Identification of a thermodynamically stable polymorph
is an important
step in the early stage of drug development. Ritonavir (RIT) is a
well-known case where the most stable polymorph II emerged after being
marketed, leading to a loss of $250 million. Herein, we report the
findings that routine melt crystallization can reveal the late-appearing
polymorph II of RIT at small supercooling, but the probability of
nucleation is very low. The addition of 30–50% polyethylene
glycol (PEG) promotes the crystallization of Form II as the only phase
at low supercooling, making it easier to detect in polymorphism screening.
During the course of our research, a new polymorph, denoted Form III,
was unexpectedly discovered, crystallizing as the major phase from
neat RIT melts. Single crystals of Form III were grown from melt microdroplets.
Benefiting from the ability of synchrotron radiation to detect weak
diffraction signals that cannot be accessible by a laboratory diffractometer,
a reasonable structure of Form III was solved with slight disorder
relative to thiazole groups (P1 space group and Z′ = 4). The thermodynamic stability ranking of the
three true polymorphs is Form II > Form I > Form III, as opposed
to
the order of solubility. The capacity to efficiently reveal rich polymorphs,
especially the kinetically hindered polymorph, and rapidly grow single
crystals of a new phase for structure determination together highlights
the necessity of incorporating melt crystallization into routine methods
for pharmaceutical polymorphism screening.