X-ray crystallography provides a distinctive view on the three-dimensional structure of crystals. To reconstruct the electron density map, the complex structure factors of a sufficiently large number of diffracted reflections must be known. In a conventional experiment, only the amplitudes are obtained, and the phases ϕ are lost. This is the crystallographic phase problem. In this work, we show that a neural network, trained on millions of artificial structure data, can solve the phase problem at a resolution of only 2 angstroms, using only 10 to 20% of the data needed for direct methods. The network works in common space groups and for modest unit-cell dimensions and suggests that neural networks could be used to solve the phase problem in the general case for weakly scattering crystals.
Dehydration of crystalline solids is a widespread phenomenon, yet the fundamental mechanisms by which dehydration occurs are not properly understood. This arises due to technical limitations in studying such fast processes with sufficient sensitivity; nevertheless, understanding dehydration pathways is critical for designing optimal properties for materials, particularly in the case of pharmaceutical solids. The computational methods presented here allow for accurate determination of the dehydrated species’ crystal structure and to develop an understanding of the mechanism of dehydration at the molecular level. This work also highlights the critical role of explicitly taking into account the dynamical aspect of molecules using computational techniques, rather than relying on static energy minimization approaches. Specifically, the crystalline active pharmaceutical agent naproxen sodium, and its hydrates, is studied in silico using density functional theory and molecular dynamics, ultimately elucidating the face-specific dehydration mechanisms and revealing highly complex diffusion and nucleation behavior. Additionally, the results indicate that the method is a viable way to explore dehydration pathways and predict new dehydrated crystal structures.
Co-crystals offer a promising strategy to create drug constructs for which the pharmacokinetic properties of an active pharmaceutical ingredient can be improved without compromising its pharmacological effectiveness.
Atomic force microscopy and molecular dynamics simulations probed the crystallinity and hydrophobicity of a paracetamol crystal surface in water–ethanol mixtures. We observe the formation of a dynamic heterogenous disordered surface (DHDS) layer.