Executive Summary : | Pharmaceutical industries require reliable quantitative analysis of polymorphs (different crystalline forms of the same drug) for obtaining drug product approvals by regulatory agencies. Powder X-Ray Diffraction (XRD) is the most reliable technique to distinguish polymorphs and estimate their percentage. The desirable criteria are that the drug must be phase pure and the industry must prove that no other polymorph exists or should transform during storage or stability studies. Reflection method, also known as Bragg-Brentano geometry, is commonly used in this task, but major problems such as sample homogeneity, flat specimen error, particle size, and preferred orientation of crystallites lead to large intensity variations and inconsistent results. A single non-overlapping peak is often used, and the information from rest other peaks is neglected. Many pharmaceutical industries need support in this specialized area. To address these issues, X-ray diffraction studies in transmission geometry are proposed. Transmission geometry minimizes preferred orientations in the bulk sample, providing correct relative intensities. Reflection geometry, on the other hand, scatters X-rays from the surface of the sample, masking some parts of the sample to be analyzed, affecting the analysis and quality of results. Big data statistical routines such as Principal Component Analysis (PCA), Partial Least Square Regression (PLSR), and Hierarchical Clustering algorithms (HCA) are proposed for multivariate data analysis. Four challenging polymorphic systems of drug molecules have been identified, and the work will focus on these systems to better understand the merits and demerits of transmission X-ray studies over reflection geometry and develop a novel and reliable method of analysis for various pharmaceutical industries in India. |