A classification system to assess the crystallization tendency of organic molecules from undercooled melts.

Assessing the viability of an amorphous formulation strategy is of great importance in an era of drug discovery where a large percentage of new molecules have solubility limited dissolution rates, and disruption of the crystal lattice is a potential strategy to improve this process. The objective of the current study was to evaluate the glass forming ability (GFA) of a large data set of organic molecules and also to evaluate potential links between GFA and glass stability (GS). The crystallization tendency from the undercooled melt was evaluated for a group of 51 organic molecules and separated into three separate classes [class (I), class (II), class (III)] based upon the presence/absence of observable crystallization during a heating/cooling/heating cycle, as measured using differential scanning calorimetry (DSC). Class (I) molecules were further delineated based upon the observation of a crystalline [class (I-A)] or amorphous [class (I-B)] solid after quench cooling in liquid N(2). Principal component analysis (PCA) of various physiochemical descriptors suggested that molecules with low GFA tended to be low molecular weight (MW), rigid structures while class (III) molecules tended to be higher MW, more complex structures. For select compounds, it was observed that crystallization from the glassy state was much faster for compounds with a lower GFA. It is believed that nuclei are quenched into the glass during cooling for class (I-B) and (II) molecules, leading to more facile crystallization below T(g). In addition, these quenched in nuclei are also thought to be responsible for the recrystallization observed for these classes of molecules upon heating above T(g). In conclusion, the DSC screening method and classification scheme may be a useful tool to quickly assess the GFA and potential GS of new chemical entities during early drug development.