Reversible solid-state transformations are important in stimuli-responsive materials. The current understanding is limited on what kind of structure enables reversible transitions. We report that for molecular solids containing nitro groups, reversible phase transitions can occur if the nitro group is a hydrogen bond (HB) acceptor and has torsional freedom. We use the polymorphs of nifedipine (NIF) to illustrate this phenomenon taking advantage of their different molecular packing but identical chemical structure. NIF has six known polymorphs with four being kinetically stable at 100 K. Upon heating, two polymorphs undergo reversible phase transitions with large volume change, while the others do not. In the transforming structures, the nitro group is an HB acceptor and rotates to optimize HBs to offset loss of close packing, while in the inactive structures, the nitro group has similar torsional freedom but is not engaged in HBs. We test the generality of this phenomenon using all available systems in the Cambridge Structural Database, including NIF's derivative nisoldipine, and suggest possible applications in designing materials with controlled mechanical response.

The molecules at the surface of a liquid have different organization and dynamics from those in the bulk, potentially altering the rate of crystal nucleation and polymorphic selection, but this effect remains poorly understood. Here we demonstrate that nucleation at the surface of a pure liquid, d-arabitol, is vastly enhanced, by 12 orders of magnitude, and selects a different polymorph. The surface effect intensifies with cooling and can be inhibited by a dilute, surface-active second component. This phenomenon arises from the anisotropic molecular packing at the interface and its similarity to the surface-nucleating polymorph. Our finding is relevant for controlling the crystallization and polymorphism in any system with a significant interface such as nanodroplets and atmospheric water.

Crystal nucleation plays a critical role in the stability of supercooled liquids and glasses and is often controlled through addition of polymers. A dissolved polymer alters both the thermodynamics and the kinetics of nucleation, but the current understanding of these effects is limited. The rate of crystal nucleation has been measured in two molecular liquids, d-sorbitol and d-arabitol, containing polyvinylpyrrolidone (PVP) at different concentrations (0–15 wt %) and molecular weights (224 g/mol for the dimer up to 2 Mg/mol). We observe a significant inhibitory effect of PVP on crystal nucleation. Near the peak temperature for the nucleation rate (∼20 K above the glass transition temperature), 10 wt % PVP can slow down nucleation by approximately 1 order of magnitude, and the effect increases with polymer concentration exponentially and with molecular weight. Remarkably, the polymer effect on the nucleation rate is nearly the same as that on the crystal growth rate so that the ratio of the two rates is nearly constant at a given temperature independent of polymer concentration and molecular weight. This “master curve” behavior can be used to predict nucleation rates in multicomponent systems from more easily measured growth rates. It argues that nucleation and growth in these viscous liquids are both mobility-limited and that a polymer solute functions mainly as a mobility modifier, suppressing nucleation and growth to a similar degree.

We report that the stability of amorphous clofazimine (CFZ) against crystallization is vastly improved by salt formation with a polymer without sacrificing dissolution rate. A simple slurry method was used to produce the amorphous salt of CFZ with poly(acrylic acid) (PAA) at 75 wt % drug loading. The synthesis was performed under a mild condition suitable for thermally unstable drugs and polymers. Salt formation was confirmed by visible spectroscopy and glass temperature elevation. The amorphous salt at 75 wt % drug loading is remarkably stable against crystallization at 40 °C and 75% RH for at least 180 days. In contrast, the amorphous solid dispersion containing the un-ionized CFZ dispersed in poly(vinylpyrrolidone) crystallized in 1 week under the same condition. The high stability of the amorphous drug–polymer salt is a result of the absence of a drug–polymer crystalline structure, reduced driving force for crystallizing the free base, and reduced molecular mobility. Despite the elevated stability, the amorphous drug–polymer salt showed fast dissolution and high solution concentration in two biorelevant media (SGF and FaSSIF). Additionally, the amorphous CFZ–PAA salt has improved tabletability and powder flow relative to crystalline CFZ. The CFZ–PAA example suggests a general method to prepare amorphous drugs with high physical stability under tropical conditions and fast dissolution.

No abstract