Introduction
X-Ray Powder Diffraction (XRPD) continues to be an indispensable analytical technique supporting a wide variety of pharmaceutical development activities. This Perspective will highlight some of the current developments and emerging technologies of XRPD. For the interested reader, a comprehensive review of pharmaceutical applications of XRPD is available elsewhere [1].
Characterization of Pharmaceuticals by XRPD in Solid Form Screening
Solid form screening will typically involve preparation of hundreds if not thousands of samples, which are then analyzed by XRPD [2]. A sample exhibiting a novel XRPD pattern is taken as potential evidence of the existence of a new solid phase for a given material, which can then be investigated further [3].
The push to generate a maximum number samples in “scorched Earth” solid form screening using a minimum amount of material has been facilitated by advances in both crystallization technologies as well as XRPD instrumentation. It is now possible to conduct a comprehensive polymorph screen with a few grams or less of starting material. The use of automated XRPD sample stages has increased throughput such that the bottleneck in a screening program is no longer at the samples analysis stage. The requirement to interpret and classify the multitude of diffraction patterns that are generated in automated solid form screens has led to the development of algorithms and software to classify novel XRPD patterns, thereby further streamlining the steps needed to identify novel solid forms [4,5]. Various instrument vendors and other providers now offer XRPD pattern cluster analysis software for this task.
While automated screening processes have resulted in larger numbers of samples, the analysis of a small quantity of material for each individual sample is frequently challenging with respect to obtaining a useful XRPD pattern. Too few crystals do not provide the necessary population of random orientations to properly define a given XRPD peak, while crystals oriented preferentially in one direction will also not provide a diffraction pattern representative of the crystalline form. A common artifact of preferred orientation is that peak intensities are distorted (both increased and diminished) when compared to the ideal diffraction pattern. Grinding the sample to overcome this issue is usually not an option due to sample size limitations and the need to analyze the sample within the crystallization vessel. For samples analyzed in capillary tubes, the best alternative is to spin the sample during analysis. XRPD instruments may also have the option of rocking the goiniometer during analysis, which also serves to minimize the effects of preferred orientation and poor particle statistics. Alternatively, samples may also be individually mounted in a Gandolfi (precession) spinner, which has the effect of providing random or nearly-random orientations of the sample. By using this sample mount, it is possible to produce a “powder” diffraction pattern from sample containing a single crystal.
Variable Temperature and Relative Humidity XRPD
Solid phase transitions such as polymorph interconversions are routinely examined by XRPD using variable temperature sample stages (VT-XRPD). Both sub-ambient and elevated temperature stages are available. While thermal techniques such as DSC and TGA are also widely used for these studies, VT-XRPD permits the direct identification of crystalline phase as a function of temperature. Most pharmaceutical laboratories rely on both technologies. Indeed, instruments are available that permit simultaneous XRPD and DSC sample analyses [6].
VT-XRPD is also widely used to study the thermal stabilities of pharmaceutical hydrates. As an example, Borghetti, et al. characterized the flavonoid quercetin obtained from three different commercial suppliers by a variety of analytical techniques including VT-XRPD [7]. All three samples were found to contain different crystalline phases. By examining the changes in the diffraction patterns as a function of temperature the authors were able to determine that one of the three commercial samples consisted of a mixture of the known crystalline hydrate and another solid phase.
XRPD is also commonly used to investigate the structure of variable hydrates, which are crystalline species that contain non-stoichiometric amounts of water held within channels present in the crystal lattice. The amount of water present in a variable hydrate is typically a function of the relative humidity (RH) environment of the sample. Since the XRPD peak positions are directly related to the dimensions of the unit cell, subtle changes in the size of the unit cell due to the presence of water may be evaluated by comparison of XRPD patterns collected for the species under different RH environments. In certain systems, an increase in a particular lattice parameter may be a direct function of the amount of water present in the hydrated structure [8]. Recently, Suzuki, et al. used XRPD along with with other techniques to determine the structure of the variable hydrate of sitafloxacin hydrate [9]. This species contains both water molecules that are integral to the crystal structure along with amounts of loosely bound water molecules that vary with RH. Vogt, et al. also demonstrated through XRPD measurements that the variable hydrate structure of topotecan hydrochloride undergoes no change in lattice dimensions upon dehydration from a pentahydrated to a trihydrated species [10]. While variable RH XRPD studies can be conducted on samples externally conditioned under various RH environments, it is also possible to conduct these analyses in situ using XRPD instruments equipped with variable RH sample stages.
XRPD may also be used to probe the hydration states of noncrystalline materials. For example, the structure of the common excipient poly(vinylpyrrolidone) (PVP) was analyzed by XRPD after equilibration under various RH environments [11]. Both the positions and the relative intensities of halos in the diffraction patterns were found to be diagnostic of the presence of free water. These observations were then analyzed along with gravimetric measurements of the PVP water content using chemometric methods, which demonstrated that that the amount of free water present in PVP was a linear function of relative humidity environment under which the sample was equilibrated.
VT-XRPD measurements are not necessarily constrained to elevated temperatures. An application of sub-ambient VT-XRPD involved measurement of the thermal expansion coefficients for urea over the temperature range of 188 – 328 K [12]. The smooth change in lattice parameters over the temperature range studied indicated the absence of phase transitions under these conditions. This study illustrates the well-known principle that XRPD peak positions will vary with temperature, typically in a non-uniform fashion due to different thermal expansion coefficients for the different axes of the unit cell.
Crystal Structure Analysis Using XRPD
Characterization of a crystalline pharmaceutical drug substance frequently involves obtaining a structural solution by single-crystal X-ray crystallography. This crystallographic data set can also be used to calculate a powder diffraction pattern for that particular crystalline form, which can serve as a reference pattern for subsequent XRPD analyses. However, it is important to recognize that such direct comparisons are only valid when both the experimental and calculated XRPD data sets are collected at the same temperature. This is frequently not the case as it is advantageous to collect single crystal diffraction data at temperatures of ~ 100 K. Depending on the thermal expansion properties of the material, the differences between the calculated XRPD pattern obtained from crystallographic data collected at low temperature and the experimental XRPD pattern collected at room temperature can be profound. If the crystal can be indexed at room temperature, then these cell parameters may be used to calculate a room temperature powder diffraction pattern.
One of the most exciting research areas involving XRPD is directed toward obtaining structural solutions directly from powder diffraction data. While it will always be advantageous to solve the crystal structure using single-crystal diffraction, this is not always possible as some crystalline substances do not possess the necessary size and quality. Traditionally, structural analyses using powder diffraction data are conducted using patterns obtained from synchrotron sources [13]. Recent advances in the performance of XRPD instrumentation (particularly with respect to peak resolution) permit the use of data obtained from conventional laboratory diffractometers. In addition, the development of direct-space methods for structure solution require only that the XRPD peak shape and width functions of the XRPD patterns be accurately defined [14]. Recently, these techniques were applied to obtain crystal structures for three out of five polymorphs of m-aminobenzoic acid [15]. An interesting aspect of this study is the discovery that some of the polymorphs m-aminobenzoic acid were shown to be zwitterionic, while others were non-zwitterionic.
At present, the methodologies for structure solution from XRPD data are highly complex and are best left to the expert practitioners in this field. Even for the m-aminobenzoic acid study cited above, the investigators were unable to obtain a crystal structure for the Form I polymorph of this compound. When a complete structural solution is not possible from XRPD, it may still be possible to use a high-quality XRPD pattern to obtain useful information such as the unit cell parameters (indexing) and the crystalline space group. A successfully indexed powder diffraction pattern is a strong indication that the sample contains a single crystalline phase [1]. Hopefully, advances in computational techniques and software will provide more user-friendly tools for structure determination from XRPD data.
Summary Advances in XRPD instrumentation and software have improved the efficiency of pharmaceutical solid form screening, such that hundreds of XRPD patterns may be rapidly analyzed and interpreted. Variable temperature and humidity XRPD techniques are especially powerful in understanding structural changes of pharmaceutical substances under these conditions. Ongoing developments involving structure determination directly from XRPD are encouraging in that eventually such analyses may be commonly used to determine the crystal structures of pharmaceuticals.
References
1. I. Ivanisevic, R. B. McClurg, P J. Schields, Uses of X-Ray Powder Diffraction In the Pharmaceutical Industry, Pharmaceutical Sciences Encyclopedia: Drug Discovery, Development, and Manufacturing, Wiley (2010) pp 1-42.
2. J. Aaltonen, M. Alleso, S. Mirza, V. Koradia, K. C. Gordon, J. Rantanen Solid form screening – A review Eur. J. Pharm. Biopharm. 71, 23–37 (2009).
3. A. Newman, X-ray Powder Diffraction in Solid Form Screening and Selection, Am. Pharm. Rev. 14, 44-46,48-51, (2011).
4. R. Storey, R. Docherty, P. Higginson, C. Dallman, C. Gilmore, G. Barr, W. Dong, Automation of solid form screening procedures in the pharmaceutical industry – how to avoid the bottlenecks, Crystallogr. Rev. 10 45–56 (2004).
5. Ivanisevic, D. Bugay, S. Bates, On pattern matching of X-ray powder diffraction data J. Phys. Chem B, 109, 7781-7787 (2005).
6. A. Kishi, Detailed observations of dynamic changes such as phase transitions, melting and crystallization using an XRD-DSC with a high-speed, high-sensitivity two-dimensional PILATUS detector Rigaku J. 27, 9-14 (2011).
7. G. S. Borghettia, J. P. Carinia, S. B. Honoratob, A. P. Ayalab, J. C. F. Moreirac, V. L. Bassania, Physicochemical properties and thermal stability of quercetin hydrates in the solid state, Thermochim. Acta 539, 109–114 (2012).
8. F. G. Vogt, G. R. Williams, Advanced approaches to effective solid-state analysis: X-ray diffraction, vibrational spectroscopy, and solid-state NMR, Am. Pharm. Rev. 13, 58-65 (2010).
9. T. Suzuki, T. Araki, H. Kitaoka, K. Teradab, Characterization of non-stoichiometric hydration and the Dehydration Behavior of Sitafloxacin Hydrate Chem. Pharm. Bull. 60, 45-55 (2012).
10. F. G. Vogt, P. C. Dell’Orco, A. M. Diederich, Q. Su, J. L. Wood, G. E. Zuber, L. M. Katrincic, R. L. Mueller, D.J. Busby, C. W. DeBrosse, A study of variable hydration states in topotecan hydrochloride, J. Pharm. Biomed. Anal. 40, 1080-1088 (2006).
11. J. Teng, S. Bates, D. A. Engers, K. Leach, P. Schields, Y. Yang, Effect of water vapor sorption on local structure of poly(vinylpyrrolidone), J. Pharm. Sci. 99, 3815-3825 (2010).
12. R. Hammond, K. Pencheva, K. J. Roberts, P. Mougin, D. Wilkinson, An examination of the thermal expansion of urea using high-resolution  varible temperature  X-ray powder diffraction, Appl. Crystallogr. 38, 1038 -1039 (2005).
13. M. Hušak, A. Jegorov, J. Brus, W. van Beek, P. Pattison, M. Christensen, V. Favre-Nicolin, J. Maixner, Metergoline II: structure solution from powder diffraction data with preferred orientation and from microcrystal, Struct. Chem. 19, 517-525 (2008).
14. K. D. M. Harris, Powder diffraction crystallography of molecular solids, Top. Curr. Chem., 315, 133–178 (2012).
15. P. A. Williams, C. E. Hughes, G. K. Lim, B. M. Kariuki, K. D. M. Harris, Discovery of a New System Exhibiting Abundant Polymorphism: m-Aminobenzoic Acid Cryst. Growth Des. 12, 3104−3113 (2012).
Author Biography
Dr. Leonard Chyall is the President of Chyall Pharma where he provides scientific consulting services related to the CMC aspects of the drug development process. Dr. Chyall is an expert in the identification and characterization of polymorphs, salts, cocrystals and amorphous materials. Dr. Chyall began his career as a pharmaceutical chemist at SSCI, Inc. and subsequently at Aptuit, Inc. He is a coauthor of over 20 publications and holds 4 patents granted by the US Patent and Trademark Office. Dr. Chyall received an A.B. degree in Chemistry from Oberlin College and a Ph.D. in Chemistry from University of Minnesota.