Making Sense of Disorder in Molecular Solids Using Solid-State NMR

Abstract

Crystalline materials are typically characterised by diffraction-based techniques which excel at solving the structures of well-ordered materials. However, these techniques have difficulty characterising disorder, as it disrupts the long-range order needed for defined Bragg peaks. Solid-state nuclear magnetic resonance (NMR) is an excellent tool for studying disorder, as it is sensitive to local environment and can detect motion over many orders of magnitude. In this work, NMR-based techniques, primarily lineshape and relaxation analyses, are used in combination with diffraction and molecular dynamics (MD) simulations to uncover the dynamic disorder in three types of system. Firstly, despite limited amounts of some samples, NMR is able to characterise the dynamics of diamondoids, a series of rigid hydrocarbon cages. In diamantane and 1(2)3 tetramantane, 13C relaxation shows evidence of C3 rotations with activation energies of 21.1(4) kJ mol−1 and 15(2) kJ mol−1 respectively. For triamantane, second moments, a historical method of summarising static 1H lineshapes, show that the molecules undergo multi-axis jumps, information that could not be obtained through modern NMR or diffraction-based techniques. Secondly, the phase transition in the relaxor ferroelectric material, hydrazinium magnesium formate, is shown to be caused by reorientation of hydrazinium ions from perpendicular to parallel in the channels. Here, a new mechanism is proposed whereby the relaxor ferroelectric response arises directly because of the molecular motion, not despite it. Finally, MD simulations reveal that the dynamics of the solvent molecules in two cocrystal solvates of furosemide-picolinamide have significant librational character. This explains why lineshape and relaxation analyses, which assume a constant amplitude with temperature, are unable to provide a coherent dynamic model. In summary, the NMR techniques used herein, along with supporting diffraction and computational tools, supports their utilisation in the future to better understand disordered materials.