Key points are not available for this paper at this time.
A general protocol is demonstrated for determining the structures of molecularly ordered but noncrystalline solids, which combines constraints provided by X-ray diffraction (XRD), one- and two-dimensional solid-state nuclear magnetic resonance (NMR) spectroscopy, and first-principles quantum chemical calculations. The approach is used to determine the structure(s) of a surfactant-directed layered silicate with short-range order in two dimensions but without long-range periodicity in three-dimensions (3D). The absence of long-range 3D molecular order and corresponding indexable XRD reflections precludes determination of a space group for this layered silicate. Nevertheless, by combining structural constraints obtained from solid-state (29)Si NMR analyses, including the types and relative populations of distinct (29)Si sites, their respective (29)Si-O-(29)Si connectivities and separation distances, with unit cell parameters (though not space group symmetry) provided by XRD, a comprehensive search of candidate framework structures leads to the identification of a small number of candidate structures that are each compatible with all of the experimental data. Subsequent refinement of the candidate structures using density functional theory calculations allows their evaluation and identification of "best" framework representations, based on their respective lattice energies and quantitative comparisons between experimental and calculated (29)Si isotropic chemical shifts and (2)J((29)Si-O-(29)Si) scalar couplings. The comprehensive analysis identifies three closely related and topologically equivalent framework configurations that are in close agreement with all experimental and theoretical structural constraints. The subtle differences among such similar structural models embody the complexity of the actual framework(s), which likely contain coexisting or subtle distributions of structural order that are intrinsic to the material.
Brouwer et al. (Fri,) studied this question.