Control of polymorphism, crystal size and habit in pharmaceuticals

Crystallization is widely used as a purification method of Active Pharmaceutical Ingredients, APIs. During the process a crystalline compound is generally obtained, and it can display general properties and characteristics based on the way molecules organize in the space. This molecule organization can be responsible for several characteristics such as polymorphism, morphology, particle size distribution, and consequently bioavailability. For these reasons we selected APIs subject to polymorphism and displaying needle-like morphology. Needle-like crystals are difficult to manufacture due to poor flow property, compressibility and differential dissolution, as long as particle breakage. These two characteristics can mainly affect solubility, processability and bioavailability of the API. Diflunisal was selected for its propensity to generate several polymorphic forms, all of which happen to be needle like crystals. For these reasons we studied how diflunisal can generate and modify its polymorphic forms when subject to a variety of methods. In order to analyse the outcomes, spectroscopic methods were used, because they are sensitive detections tools towards small change in functional group interactions, therefore useful for the identifications of a specific polymorphic form. It is also know how solvents can modify the crystals morphology, therefore several solvents where used in the process to try and change the habit of the polymorphs but no solvent was able to effectively modify the crystal habit. Because of this ineffective behaviour different approaches were taken in consideration. Upon evaluation of the driving force of the crystallization we developed several methods capable of stopping needle growth in this system. A co-crystallization approach was developed. Several co-formers were successfully identified with the aim to create a co-crystal and interrupt the van der Waals contact stacking interaction responsible for the needle growth. A combination of molecular searching tools from the Mercury® program combined with principal component analysis was used in the process to narrow down the number of possible candidates. Using this approach, a small library of molecules was isolated and used in different stoichiometric ratios. Despite the fact several molecules were founds to be effective in create co-crystals this remain a trial and error approach which remain time consuming and not always successful. The needle growth formation was successfully stopped. Regrettably not all the 3 isolated molecules were classified as GRAS. Furthermore, we designed and developed a method based on thermodynamic principle, successfully used to modify the aspect ratio of two different APIs, diflunisal and isonicotinohydrazide, both of which grow as needle like crystals. Using a High-Shear Ultralow Attrition Agitation apparatus we promoted the crystal growth via two factors, Ostwald ripening and maximization of the mass transfer. Experiment shows a significant decrease in aspect ratio after 14 days of constant agitation. Based on the successful results of this approach we identified a co-crystal (BZA-INA) capable of growing as needle-like crystals. We developed a temperature cycling approach capable of greatly changing the aspect ratio in a reduced amount of temperature cycles. To maximise\ the temperature cycling effect we selected a small library of compounds used to modify the growth kinetics of a specific face. Two molecules in particular were able to amplify the already significant effect of the temperature cycling further affecting the aspect ratio of the final compounds. During this work we demonstrate how to perform a polymorphism interconversion analysis followed by stress test to evaluate the stability of API polymorphic form.

Funder

Science Foundation Ireland

Publisher

NUI Galway

Rights

Attribution-NonCommercial-NoDerivs 3.0 Ireland

cbn

Collections

University of Galway Theses (PhD Theses)