment which may lead to either drug degradation or cause handling and processability issues. Platy and needle-like crystal exhibit enhanced hygroscopicity due to higher surface area to volume ratio. For instance, Prestidge and Tsatouhas studied the wettability of morphine sulphate crystallized from different solvent systems. It was found that the morphine sulphate samples showed different crystal habits in different solvents and as a result, its wettability profile also changed due to difference in the elemental composition of the different faces of morphine sulphate.¹⁵¹ Also, the degradation of crystals is known to be dependent on the crystal habit. For instance, the crystals of poly(hexamethylene adipate) were known to undergo enzymatic degradation mainly from the (110) face.¹⁵²

Characterization studies after crystal habit modification experiments

Morphological characterization

Optical microscopy. Optical microscopy using a compound, digital, or polarized light microscope is a preliminary evaluation carried out to characterize the habit and crystal growth kinetics of compounds. The camera-mounted systems can be used to capture digital images of the crystals. In combination with image analysis software (e.g., ImageJ, L-Measure, etc.) and calibrated micrometres, the size analysis of crystals can also be carried out efficiently. The sample preparation involves easy steps of spreading the crystal sample (∼10 mg) onto a clean glass slide and covering the sample using a coverslip. The prepared sample can be mounted onto the microscope stage and visualized through an eyepiece or live projection on the computer. The limitation of optical microscopy is that the samples of size <100 μm cannot be visualized clearly and necessitates electron microscopy for samples with small particle size.⁵⁷

Electron microscopy. Advances in microscopic techniques such as scanning electron microscopy (SEM) analysis can be carried out to obtain much clearer and high-resolution pictures of crystal samples. In SEM the samples are scanned using a focused electron beam, and the scattered electrons are detected, revealing the powder sample's topography. The samples for this purpose are required to be conductive in nature and are sputter coated with conductive materials such as gold to achieve a coating thickness of ∼10 nm. The digital micrographs are then captured using a suitable computer to reveal the habit of crystal samples. Further image analysis can be carried out to measure the dimensions of the samples (length and width of the crystals, facet dimensions, etc.).¹⁵³ Additionally, SEM analysis also allows for topographical analysis (such as surface smoothness, presence of pores, etc.) of crystals. The identification of crystal facets can be performed in combination with powder X-ray diffraction, nanoindentation and molecular dynamics simulation studies.^30,154

Solid state characterization

Solids are made of repetitive molecules known as unit cells which pack/arrange in a specific manner, resulting in a particular solid form. The specific intramolecular and intermolecular forces guide this packing/arrangement; changes in these forces also result in changes in the packing behavior, which results in changes in the solid-state form (polymorphs or amorphous form). Though exhibiting the same chemical composition, the polymorphs of compounds show remarkable changes in their physicochemical properties.¹⁵⁵ Hence, during crystallization experiments, it is necessary to confirm the polymorphic form of a compound. The polymorphic confirmation of compounds can be made by estimating the unit cell parameters of compounds before and after the habit modification experiments. Any changes in the unit cell parameters is an indicative of polymorphic transformation which is undesired during habit modification experiments. Studies such as powder X-ray diffraction (PXRD), differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), Fourier-transform infra-red spectroscopy (FTIR) can be carried out to confirm the polymorphic identity of APIs.¹⁵⁶

Powder X-ray diffraction (PXRD). Most pharmaceutical compounds are crystalline in nature, and powder X-ray diffraction (PXRD) is a non-destructive method used to characterize crystalline compounds. When a beam of X-rays is made incident on a powder sample, the beam is scattered at specific angles by the lattice planes of a crystalline compound. The intensity of the scattered X-rays at different angles depends on the crystal lattice's atomic distribution. Each peak in PXRD spectra is indicative of a particular crystal facet. Thus, the X-ray diffraction pattern can be regarded as the fingerprint of the periodic arrangement of atoms in a material. The plot between the diffracted (scattered) X-ray intensity and the angle (2-theta) gives a PXRD diffractogram. The position of the peaks shall be the same for a given polymorphic form of material. For example, if a compound exists in two different polymorphic forms, both samples will show different PXRD patterns.¹⁵⁷ However, the peak intensity of the same polymorphic form varies if the crystal habit is changed. As said earlier, the peaks represent crystal facets, and when a particular facet grows, the intensity of X-rays scattered by this facet also increases. Thus, the PXRD pattern can also be used to confirm the habit modification.⁵⁷ For analysis, around 100 mg to 1 g of the sample is neatly spread on a glass slide or low-background silicon sample holder to form a thin layer. The slide is then placed in the instrument sample holder, and the intensity of diffracted X-rays is recorded at different diffraction angles. For most pharmaceutical compounds, diffraction peaks appear at the 0–50° scanning range.

Sometimes, a phenomenon called as “Preferred orientation” is seen during PXRD studies, where certain facets of a crystal tend to align parallel to the sample surface which then shows enhanced diffraction intensities from these facets. In such cases, certain peaks appear with high intensity and certain peaks may not appear in the diffractogram leading to erroneous interpretation of the PXRD results. Such effects are mostly seen in acicular and platy crystals and can be avoided by using a rotating sample holder or spinning capillary method solves the problem of preferred orientation.^158,159

Differential scanning calorimetry (DSC). In DSC instrument the difference in heat flow between the sample pan (containing compound under study) to that of a reference pan (either empty or with an inert material) is measured and involves a plot between heat flow and temperature. The heat flow patterns indicate endothermic (negative peak) and exothermic (positive peak) processes. Accordingly, the melting point (T_m), heat capacity (C), enthalpy of fusion (ΔH_fus), temperature-dependent polymorphic changes, etc. of a compound can be characterized.^16,160 For analysis, the sample (∼5 mg) is placed inside an aluminum pan (∼40 μL volume), and a thermogram is recorded at a temperature range to include all the thermal events occurring in the sample (for example, 0–400 °C) at a specific heating rate (for example 20 °C min⁻¹) under constant nitrogen flow (for example 50 ml min⁻¹).

Thermogravimetric analysis (TGA). TGA is used to study the relation between changes in the mass (weight) of the sample and temperature changes. It is employed to assess the thermal stability of drugs, differentiate the bound and unbound water and its quantification, identify solvates, determine solvent content, etc. In TGA, the accurately weighed sample (5–50 mg) is placed inside the instrument and heated either in a linear manner (dynamic mode) or maintained at a particular temperature (static mode). The change in the weight of the sample is recorded against the temperature/time. The plot reveals the sample's weight change at increasing temperatures.¹⁶¹

Fourier-transform infra-red spectroscopy (FTIR). FTIR spectroscopy gives information about the functional groups present in a compound. The compound absorbs incident IR radiation and undergoes vibrational changes depending on the types of functional groups present. The transmitted (unabsorbed) radiation is detected as % transmittance and plotted against wavenumber (cm⁻¹) to obtain FTIR spectra. During crystal habit modification experiments, the additives sometimes appear in the final product if not removed during the washing phase, which can be detected through FTIR spectroscopy. Also, any changes in the structure of the molecule due to high energy (sonocrystallization) can be detected using FTIR spectroscopy. Sample preparation involves grinding the dried API powder with dry potassium bromide (taken in ratio 1 : 100 w/w), followed by preparation of a thin transparent compact using a hydraulic press. The compact can then be placed in the sample holder to record FTIR spectra. Alternatively, if attenuated total reflectance (ATR) mode is available, the powder sample can directly be placed on the ATR crystal to record IR spectra (between 400–4000 cm⁻¹).⁵⁷

Powder rheology

Angle of repose (AOR). AOR is the most basic and simple method to assess the powder flow behavior of a powder sample. The fixed funnel method is widely used in the pharmaceutical industry to measure AOR. Herein; a measured quantity of powder sample is poured through a funnel fixed at a definite height. The powder then forms a heap, based on its flow properties. The height and base diameter of the powder pile can be used to calculate the AOR of the powder sample. Generally, free-flowing powders form a heap of lower height and wider diameter as they tend to flow down easily.¹⁶² The angle of repose of powders can be calculated using eqn (1).where ‘h’ is the height and ‘r’ is the radius of the powder heap.

The United States Pharmacopoeia (USP) has defined the flow behavior of powders based on their AOR values, as shown in Table 5.

Table 5 Powder flow properties and their corresponding AOR, CI and HR values

Flow property	Angle of repose (°)	Carr's index (%)	Hausner ratio
Excellent	25–30	≤10	1.00–1.11
Good	31–35	11–15	1.12–1.18
Fair	36–40	16–20	1.19–1.25
Passable	41–45	21–25	1.26–1.34
Poor	46–55	26–31	1.35–1.45
Very poor	56–65	32–37	1.46–1.59
Very, very poor	>66	>38	>1.60

---

Carr's index (CI) and Hausner's ratio (HR). CI and HR measure the inter-particular interactions in a powder sample and indicate the compressibility of the powder. Carr's index provides information on powder flowability by considering the differences between a powder sample's bulk and tapped densities. A greater difference between bulk and tapped densities correlates to higher interparticular interactions and thus denotes poor powder flowability. CI and HR can be calculated using the following eqn (2) and (3).¹⁶³where ρ_t is tapped density (g mL⁻¹) and ρ_b is bulk density (g mL⁻¹). The flowability character of powder and its corresponding CI and HR values are given in Table 5.

The bulk density of a sample can be determined by pouring a weighted quantity of the powder sample in a 100 mL measuring cylinder and noting down the volume (called bulk volume). Tapped density can be determined by subjecting the powder sample to around 1250 tapping using suitable apparatus (such as tapped density testers). The volume after the tapping is noted down as tapped volume. Bulk and tapped densities can be calculated using the following formulae (eqn (4) and (5)).

where ‘W’ is the weight (g) of the powder sample, V_b is the bulk volume (mL), and V_t is the tapped volume (mL). For bulk and tapped density measurements, approx. 50–60 g of the powder sample is required.

Shear cell testing. In a shear cell tester, the powder sample is subjected to consolidation stress (σ₁) followed by the application of horizontal stress, leading to the failure (breakage) of the sample at a certain stress, which is called the unconfined yield strength (σ_c). At the point of failure/breakage, the sample starts to flow and is termed as incipient flow. Graphically, the flow of powder is indicated by the yield loci in a plot between normal stress (σ) and shear stress (τ). Horizontal stress is applied to the sample at different load steps and is represented by the stress circles in Fig. 9A (also called Mohr stress circles which is a two-dimensional graphical representation that depicts the relationship between stress states at different orientations) of increasing diameters which forms due to the increasing load steps. At a certain load, the Mohr stress circle touches the yield loci indicating that the applied shear stress is high enough to cause movement of particles across each other. Through these stress circles, the yield limit of the sample can be calculated. Further, using the σ₁ and σ_c, it is possible to characterize the powder flowability behavior. The flow function (ff_c) is the ratio of consolidation stress (σ₁) to confined yield strength (σ_c).^164,165 The flow behavior of powder samples is represented in Fig. 9B and can be defined by ff_c as follows:

Fig. 9 A) Measurement of yield strength of sample through Mohr stress circles using a σ–τ diagram, B) flow function and lines of constant flowability.

ff_c < 1 not flowing.

1 < ff_c < 2 very cohesive.

2 < ff_c < 4 cohesive.

4 < ff_c < 10 easy flowing.

10 < ff_c free-flowing.

Surface changes

Contact angle studies. Post-habit modification, the surface composition of the crystals is altered due to surface anisotropy. The surface free energies of an organic (pharmaceutical) compound can be determined by contact angle studies using the van Oss, Chaudhury, and Good model. In this model, the surface free energy is expressed in terms of Lifshitz–van der Waals (LW) and Lewis acid–base components (AB). The contact angle (θ) between the surface of the compound and different liquids can, therefore, be measured and employed to determine the total surface free energy (γ^S), polar surface energy (γ^AB), and dispersive surface energy (γ^LW) of the solid compound. Further, the acidic (γ^A) and basic elements (γ^B) constitute the polar component. These elements explain the propensity of the surface towards polar interactions, through donating or accepting electrons.¹⁶⁶

In this method, a minimum of three probe liquids must be used for contact angle studies and to determine the above-mentioned parameters. Different polarity liquids (for instance, water as a polar liquid, diiodomethane (DIM) as a non-polar liquid, and ethylene glycol as a semi-polar liquid) are commonly used in the study. The total surface energy, dispersive and polar energies, and acidic and basic components are calculated using eqn (6).

where, γ_LV is the surface tension, γ^LW_S and γ^LW_L are the dispersive surface energies of solid and liquid, respectively, γ^A_S and γ^A_L are acidic components of solid and liquid, respectively, and γ^B_S and γ^B_L are basic components of solid and liquid, respectively.^125,167 The procedure involves the preparation of compacts of the solid compound under study using a hydraulic press. Further, the contact angle measurements are carried out using three different liquids, as mentioned above, using a suitable instrument with a high-speed camera. The measured contact angle values are used to calculate the parameters of eqn (6).

X-ray photoelectron spectroscopy (XPS)

XPS is a surface-sensitive analytical technique that is based on the photoelectric effect. In XPS, the surface of the material is bombarded with X-rays, which emit electrons. This method is highly surface sensitive (∼10 nm), and the chemical state information of the sample can be studied. Through this method, all elements except hydrogen and helium can be detected. Habit modification of APIs changes its surface chemistry (anisotropy), and thus XPS is a suitable technique to study the surface changes occurring in the sample. Changes in the hydrophilicity of the samples can be supported through XPS studies. For instance, after habit modification, the surface composition of hydrophilic moieties such as –OH, –NH increases or decreases. This change can be studied through XPS, wherein the % elemental composition of oxygen and nitrogen is increased in the modified samples. Thus, XPS serves as an important analytical tool to characterize the surface chemistry of samples.¹⁶⁸

Surface area analysis. The surface area to mass ratio of different crystal habits of the same compound differs, which might also increase or decrease the compound's dissolution rate. As explained in Table 1, the acicular (needle-shaped) crystals have the highest surface area to mass ratio, while the spherical crystals have the lowest surface area to mass ratio. Surface area analysis can provide a deeper insight into the mechanism of dissolution rate enhancement in APIs that exist in different habits. The most common method is the BET (Brunauer–Emmett–Teller) surface area analysis. This method is based on the physisorption of a gas onto the powder sample to determine the specific surface area, based on an evaluation of the amount of gas adsorbed when a gas monolayer has adsorbed onto the surface of the solid.¹⁶⁹ For instance, Singh et al., employed BET surface area analysis to characterize sonocrystallized samples of dapagliflozin propanediol monohydrate which had shown changes both in habit as well as crystal size.⁵⁷

Dissolution rate/solubility analysis

Intrinsic dissolution rate (IDR) studies. Intrinsic dissolution rate (IDR) is defined as the rate at which the drug substance (only API) dissolves from the surface of the fixed area at constant temperature, agitation, degree of ionization, and pH of the solvent. Herein, pure drug substances without any excipients, such as disintegrating agents, are compacted into pellet form, and the dissolution rate from the fixed surface area (mass per cm⁻² min⁻¹) is reported. Crystal habit modification can alter (either increase or decrease) the dissolution rate of a compound. Hence, through IDR studies, it is possible to assess the changes in the dissolution rate of a drug under constant conditions. U.S. Pharmacopoeia recommends two types of apparatus for IDR studies: the rotating-disk apparatus and the stationary-disk apparatus. In the rotating-disk method, the drug is compressed inside the cavity of the apparatus to form a pellet with a single face exposed to the dissolution media at the bottom of the apparatus, which further rotates at set RPM. In the stationary disk type, the die assembly containing the drug pellet is placed in an upright position at the bottom of the dissolution vessel (flat-bottomed). The stirring element (paddle) is positioned above the die assembly (typically 2.54 cm) to form the compact surface. Appropriate buffer systems can be selected, and the volume can vary between 500–900 mL.¹⁷⁰ The intrinsic dissolution rate study is conducted at a fixed temperature, and the sampling is performed similarly to dissolution studies. The samples are analyzed using any of the validated analytical technique, and the amount of drug dissolved per cm² at particular time points is plotted. The slope of the linear plot between the amount of drug dissolved per cm²vs. time gives the IDR in mg cm⁻² min⁻¹.

When conventional IDR studies are not possible (in cases where API quantity available is too low), powder dissolution studies can be carried out which utilize very less amount of API sample.¹⁷¹ In contrast to IDR studies, powder dissolution studies use powder sample without compression in a USP Type-IV “Flow through cell” apparatus. It is possible to monitor the dissolution rate of individual crystal through this method in combination with shadowgraph imaging technique. Herein, a high-resolution camera is utilized to capture video/images and tag and track individual crystals/particles in the flow through cell during dissolution studies. Further, the video analysis can be carried out using software such as Matlab to track the surface area of dissolving crystal/particle.^172,173 The results obtained by powder dissolution studies can also be correlated with that of pellet IDR studies.¹⁷¹

Compaction behaviour

CTC profiling. CTC profile refers to the compressibility–tabletability–compactability profile of powder samples. Different crystal habits of the same compound show differences in their ability to form a solid compact when subjected to compression force. Hence, it is important to study the CTC behavior. Evaluation of compressibility involves plotting tablet porosity (%) (or soldi fraction) versus applied compression pressure (MPa). The tabletability profile is a plot of tablet tensile strength (MPa) versus applied compression pressure (MPa). The compaction profile is a plot of tablet tensile strength (MPa) versus the porosity (%) (or solid fraction) of the solid compact. The tensile strength of the solid compact influences the disintegration and dissolution profiles of the tablets, and CTC profiling helps the formulation scientist select the appropriate compression force for preparing the tablets. Comprehensive CTC profiling also helps minimize defects in a formulation, such as capping, sticking, etc.^174,175

In CTC profiling, tablet porosity and solid fraction can be used interchangeably, and the solid fraction is calculated using eqn (7).

True density measurement can be carried out using a helium pycnometer. Tablet density calculations depend upon the shape and dimensions of the tablet and are given by eqn (8) and (9) for flat and oblong tablets, respectively.

For flat round tablets:

For oblong convex shape tablets:

The tensile strength can be calculated using the measured hardness and tablet dimensions using eqn (10).where σ is the tablet tensile strength and x is the tablet hardness in Newton.

The experimental procedure involves the preparation of compacts/tablets of fixed weight and dimensions at different compression pressures using a hydraulic press or tablet compression machine. The tablet hardness and dimensions are then measured for each tablet, and CTC plots are prepared.

Heckel analysis. Heckel analysis is the most widely used method to analyze the change in volume under compression pressure for pharmaceutical powders. It is based on the assumption that the change in porosity upon application of pressure follows first-order kinetics. Heckel demonstrated that the compaction pressure on the API is directly proportional to the natural logarithm of the reciprocal of tablet porosity by eqn (11).where D is the relative density (which makes 1 − D the porosity of the tablet), K is the slope of the Heckel plot, P is the compaction pressure, and A is the natural logarithm of the reciprocal of the tablet porosity at zero compaction pressure.

In a Heckel plot the natural logarithm of the reciprocal of tablet porosity is plotted against the applied compaction pressure.¹⁷⁶ Such plots can be used to determine and compare the compaction properties of the API. Heckel plots are linear for intermediate pressure ranges, and there are significant deviations when the pressure is either too high or too low. In the context of Heckel plots, API powders are classified into three types based on their compaction behavior: type A (undergoes plastic deformation and has a linear Heckel plot), type B (undergoes fracture at lower pressures resulting in initial curvature followed by a linear region in the plot), and type C (undergoes particle fragmentation with little plastic deformation). Larger values of ‘K’ indicate increased hardness of tablets. Likewise, K′ decreases when the concentration of lubricants in the tablet increases. Thus, Heckel plots can also aid in choosing the appropriate binding agents.¹⁰⁹

Regulatory perspective on control over crystal habit

Crystal habit modification can lead to changes in the processability (powder flowability, compressibility, filterability), stability (hygroscopicity, degradation rate, etc.) and performance (solubility/dissolution rate, bioavailability, etc.). It is absolute necessity for the manufacturers to maintain these quality attributes in the drugs to avoid regulatory repercussion. The United States Food and Drug Administration (USFDA) in its guidance documents mentions the need for the analysis of the particle size distribution and crystal habit of drugs wherever such factors are found to affect its performance.¹⁷⁷ From the discussion presented in this tutorial, it is evident that to control the habit and size distribution of drugs, the crystallization process must be well understood, and adequate control measures must be in place. Additionally, the regulatory agencies recommend to use quantitative statistical assessment tools such as design of experiment (DoE) to study the effect of process parameters on the process robustness and the product (drug) quality. For thorough understanding of the DoE topic, the readers are advised to refer standard texts by Montgomery,¹⁷⁸ Snee and Hoerl,¹⁷⁹etc.

Summary

As the pharmaceutical industry continues to strive for innovation, and regulating crystal habit remains a critical aspect in medication development and production, the factors that influence crystal habit modification are being investigated in detail. Modifying the crystal habit to improve physicochemical properties of an API can be considered as a cost-effective and simple alternative to other techniques such as the preparation of cocrystals, milling, amorphous solid dispersion formulation, etc. With appropriate process controls, crystal habit modification can yield excellent results in terms of improved processability (e.g. filterability, flowability, compaction) and dissolution performance of an API. In summary, the present tutorial details a procedure for carrying out crystal habit modification experiments and highlights the possible effects of different critical process parameters on crystal habit modification experiments. Various methods of crystallization (evaporative, cooling and anti-solvent crystallization) are also explained in brief, as is the effect of additives on crystal habit during crystallization. Further characterization techniques (morphological, physicochemical, rheological, and surface characterization), which can be employed to characterize the modified API, are discussed.

Data availability

No primary research results and no new data were generated or analysed as a part of this review.

Author contributions

Niranj H. Ram: conceptualization, literature survey, writing – original draft; Madhukiran R. Dhondale: conceptualization, content structuring, visualization, writing – original draft; Maan Singh: writing – review and editing; Brahmeshwar Mishra: supervision, Ashish Kumar Agrawal: writing – review and editing; Anne Marie Healy: writing – review and editing; Dinesh Kumar: conceptualization, visualization, writing – review and editing.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Madhukiran R. Dhondale acknowledges the funding from PMRF, India under the grant number 1102018.

References

1. A. Tiwary, Drug Dev. Ind. Pharm., 2001, 27, 699–709.
2. J. Prywer, Prog. Cryst. Growth Charact. Mater., 2005, 50, 1–38.
3. R. Ho, D. A. Wilson and J. Y. Heng, Cryst. Growth Des., 2009, 9, 4907–4911.
4. S. J. Satyawati, in Modern Aspects of Bulk Crystal and Thin Film Preparation, ed. K. Nikolai and B. Elena, IntechOpen, Rijeka, 2012, ch. 18,  DOI:10.5772/28451.
5. M. Okayasu, S. Kikkawa, H. Hikawa and I. Azumaya, CrystEngComm, 2021, 23, 7760–7770.
6. F. M. Mahdi, A. P. Shier, I. S. Fragkopoulos, J. Carr, P. Gajjar and F. L. Muller, Pharm. Res., 2020, 37, 231.
7. C. Y. Ma and X. Z. Wang, J. Process Control, 2012, 22, 72–81.
8. M. R. Dhondale, A. G. Nambiar, M. Singh, A. R. Mali, A. K. Agrawal, N. R. Shastri, P. Kumar and D. Kumar, J. Pharm. Sci., 2023, 112, 2010–2028.
9. Y. Kawashima, M. Imai, H. Takeuchi, H. Yamamoto, K. Kamiya and T. Hino, Powder Technol., 2003, 130, 283–289.
10. O. Gyulai and Z. Aigner, Powder Technol., 2018, 336, 144–149.
11. H. A. Garekani, J. L. Ford, M. H. Rubinstein and A. R. Rajabi-Siahboomi, Int. J. Pharm., 1999, 187, 77–89.
12. Y. Wang, H. Zhang, L. Cai, F. Xue, H. Chen, J. Gong and S. Du, Ultrason. Sonochem., 2023, 97, 106475.
13. A. Maharana, A. Das, J. Kumar and D. Sarkar, Comput. Chem. Eng., 2024, 184, 108651.
14. T. Munk, S. Baldursdottir, S. Hietala, T. Rades, S. Kapp, M. Nuopponen, K. Kalliomaki, H. Tenhu and J. Rantanen, Mol. Pharmaceutics, 2012, 9, 1932–1941.
15. S. R. Modi, A. K. R. Dantuluri, V. Puri, Y. B. Pawar, P. Nandekar, A. T. Sangamwar, S. R. Perumalla, C. C. Sun and A. K. Bansal, Cryst. Growth Des., 2013, 13, 2824–2832.
16. M. K. Bommaka, M. K. Chaitanya Mannava, S. K. Rai, K. Suresh and A. K. Nangia, Cryst. Growth Des., 2021, 21, 5573–5585.
17. D. Liu and M. Z. Yates, Langmuir, 2006, 22, 5566–5569.
18. Y. Masui, Y. Kitaura, T. Kobayashi, Y. Goto, S. Ando, A. Okuyama and H. Takahashi, Org. Process Res. Dev., 2003, 7, 334–338.
19. T. Wu, L. Valencia, T. Pfohl, B. Heck, G. Reiter, P. J. Lutz and R. Mülhaupt, Macromolecules, 2019, 52, 4839–4846.
20. I. Wünsch, J. H. Finke, E. John, M. Juhnke and A. Kwade, Int. J. Pharm., 2022, 626, 122117.
21. G. Perini, C. Avendano, W. Hicks, A. R. Parsons and T. Vetter, Chem. Eng. Sci., 2021, 230, 116151.
22. H. Kalman, Powder Technol., 2021, 393, 582–596.
23. D. Geldart, E. Abdullah, A. Hassanpour, L. Nwoke and I. Wouters, China Particuol., 2006, 4, 104–107.
24. M. Maghsoodi, Adv. Pharm. Bull., 2015, 5, 13.
25. C. Noiriel, M. Oursin, G. Saldi and D. Haberthür, ACS Earth Space Chem., 2018, 3, 100–108.
26. Y. Tsume, D. M. Mudie, P. Langguth, G. E. Amidon and G. L. Amidon, Eur. J. Pharm. Sci., 2014, 57, 152–163.
27. S. Mirza, I. Miroshnyk, J. Heinämäki, O. Antikainen, J. Rantanen, P. Vuorela, H. Vuorela and J. Yliruusi, AAPS PharmSciTech, 2009, 10, 113–119.
28. S. Clercq, A. Mouahid, G. Pèpe and E. Badens, Cryst. Growth Des., 2020, 20, 6863–6876.
29. H. Gu, R. Li, Y. Sun, S. Li, W. Dong and J. Gong, J. Cryst. Growth, 2013, 373, 146–150.
30. D. Kumar and N. R. Shastri, Cryst. Growth Des., 2014, 14, 326–338.
31. K. Chadwick, J. Chen, E. E. Santiso and B. L. Trout, in Handbook of Industrial Crystallization, ed. A. S. Myerson, D. Erdemir and A. Y. Lee, Cambridge University Press, Cambridge, 3rd edn, 2019, pp. 136–171,  DOI:10.1017/9781139026949.005.
32. T. Tari, P. Szabó-Révész and Z. Aigner, Crystals, 2019, 9, 295.
33. A. Bártová, R. Gabriel, B. B. Prudilová, E. Otyepková, L. Malina and M. Otyepka, Powder Technol., 2022, 402, 117334.
34. R.-Q. Song and H. Cölfen, CrystEngComm, 2011, 13, 1249–1276.
35. K. N. Olafson, R. Li, B. G. Alamani and J. D. Rimer, Chem. Mater., 2016, 28, 8453–8465.
36. A. V. Tulcidas, Ph.D. thesis, Universidade Do Porto, 2019.
37. Z. Gao, S. Rohani, J. Gong and J. Wang, Engineering, 2017, 3, 343–353.
38. M. Shahid, G. Sanxaridou, S. Ottoboni, L. Lue and C. Price, Org. Process Res. Dev., 2021, 25, 969–981.
39. T. Lee, H. Y. Lin and H. L. Lee, Org. Process Res. Dev., 2013, 17, 1168–1178.
40. S. Pu and K. Hadinoto, Chem. Eng. Res. Des., 2024, 201, 45–66.
41. I. Nikolakakis, K. Kachrimanis and S. Malamataris, Int. J. Pharm., 2000, 201, 79–88.
42. E. M. Soper, R. Y. Penchev, S. M. Todd, F. Eckert and M. Meunier, J. Cryst. Growth, 2022, 591, 126712.
43. A. T. Karunanithi, C. Acquah, L. E. K. Achenie, S. Sithambaram and S. L. Suib, Comput. Chem. Eng., 2009, 33, 1014–1021.
44. A. T. Karunanithi, L. E. Achenie and R. Gani, Chem. Eng. Sci., 2006, 61, 1247–1260.
45. L. E. Hatcher, W. Li, P. Payne, B. Benyahia, C. D. Rielly and C. C. Wilson, Cryst. Growth Des., 2020, 20, 5854–5862.
46. T. D. Turner, L. E. Hatcher, C. C. Wilson and K. J. Roberts, J. Pharm. Sci., 2019, 108, 1779–1787.
47. Z. Liang, M. Zhang, F. Wu, J.-F. Chen, C. Xue and H. Zhao, Comput. Chem. Eng., 2017, 99, 296–303.
48. P. Neugebauer, J. Cardona, M. O. Besenhard, A. Peter, H. Gruber-Woelfler, C. Tachtatzis, A. Cleary, I. Andonovic, J. Sefcik and J. G. Khinast, Cryst. Growth Des., 2018, 18, 4403–4415.
49. C.-H. Chang, C.-M. Hsieh and C.-S. Su, Cryst. Res. Technol., 2021, 56, 2000182.
50. C. J. Callahan and X.-W. Ni, CrystEngComm, 2014, 16, 690–697.
51. Z. Yu, R. Tan and P. Chow, J. Cryst. Growth, 2005, 279, 477–488.
52. L. Jia, D. Wu, P. Cui, L. Zhou and Q. Yin, Particuology, 2023, 82, 146–156.
53. E. Kougoulos, C. Chadwick and M. Ticehurst, Powder Technol., 2011, 210, 308–314.
54. R. Chow, R. Blindt, R. Chivers and M. Povey, Ultrasonics, 2003, 41, 595–604.
55. J. R. G. Sander, B. W. Zeiger and K. S. Suslick, Ultrason. Sonochem., 2014, 21, 1908–1915.
56. R. Prasad and S. V. Dalvi, Chem. Eng. Sci., 2020, 226, 115911.
57. M. Singh, M. R. Dhondale, A. K. Agrawal, S. Panda and D. Kumar, Chem. Eng. Res. Des., 2024, 207, 308–319.
58. G. Ruecroft, D. Hipkiss, T. Ly, N. Maxted and P. W. Cains, Org. Process Res. Dev., 2005, 9, 923–932.
59. A. G. Shtukenberg, M. D. Ward and B. Kahr, J. Cryst. Growth, 2022, 597, 126839.
60. S. Xu, D. Cao, Y. Liu and Y. Wang, Cryst. Growth Des., 2022, 22, 2001–2022.
61. N. Rodríguez-hornedo and D. Murphy, J. Pharm. Sci., 1999, 88, 651–660.
62. A. L. Sarode, P. Wang, S. Obara and D. R. Worthen, Eur. J. Pharm. Biopharm., 2014, 86, 351–360.
63. Y. Wang, F. Xue, S. Yu, Y. Cheng, M. Yin, S. Du and J. Gong, J. Mol. Liq., 2021, 343, 116967.
64. D. Kumar, R. Thipparaboina, S. R. Modi, A. K. Bansal and N. R. Shastri, CrystEngComm, 2015, 17, 1615–1624.
65. M. A. Bellucci, A. Marx, B. Wang, L. Fang, Y. Zhou, C. Greenwell, Z. Li, A. Becker, G. Sun and J. G. Brandenburg, Small Methods, 2023, 7, 2201692.
66. W. Kaialy, H. Larhrib, B. Chikwanha, S. Shojaee and A. Nokhodchi, Int. J. Pharm., 2014, 464, 53–64.
67. L. Y. Pfund, C. P. Price, J. J. Frick and A. J. Matzger, J. Am. Chem. Soc., 2015, 137, 871–875.
68. M. N. Femi-Oyewo and M. Spring, Int. J. Pharm., 1994, 112, 17–28.
69. S. K. Poornachary, V. D. Chia, Y. Yani, G. Han, P. S. Chow and R. B. Tan, Cryst. Growth Des., 2017, 17, 4844–4854.
70. S. Mirza, I. Miroshnyk, J. Heinämäki, J. Rantanen, O. Antikainen, P. Vuorela, H. Vuorela and J. Yliruusi, Cryst. Growth Des., 2008, 8, 3526–3531.
71. K. Smiti and C. Garima, Yakugaku Zasshi, 2008, 128, 281–289.
72. F. Tian, S. Baldursdottir and J. Rantanen, Mol. Pharmaceutics, 2009, 6, 202–210.
73. Y. Wang, H. Zhang, L. Cai, F. Xue, H. Chen, J. Gong and S. Du, Ultrason. Sonochem., 2023, 106475.
74. A. R. Klapwijk, E. Simone, Z. K. Nagy and C. C. Wilson, Cryst. Growth Des., 2016, 16, 4349–4359.
75. W. Zhu, F. S. Romanski, X. Meng, S. Mitra and M. S. Tomassone, Eur. J. Pharm. Sci., 2011, 42, 452–461.
76. A. Kuldipkumar, Y. T. Tan, M. Goldstein, Y. Nagasaki, G. G. Zhang and G. S. Kwon, Cryst. Growth Des., 2005, 5, 1781–1785.
77. J. Jennings, M. F. Butler, M. McLeod, E. Csányi, A. J. Ryan and O. O. Mykhaylyk, Cryst. Growth Des., 2018, 18, 7094–7105.
78. S. Zhang, L. Zhou, W. Yang, C. Xie, Z. Wang, B. Hou, H. Hao, L. Zhou, Y. Bao and Q. Yin, Cryst. Growth Des., 2020, 20, 1609–1617.
79. X. Cheng, X. Huang, Y. Hao, B. Wang, C. Sun, J. Shu and H. Hao, Ind. Eng. Chem. Res., 2022, 61, 7193–7203.
80. D. Prasad, H. Chauhan and E. Atef, Mol. Pharmaceutics, 2016, 13, 756–765.
81. F. Liu, L. Wang, W. Li, M. Li, J. Gong, Y. Wang and D. Han, Cryst. Growth Des., 2021, 21, 3818–3830.
82. E. Simone, M. V. Cenzato and Z. K. Nagy, J. Cryst. Growth, 2016, 446, 50–59.
83. S. Xie, S. K. Poornachary, P. S. Chow and R. B. Tan, Cryst. Growth Des., 2010, 10, 3363–3371.
84. V. Chikhalia, R. T. Forbes, R. A. Storey and M. Ticehurst, Eur. J. Pharm. Sci., 2006, 27, 19–26.
85. K. Patchigolla and D. Wilkinson, Ind. Eng. Chem. Res., 2008, 47, 804–812.
86. M. Sultana and K. F. Jensen, Cryst. Growth Des., 2012, 12, 6260–6266.
87. S. Saifoori, W.-P. Goh, M. Ali and M. Ghadiri, Powder Technol., 2020, 361, 651–662.
88. B. Li, H. Zhang, K. Saranteas and M. A. Henson, Sep. Purif. Technol., 2021, 278, 119462.
89. D. Acevedo, R. Peña, Y. Yang, A. Barton, P. Firth and Z. K. Nagy, Chem. Eng. Process.: Process Intesif., 2016, 108, 212–219.
90. K. S. Gouthami, D. Kumar, R. Thipparaboina, R. B. Chavan and N. R. Shastri, Int. J. Pharm., 2015, 491, 26–34.
91. F. Podczeck and Y. Mia, Int. J. Pharm., 1996, 144, 187–194.
92. M. Manish, J. Harshal and P. Anant, Eur. J. Pharm. Sci., 2005, 25, 41–48.
93. D. Wilson, M. Bunker, D. Milne, A. Jawor-Baczynska, A. Powell, J. Blyth and D. Streather, Powder Technol., 2018, 339, 641–650.
94. M. R. Dhondale, M. Singh, A. K. Agrawal and D. Kumar, Powder Technol., 2025, 457, 120885.
95. K. Kedia and S. Wairkar, Powder Technol., 2019, 344, 665–672.
96. K. K. Moravkar, D. S. Shah, A. G. Magar, B. A. Bhairav, S. D. Korde, K. M. Ranch and S. S. Chalikwar, J. Drug Delivery Sci. Technol., 2022, 71, 103265.
97. A. Nokhodchi, A. Homayouni, R. Araya, W. Kaialy, W. Obeidat and K. Asare-Addo, RSC Adv., 2015, 5, 46119–46131.
98. V. R. Nalluri and M. Kuentz, Eur. J. Pharm. Biopharm., 2010, 74, 388–396.
99. N. Pudasaini, P. P. Upadhyay, C. R. Parker, S. U. Hagen, A. D. Bond and J. Rantanen, Org. Process Res. Dev., 2017, 21, 571–577.
100. O. A. Odeku, Pharm. Rev., 2007, 5, 1–20.
101. United States Pharmacopoeia (USP), 〈1062〉 Tablet Compression Characterization, USP, 2022, pp. 1–15,  DOI:10.31003/USPNF_M99395_02_01.
102. L. Keshavarz, M. Pishnamazi, U. R. Khandavilli, S. Shirazian, M. N. Collins, G. M. Walker and P. J. Frawley, Arab. J. Chem., 2021, 14, 103089.
103. A. Nokhodchi, N. Bolourtchian and R. Dinarvand, Int. J. Pharm., 2003, 250, 85–97.
104. S. R. Modi, K. S. Khomane and A. K. Bansal, Int. J. Pharm., 2014, 460, 189–195.
105. S. R. Datir and M. H. Bele, J. Cryst. Growth, 2023, 606, 127047.
106. N. Rasenack and B. W. Müller, Int. J. Pharm., 2002, 244, 45–57.
107. P. Di Martino, M. Beccerica, E. Joiris, G. F. Palmieri, A. Gayot and S. Martelli, J. Cryst. Growth, 2002, 243, 345–355.
108. S. R. Datir, D. Kumar and M. H. Bele, J. Cryst. Growth, 2022, 592, 126711.
109. S. Jain, Pharm. Sci. Technol. Today, 1999, 2, 20–31.
110. S. Pundir and A. Badola, Int. J. Pharm. Life Sci., 2013, 2, 141–157.
111. H. Chen, S. Paul, H. Xu, K. Wang, M. K. Mahanthappa and C. C. Sun, Mol. Pharmaceutics, 2020, 17, 1387–1396.
112. V. Waknis, E. Chu, R. Schlam, A. Sidorenko, S. Badawy, S. Yin and A. S. Narang, Pharm. Res., 2014, 31, 160–172.
113. M. Roberts, J. L. Ford, G. S. MacLeod, J. T. Fell, G. W. Smith, P. H. Rowe and A. M. Dyas, J. Pharm. Pharmacol., 2004, 56, 947–950.
114. L. Samiei, K. Kelly, L. Taylor, B. Forbes, E. Collins and M. Rowland, Powder Technol., 2017, 305, 509–517.
115. M. U. Ghori, E. Šupuk and B. R. Conway, Eur. J. Pharm. Sci., 2014, 65, 1–8.
116. D. Hooper, F. Clarke, R. Docherty, J. Mitchell and M. Snowden, Int. J. Pharm., 2017, 531, 266–275.
117. A. A. Moldovan and A. G. P. Maloney, Cryst. Growth Des., 2024, 24, 4160–4169.
118. S. Paul, L. J. Taylor, B. Murphy, J. F. Krzyzaniak, N. Dawson, M. P. Mullarney, P. Meenan and C. C. Sun, Mol. Pharmaceutics, 2020, 17, 1148–1158.
119. C. A. Gunawardana, A. Kong, D. Wanapun, D. O. Blackwood, C. T. Powell, J. F. Krzyzaniak, M. C. Thomas, J. E. Kresevic and C. C. Sun, Int. J. Pharm., 2023, 123016.
120. J. Li and Y. Wu, Lubricants, 2014, 2(1), 21–43.
121. H. S. Purohit, G. G. Zhang and Y. Gao, J. Pharm. Sci., 2023, 112, 290–303.
122. C. U. Phan, J. Shen, K. Yu, J. Mao and G. Tang, Molecules, 2021, 26, 3469.
123. H.-H. Chen, C.-S. Su, J.-J. Liu and M.-T. Sheu, J. Supercrit. Fluids, 2015, 101, 17–23.
124. G. Kuminek, G. S. Rauber, M. K. Riekes, C. E. M. de Campos, G. A. Monti, A. J. Bortoluzzi, S. L. Cuffini and S. G. Cardoso, J. Pharm. Biomed. Anal., 2013, 78, 105–111.
125. D. Kumar, R. Thipparaboina, S. R. Modi, A. K. Bansal and N. R. Shastri, J. Cryst. Growth, 2015, 422, 44–51.
126. Y. Gao, B. Glennon, V. K. Kamaraju, G. Hou and P. Donnellan, Org. Process Res. Dev., 2018, 22, 328–336.
127. P. Bukovec, P. Benkič, M. Smrkolj and F. Vrečzer, Pharmazie, 2016, 71, 263–268.
128. R. A. Keraliya, T. G. Soni, V. T. Thakkar and T. R. Gandhi, Dissolution Technol., 2010, 17, 16–21.
129. D. Kumar, R. Thipparaboina and N. R. Shastri, Org. Process Res. Dev., 2015, 19, 1912–1917.
130. R. Adhiyaman and S. K. Basu, Int. J. Pharm., 2006, 321, 27–34.
131. S. R. Datir, D. Kumar, P. Kumar, S. Jain, A. K. Bansal, B. Nallamothu, S. D. Thakore and M. H. Bele, Appl. Sci., 2021, 11, 5604.
132. S. R. Modi, A. K. R. Dantuluri, S. R. Perumalla, C. C. Sun and A. K. Bansal, Cryst. Growth Des., 2014, 14, 5283–5292.
133. M. K. Raval, J. M. Patel, R. K. Parikh and N. R. Sheth, Part. Sci. Technol., 2014, 32, 431–444.
134. J. Hansen and P. Kleinebudde, Eur. J. Pharm. Biopharm., 2021, 159, 170–176.
135. P. P. Upadhyay, N. Pudasaini, M. K. Mishra, U. Ramamurty and J. Rantanen, Chem. Eng. Res. Des., 2018, 136, 447–455.
136. S. Banga, G. Chawla, D. Varandani, B. Mehta and A. K. Bansal, J. Pharm. Pharmacol., 2007, 59, 29–39.
137. V. Verma, R. V. Peddapatla, C. M. Crowley, A. M. Crean, P. Davern, S. Hudson and B. K. Hodnett, Cryst. Growth Des., 2018, 18, 338–350.
138. D. S. Bindra and S. Desikan, Pharm. Dev. Technol., 2015, 20, 129–138.
139. C. Thompson, M. C. Davies, C. J. Roberts, S. J. Tendler and M. J. Wilkinson, Int. J. Pharm., 2004, 280, 137–150.
140. C. N. Bolour, A. Nokhoudchi and R. Dinarvand, Daru, J. Pharm. Sci., 2001, 9, 12–22.
141. A. H. Chow, J. D. Gordon, A. Szeitz and J. W. Young, Int. J. Pharm., 1995, 126, 11–19.
142. A. Watanabe, Y. Yamaoka and K. Takada, Chem. Pharm. Bull., 1982, 30, 2958–2963.
143. N. Bolourchian, M. Nili, S. M. Foroutan, A. Mahboubi and A. Nokhodchi, J. Drug Delivery Sci. Technol., 2020, 55, 101485.
144. P. Bukovec, A. Meden, M. Smrkolj and F. Vrečer, Acta Chim. Slov., 2015, 62, 958–966.
145. L. Lyn, H. Sze, A. Rajendran, G. Adinarayana, K. Dua and S. Garg, Acta Pharm., 2011, 61, 391.
146. Y. Javadzadeh, A. Mohammadi, N. S. Khoei and A. Nokhodchi, Acta Pharm., 2009, 59, 187–197.
147. Y. Ren, J. Shen, K. Yu, C. U. Phan, G. Chen, J. Liu, X. Hu and J. Feng, Crystals, 2019, 9, 556.
148. R. Adhiyaman and S. K. Basu, Int. J. Pharm., 2006, 321, 27–34.
149. A. K. Tiwary and G. M. Panpalia, Pharm. Res., 1999, 16, 261–265.
150. A. K. Tiwary and G. M. Panpalia, Pharm. Res., 1999, 16, 261–265.
151. C. A. Prestidge and G. Tsatouhas, Int. J. Pharm., 2000, 198, 201–212.
152. S. Gestí, A. Almontassir, M. T. Casas and J. Puiggalí, Biomacromolecules, 2006, 7, 799–808.
153. C. G. Jones, in Forensic Microscopy for Skeletal Tissues: Methods and Protocols, ed. L. S. Bell, Humana Press, Totowa, NJ, 2012, pp. 1–20,  DOI:10.1007/978-1-61779-977-8_1.
154. M. S. R. N. Kiran, S. Varughese, C. M. Reddy, U. Ramamurty and G. R. Desiraju, Cryst. Growth Des., 2010, 10, 4650–4655.
155. Y. Zhou, J. Wang, Y. Xiao, T. Wang and X. Huang, Curr. Pharm. Des., 2018, 24, 2375–2382.
156. N. Salunke, R. Thipparaboina, R. B. Chavan, A. Lodagekar, S. Mittapalli, A. Nangia and N. R. Shastri, J. Pharm. Biomed. Anal., 2018, 149, 185–192.
157. A. A. Bunaciu, E. G. Udriştioiu and H. Y. Aboul-Enein, Crit. Rev. Anal. Chem., 2015, 45, 289–299.
158. S. Halder and S. Ray, Int. J. Ceram. Eng. Sci., 2022, 4, 309–314.
159. C. F. Holder and R. E. Schaak, ACS Nano, 2019, 13, 7359–7365.
160. D. Faroongsarng, AAPS PharmSciTech, 2016, 17, 572–577.
161. N. Saadatkhah, A. Carillo Garcia, S. Ackermann, P. Leclerc, M. Latifi, S. Samih, G. S. Patience and J. Chaouki, Can. J. Chem. Eng., 2020, 98, 34–43.
162. H. M. Beakawi Al-Hashemi and O. S. Baghabra Al-Amoudi, Powder Technol., 2018, 330, 397–417.
163. R. B. Shah, M. A. Tawakkul and M. A. Khan, AAPS PharmSciTech, 2008, 9, 250–258.
164. D. Schulze, in Powders and Bulk Solids: Behavior, Characterization, Storage and Flow, ed. D. Schulze, Springer International Publishing, Cham, 2021, pp. 57–100,  DOI:10.1007/978-3-030-76720-4_3.
165. D. Schulze, Annu. Trans. – Nord. Rheol. Soc., 2013, 21, 99–106.
166. A. R. Balkenende, H. J. A. P. van de Boogaard, M. Scholten and N. P. Willard, Langmuir, 1998, 14, 5907–5912.
167. F. M. Etzler, Rev. Adhes. Adhes., 2013, 1, 3–45.
168. F. A. Stevie and C. L. Donley, J. Vac. Sci. Technol., A, 2020, 38, 063204.
169. K. Sing, Colloids Surf., A, 2001, 187–188, 3–9.
170. United States Pharmacopoeia (USP), 〈1087〉 Intrinsic Dissolution—Dissolution Testing Procedures for Rotating Disk and Stationary Disk, USP, 2020, pp. 1–7,  DOI:10.31003/USPNF_M99806_02_01.
171. K. Tsinman, A. Avdeef, O. Tsinman and D. Voloboy, Pharm. Res., 2009, 26, 2093–2100.
172. D. R. Serrano, T. Persoons, D. M. D'Arcy, C. Galiana, M. A. Dea-Ayuela and A. M. Healy, Eur. J. Pharm. Sci., 2016, 89, 125–136.
173. D. M. D'Arcy and T. Persoons, J. Pharm. Sci., 2011, 100, 1102–1115.
174. S. Patel, A. M. Kaushal and A. K. Bansal, Crit. Rev. Ther. Drug Carrier Syst., 2006, 23, 1–66.
175. H. Triboandas, K. Pitt, M. Bezerra, D. Ach-Hubert and W. Schlindwein, Pharmaceutics, 2022, 14, 2398.
176. G. Vreeman, C. Wang, C. M. Reddy and C. C. Sun, Cryst. Growth Des., 2021, 21, 6655–6659.
177. USFDA, Drug Substance Chemistry, Manufacturing, and Controls Information, Guidance for Industry, 2010, pp. 1–59.
178. D. C. Montgomery, Design and analysis of experiments, John Wiley & Sons, 2017.
179. R. Snee and R. Hoerl, Strategies for formulations development: a step-by-step guide using JMP, SAS Institute, 2016.

---

Footnote

† Joint first authors.

---