Division of Pharmaceutical Technology Faculty of Pharmacy University of Helsinki From Polymorph Screening to Dissolution Testing Solid Phase Analysis during Pharmaceutical Development and Manufacturing by Jaakko Aaltonen ACADEMIC DISSERTATION To be presented, with the permission of the Faculty of Pharmacy of the University of Helsinki, for public criticism in auditorium 2 at Viikki Infocentre (Viikinkaari 11), on March 30 th 2007, at 12 noon. Helsinki 2007
Transcript
Division of Pharmaceutical Technology Faculty of Pharmacy University of Helsinki
From Polymorph Screening
to Dissolution Testing
Solid Phase Analysis during Pharmaceutical Development and Manufacturing
by
Jaakko Aaltonen
ACADEMIC DISSERTATION
To be presented, with the permission of the Faculty of Pharmacy of the University of Helsinki, for public criticism in auditorium 2 at Viikki Infocentre (Viikinkaari 11), on
March 30th 2007, at 12 noon.
Helsinki 2007
Supervisors: Professor Jukka Rantanen1, 2
1Department of Pharmaceutics and Analytical Chemistry Faculty of Pharmaceutical Sciences
University of Copenhagen Denmark
2Drug Discovery and Development Technology Center Faculty of Pharmacy
University of Helsinki Finland Professor Jouko Yliruusi Division of Pharmaceutical Technology Faculty of Pharmacy University of Helsinki Finland
Reviewers: Associate Professor Lars Hovgaard Department of Pharmaceutics and Analytical Chemistry
Faculty of Pharmaceutical Sciences University of Copenhagen Denmark
Professor Jari Yli-Kauhaluoma
Division of Pharmaceutical Chemistry Faculty of Pharmacy
University of Helsinki Finland
Opponent: Professor Thomas Rades School of Pharmacy
Helsinki University Printing House Helsinki, Finland, 2007
Abstract
Aaltonen, J.J., 2007. From Polymorph Screening to Dissolution Testing – Solid Phase Analysis during Pharmaceutical Development and Manufacturing.
Dissertationes bioscientiarum molecularium Universitatis Helsingiensis in Viikki, 9/2007, 41 pp. ISBN 978-952-10-3806-8 (paperback), ISBN 978-952-10-3807-5 (pdf), ISSN 1795-7079
Solid materials can exist in different physical structures without a change in chemical composition. This phenomenon, known as polymorphism, has several implications on pharmaceutical development and manufacturing. Various solid forms of a drug can possess different physical and chemical properties, which may affect processing characteristics and stability, as well as the performance of a drug in the human body. Therefore, knowledge and control of the solid forms is fundamental to maintain safety and high quality of pharmaceuticals. During manufacture, harsh conditions can give rise to unexpected solid phase transformations and therefore change the behavior of the drug. Traditionally, pharmaceutical production has relied on time-consuming off-line analysis of production batches and finished products. This has led to poor understanding of processes and drug products. Therefore, new powerful methods that enable real time monitoring of pharmaceuticals during manufacturing processes are greatly needed.
The aim of this thesis was to apply spectroscopic techniques to solid phase analysis within different stages of drug development and manufacturing, and thus, provide a molecular level insight into the behavior of active pharmaceutical ingredients (APIs) during processing. Applications to polymorph screening and different unit operations were developed and studied. A new approach to dissolution testing, which involves simultaneous measurement of drug concentration in the dissolution medium and in-situ solid phase analysis of the dissolving sample, was introduced and studied.
Solid phase analysis was successfully performed during different stages, enabling a molecular level insight into the occurring phenomena. Near-infrared (NIR) spectroscopy was utilized in screening of polymorphs and processing-induced transformations (PITs). Polymorph screening was also studied with NIR and Raman spectroscopy in tandem. Quantitative solid phase analysis during fluidized bed drying was performed with in-line NIR and Raman spectroscopy and partial least squares (PLS) regression, and different dehydration mechanisms were studied using in-situ spectroscopy and partial least squares discriminant analysis (PLS-DA). In situ solid phase analysis with Raman spectroscopy during dissolution testing enabled analysis of dissolution as a whole, and provided a scientific explanation for changes in the dissolution rate. It was concluded that the methods applied and studied provide better process understanding and knowledge of the drug products, and therefore, a way to achieve better quality.
-
i
Table of contents
Abstract.................................................................................................. i
Table of contents ................................................................................. ii
Abbreviations...................................................................................... iv
List of original publications .................................................................v
4.1 Materials .............................................................................................. 13 4.1.1 Raw materials.......................................................................................................... 13 4.1.2 Preparation of solid forms used in the studies ..................................................... 13 4.1.3 Preparation of granules for fluidized bed drying studies (III) ............................. 13 4.1.4 Preparation of powder compacts for dissolution studies (V) .............................. 14
4.2 Methods of analysis ............................................................................ 14 4.2.1 Near-infrared (NIR) spectroscopy (I-IV).................................................................. 14 4.2.2 Raman spectroscopy (II-V) ...................................................................................... 14 4.2.3 Terahertz pulsed spectroscopy (TPS) (II) ................................................................ 14 4.2.4 X-ray powder diffraction and crystal structure verification (I-V) ........................ 15 4.2.5 Thermal analysis (I, III, IV) ....................................................................................... 15 4.2.6 Water content analysis (III, IV) ............................................................................... 15 4.2.7 Scanning electron microscopy (V).......................................................................... 16
4.3 Polymorph screening, processing, and dissolution testing............. 16 4.3.1 Polymorph screening (I, II)...................................................................................... 16 4.3.2 Milling and compression (I) .................................................................................... 16 4.3.3 Fluidized bed drying (III)......................................................................................... 17 4.3.4 Hot stage dehydration (IV)..................................................................................... 17 4.3.5 Dissolution testing (V) ............................................................................................ 17
4.4 Data processing ................................................................................... 17 4.4.1 Spectral treatment .................................................................................................. 17 4.4.2 Qualitative and quantitative methods .................................................................. 18
ii
5. Results and discussion ................................................................... 19
5.1 Solid phase analysis using spectroscopy ...........................................19
5.2 Spectroscopic methods as polymorph screening tools ....................21
5.3 Spectroscopic analysis of processing-induced transformations......23 5.3.1 The effect of milling and compression.................................................................. 23 5.3.2 Solid phase analysis during fluidized bed drying ................................................. 24 5.3.3 Investigation of dehydration mechanisms............................................................ 26
5.4 In-situ solid phase analysis during dissolution testing....................27
5.5 Interfacing spectroscopic tools with different environments.........29
API active pharmaceutical ingredient CBZ carbamazepine CSD Cambridge Structural Database FDA Food and Drug Administration ICH International Conference on Harmonisation IDR intrinsic dissolution rate MSC multiplicative scatter correction NF nitrofurantoin NIR near-infrared spectroscopy PAT process analytical technology PIT processing-induced transformation PCA principal component analysis PLS partial least squares PLS-DA partial least squares discriminant analysis PRX piroxicam SNV standard normal variate STZ sulfathiazole TP theophylline TPS terahertz pulsed spectroscopy VT-XRPD variable temperature X-ray powder diffraction XRPD X-ray powder diffraction
iv
List of original publications
This thesis is based on the following publications, which are referred to in the text by their respective Roman numerals (I-V).
I Aaltonen J, Rantanen J, Siiriä S, Karjalainen M, Jørgensen A, Laitinen N,
Savolainen M, Seitavuopio P, Louhi-Kultanen M, Yliruusi J. Polymorph screening using near-infrared spectroscopy. Analytical Chemistry 75 (2003) 5267-5273. doi: 10.1021/ac034205c
II Aaltonen J, Strachan CJ, Pöllänen K, Yliruusi J, Rantanen J. Hyphenated
spectroscopy as a polymorph screening tool. Journal of Pharmaceutical and Biomedical Analysis. In press. doi: 10.1016/j.jpba.2007.02.009
III Aaltonen J, Kogermann K, Strachan CJ, Rantanen J. In-line monitoring of solid-
state transitions during fluidisation. Chemical Engineering Science 62 (2007) 408-415. doi: 10.1016/j.ces.2006.08.061
Yliruusi J, Rantanen J. Qualitative in situ analysis of multiple solid-state forms using spectroscopy and partial least squares discriminant modeling. Journal of Pharmaceutical Sciences. In press. doi: 10.1002/jps.20840
Hirvonen J, Yliruusi J, Rantanen J. In-situ measurement of solvent-mediated phase transformations during dissolution testing. Journal of Pharmaceutical Sciences 95 (2006) 2730-2737. doi: 10.1002/jps.20725
Reprinted with permission from the American Chemical Society (I), Elsevier Ltd. (II, III), and John Wiley & Sons, Inc. (IV, V).
v
INTRODUCTION
1. Introduction
Research and development of a new drug is an expensive and lengthy process. On average it takes 10-15 years for a drug to progress from laboratory to pharmacy, and the success rate is low: only five per 5000 compounds screened enter clinical trials and one of those five eventually enters the market (PhRMA, 2005). Research and development costs of a drug have been estimated to exceed US $800 million (DiMasi et al., 2003). Considering the size of the pharmaceutical industry – in 2005 the global sales of pharmaceuticals was worth US $602 billion (IMS Health, 2006) – and the fact that medication safety is of extreme importance, it is somewhat surprising that quality control and assurance still relies quite heavily on off-line analysis of production batches and finished products (Abboud and Hensley, 2003). Such analyses are time-consuming and provide very little information on product behavior during processing. If a production batch does not meet specifications it has to be discarded. In such a case, information lacking off-line analyses provide little help for troubleshooting. Therefore, new and efficient methods of process analysis are truly needed within the pharmaceutical industry.
Decisions determining the directions of further development of the drug product are made at an early stage of drug development. During preformulation stage physicochemical properties of candidate drugs and the conditions under which the candidate drug should be formulated are investigated. Key issues include investigation of polymorphism, the ability of a compound to exist in more than one crystalline form, and careful selection of the solid form for further development. The term polymorphism is often used broadly to refer to, in addition to true crystalline polymorphs, also to other solid forms (hydrates/solvates and amorphous form) (Hilfiker, 2006). The selection of the solid form of the active pharmaceutical ingredient (API) is made after polymorph screening, during which several solid forms of the candidate drug are generated and analyzed. Even though it is called polymorph screening, it is understood that other solid forms cannot be neglected, since they may possess properties which make them the most suitable for development.
During various unit operations used in pharmaceutical production the materials are subject to mechanical and/or thermal stress, high moisture content and contact with solvents, which set the scene for solid phase transformations from one solid form to another. Such processing-induced transformations (PITs) are well known but difficult to predict or control (Morris et al., 2001). On these accounts it should be obvious that monitoring of the solid forms should not be stopped once the desired solid form is chosen after polymorph screening. In order to fully understand the behavior of a given drug the solid state of the API should be continuously monitored and analyzed as long as it exists as a solid – from crystallization, via various process steps and shelf-life, until the drug is dissolved after administration (in vitro dissolution testing) (figure 1).
1
INTRODUCTION
Figure 1. A simplified graph of various steps an API may be subjected to during pharmaceutical development and manufacturing. To fully understand the behavior of the API and the final dosage form, solid phase analysis should be carried out during every step.
Spectroscopic techniques enable continuous and real time solid phase analysis during processing. Despite recent advances in manufacturing and analytical technologies the industry has been reluctant to make use of novel methods that would enable better understanding of products and processes, and hence better quality. This has been due to a strict regulatory environment unfavorable to introduction of innovative systems (Abboud and Hensley, 2003), and also the high cost of introducing new methods, even though in long term it would probably be beneficial. Lately, however, regulatory authorities have taken the initiatives and published guidelines to encourage pharmaceutical manufacturers to utilize advanced methods to ensure high drug product quality (FDA, 2004a; ICH, 2005a; ICH, 2005b). The process analytical technology (PAT) guideline emphasizes the importance of real time monitoring of manufacturing processes (FDA, 2004b). In addition to above-mentioned economical and regulatory issues, there is still one more obstacle on the way to successful implementation of these advanced techniques: the cultural change needed in the laboratories. This may, in fact, be the most important one.
In this thesis, approaches to solid phase analysis using spectroscopic methods are taken. Case studies of polymorph screening, pharmaceutical unit operations, as well as dissolution testing are provided. The literature review focuses on the solid phase, importance of solid-state properties, and generation of different solid forms. Basic principles of the measurement techniques are well established, and therefore left out.
2
THEORY AND LITERATURE REVIEW
2. Theory and literature review
2.1 Solid phase
A phase is defined as a homogeneous, physically distinct and mechanically separable portion of a system. Examples are gases, liquids and solids. Solid phase can be crystalline or amorphous. In the crystalline state (polymorphs) the constituent molecules are regularly arranged into a fixed and rigid repeating array known as a lattice, whereas the amorphous state lacks a definite long-range order (Byrn et al., 1999b; Grant, 1999). The smallest three-dimensional volume element from which crystals can be built is the unit cell. When a volatile solvent molecule occupies regular positions in the crystal lattice, the solid is called a solvate, and if the volatile solvent molecule is water, the term hydrate is used. Figure 2 is a schematic representation of the solid forms mentioned.
Figure 2. Arrangement of molecules in different types of solid material.
2.1.1 Polymorphs
Polymorphism in molecular crystals is defined as the ability of a compound to crystallize in more than one crystalline phase (Bernstein, 2002b). In the strictest sense polymorphism refers only to crystalline solid forms that accompany molecules of only one chemical species in the crystal lattice. According to McCrone, every compound has different polymorphic forms, and the number of forms known for a given compound is proportional to the time and money spent in research on that compound (McCrone, 1965). A good example of this is ROY (Chen et al., 2005) a compound that, at present, has seven reported crystal structures in Cambridge Structural Database (CSD) (Allen, 2002).
3
THEORY AND LITERATURE REVIEW
There are two ways in which different crystal lattices can be formed: packing polymorphism and conformational polymorphism. Packing polymorphism denotes crystal forms of a compound where rigid molecules with uniform conformation are packed in different arrangements, whereas conformational polymorphism refers to crystal forms where flexible molecules with different conformations are packed in different arrangements (Grant, 1999). Known examples of packing and conformational polymorphism are paracetamol (Boldyreva et al., 2002) and l-glutamic acid, respectively (Davey et al., 1997).
The Gibbs free energy of a polymorph is defined as
G = H – TS (1)
where G is the Gibbs free energy, H is the enthalpy and TS is the entropy term (fig. 3). Thermodynamically, only one polymorph is stable at a given temperature and a constant pressure. The stable form is that with the lowest Gibbs free energy in the given conditions and is the least soluble form. The order of polymorph stability is defined by the difference in Gibbs free energies
∆G = ∆H – T∆S (2)
In an enantiotropic polymorphic system the stable polymorph is different above and below a transition temperature. At the transition temperature, ∆G = 0 (fig. 4a). In a monotropic system one polymorph is always more stable than the other below the melting point of either form (fig. 4b). In addition to thermodynamics the existence of different polymorphs is controlled by kinetics. Therefore, other thermodynamically unstable forms with a higher Gibbs free energy can exist in the given conditions, and are known as metastable forms.
Figure 3. Energy-temperature plot of a crystalline solid under constant pressure. Modified from Lohani and Grant (2006).
4
THEORY AND LITERATURE REVIEW
Figure 4. Energy–temperature diagram of (a) an enantiotropic system and (b) a monotropic system where G is the Gibbs free energy; T is the temperature; A, B and Liq refer to polymorph A, polymorph B, and the liquid phase, respectively; Tt is the transition point between A and B, and TmA and TmB are the melting points of A and B, respectively. Tt of the monotropic system is virtual since it is above TmA and TmB. Modified from Burger and Ramberger (1979).
2.1.2 Hydrates
Hydrates are the most important subclass of solvates. This is due to water being the most common solvent used in processing, and also because of the omnipresence of water in the environment. In addition, the small size of the water molecule and its multidirectional hydrogen bonding capability facilitate incorporation into the crystal lattice and stabilization of the crystal structure (Byrn et al., 1999a; Gillon et al., 2003). It has been estimated that every third drug molecule is capable of forming a hydrate (Stahl, 1980). Hydrates and solvates also exhibit polymorphism.
Hydrates can be classified into three categories: 1) isolated site hydrates, 2) channel hydrates and 3) ion associated hydrates (Morris, 1999). In the crystal structure of an isolated site hydrate the water molecules are isolated from direct contact with other water molecules by the API molecules, whereas in channel hydrates the water molecules are located next to each other along one direction in the lattice. Hydrates can contain stoichiometric or nonstoichiometric amounts of water molecules per API molecule. An expanded channel hydrate can take up water into the channels when exposed to high humidity and release water when exposed to relatively low humidity (Datta and Grant, 2004). The crystal lattice of such hydrates can expand or contract as hydrate formation or dehydration proceeds, changing the dimensions of the unit cell. In ion associated hydrates, the water molecules are coordinated by ions incorporated in the crystal lattice.
5
THEORY AND LITERATURE REVIEW
The stability of hydrates is governed by the water activity of the environment, and the intermolecular bonding and arrangement of molecules in the lattice. Therefore, the critical water activity that determines whether the stable form is the anhydrate or the hydrate (or lower hydrate / higher hydrate) is drug-specific. Hydrates are more stable than their anhydrous counterparts at conditions below dehydration temperature and above critical water activity, such as high humidity or saturated aqueous solution. Isolated site hydrates usually dehydrate at relatively high temperatures. The dehydration process of such hydrates is destructive for the crystal structure since it requires rearrangement of the molecules in the unit cell in order to allow water molecules to escape the lattice. An example is dehydration process of nitrofurantoin monohydrate, during which a partially amorphous intermediate form is detected (Kishi et al., 2002; Karjalainen et al., 2005). Channel hydrates usually dehydrate at lower temperatures than isolated site hydrates. Theophylline monohydrate is an example of a channel hydrate. It dehydrates (via a metastable phase) at lower temperatures than nitrofurantoin monohydrate, and even though an intermediate metastable form is detected, the dehydration process does not involve a collapse of the crystal structure (Phadnis and Suryanarayanan, 1997). In ion associated hydrates the bonding between the metal ion and the water molecule can be strong, which results in very high dehydration temperatures (Morris, 1999).
2.1.3 Amorphous solids
Solid materials can be intrinsically amorphous or intentionally made amorphous by physical manipulation (Hancock and Zografi, 1997). Amorphous forms are metastable, higher-energy forms than their crystalline counterparts and tend to release their energy through crystallization. They are hygroscopic and absorb moisture which acts as a plasticizer and promotes molecular mobility, which leads to crystallization. Thus humidity and aqueous environments further reduce the stability of amorphous solids. Even though amorphous forms lack definite long-range order, they typically exhibit short-range order over a few molecular dimensions. Organic compounds have not been shown to exhibit polyamorphism (characterized by a transition between two amorphous states), which has been reported for inorganic compounds (Petit and Coquerel, 2006).
2.2 Importance of solid-state properties
Different solid forms have different physicochemical properties that may significantly affect the performance of an API (Haleblian and McCrone, 1969; Threlfall, 1995). Examples of these properties are listed in table 1.
6
THEORY AND LITERATURE REVIEW
Table 1. Physical properties that differ among various solid forms. Modified from Grant (1999).
Packing properties Unit cell volume (crystalline forms only), density, refractive
index, hygroscopicity
Thermodynamic properties Melting point, enthalpy, entropy, free energy, solubility
The differences in the mechanical properties can affect processing behavior, which is the case with paracetamol polymorphs (Di Martino et al., 1996), where form II is suitable for direct tablet compression, but form I needs to be mixed with binding agents prior to compression (Nichols and Frampton, 1998). Polymorphs I and II of sulfamerazine have also been reported to show different tableting properties (Sun and Grant, 2001), as well as different forms of theophylline (Suihko et al., 2001).
The performance of the drug in the human body can depend on the solid form, if solubility and dissolution rate are dependent on the solid state of the API. Well-known examples are carbamazepine (Kahela et al., 1983; Meyer et al., 1992) and ritonavir (Chemburkar et al., 2000).
Physical stability may depend on the solid form. This is an important issue since solid dosage forms may be stored for prolonged periods and naturally, the solid state of the API must not change during its shelf-life. The poor stability of amorphous forms is one of the biggest hurdles to developing formulations with amorphous APIs (Kaushal et al., 2004). In addition to physical stability, chemical stability can also be different because chemical reactivity can vary between different solid forms (Byrn et al., 2001). For example, chemical stability differences between polymorphs of prednisolone tert-butylacetate (Byrn et al., 1988), and between crystalline and amorphous form of quinapril hydrochloride (Guo et al., 2000) have been published.
Different solid forms of an API also have major economic implications, since they can be patented. Different interpretations of polymorph patents have caused controversy between drug manufacturers. The best-known case is polymorphism of ranitidine hydrochloride, which has given rise to several patent trials (Bernstein, 2002a). Unexpected solid phase transformations, sudden appearance or disappearance of polymorphs, and sometimes, differences in determining polymorphic purity, can make patent interpretation everything but simple.
7
THEORY AND LITERATURE REVIEW
2.3 Generation of different solid forms
2.3.1 Polymorph screening
The purpose of polymorph screening is to, first, find the different solid forms the API may exhibit, and second, choose the form most suitable for further development (Hilfiker et al., 2006). Even though the stable form would normally be the safest choice, there are cases in which other forms are more suitable (Singhal and Curatolo, 2004). In this situation, a metastable form is usually selected for an enhanced solubility or dissolution rate.
Polymorph screening can be approached either experimentally or by computational methods (Price, 2004). Even though computational methods have advanced markedly during the last years, experimental work is still needed since the difference in lattice energy between polymorphs can be small making polymorph prediction very challenging. Furthermore, crystallization phenomena, especially nucleation, are not yet fully understood (Davey, 2004). A combination of crystal structure prediction and targeted experimental crystallizations has been published (Cross et al., 2003). However, since properties are compound-specific, there are no common guidelines how to perform experimental polymorph screening. Some general suggestions and guidelines for systematic screening have been published (Byrn et al., 1995; ICH, 1999; Yu et al., 2003).
There are several approaches to produce different solid forms of an API (table 2). During polymorph screening different methods with varying parameters should be trialed. Crystallization from solution is the most widely used method (Mullin, 2001), with the solvent the most influential variable. Various solvents with different properties are used in polymorph screening experiments, but unfortunately publications on rational solvent selection are rare. Principal component analysis (Carlson et al., 1985) and cluster analysis (Gu et al., 2004) basing on molecular descriptors have been used for solvent classification.
High-throughput polymorph screening
High-throughput methods for polymorph screening have been developed to perform polymorph screening rapidly with a small amount of drug. High-throughput polymorph screening is usually performed with automated crystallization stations that include robotic systems for liquid handling and solid phase analysis (Storey et al., 2004). Possible bottlenecks during high-throughput polymorph screening are the experimental crystallizations, solid form analysis, and data analysis. High-throughput crystallizations can be performed on well-plate systems, or at an even smaller scale such as on a chip (Lee et al., 2006), or self-assembled monolayers (Lee et al., 2005). Crystallizations on small scale certainly bring saving in time and material, but scale-up problems are probable, since small-scale crystallizations cannot always be reproduced at the production scale. The gold standard in solid phase analysis is X-ray powder diffraction (XRPD), but Raman spectroscopy is receiving more attention because much smaller scale solid phase analysis is feasible (Peterson et al., 2002;
8
THEORY AND LITERATURE REVIEW
Anquetil et al., 2003; Remenar et al., 2003). Regardless of the method of analysis a large amount of data is produced, and therefore different multivariate methods for data analysis and pattern identification have been suggested (Peterson et al., 2002; Remenar et al., 2003; Barr et al., 2004a; Barr et al., 2004c; Barr et al., 2004b; Gilmore et al., 2004; Ivanisevic et al., 2005).
Table 2. Methods to prepare various solid forms (Guillory, 1999; Mullin, 2001; Yu, 2001; Hilfiker et al., 2006). Crystalline forms can be prepared by alllisted methods and those marked with (A) can also be used to prepare amorphous forms.
Suspension equilibration Solvent, temperature, solubility, temperature
programs, mixing, equilibration time
Crystallization from the melt Temperature changes (min, max, gradients)
Quench cooling the melt (A) Cooling rate
Heat induced transformations Temperature changes
Sublimation Temperature gradient, pressure, surface type
Desolvation of solvates (A) Temperature, pressure
pH change (A) Temperature, rate of change, acid / conjugate base ratio
Mechanical treatment (A)
(i.e. milling, cryo-grinding)
Milling time, mill type
Freeze-drying (A) Solvent, concentration, temperature programs
Spray drying (A) Solvent, concentration, drying temperature
2.3.2 Solid phase transformations during processing
In this chapter, examples of pharmaceutical PITs are reviewed. Process steps that follow the preparation step of the initial solid form are discussed, but preparative process steps such as crystallization, freeze drying and spray drying are not. Dissolution testing is also included in this chapter even though it is not a manufacturing process. However, all solid dosage forms that are administered orally
9
THEORY AND LITERATURE REVIEW
are subject to aqueous environment of the gastrointestinal tract in which solid phase transformations can occur, as well as during in vitro dissolution testing. In addition to processing, APIs may undergo solid phase transformations during storage and shelf-life, depending on the storage conditions.
During various process steps materials are subject to mechanical stress, temperature changes and different humidities. The different PITs that can occur in the solid state are polymorphic transformation, dehydration/desolvation, hydrate/solvate formation, and alteration of the crystallinity (crystallization of amorphous form and vice versa) (table 3). Multiple solid phase transformations during one process step, such as wet granulation, are also possible. Solid phase transformations can be solid-solid transformations or solvent-mediated transformations. Solid-solid transformations can occur towards higher or lower energy forms, but solvent-mediated transformations can only proceed towards lower energy and the stable form. The solvent-mediated transformation requires dissolution of the metastable phase, and, after the solution has become supersaturated with respect to the stable phase (having lower solubility), nucleation and growth of the stable phase (Cardew and Davey, 1985). The kinetics of such transformations is determined by the dissolution rate of the metastable phase and the growth rate of the stable phase (Davey et a ., 1986). l
Table 3. Types of processing-induced solid phase transformations. Transformations of solvates other than hydrates are not mentioned (Morris et al., 2001; Zhang et al., 2004; Govindarajan and Suryanarayanan, 2006).
Phase transformation Mechanism
Transformations in the crystalline state
(polymorphic transformation,
hydrate formation, dehydration)
Solid-solid transformation
Solvent-mediated transformation
Transformation via melt
Decrease in the long range order
(amorphization)
Solid-solid transformation (i.e. mechanical
processing induced disorder)
Transformation via melt
Increase in the long range order
(crystallization of amorphous form)
Solid-solid transformation (i.e. heat or plasticizer
induced crystallization)
Solvent-mediated crystallization
Polymorphic transformations have been reported to occur during compression (Chan and Doelker, 1985; Otsuka et al., 1989; Matsumoto et al., 1991), milling (Otsuka and Kaneniwa, 1986; Shakhtshneider and Boldyrev, 1993; Bauer-Brandl, 1996), wet granulation (Davis et al., 2003; Otsuka et al., 2004), the drying phase of wet granulation (Davis et al., 2004) and dissolution testing (Lagas and Lerk, 1981). Hydrate formation during wet granulation has been reported (Herman et al., 1988; Wong and Mitchell, 1992; Otsuka et al., 1999a; Räsänen et al., 2001; Jørgensen et al., 2002; Airaksinen et al., 2003; Jørgensen et al., 2004; Airaksinen et al., 2005; Sandler
10
THEORY AND LITERATURE REVIEW
et al., 2005; Wikström et al., 2005), as well as during dissolution (Shefter and Higuchi, 1963; Kahela et al., 1983; De Smidt et al., 1986; Leung et al., 1998). Dehydration during different drying processes has been observed (Räsänen et al., 2003; Airaksinen et al., 2004; Hausman et al., 2005), and also dehydration induced by mechanical treatment (milling and compression) (Ketolainen et al., 1995). Milling and grinding have been shown to induce formation of amorphous material (Ward and Schultz, 1995; Ticehurst et al., 2000; Mackin et al., 2002; Brodka-Pfeiffer et al., 2003; Steckel et al., 2003; Ohta and Buckton, 2004; Chikhalia et al., 2006). Humidity and temperature affect the solid state of the end product of milling: milling of carbamazepine induces formation of amorphous regions, but in humid conditions the end product is predominantly the dihydrate (Otsuka et al., 1999b). Low temperatures promote low molecular mobility and hence stabilize amorphous solids (Andronis and Zografi, 1998). Crystallization of the amorphous form during dissolution testing has been observed (Stagner and Guillory, 1979). The transformation from the amorphous form to the crystalline form can occur very rapidly after administration and therefore the solubility advantage of amorphous solids may not always be realized (Hancock and Parks, 2000). The various cases listed above show that the occurrence of PITs is by no means rare, and therefore, should be continuously monitored.
11
AIMS OF THE STUDY
3. Aims of the study
The aim of this thesis was to apply spectroscopic techniques to solid phase analysis within different stages of drug development and manufacturing, and thus, provide a molecular level insight into the behavior of APIs during processing. The specific aims of the publications included in the thesis were:
to use near-infrared (NIR) spectroscopy and multivariate methods as tools in
screening of polymorphs and processing-induced transformations to use complementary methods, NIR and Raman spectroscopy, in tandem
instead of using them in parallel during polymorph screening to quantitatively analyze solid phase transformations during fluidized bed
drying with in-line NIR and Raman spectroscopy
to investigate dehydration mechanisms with NIR and Raman spectroscopy in situ
to simultaneously measure the drug concentration in the dissolution medium
and quantitate solid phase transformations in situ with Raman spectroscopy during dissolution testing, and use the direct solid phase analysis to explain the changes in the dissolution rate
12
EXPERIMENTAL
4. Experimental
Experimental details are reported in the original publications (I-V). Polymorphs of sulfathiazole are referred to using their corresponding CSD refcodes to avoid confusion due to complicated nomenclature in the literature.
4.1 Materials
4.1.1 Raw materials
The model compounds used in the studies were sulfathiazole (STZ) (Orion Pharma, Espoo, Finland) (I), theophylline (TP) (BASF, Ludwigshafen, Germany) (III, V), nitrofurantoin (NF) (Sigma-Aldrich, Seelze, Germany) (II, V), carbamazepine (CBZ) and piroxicam (PRX) (Hawkins, Minneapolis, MN, USA) (IV). The solvents used were purified water (I-V), acetone (Sigma-Aldrich, Seelze, Germany) (I, II), 1-propanol (Sigma-Aldrich, Seelze, Germany) (I), ethanol (99.5%; Altia, Rajamäki, Finland) (I), and 2-propanol (Rathburn Chemicals, Peebleshire, UK) (I). Microcrystalline cellulose (MCC) (Penwest Pharmaceuticals, Nastola, Finland) was used as a tablet excipient (V).
4.1.2 Preparation of solid forms used in the studies
In the polymorph screening studies STZ and NF were used as received. PRX monohydrate, CBZ dihydrate, TP monohydrate, and NF monohydrate were crystallized from aqueous solutions. TP anhydrate, NF anhydrate (form β), PRX anhydrate (form I), and CBZ anhydrate (form III) were obtained by dehydration of their respective monohydrate forms at 100 ºC (TP, PRX, and CBZ) or 130 ºC (NF) under reduced pressure (72 mbar), for 24 hours. The high temperature form of CBZ (form I) was obtained by heating the raw material at 170 ºC for 2 h. Amorphous CBZ was prepared by quench cooling of the melt in liquid nitrogen.
4.1.3 Preparation of granules for fluidized bed drying studies (III)
During wet granulation, TP anhydrate undergoes a solvent-mediated solid-state transformation into TP monohydrate (Herman et al., 1988). To obtain monohydrate granules, raw material was granulated by adding 100 g of purified water to 300 g of the dry TP powder at a constant rate (20 g/min) in a planetary mixer (Kenwood KM400, Kenwood Ltd., UK). After the addition of water, the wet mass was mixed for five minutes. The resulting granules were dried in ambient conditions for 24 h, and a sieved fraction between 1 and 2 mm was used in the fluidized bed drying.
13
EXPERIMENTAL
4.1.4 Preparation of powder compacts for dissolution studies (V)
Powder compacts for dissolution tests were compressed with a single punch tablet machine (Korsch EK0, Erweka Apparatebau, Germany) using flat-faced punches (punch diameter = 9 mm). The compacts (m = 150 mg) were compressed to a crushing strength of 70 N. The compositions of the powder compacts were composed of: 100% TP anhydrate, 100% TP monohydrate, 1:1 (w/w) mixture of TP anhydrate and MCC, or 100% NF anhydrate.
4.2 Methods of analysis
4.2.1 Near-infrared (NIR) spectroscopy (I-IV)
Two different NIR spectrometers were used in the studies. In paper I an FT-NIR spectrometer (MB-160 DX, Bomem, Quebec, Canada) was used, whereas in papers II-IV NIR spectrometry was performed with a process-type NIR spectrometer (Control Development, South Bend, IN, USA) with a fiber optic probe. NIR spectra were recorded of samples in glass vials through the bottom of the vial (diameter of the sampling area = 5 mm) (I), crystals on a rotating sample holder (II), granules in a fluidized bed dryer through a sight window during drying process (III), and crystals on a hot stage during dehydration (IV).
4.2.2 Raman spectroscopy (II-V)
A process-type Raman spectrometer (Control Development Inc., South Bend, IN, USA) equipped with a fiber optic probe (RamanProbe, InPhotonics, Norwood, MA, USA) and a 500 mW laser source (Starbright 785S, Torsana Laser Technologies, Skodsborg, Denmark) working at a wavelength of 785 nm was used. Raman spectra were recorded of crystals on a rotating sample holder (II), granules in a fluidized bed dryer through a sight window during drying process (III), crystals on a hot stage during dehydration (IV), and powder compacts in dissolution vessel through a sight window during dissolution testing (V).
4.2.3 Terahertz pulsed spectroscopy (TPS) (II)
In paper II, terahertz pulsed spectra were recorded using a TPSspectra1000 V spectrometer (TeraView, Cambridge, UK). NF samples were mixed with polyethylene (Induchem, Volketwil, Switzerland, particle size < 10 µm) and pressed into discs. Spectra were recorded at room temperature while the sample chamber was purged with dry nitrogen.
14
EXPERIMENTAL
4.2.4 X-ray powder diffraction and crystal structure verification (I-V)
XRPD patterns were measured with a theta-theta X-ray powder diffractometer (D8 Advance, Bruker AXS GmbH, Karlsruhe, Germany). Measurements were performed in symmetrical reflection mode with CuKα radiation (λ=1.54 Å) using Göbel mirror bent multilayer optics. Crystal structures and all phase transformations were verified by comparing the experimental XRPD patterns to the reported structures of the crystal forms (I-V). Variable temperature X-ray powder diffraction (VT-XRPD) was used to study the structures at different temperatures (I, III, IV).
The CSD refcodes used for STZ polymorphs were SUTHAZ01 and SUTHAZ02 (Kruger and Gafner, 1972), SUTHAZ (Kruger and Gafner, 1971), SUTHAZ04 (Babilev et al., 1987), and SUTHAZ05 (Hughes et al., 1999). For NF the anhydrate forms α and β with the corresponding CSD refcodes LABJON01 and LABJON02 respectively (Pienaar et al., 1993a), and monohydrate forms I and II with the CSD refcodes HAXBUD01 and HAXBUD respectively (Pienaar et al., 1993b), were used. For PRX, the crystal forms were PRX anhydrate (form I) and PRX monohydrate, with CSD refcodes BIYSEH (Kojic-Prodic and Ruzic-Toros, 1982), and CIDYAP01 (Reck et al., 1988), respectively. The crystal forms and corresponding CSD refcodes of CBZ were anhydrate form I, CBMZPN11 (Grzesiak et al., 2003) and anhydrate form III, CBMZPN01 (Reboul et al., 1981). The dihydrate form of CBZ was compared to a recently reported crystal structure (Harris et al., 2005). Amorphous CBZ was verified by the characteristic halo in the XRPD pattern. For TP anhydrate and TP monohydrate the CSD refcodes were BAPLOT01 (Ebisuzaki et al., 1997) and THEOPH01 (Sun et al., 2002), respectively.
4.2.5 Thermal analysis (I, III, IV)
Differential scanning calorimetry (DSC) was performed with a TA Instruments 910S differential scanning calorimeter (TA Instruments, New Castle, DE, USA) (I, III, IV). Two different apparatuses were used for thermogravimetry (TG): a TGA7 by Perkin Elmer (Norwalk, CT, USA) (I) and a TGA 850 by Mettler Toledo (Greifensee, Switzerland) (IV).
4.2.6 Water content analysis (III, IV)
The water content before and after dehydration on hot stage and fluidized bed drying was determined using a Karl Fischer titrator (DL 35, Mettler Toledo, Greifensee, Switzerland).
15
EXPERIMENTAL
4.2.7 Scanning electron microscopy (V)
The solid-state transformations on the surfaces of the powder compacts were investigated by scanning electron microscopy (SEM): powder compacts of TP anhydrate and NF anhydrate were immersed in the dissolution medium on Petri dishes. Thereafter, the compacts were withdrawn from the dishes at various time points and excess dissolution medium was immediately removed from the surface of the compact with a tissue. SEM micrographs were recorded with a JSM-840A scanning electron microscope (Jeol, Tokyo, Japan).
4.3 Polymorph screening, processing, and dissolution testing
4.3.1 Polymorph screening (I, II)
The polymorph screening crystallizations were performed using a 24-well crystallization station (H+P Labortechnik, Oberschleissheim, Germany) with a Huber cc 250 cryostat (Peter Huber Kältemaschinenbau, Offenburg, Germany). STZ (I) was crystallized from 100 ml of 1-propanol, 2-propanol, acetone, ethanol and purified water, with a cooling rate of 35 ºC/h. Additional crystallizations for processing studies were performed from 2 L of 1-propanol and purified water (to prepare polymorphs SUTHAZ01 and SUTHAZ02, respectively). The crystals were harvested by filtration, and excess solvent was evaporated from the crystals at 35 ºC and 35% RH, for 24 h. Thereafter, the crystals were analyzed with NIR spectroscopy, XRPD, DSC, and TG. NF (II) was crystallized from 50 ml of acetone-water mixtures. The water contents of the mixtures were 2%, 4%, 10%, 17%, 20%, 25%, 33%, 50%, 67%, and 75% (v/v). Solutions were heated to 55 ºC (at which all nitrofurantoin had dissolved) and cooled to 10 ºC at two different cooling rates (1 ºC/min and 0.0375 ºC/min), with and without stirring. In addition to the automated crystallizations, crystallization from pure acetone was performed as previously reported (Pienaar et al., 1993a). Filtered crystals were dried overnight in ambient conditions and analyzed thereafter with NIR spectroscopy, Raman spectroscopy, TPS and XRPD.
4.3.2 Milling and compression (I)
STZ crystals were milled and compressed. Milling was performed using a planetary ball mill (Pulverisette 6, Fritsch, Idar-Oberstein, Germany). An 80 ml steel bowl was used with two different rotation speeds (100 and 400 rpm) and six different milling times (1, 2, 5, 10, 20 and 40 min). Compression of the crystals was performed with a hydraulic press using three different compression pressures (55, 90 and 180 MPa). The milled and compressed samples were analyzed with NIR spectroscopy and XRPD.
16
EXPERIMENTAL
4.3.3 Fluidized bed drying (III)
TP monohydrate granules were dried in a multichamber microscale fluidized bed dryer (MMFD, Ariacon Oy, Turku, Finland) (Räsänen et al., 2003). A fluidization chamber made of glass was modified with a quartz sight window for spectroscopic analysis. Two different inlet air temperatures (328 and 333 K) were used in the experiments. The batch size (4.5 g) and the flow rate of the inlet air (2.15×10-4 m3 s-1) were kept constant throughout the experiments. Prior to the experiments, the chamber was heated to 313 K. Process data and in-line NIR and Raman spectra were recorded.
4.3.4 Hot stage dehydration (IV)
PRX monohydrate and CBZ dihydrate were dehydrated isothermally on a hot stage (Mettler Toledo, Greifensee, Switzerland) at different temperatures. The temperatures were 381 K, 389 K, 397 K, and 405 K for PRX, and 313 K, 318 K, 323 K, 328 K, 333 K and 338 K for CBZ. In-situ NIR and Raman spectroscopy were performed during dehydration.
4.3.5 Dissolution testing (V)
The dissolution tests were conducted with a channel flow intrinsic dissolution test apparatus in which only one surface of the powder compact is in contact with the dissolution medium (Peltonen et al., 2003). The apparatus was modified with a quartz sight window for the Raman probe. The dissolution medium used in the study was purified water (t = 25 ºC, V = 500 ml), and the flow rates of the dissolution medium were 9 ml/min and 20 ml/min for TP and NF, respectively. The flow rate in NF dissolution tests was increased to ensure sink conditions. The concentration of the dissolution medium was measured with a UV-Vis spectrophotometer (Ultrospec III, Pharmacia LKB Biotechnology, Sweden) and a flow-through cuvette. The solid state of the dissolving compact was measured in situ with Raman spectroscopy.
4.4 Data processing
4.4.1 Spectral treatment
Different treatment methods were used to improve the signal/noise ratio, remove the baseline and scale differences between the spectra. The methods applied were Savitzky-Golay 2nd derivative treatment (Savitzky and Golay, 1964) (I), standard normal variate (SNV) transformation (Barnes et al., 1989) (I-V), and multiplicative scatter correction (MSC) (Geladi et al., 1985) (I, IV). In paper II, NIR and Raman
17
EXPERIMENTAL
spectra were first treated with SNV transformation, and then merged into one array prior to further processing.
4.4.2 Qualitative and quantitative methods
The following multivariate methods were used to qualitatively or quantitatively analyze the spectral data: principal component analysis (PCA) (I, II), dendrogram cluster analysis (I), partial least squares discriminant analysis (PLS-DA) (IV), and partial least squares regression (PLS) (III). In paper V the solid phase was quantified by ratioing peak intensities in the Raman spectra characteristic of each solid form involved. Quantification models were composed and spectra were pre-processed and analyzed using MatLab (Mathworks, Natick, MA, USA) and Simca-P software (Umetrics AB, Umeå, Sweden).
In paper III, PLS quantification was based on calibration NIR and Raman spectra recorded of binary powder mixtures with known ratios of anhydrous TP and TP monohydrate. The number of binary mixture samples used in the calibration sets was 59 and 22 for NIR and Raman, respectively. The quantification models were tested using independent test sets of 10 binary mixture samples of known ratios.
In paper IV, PLS-DA models were constructed using NIR and Raman spectra of pure solid forms of PRX (form I and monohydrate) and CBZ (form I, form III, dihydrate and amorphous form). The PLS-DA model of PRX was constructed using 12 samples of each form, and tested using 3 independent samples of each form. CBZ models were constructed using 8 samples of each form, and tested with 1 sample of each form.
In paper V, quantification of the solid forms was based on calibration curves constructed using Raman spectra of anhydrate and monohydrate powder mixtures of TP and NF. Two peaks characteristic of the corresponding anhydrate form, and one of the corresponding monohydrate were used. A ratio of peak intensities was calculated for each calibration sample (the monohydrate peak intensity divided by the sum of all three characteristic peak intensities) and thereafter, the calculated ratios were correlated to the hydrate form contents of the samples. Twenty-two samples were used to build the calibration curve and 11 independent samples were used as a test set.
18
RESULTS AND DISCUSSION
5. Results and discussion
5.1 Solid phase analysis using spectroscopy
Figure 5 shows NIR spectra, terahertz pulsed spectra and Raman spectra of three solid forms of nitrofurantoin (NF) (II). The different characteristics of the three methods are visible in the spectra.
Spectral information in the NIR region is composed of combinations and overtones of the fundamental vibrations associated with C-H, O-H, and N-H bonds. Hence, NIR spectroscopy is sensitive to differences in hydrogen bonding between the molecules in the crystal lattice (intermolecular bonding). Also, the presence of water molecules and hydrogen bonding interactions between the API and water molecules strongly influences the NIR spectra of solid drugs. These features can be seen in the NIR spectra of NF (figure 5a). The NIR spectrum of NF monohydrate form II is dominated by features due to the presence of water. The NIR spectra of all three forms contain bands in the region 1600-1750 nm indicating similarities in the intramolecular bonding. Regardless of the common bands mentioned, the anhydrate forms α and β have differences in the intermolecular bonding. These are indicated by bands in the 1st overtone and combination band region in the NIR spectrum of form β. NIR spectroscopy has been used for a wide variety of pharmaceutical analyses (Ciurczak and Drennen, 2002) including determination of crystallinity (Blanco et al., 2000; Seyer et al., 2000; Seyer and Luner, 2001), polymorphic forms (Gimet and Luong, 1987; Dreassi et a ., 1995; Norris et al., 1997; Luner et al., 2000; Patel et al., 2001), and hydrates (Räsänen et al., 2001; Airaksinen et al., 2003; Airaksinen et al., 2005).
l
l
TPS is most sensitive to intermolecular bonding since it mainly probes intermolecular vibrations including crystalline phonon vibrations. The three NF forms exhibited uniform and easily separable terahertz pulsed spectra (figure 5b). At present these modes cannot be assigned to specific phonon vibrations, but despite the present difficulties in interpreting terahertz spectra, the differences between the spectra of the NF forms show that TPS is another potential spectroscopic tool for polymorph screening. Like NIR and Raman spectroscopy, TPS has been used for analysis of polymorphism (Taday et al., 2003; Zeitler et al., 2006b) and crystallinity (Strachan et a ., 2004b; Strachan et al., 2005), as well as hydrates (Liu et al., 2006; Zeitler et al., 2006a).
Raman spectroscopy predominantly detects symmetric vibrations of nonpolar (mainly C-C) and aromatic groups. These vibrations can be affected by intramolecular events like molecular conformation and bonding changes. Each NF form exhibited distinct Raman spectra (figure 5c). Raman spectroscopy has been used to analyze crystallinity (Taylor and Zografi, 1998; Schmidt et al., 2003; Okumura and Otsuka, 2005), polymorphic forms (Tudor et al., 1993; Langkilde et al., 1997; Taylor and Langkilde, 2000; Strachan et al., 2004a), and hydrates (Rantanen et al., 2005;
19
RESULTS AND DISCUSSION
Wikström et a ., 2005). Unlike NIR spectroscopy, Raman spectroscopy can be used in aqueous media, since water is a poor Raman scatterer.
l
Figure 5. Spectroscop c characterization of three NF forms (monohydrate II, and anhydrates α and β). (a) NIR spectra, (b) terahertz pulsed spectra, and (c) Raman spectra. Modified from paper II.
i
20
RESULTS AND DISCUSSION
5.2 Spectroscopic methods as polymorph screening tools
Sulfathiazole (STZ) crystallizations yielded three polymorphs (SUTHAZ01, SUTHAZ02, and SUTHAZ), which were verified by XRPD and DSC (I). It was confirmed by TG that no solvates had been formed. The XRPD patterns of STZ polymorphs resemble each other and therefore, identification with XRPD alone would not have been reliable. As a further complication, the samples were mixtures of different polymorphs. It has been reported that STZ crystallizes according to the Ostwald’s rule of stages (Ostwald, 1896), and accordingly, initially crystallizes in the least stable form, and stepwise, converts towards the stable form (Blagden et al., 1998a; Blagden et al., 1998b). Based on this rule, the time point of crystal harvesting partially determines the final crystal form. The main polymorph from water was SUTHAZ02, and a mixture of SUTHAZ and SUTHAZ02 was obtained from ethanol. From 1-propanol the polymorphic content was mainly SUTHAZ01, and crystallization from 2-propanol yielded a mixture of the three forms mentioned. Similar results have been published previously (Khoskhoo and Anwar, 1993; Blagden et al., 1998a; Blagden et al., 1998b).
NIR spectroscopy helped to differentiate between the crystallized samples. Second derivative treatment was more efficient than SNV transformation or MSC. The difference between NIR spectra of SUTHAZ and SUTHAZ02 is very small, and this difference was most obvious after second derivative treatment. Spectral differences between SUTHAZ01 and the other forms were obvious due to distinctly different hydrogen bonding between the STZ molecules. Initially, the distances between the second derivative NIR spectra were visualized using dendrogram cluster analysis. This method was able to cluster the samples that contained mainly SUTHAZ01 in one group but the other forms were clustered together in another group, so only a partial clustering was achieved. The second method applied, PCA, was able to differentiate between all three polymorphs. The scores of the PCA were plotted, and two principal components were not enough to separate polymorphs SUTHAZ and SUTHAZ02 into their own clusters. The third principal component explained only 0.6% of the spectral variation, but was essential, since it separated the scores of forms SUTHAZ and SUTHAZ02 (figure 6).
21
RESULTS AND DISCUSSION
Figure 6. PCA score p ot of the second derivative NIR spectra of the crystallizedsamples. The legend refers to the crystallization solvents. The principal components PC1, PC2, and PC3 together explained 99.4% of the NIR spectral variation. Polymorphic purity increases in the direction of the arrows. Modified from paper I.
l
Two anhydrate forms (α and β) and the monohydrate form II were crystallized in the polymorph screen of NF (II). The three forms were identified with three complementary spectroscopic techniques (NIR, Raman and TPS, figure 5) and verified with XRPD. As expected, the water content of the solvent most affected the outcome of the crystallization. With water contents above 4%, the resulting crystal form was always form II, regardless of the cooling rate or whether stirring was used. Interestingly, at low water contents (2% and 4%), hydrate formation could be controlled by the stirring and cooling rate. At a water content of 2%, the unstirred crystallizations yielded form II crystals, but stirring led to form α or β. With a water content of 4%, a combination of slow cooling and stirring produced crystals of form α, whereas crystallizations with fast cooling and stirring resulted in form II. These results underline the need to consider not only the solvent, but also various processing parameter effects.
The rationale behind the combination of NIR and Raman spectroscopy is that they are fast methods that require no or very little sample preparation and are suitable for high throughput polymorph screening. Furthermore, they are complementary. Instead of comparing the results obtained with various techniques, coupling complementary techniques may be useful. In the PCA of the merged spectra, two principal components were needed to separate the scores of the three solid forms (figure 7a). The first principal component separated the monohydrate
22
RESULTS AND DISCUSSION
from the anhydrate forms, while the second principal component separated the two anhydrate forms from each other. The first principal component had the largest loading values in the NIR spectral region (at 1416 nm and 1920 nm), which corresponds to the presence of water molecules in form II (figure 7b). The most important loadings of PC2 were in the Raman spectral region (between 1343 cm-1 and 1349 cm-1). Thus, NIR spectroscopy effectively separated the hydrate form from the anhydrates, and Raman spectroscopy differentiated the anhydrate forms α and β.
Figure 7. a) PCA score plot of the SNV corrected and merged NIR/Raman spectra. The three clusters rep esent crystal forms II, α and β of nitrofurantoin. The principal components PC1 and PC2 together explained 94.2% of the NIR/Raman spectral variation. b) The loadings of the principal components PC1 and PC2. Modified from paper II.
r
5.3 Spectroscopic analysis of processing-induced transformations
5.3.1 The effect of milling and compression
Since the conditions APIs are likely to meet during pharmaceutical production set the stage for PITs, analysis of the solid phase should be carried out also during processing. In paper I, PIT screening of STZ was performed. During milling, a decrease in crystallinity of STZ was observed as peak broadening in the XRPD patterns. Crystallinity decreased as a function of milling speed and time due to the higher energy input. Preferred orientation of the samples with slower milling speed made interpretation of XRPD patterns difficult, particularly with crystals of form SUTHAZ02. Therefore, complementary methods were needed. NIR analysis showed the decrease of crystallinity, but also a transformation from SUTHAZ02 to SUTHAZ01. This polymorphic transformation was not visible in the XRPD (due to the preferred orientation).
The NIR spectra of the processed samples were analyzed with PCA, and the milling-induced transformations were visualized using a PCA score plot. In the score
23
RESULTS AND DISCUSSION
plot, the polymorphic transformation was mainly seen in the direction of the first principal component, and the decrease in crystallinity was seen in the direction of the second principal component. Even though the pressures used were rather high, no phase transformations were detected after compression of STZ crystals.
5.3.2 Solid phase analysis during fluidized bed drying
In paper III, fluidized bed drying of theophylline (TP) monohydrate granules was studied. During drying, TP monohydrate dehydrated to TP anhydrate at both temperatures investigated. The dehydration could be identified by following either the absolute humidity of outlet air or pressure difference over the fluidized bed. These measurements indicate the water loss from the crystalline lattice and the weight loss due to dehydration, respectively. However, no molecular level insight into the system is achieved and possible underlying structural changes cannot be identified with these data.
The dehydration was directly analyzed with in-line NIR and Raman spectroscopy: NIR spectroscopic quantification results showed that the TP monohydrate content of the granules started to decrease immediately after the fluidization had started (figure 8a). This was also seen as an increase in outlet air humidity and a decrease in pressure difference over the bed. Dehydration was completed and, according to NIR spectroscopy, reached a plateau after fluidization for 7 minutes at 328 K and 6 minutes at 333 K. At these time points humidity of the outlet air was highest. A change in the Raman spectroscopic quantification of hydrate form was detected after a 3-minute lag time (figure 8b). This change is related to differences in molecular packing of TP solid-state forms (figure 9).
Figure 8. Dehydration profiles during fluidized bed drying of theophylline monohydra e granules. Quantification of hydrate form by in-line (a) NIR spectroscopy and (b) Raman spectroscopy with PLS regression. Modified from paper III.
t
The crystal structure started to change after water molecules had left the water channels (figure 9d), thereafter the amount of hydrate form decreased rapidly until
24
RESULTS AND DISCUSSION
no more water could be removed from the structure. Unlike with NIR spectroscopy, quantification with Raman spectroscopy did not reach a plateau when the outlet air humidity was highest. A clear difference between experiments at different temperatures was noticed in the Raman spectroscopic results (figure 8b). The amount of hydrate decreased faster at 333 K than at 328 K and reached a lower level, after which the amount of hydrate decreased much more slowly. The slight downward trend seen after the rapid decrease in the amount of hydrate in the Raman spectroscopic quantification is probably related to the existence of the metastable anhydrate form (Phadnis and Suryanarayanan, 1997), not to the presence of monohydrate form in the granules. The occurrence of the metastable form has been linked to dehydration in open conditions in DSC studies (Suihko et al., 1997), and the relative amount of the metastable form during drying has been found to be dependent on dehydration conditions: fluidized bed drying promotes metastable anhydrate formation compared to tray drying. (Airaksinen et al., 2004). Obviously, the amount of metastable form could not be quantified since it was not included in the PLS calibration model. All intermediate forms that are detected during solid phase transformations should be included in the calibration samples. This aspect was studied in paper IV.
Figure 9. Molecular structure of theophylline (TP) (a), and arrangement of molecules and hydrogen bonding in TP anhydrate (b), TP monohydrate (c), and TP monohydrate viewed along a water channel (d). Dotted lines represent hydrogen bonds, water molecules are highlighted. Modified from paper III.
25
RESULTS AND DISCUSSION
5.3.3 Investigation of dehydration mechanisms
In paper IV, the dehydration of piroxicam (PRX) monohydrate and carbamazepine (CBZ) dihydrate was investigated. PRX and CBZ showed very different dehydration behavior. The dehydration of PRX monohydrate was very straightforward, whereas the dehydration of CBZ dihydrate was complicated. A heating-induced dehydration from PRX monohydrate straight to PRX form I was observed very successfully with NIR and Raman spectroscopy. No intermediate solid forms were observed during the dehydration. This finding was supported by VT-XRPD analysis of non-isothermal dehydration. Since the dehydration involved only two solid forms, one component alone explained the whole variation in PLS-DA analysis.
DSC analysis of CBZ dihydrate showed one broad endotherm, but VT-XRPD analysis revealed that the dehydration was not as simple as one might have expected from the DSC trace. The removal of the water from the crystal lattice of CBZ dihydrate initially resulted in a decrease in crystallinity. This was seen in the XRPD pattern as a sudden drop in peak intensities. After additional heating form III peaks appeared, which were eventually followed by form I peaks. The PLS-DA of the in-situ NIR and Raman spectra recorded during dehydration showed differences due to the intrinsic nature of the methods. With both methods, a three-component model was used. In NIR spectral PLS-DA, the first component explained 76.3% of the spectral variation and separated the dihydrate from anhydrate forms (form I, form III, and amorphous). The second component, explaining 17.8%, separated the amorphous form, and third component, explaining 5.5%, discriminated between form I and form III. Like in the fluidized bed drying paper (III), NIR spectroscopy was a powerful method for determining interactions between the API and water. The differences between the spectra of the anhydrous forms (form I, form III and amorphous) were smaller and the intermediate forms during dehydration could not be as accurately followed, although differences in the dehydration paths could be observed. In Raman spectral PLS-DA, the first component explained 73.0% and separated crystalline forms (dihydrate, form I, and form III) from the amorphous form (figure 10a and 10b). The second component, explaining 19.9%, separated the dihydrate, and third component, explaining 6.3%, discriminated between forms I and III. The paths of dehydration were more accurately followed by Raman spectroscopy than NIR spectroscopy, since Raman spectroscopy is more sensitive to structural changes. Regardless of CBZ being classified as a channel hydrate, removal of water involves reorganization of the CBZ molecules in the lattice and can be well observed with Raman spectroscopy. At higher temperatures dehydration was fast and proceeded via the amorphous form, whereas at lower temperatures dehydration was slower and not as destructive and the scores of in-situ Raman spectra were located further from the scores of the pure amorphous form (figure 10a).
26
RESULTS AND DISCUSSION
Figure 10. (a) PLS-DA score plot of dehydration of CBZ dihydrate investigated by in-situ Raman spectroscopy. Squares and triangles represent scores of in-situ Raman spectra at d fferent temperatures, hexagons are scores ofpure solid forms (calibration samples). (b) Weight vectors of the Raman spectroscopic PLS-DA model. Modified from paper IV.
i
l
5.4 In-situ solid phase analysis during dissolution testing
The effect of solid form on the dissolution rate of APIs can be investigated by measuring the intrinsic dissolution rate (IDR). However, there are also several properties other than the solid form that can affect the dissolution rate, and furthermore the dissolution rates of different forms are not always in line with corresponding solubilities (Allen et a ., 1978; Kahela et al., 1983; Kristl et al., 1996; Laihanen et al., 1996). A change in the IDR is not a reliable measure of a solid phase transformation during dissolution test. Therefore, direct solid phase analysis was applied to dissolution testing (Rantanen et al., 2006) (figure 11). The combination of liquid and solid phase analysis enables understanding of dissolution as a whole.
Figure 11. Dissolution testing cheme. s
27
RESULTS AND DISCUSSION
During dissolution TP anhydrate underwent a transformation to TP monohydrate. Almost immediately after the compact prepared from 100% TP anhydrate was exposed to water, transformation to TP monohydrate started. The dissolution rate of the initially anhydrous compact decreased as the amount of monohydrate form increased (figure 12). The transformation was seen with SEM as rapid growth of needle-like TP monohydrate crystals on the surface of the compact (figure 10a-c). During the transformation dissolution of both forms and crystallization of the monohydrate form occurred. Consequently, the larger the amount of monohydrate the slower the dissolution rate of the compact. Once the solid phase transformation was complete, the dissolution rate of the initially anhydrous surface became constant (at ~6 min, figure 12c). At this stage the anhydrate form was no longer present and therefore only the monohydrate phase dissolved. Even though the surface of the compact was fully transformed to monohydrate, the dissolution rate remained higher than that of a compact prepared from TP monohydrate. The dissolution rate difference between the transformed, initially anhydrous compact and the compact prepared from TP monohydrate was most likely caused by a difference in the specific surface area due to the monohydrate crystal growth on the initially anhydrous surface (figure 12a-c).
The surface of the compact consisting of a 1:1 mixture of TP anhydrate and MCC started to convert to TP monohydrate after a lag time of approximately two minutes. The presence of MCC delayed the transformation onset but had no significant effect on the transformation rate once the transformation was underway. Similar hydrate formation delaying behavior due to MCC has also been detected in an aqueous processing environment and vapor phase induced solid phase transformation studies (Airaksinen et al., 2003; Salameh and Taylor, 2006).
Analogously to TP, NF anhydrate (form β) transformed to NF monohydrate (form II) during dissolution. The transformation was an order of magnitude slower than the transformation of TP anhydrate. This can be explained by the considerable difference between the solubilities of TP and NF. Unlike with TP, the correlation between the solid phase transformation and IDR changes was not evident with NF. There was no clear decrease in the dissolution rate during the transformation stage. The dissolution rate of compacts prepared from NF anhydrate started to decrease immediately after the dissolution test began and became constant after approximately 8 minutes. This result is consistent with dissolution studies performed with the rotating disk method (Otsuka et al., 1992). The fast dissolution rate at the very beginning can be partly explained by the dissolution of NF anhydrate before the monohydrate began to crystallize. Further, the dissolution of loose NF particles on the surface can promote the burst effect during the first minutes. The dissolution rate then increased after approximately 50 minutes and decreased again after 200 minutes. The variation of the dissolution rate can be due to several overlapping factors, e.g., the changing anhydrate/monohydrate ratio and the decreasing rate of hydrate formation. Also, the increase in the specific surface area may partly compensate for the lower dissolution rate of the monohydrate form.
28
RESULTS AND DISCUSSION
Figure 12. Results of the dissolution test of TP anhydrate. a), b), and c) are timepoints at which the surface of the compact was investigated with (off-line) SEM. For comparison, dissolution rate of TP monohydrate is shown (dashed line). Modified from paper V.
5.5 Interfacing spectroscopic tools with different environments
With fiber optic probes spectrometers can be interfaced with a variety of environments, but there are several pitfalls that have to be avoided. Measurements can be performed non-invasively or in situ using immersion probes or through a spectroscopic window. Possible pitfalls include poor sampling and sticking of material to the probe/window. Sampling errors can be avoided by increasing the effective sample volume and by moving the sample or the probe during measurements (Bell et al., 2004). In this thesis, a large sampling area (I) and a rotating sample holder (II) were used to ensure representative sampling. In process measurements the samples are rarely static, so sampling area is usually not an issue.
29
RESULTS AND DISCUSSION
In the fluidized bed drying study (III), it was visually confirmed that the granules were well fluidized and the window remained clean during the experiments. If spectra are measured through a window, the material of the window is of importance. In this thesis, quartz windows were used to avoid unwanted spectral features (III, V).
When quantitative analysis is performed, development and choice of the suitable quantitative method is important. One can use either univariate or multivariate methods. Univariate methods are straightforward and easier to apply than multivariate methods. If good quality spectra are obtained and peaks characteristic of each form to be quantified can be identified, simple univariate quantification is advisable. In paper V, where static samples were under investigation, ratioing intensities of three Raman peaks enabled reliable quantification, and thus, complicated methods were not needed. Multivariate methods can extract information from regions of or entire spectra and are particularly useful for coping with cross-correlated and noisy data (Beebe et al., 1988). In paper III, PLS regression was used to obtain quantitative information from in-line spectra that contained considerable amounts of process-related noise.
30
CONCLUSIONS
6. Conclusions
Near-infrared (NIR) spectroscopy can be utilized in screening of polymorphs and processing-induced transformations (PITs). With principal component analysis (PCA) it is possible to analyze large amounts of samples and differentiate between spectra that contain very slight differeces. By using NIR and Raman spectroscopy in tandem a clear classification of solid forms during polymorph screening was achieved. The complementary nature of the methods is best realized when analyzing complicated systems that exist as several solid forms.
Quantitative solid phase analysis can be performed during fluidized bed drying using in-line NIR and Raman spectroscopy and partial least squares (PLS) regression. To quantitate all solid forms upon dehydration, the dehydration mechanism has to be investigated first. This can be carried out with in situ spectroscopy and partial least squares discriminant analysis (PLS-DA).
-
Simultaneous analysis of the liquid and the solid phase in situ during dissolution testing enables analysis of dissolution as a whole. Direct solid phase analysis of the dissolving solid sample provides a science-based explanation for changes in the dissolution rate.
Spectroscopic methods are well suitable for solid phase analysis during pharmaceutical development and manufacturing, and provide means for analyses throughout the life cycle of the drug product. In-line measurements can be performed, providing molecular level information on API behavior during processing. This enables better understanding and knowledge of the drug products and processes, and therefore, a way to achieve better quality.
31
ACKNOWLEDGEMENTS
7. Acknowledgements
This study was carried out at the Division of Pharmaceutical Technology, Faculty of Pharmacy, University of Helsinki. I would like to thank all the people involved in this project.
The biggest thank you goes to my supervisor, Professor Jukka Rantanen. His inspiring guidance has definitely made my work a lot easier, and also much more interesting. My co-supervisor, Professor Jouko Yliruusi, is respectfully acknowledged for advice and sharing his experience in physical pharmacy, and of course, for providing excellent facilities to conduct research.
The reviewers of this thesis, Professor Jari Yli-Kauhaluoma and Associate Professor Lars Hovgaard, are thanked for prompt review process and their constructive comments on the manuscript.
I am sincerely grateful to all my co-authors and the whole solid-state research group at the division of Pharmaceutical Technology for scientific contributions, fruitful discussions, technical assistance, guidance, and friendship throughout the course of this work. In particular, the contributions of Clare Strachan, Karin Kogermann and Niklas Sandler are greatly appreciated.
Finnish Cultural Foundation (Elli Turunen fund), the Academy of Finland, Finnish Pharmaceutical Society, the Graduate School in Pharmaceutical Research, and Drug Research Academy (University of Copenhagen) are acknowledged for funding my research.
I would also like to express my gratitude to my parents and my sister for endless support and love. Special thanks to my grandfather for introducing me to the field of pharmacy. Finally, I want to thank the most loved ones, Katri and Aimo for just being there. Helsinki, March 2007
32
REFERENCES
8. References
Abboud, L. and Hensley, S. 2003. Factory Shift: New prescription for drug makers: Update the
plants. Wall Street Journal, September 3.
Airaksinen, S., Luukkonen, P., Jørgensen, A., Karjalainen, M., Rantanen, J., and Yliruusi, J. 2003. Effect of excipients on hydrate formation in wet masses containing theophylline. Journal of Pharmaceutical Sciences, 92, 516-528.
Airaksinen, S., Karjalainen, M., Räsänen, E., Rantanen, J., and Yliruusi, J. 2004. Comparison of the effects of two drying methods on polymorphism of theophylline. International Journal of Pharmaceutics, 129-141.
Airaksinen, S., Karjalainen, M., Kivikero, N., Westermarck, S., Shevchenko, A., Rantanen, J., and Yliruusi, J. 2005. Excipient selection can significantly affect solid-state phase transformation in formulation during wet granulation. AAPS PharmSciTech, 6, 311-322.
Allen, F.H. 2002. The Cambridge Structural Database: a quarter of a million crystal structures and rising. Acta Crystallographica B, 58, 380-388.
Allen, P.V., Rahn, P.D., Sarapu, A.C., and Vanderwielen, A., J. 1978. Physical characterization of erythromycin: anhydrate, monohydrate, and dihydrate crystalline solids. Journal of Pharmaceutical Sciences, 67, 1087-1093.
Andronis, V. and Zografi, G. 1998. The molecular mobility of supercooled amorphous indomethacin as a function of temperature and relative humidity. Pharmaceutical Research, 15, 835-842.
Anquetil, P.A., Brenan, C.J.H., Marcolli, C., and Hunter, I.W. 2003. Laser Raman spectroscopic analysis of polymorphic forms in microliter fluid volumes. Journal of Pharmaceutical Sciences, 92, 149-160.
Babilev, F.V., Bel'skii, V.K., Simonov, A., and Arzamastsev, A.P. 1987. Khimiko Farmatsevticheskii Zhurnal, 21, 1275-1280.
Barnes, R.J., Dhanoa, M.S., and Lister, S.J. 1989. Standard normal variate transformation and de-trending of near-infrared diffuse reflectance spectra. Applied Spectroscopy, 43, 772-777.
Barr, G., Dong, W., Gilmore, C., and Faber, J. 2004a. High-throughput powder diffraction. III: The application of full-profile pattern matching and multivariate statistical analysis to round-robin-type data sets. Journal of Applied Crystallography, 37, 635-642.
Barr, G., Dong, W., and Gilmore, C.J. 2004b. High-throughput powder diffraction. IV. Cluster validation using silhouettes and fuzzy clustering. Journal of Applied Crystallography, 37, 874-882.
Barr, G., Dong, W., and Gilmore, C.J. 2004c. High-throughput powder diffraction. II. Application of clustering methods and multivariate data analysis. Journal of Applied Crystallography, 37, 243-252.
Bauer-Brandl, A. 1996. Polymorphic transitions of cimetidine during manufacture of solid dosage forms. International Journal of Pharmaceutics, 140, 195-206.
Beebe, K.R., Pell, R.J., and Seasholtz, M.B., Chemometrics: A Practical Guide. 1988, New York: John Wiley & Sons, Inc. p. 348.
Bell, S.E.J., Beattie, J.R., McGarvey, J.J., Peters, K.L., Sirimuthu, N.M.S., and Speers, S.J. 2004. Development of sampling methods for Raman analysis of solid dosage forms of therapeutic and illicit drugs. Journal of Raman Spectroscopy, 35, 409-417.
Bernstein, J., Polymorphism and patents, in Polymorphism in Molecular Crystals. 2002a, Clarendon Press: Oxford. p. 297-307.
Bernstein, J., Introduction and historical background, in Polymorphism in Molecular Crystals. 2002b, Clarendon Press: Oxford. p. 1-28.
33
REFERENCES
Blagden, N., Davey, R.J., Lieberman, H.F., Williams, L., Payne, R., Roberts, R., Rowe, R., and Docherty, R. 1998a. Crystal chemistry and solvent effects in polymorphic systems. Sulfathiazole. Journal of the Chemical Society, Faraday Transactions, 94, 1035-1044.
Blagden, N., Davey, R.J., Rowe, R., and Roberts, R. 1998b. Disappearing polymorphs and the role of reaction by-products: the case of sulphathiazole. International Journal of Pharmaceutics, 172, 169-177.
Blanco, M., Coello, J., Iturriaga, H., Maspoch, S., and Pérez-Maseda, C. 2000. Determination of polymorphic purity by near infrared spectrometry. Analytica Chimica Acta, 407, 247-254.
Boldyreva, E.V., Shakhtshneider, T.P., Ahsbahs, H., Sowa, H., and Uchtmann, H. 2002. Effect of high pressure on the polymorphs of paracetamol. Journal of Thermal Analysis and Calorimetry, 68, 437-452.
Brodka-Pfeiffer, K., Langguth, P., Gra, P., and Hausler, H. 2003. Influence of mechanical activation on the physical stability of salbutamol sulphate. European Journal of Pharmaceutics and Biopharmaceutics, 56, 393-400.
Burger, A. and Ramberger, R. 1979. On the polymorphism of pharmaceuticals and other molecular crystals. Microchimica Acta, 2, 259-271.
Byrn, S.R., Sutton, P.A., Tobias, B., Frye, J., and Main, P. 1988. Crystal structure, solid-state NMR spectra, and oxygen reactivity of five crystal forms of prednisolone tert-butylacetate. Journal of the American Chemical Society, 110, 1609-1614.
Byrn, S.R., Pfeiffer, R.R., Ganey, M., Hoiberg, C., and Poochikian, G. 1995. Pharmaceutical solids: a strategic approach to regulatory considerations. Pharmaceutical Research, 12, 945-954.
Byrn, S.R., Pfeiffer, R.R., and Stowell, J.G., Hydrates and solvates, in Solid-State Chemistry of Drugs. 1999a, SSCI Inc.: West Lafayette. p. 233-258.
Byrn, S.R., Pfeiffer, R.R., and Stowell, J.G., Polymorphs, in Solid-State Chemistry of Drugs. 1999b, SSCI Inc.: West Lafayette. p. 143-231.
Byrn, S.R., Xu, W., and Newman, A.W. 2001. Chemical reactivity in solid-state pharmaceuticals: formulation implications. Advanced Drug Delivery Reviews, 48, 115-136.
Cardew, P.T. and Davey, R.J. 1985. The kinetics of solvent-mediated phase transformations. Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, 398, 415-428.
Carlson, R., Lundstedt, T., and Albano, C. 1985. Screening of suitable solvents in organic synthesis. Strategies for solvent selection. Acta Chemica Scandinavica B, 39, 79-91.
Chan, H.K. and Doelker, E. 1985. Polymorphic transformation of some drugs under compression. Drug Development and Industrial Pharmacy, 11, 315-332.
Chemburkar, S.R., Bauer, J., Deming, K., Spiwek, H., Patel, K., Morris, J., Henry, R., Spanton, S., Dziki, W., Porter, W., Quick, J., Bauer, P., Donaubauer, J., Narayanan, B.A., Soldani, M., Riley, D., and McFarland, K. 2000. Dealing with the impact of ritonavir polymorphs on the late stages of bulk drug process development. Organic Process Research & Development, 4, 413-417.
Chen, S., Guzei, I.A., and Yu, L. 2005. New polymorphs of ROY and new record for coexisting polymorphs of solved structures. Journal of the American Chemical Society, 127, 9881-9885.
Chikhalia, V., Forbes, R.T., Storey, R.A., and Ticehurst, M. 2006. The effect of crystal morphology and mill type on milling induced crystal disorder. European Journal of Pharmaceutical Sciences, 27, 19-26.
Ciurczak, E.W. and Drennen, J.K., Pharmaceutical assays, in Pharmaceutical and Medical Applications of Near-Infrared Spectroscopy. 2002, Marcel Dekker, Inc.: New York. p. 73-105.
34
REFERENCES
Cross, W.I., Blagden, N., Davey, R.J., Pritchard, R.G., Neumann, M.A., Roberts, R.J., and Rowe, R.C. 2003. A whole output strategy for polymorph screening: combining crystal structure prediction, graph set analysis, and targeted crystallization experiments in the case of diflunisal. Crystal Growth & Design, 3, 151-158.
Datta, S. and Grant, D.W.J. 2004. Crystal structures of drugs: advances in determination, prediction and engineering. Nature Reviews Drug Discovery, 3, 42-57.
Davey, R.J., Cardew, P.T., McEvan, D., and Sadler, D.E. 1986. Rate controlling processes in solvent-mediated phase transformations. Journal of Crystal Growth, 79, 648-653.
Davey, R.J., Blagden, N., Potts, G.D., and Docherty, R. 1997. Polymorphism in molecular crystals: stabilization of a metastable form by conformational mimicry. Journal of the American Chemical Society, 119, 1767-1772.
Davey, R.J. 2004. Crystallization: How come you look so good? Nature, 428, 374-375.
Davis, T.D., Morris, K.R., Huang, H., Peck, G.E., Stowell, J.G., Eisenhauer, B.J., Hilden, J.L., Gibson, D., and Byrn, S.R. 2003. In situ monitoring of wet granulation using online X-ray powder diffraction. Pharmaceutical Research, 20, 1851-1857.
Davis, T.D., Peck, G.E., Stowell, J.G., Morris, K.R., and Byrn, S.R. 2004. Modeling and monitoring of polymorphic transformations during the drying phase of wet granulation. Pharmaceutical Research, 21, 860-866.
De Smidt, J.H., Fokkens, J.G., Grijseels, H., and Crommelin, D.J.A. 1986. Dissolution of theophylline monohydrate and anhydrous theophylline in buffer solutions. Journal of Pharmaceutical Sciences, 75, 497-501.
Di Martino, P., Guyot-Hermann, A.M., Conflant, P., Drache, M., and Guyot, J.C. 1996. A new pure paracetamol for direct compression: The orthorhombic form. International Journal of Pharmaceutics, 128, 1-8.
DiMasi, J.A., Hansen, R.W., and Grabowski, H.G. 2003. The price of innovation: new estimates of drug development costs. Journal of Health Economics, 22, 151-185.
Dreassi, E., Ceramelli, G., Corti, P., Lonardi, S., and Perruccio, P.L. 1995. Near-infrared reflectance spectrometry in the determination of the physical state of primary materials in pharmaceutical production. Analyst, 120, 1005-1008.
Ebisuzaki, Y., Boyle, P.D., and Smith, J.A. 1997. Methylxanthines. I. Anhydrous theophylline. Acta Crystallographica C, 53, 777-779.
FDA, Pharmaceutical CGMPs for the 21st Century - A Risk-Based Approach. 2004a, U.S. Food and Drug Administration.
FDA, PAT - A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance. 2004b, U.S. Food and Drug Administration.
Geladi, P., MacDougall, D., and Martens, H. 1985. Linearization and scatter correction for near-infrared reflectance spectra of meat. Applied Spectroscopy, 39, 491-500.
Gillon, A.L., Feeder, N., Davey, R.J., and Storey, R. 2003. Hydration in molecular crystals - a Cambridge Structural Database analysis. Crystal Growth & Design, 3, 663-673.
Gilmore, C., Barr, G., and Paisley, J. 2004. High-throughput powder diffraction. I. A new approach to qualitative and quantitative powder diffraction pattern analysis using full pattern profiles. Journal of Applied Crystallography, 231-242.
Gimet, R. and Luong, A.T. 1987. Quantitative determination of polymorphic forms in a formulation matrix using the near infra-red reflectance analysis technique. Journal of Pharmaceutical and Biomedical Analysis, 5, 205-211.
Govindarajan, R. and Suryanarayanan, R., Processing-induced phase transformations and their implications on pharmaceutical product quality, in Polymorphism in the Pharmaceutical Industry, R. Hilfiker, Editor. 2006, WILEY-VCH Verlag GmbH: Weinheim. p. 333-364.
Grant, D.W.J., Theory and origin of polymorphism in Polymorphism in Pharmaceutical Solids, H.G. Brittain, Editor. 1999, Marcel Dekker, Inc.: New York. p. 1-33.
35
REFERENCES
Grzesiak, A.L., Lang, M., Kim, K., and Matzger, A.J. 2003. Comparison of the four anhydrous polymorphs of carbamazepine and the crystal structure of form I. Journal of Pharmaceutical Sciences, 92, 2260-2271.
Gu, C.-H., Li, H., Gandhi, R.B., and Raghavan, K. 2004. Grouping solvents by statistical analysis of solvent property parameters: implication to polymorph screening. International Journal of Pharmaceutics, 283, 117-125.
Guillory, J.K., Generation of polymorphs, hydrates, solvates and amorphous solids, in Polymorphism in Pharmaceutical Solids, H.G. Brittain, Editor. 1999, Marcel Dekker, Inc.: New York. p. 183-226.
Guo, Y., Byrn, S.R., and Zografi, G. 2000. Physical characteristics and chemical degradation of amorphous quinapril hydrochloride. Journal of Pharmaceutical Sciences, 89, 128-143.
Haleblian, J. and McCrone, W. 1969. Pharmaceutical applications of polymorphism. Journal of Pharmaceutical Sciences, 58, 911-929.
Hancock, B.C. and Zografi, G. 1997. Characteristics and significance of the amorphous state in pharmaceutical systems. Journal of Pharmaceutical Sciences, 86, 1-12.
Hancock, B.C. and Parks, M. 2000. What is the true solubility advantage for amorphous pharmaceuticals? Pharmaceutical Research, 17, 397-404.
Harris, R.K., Ghi, P.Y., Puschmann, H., Apperley, D.C., Griesser, U.J., Hammond, R.B., Ma, C., Roberts, K.J., Pearce, G.J., Yates, J.R., and Pickard, C.J. 2005. Structural studies of the polymorphs of carbamazepine, its dihydrate, and two solvates. Organic Process Research & Development, 9, 902-910.
Hausman, D.S., Cambron, R.T., and Sakr, A. 2005. Application of on-line Raman spectroscopy for characterizing relationships between drug hydration state and tablet physical stability. International Journal of Pharmaceutics, 299, 19-33.
Herman, J., Remon, J.P., Visavarungroj, N., Schwartz, J.B., and Klinger, G.H. 1988. Formation of theophylline monohydrate during the pelletization of microcrystalline cellulose-anhydrous theophylline blends. International Journal of Pharmaceutics, 42, 15-18.
Hilfiker, R., ed. Polymorphism in the Pharmaceutical Industry. 2006, Weinheim: WILEY-VCH Verlag GmbH.
Hilfiker, R., De Paul, S.M., and Szelagiewcz, M., Approaches to polymorphism screening, in Polymorphism in the Pharmaceutical Industry, R. Hilfiker, Editor. 2006, WILEY-VCH Verlag GmbH: Weinheim. p. 287-308.
Hughes, D.S., Hursthouse, M.B., Threlfall, T., and Tavener, S. 1999. A new polymorph of sulfathiazole. Acta Crystallographica C, 55, 1831-1833.
ICH, Q6A: Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products: Chemical Substances. 1999, International Conference on Harmonisation.
ICH, Q8: Pharmaceutical Development. 2005a, International Conference on Harmonisation.
ICH, Q9: Quality Risk Management. 2005b, International Conference on Harmonisation.
IMS Health. 2006. Total market estimates and global pharma forecasts.
Ivanisevic, I., Bugay, D.E., and Bates, S. 2005. On pattern matching of X-ray powder diffraction data. Journal of Physical Chemistry B, 109, 7781-7787.
Jørgensen, A., Rantanen, J., Karjalainen, M., Khriachtchev, L., Räsänen, E., and Yliruusi, J. 2002. Hydrate formation during wet granulation studied by spectroscopic methods and multivariate analysis. Pharmaceutical Research, 19, 1285-1291.
Jørgensen, A.C., Airaksinen, S., Karjalainen, M., Luukkonen, P., Rantanen, J., and Yliruusi, J. 2004. Role of excipients in hydrate formation kinetics of theophylline in wet masses studied by near-infrared spectroscopy. European Journal of Pharmaceutical Sciences, 23, 99-104.
Kahela, P., Aaltonen, R., Lewing, E., Anttila, M., and Kristoffersson, E. 1983. Pharmacokinetics and dissolution of two crystalline forms of carbamazepine. International Journal of Pharmaceutics, 14, 103-112.
36
REFERENCES
Karjalainen, M., Airaksinen, S., Rantanen, J., Aaltonen, J., and Yliruusi, J. 2005. Characterization of polymorphic solid-state changes using variable temperature X-ray powder diffraction. Journal of Pharmaceutical and Biomedical Analysis, 39, 27-32.
Kaushal, A.M., Gupta, P., and Bansal, A.K. 2004. Amorphous drug delivery systems: Molecular aspects, design, and performance. Critical Reviews in Therapeutic Drug Carrier Systems, 21, 133-193.
Ketolainen, J., Poso, A., Viitasaari, V., Gynther, J., Pirttimäki, J., Laine, E., and Paronen, P. 1995. Changes in solid-state structure of cyclophosphamide monohydrate induced by mechanical treatment and storage. Pharmaceutical Research, 12, 299-304.
Khoskhoo, S. and Anwar, J. 1993. Crystallization of polymorphs: the effect of solvent. Journal of Physics D: Applied Physics, 26, 90-93.
Kishi, A., Otsuka, M., and Matsuda, Y. 2002. The effect of humidity on dehydration behavior of nitrofurantoin monohydrate studied by humidity controlled simultaneous instrument for X-ray diffractometry and differential scanning calorimetry (XRD-DSC). Colloids and Surfaces B: Biointerfaces, 25, 281-291.
Kojic-Prodic, B. and Ruzic-Toros, Z. 1982. Structure of the anti-inflammatory drug 4-hydroxy-2-methyl-N-2-pyridyl-2H-1λ6, 2-benzothiazine-3-carboxamide 1,1-dioxide (piroxicam). Acta Crystallographica B, 38, 2948-2951.
Kristl, A., Srcic, S., Vrecer, F., Sustar, B., and Vojnovic, D. 1996. Polymorphism and pseudopolymorphism: Influencing the dissolution properties of the guanine derivative acyclovir. International Journal of Pharmaceutics, 139, 231-235.
Kruger, G.J. and Gafner, G. 1971. The crystal structure of sulphathiazole II. Acta Crystallographica B, 27, 326-333.
Kruger, G.J. and Gafner, G. 1972. The crystal structures of polymorphs I and III of sulphathiazole. Acta Crystallographica B, 28, 272-283.
Lagas, M. and Lerk, C.F. 1981. The polymorphism of sulphathiazole. International Journal of Pharmaceutics, 8, 11-24.
Laihanen, N., Muttonen, E., and Laaksonen, M. 1996. Solubility and intrinsic dissolution rate of alprazolam crystal modifications. Pharmaceutical Development and Technology, 1, 373-380.
Langkilde, F.W., Sjöström, M., Tekenbergs-Hjelte, L., and Mrak, J. 1997. Quantitative FT-Raman analysis of two crystal forms of a pharmaceutical compound. Journal of Pharmaceutical and Biomedical Analysis, 15, 687-696.
Lee, A.Y., Lee, I.S., Dette, S.S., Boerner, J., and Myerson, A.S. 2005. Crystallization on confined engineered surfaces: a method to control crystal size and generate different polymorphs. Journal of the American Chemical Society, 127, 14982-14983.
Lee, T., Hung, T., and Kuo, C.S. 2006. Polymorph farming of acetaminophen and sulfathiazole on a chip. Pharmaceutical Research, 23, 2542-2555.
Leung, S.S., Padden, B.E., Munson, E.J., and Grant, D.J.W. 1998. Solid-state characterization of two polymorphs of aspartame hemihydrate. Journal of Pharmaceutical Sciences, 87, 501-507.
Liu, H.-B., Chen, Y., and Zhang, X.C. 2006. Characterization of anhydrous and hydrated pharmaceutical materials with THz time-domain spectroscopy. Journal of Pharmaceutical Sciences, In Press,
Lohani, S. and Grant, D.W.J., Thermodynamics of polymorphs, in Polymorphism in the Pharmaceutical Industry, R. Hilfiker, Editor. 2006, WILEY-VCH Verlag GmbH: Weinheim. p. 21-42.
Luner, P.E., Majuru, S., Seyer, J.J., and Kemper, M.S. 2000. Quantifying crystalline form composition in binary powder mixtures using near-infrared reflectance spectroscopy. Pharmaceutical Development and Technology, 5, 231-246.
Mackin, L., Zanon, R., Park, J.M., Foster, K., Opalenik, H., and Demonte, M. 2002. Quantification of low levels (<10%) of amorphous content in micronised active
37
REFERENCES
batches using dynamic vapour sorption and isothermal microcalorimetry. International Journal of Pharmaceutics, 231, 227-236.
Matsumoto, T., Kaneniwa, N., Higuchi, S., and Otsuka, M. 1991. Effects of temperature and pressure during compression on the polymorphic transformation and crushing strength of chlorpropamide tablets. Journal of Pharmacy and Pharmacology, 43, 74-78.
McCrone, W.C. 1965. Polymorphism. Physics and Chemistry of the Organic Solid State, 2, 725-767.
Meyer, M.C., Straughn, A.B., Jarvi, E.J., Wood, G.C., Pelsor, F.R., and Shah, V.P. 1992. The bioinequivalence of carbamazepine tablets with a history of clinical failures. Pharmaceutical Research, 9, 1612-1616.
Morris, K.R., Structural aspects of hydrates and solvates, in Polymorphism in Pharmaceutical Solids, H.G. Brittain, Editor. 1999, Marcel Dekker, Inc.: New York. p. 125-181.
Morris, K.R., Griesser, U.J., Eckhardt, C.J., and Stowell, J.G. 2001. Theoretical approaches to physical transformations of active pharmaceutical ingredients during manufacturing processes. Advanced Drug Delivery Reviews, 48, 91-114.
Mullin, J.W., Industrial techniques and equipment, in Crystallization. 2001, Butterworth Heinemann: Oxford. p. 315-402.
Nichols, G. and Frampton, C.S. 1998. Physicochemical characterization of the orthorhombic polymorph of paracetamol crystallized from solution. Journal of Pharmaceutical Sciences, 87, 684-693.
Norris, T., Aldridge, P.K., and Sekulic, S.S. 1997. Determination of end-points for polymorph conversions of crystalline organic compounds using on-line near-infrared spectroscopy. Analyst, 122, 549-552.
Ohta, M. and Buckton, G. 2004. Determination of the changes in surface energetics of cefditoren pivoxil as a consequence of processing induced disorder and equilibration to different relative humidities. International Journal of Pharmaceutics, 269, 81-88.
Okumura, T. and Otsuka, M. 2005. Evaluation of the microcrystallinity of a drug substance, indomethacin, in a pharmaceutical model tablet by chemometric FT-Raman spectroscopy. Pharmaceutical Research, 22, 1350-1357.
Ostwald, W. 1896. Studien über die Bildnung und Umwandlung fester Körper. Zeitschrift für Physicalische Chemie, 22, 289-330.
Otsuka, M. and Kaneniwa, N. 1986. Effect of seed crystals on solid-state transformation of polymorphs of chloramphenicol palmitate during grinding. Journal of Pharmaceutical Sciences, 75, 506-511.
Otsuka, M., Matsumoto, T., and Kaneniwa, N. 1989. Effects of the mechanical energy of multi-tableting compression on the polymorphic transformations of chlorpropamide. Journal of Pharmacy and Pharmacology, 41, 665-669.
Otsuka, M., Teraoka, R., and Matsuda, Y. 1992. Rotating-disk dissolution kinetics of nitrofurantoin anhydrate and monohydrate at various temperatures. Pharmaceutical Research, 9, 307-311.
Otsuka, M., Hasegawa, H., and Matsuda, Y. 1999a. Effect of polymorphic forms of bulk powders on pharmaceutical properties of carbamazepine granules. Chemical & Pharmaceutical Bulletin, 47, 852-856.
Otsuka, M., Ofusa, T., and Matsuda, Y. 1999b. Effect of environmental humidity on the transformation pathway of carbamazepine polymorphic modifications during grinding. Colloids and Surfaces B: Biointerfaces, 13, 263-273.
Otsuka, M., Kato, F., and Matsuda, Y. 2004. Effect of temperature and kneading solution on polymorphic transformation of mefenamic acid during granulation. Solid State Ionics, 172, 451-453.
38
REFERENCES
Patel, A.D., Luner, P.E., and Kemper, M.S. 2001. Low-level determination of polymorph composition in physical mixtures by near-infrared reflectance spectroscopy. Journal of Pharmaceutical Sciences, 90, 360-370.
Peltonen, L., Liljeroth, P., Heikkilä, T., Kontturi, K., and Hirvonen, J. 2003. Dissolution testing of acetylsalicylic acid by a channel flow method - correlation to USP basket and intrinsic dissolution methods. European Journal of Pharmaceutical Sciences, 19, 395-401.
Peterson, M.L., Morissette, S.L., McNulty, C., Goldsweig, A., Shaw, P., LeQuesne, M., Monagle, J., Encina, N., Marchionna, J., Johnson, A., Gonzalez-Zugasti, J., Lemmo, A.V., Ellis, S.J., Cima, M.J., and Almarsson, O. 2002. Iterative high-throughput polymorphism studies on acetaminophen and an experimentally derived structure for form III. Journal of the American Chemical Society, 124, 10958-10959.
Petit, S. and Coquerel, G., The amorphous state, in Polymorphism: in the Pharmaceutical Industry, R. Hilfiker, Editor. 2006, WILEY-VCH Verlag GmbH: Weinheim. p. 259-285.
Phadnis, N.V. and Suryanarayanan, R. 1997. Polymorphism in anhydrous theophylline - implications on the dissolution rate of theophylline tablets. Journal of Pharmaceutical Sciences, 86, 1256-1263.
PhRMA. 2005. New drug approvals in 2004.
Pienaar, E.W., Caira, M.R., and Lötter, A.P. 1993a. Polymorphs of nitrofurantoin. 2. Preparation and X-ray crystal structures of two anhydrous forms of nitrofurantoin. Journal of Crystallographic and Spectroscopic Research, 23, 785-790.
Pienaar, E.W., Caira, M.R., and Lötter, A.P. 1993b. Polymorphs of nitrofurantoin. 1. Preparation and X-ray crystal structures of two monohydrated forms of nitrofurantoin. Journal of Crystallographic and Spectroscopic Research, 23, 739-744.
Price, S.L. 2004. The computational prediction of pharmaceutical crystal structures and polymorphism. Advanced Drug Delivery Reviews, 56, 301-319.
Rantanen, J., Wikström, H., Rhea, F.E., and Taylor, L.S. 2005. Improved understanding of factors contributing to quantification of anhydrate/hydrate powder mixtures. Applied Spectroscopy, 59, 942-951.
Rantanen, J., Aaltonen, J., Hirvonen, J., and Peltonen, L., PCT application WO 2006/108908 A1: Method and apparatus for dissolving solid matter in liquid. 2006.
Reck, G., Dietz, G., Laban, G., Gunther, W., Bannier, G., and Hohne, E. 1988. X-ray studies on piroxicam modifications. Pharmazie, 43, 477-481.
Remenar, J.F., MacPhee, J.M., Larson, B.K., Tyagi, V.A., Ho, J.H., McIlroy, D.A., Hickey, M.B., Shaw, P.B., and Almarsson, O. 2003. Salt selection and simultaneous polymorphism assessment via high-throughput crystallization: the case of sertraline. Organic Process Research & Development, 7, 990-996.
Räsänen, E., Rantanen, J., Jørgensen, A., Karjalainen, M., Paakkari, T., and Yliruusi, J. 2001. Novel identification of pseudopolymorphic changes of theophylline during wet granulation using near infrared spectroscopy. Journal of Pharmaceutical Sciences, 90, 389-396.
Räsänen, E., Rantanen, J., Mannermaa, J.-P., Yliruusi, J., and Vuorela, H. 2003. Dehydration studies using a novel multichamber microscale fluid bed dryer with in-line near-infrared measurement. Journal of Pharmaceutical Sciences, 92, 2074-2081.
Salameh, A.K. and Taylor, L.S. 2006. Physical stability of crystal hydrates and their anhydrates in the presence of excipients. Journal of Pharmaceutical Sciences, 95, 446-461.
Sandler, N., Rantanen, J., Heinämäki, J., Römer, M., Marvola, M., and Yliruusi, J. 2005. Pellet manufacturing by extrusion-spheronization using process analytical technology. AAPS PharmSciTech, 6, 174-183.
39
REFERENCES
Savitzky, A. and Golay, M.J.E. 1964. Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry, 44, 1627-1639.
Schmidt, A.G., Wartewig, S., and Picker, K.M. 2003. Potential of carrageenans to protect drugs from polymorphic transformation. European Journal of Pharmaceutics and Biopharmaceutics, 56, 101-110.
Seyer, J.J., Luner, P.E., and Kemper, M.S. 2000. Application of diffuse reflectance near-infrared spectroscopy for determination of crystallinity. Journal of Pharmaceutical Sciences, 89, 1305-1316.
Seyer, J.J. and Luner, P.E. 2001. Determination of indomethacin crystallinity in the presence of excipients using diffuse reflectance near-infrared spectroscopy. Pharmaceutical Development and Technology, 6, 573-582.
Shakhtshneider, T.P. and Boldyrev, V.V. 1993. Phase transformations in sulfathiazole during mechanical activation. Drug Development and Industrial Pharmacy, 19, 2055-2067.
Shefter, E. and Higuchi, T. 1963. Dissolution behavior of crystalline solvated and nonsolvated forms of some pharmaceuticals. Journal of Pharmaceutical Sciences, 52, 781-791.
Singhal, D. and Curatolo, W. 2004. Drug polymorphism and dosage form design: a practical perspective. Advanced Drug Delivery Reviews, 56, 335-347.
Stagner, W.C. and Guillory, J.K. 1979. Physical characterization of solid iopanoic acid forms. Journal of Pharmaceutical Sciences, 68, 1005-1009.
Stahl, P.H., The problems of drug interactions with excipients, in Towards Better Safety of Drugs and Pharmaceutical Products, D.D. Braimer, Editor. 1980, Elsevier/North Holland Biomedical Press: Amsterdam.
Steckel, H., Rasenack, N., Villax, P., and Muller, B.W. 2003. In vitro characterization of jet-milled and in-situ-micronized fluticasone-17-propionate. International Journal of Pharmaceutics, 258, 65-75.
Storey, R., Docherty, R., Higginson, P., Dallman, C., Gilmore, C., Barr, G., and Dong, W. 2004. Automation of solid form screening procedures in the pharmaceutical industry - how to avoid the bottlenecks. Crystallography Reviews, 10, 45-56.
Strachan, C.J., Pratiwi, D., Gordon, K.C., and Rades, T. 2004a. Quantitative analysis of polymorphic mixtures of carbamazepine by Raman spectroscopy and principal components analysis. Journal of Raman Spectroscopy, 35, 347-352.
Strachan, C.J., Rades, T., Newnham, D.A., Gordon, K.C., Pepper, M., and Taday, P.F. 2004b. Using terahertz pulsed spectroscopy to study crystallinity of pharmaceutical materials. Chemical Physics Letters, 390, 20-24.
Strachan, C.J., Taday, P.F., Newham, D.A., Gordon, K.C., Zeitler, J.A., Pepper, M., and Rades, T. 2005. Using terahertz pulsed spectroscopy to quantify pharmaceutical polymorphism and crystallinity. Journal of Pharmaceutical Sciences, 94, 837-846.
Suihko, E., Ketolainen, J., Poso, A., Ahlgren, M., Gynther, J., and Paronen, P. 1997. Dehydration of theophylline monohydrate--a two step process. International Journal of Pharmaceutics, 158, 47-55.
Suihko, E., Lehto, V.-P., Ketolainen, J., Laine, E., and Paronen, P. 2001. Dynamic solid-state and tableting properties of four theophylline forms. International Journal of Pharmaceutics, 217, 225-236.
Sun, C. and Grant, D.J.W. 2001. Influence of crystal structure on the tableting properties of sulfamerazine polymorphs. Pharmaceutical Research, 18, 274-280.
Sun, C., Zhou, D., Grant, D.J.W., and Young, V.G. 2002. Theophylline monohydrate. Acta Crystallographica E, 58, o368-o370.
Taday, P.F., Bradley, I.V., Arnone, D.D., and Pepper, M. 2003. Using terahertz pulse spectroscopy to study the crystalline structure of a drug: a case study of the polymorphs of ranitidine hydrochloride. Journal of Pharmaceutical Sciences, 92, 831-838.
40
REFERENCES
Taylor, L.S. and Zografi, G. 1998. The Quantitative Analysis of Crystallinity Using FT-Raman Spectroscopy. Pharmaceutical Research, 15, 755-761.
Taylor, L.S. and Langkilde, F.W. 2000. Evaluation of solid-state forms present in tablets by Raman spectroscopy. Journal of Pharmaceutical Sciences, 89, 1342-1353.
Threlfall, T.L. 1995. Analysis of organic polymorphs. A review. Analyst, 120, 2435-2460.
Ticehurst, M.D., Basford, P.A., Dallman, C.I., Lukas, T., M. , Marshall, P., V., Nichols, G., and Smith, D. 2000. Characterisation of the influence of micronisation on the crystallinity and physical stability of revatropate hydrobromide. International Journal of Pharmaceutics, 193, 247-259.
Tudor, A., Church, S., Hendra, P., Davies, M., and Melia, C. 1993. The qualitative and quantitative analysis of chlorpropamide polymorphic mixtures by near-infrared fourier transform Raman spectroscopy. Pharmaceutical Research, 10, 1772-1776.
Ward, G.H. and Schultz, R.K. 1995. Process-induced crystallinity changes in albuterol sulfate and its effect on powder physical stability. Pharmaceutical Research, 12, 773-779.
Wikström, H., Marsac, P.J., and Taylor, L.S. 2005. In-line monitoring of hydrate formation during wet granulation using Raman spectroscopy. Journal of Pharmaceutical Sciences, 94, 209-219.
Wong, M.W.Y. and Mitchell, A.G. 1992. Physicochemical characterization of a phase change produced during the wet granulation of chlorpromazine hydrochloride and its effects on tableting. International Journal of Pharmaceutics, 88, 261-273.
Yu, L. 2001. Amorphous pharmaceutical solids: preparation, characterization and stabilization. Advanced Drug Delivery Reviews, 48, 27-42.
Yu, L.X., Furness, M.S., Andre, R., Woodland Outlaw, K.P., Nashed, N.E., Ramos, E., Miller, S.P.F., Adams, R.C., Fang, F., Patel, R.M., Holcombe, F.O., Chiu, Y.-Y., and Hussain, A.S. 2003. Scientific considerations of pharmaceutical solid polymorphism in abbreviated new drug applications. Pharmaceutical Research, 20, 531-536.
Zeitler, J.A., Kogermann, K., Rantanen, J., Rades, T., Taday, P.F., Pepper, M., Aaltonen, J., and Strachan, C.J. 2006a. Drug hydrate systems and dehydration processes studied by terahertz pulsed spectroscopy. International Journal of Pharmaceutics, In Press,
Zeitler, J.A., Newnham, D.A., Taday, P.F., Threlfall, T.L., Lancaster, R.W., Berg, R.W., Strachan, C.J., Pepper, M., Gordon, K.C., and Rades, T. 2006b. Characterization of temperature-induced phase transitions in five polymorphic forms of sulfathiazole by terahertz pulsed spectroscopy and differential scanning calorimetry. Journal of Pharmaceutical Sciences, 95, 2486-2498.
Zhang, G.G.Z., Law, D., Schmitt, E.A., and Qiu, Y. 2004. Phase transformation considerations during process development and manufacture of solid oral dosage forms. Advanced Drug Delivery Reviews, 56, 371-390.
