Reducing overprediction of molecular crystal structures via threshold clustering

Predictive crystallography at scale: mapping, validating, and learning from 1000 crystal energy landscapes

Computational crystal structure prediction (CSP) is an increasingly powerful technique in materials discovery, due to its ability to reveal trends and permit insight across the possibility space of crystal structures of a candidate molecule, beyond simply the observed structure(s). In this work, we demonstrate the reliability and scalability of CSP methods for small, rigid organic molecules by performing in-depth CSP investigations for over 1000 such compounds, the largest survey of its kind to-date. We show that this highly-efficient force-field-based CSP approach is superbly predictive, locating 99.4% of observed experimental structures, and ranking a large majority of these (74%) as among the most stable possible structures (to within uncertainty due to thermal effects). We present two examples of insights such large predicted datasets can permit, examining the space group preferences of organic molecular crystals and rationalising empirical rules concerning the spontaneous resolution of chiral molecules. Finally, we exploit this large and diverse dataset for developing transferable machine-learned energy potentials for the organic solid state, training a neural network lattice energy correction to force field energies that offers substantial improvements to the already impressive energy rankings, and a MACE equivariant message-passing neural network for crystal structure re-optimisation. We conclude that the excellent performance and reliability of the CSP workflow enables the creation of very large datasets of broad utility and explanatory power in materials design.

SCXRD, CSP-NMRX and microED in the quest for three elusive polymorphs of meloxicam

Crystal structure determination is a crucial aspect of almost every branch of the chemical sciences, bringing us closer to understanding crystallization, polymorphism, phase transitions, and the relationship between a structure and its physicochemical and functional properties. Unfortunately, many molecules notoriously crystallize as microcrystalline powders, providing a significant challenge in establishing their structures. In this work, we describe the crystal structure determination of three elusive polymorphs of the anti-inflammatory drug meloxicam (MLX) using three approaches, of which only one was successful for each crystal phase. Single-crystal X-ray diffraction allowed us to solve the structure of MLX-III, MLX-II was solved by a combination of NMR crystallography and crystal structure prediction (CSP) calculations, and MLX-V (Z' = 4 polymorph) was only solvable using electron diffraction. By considering the factors influencing the choice of crystal structure determination method, we showcase their strengths and weaknesses as an indication of their applicability. Additionally, we discuss the issues encountered in the CSP search for MLX-II and MLX-III (both Z' = 2 polymorphs) which turned out to be computationally elusive, in addition to being so in crystallization experiments. This indicates a complex crystal energy landscape for MLX and hints at more general challenges in CSP.

A machine learned potential for investigating single crystal to single crystal transformations in complex organic molecular systems

The packing of organic molecular crystals is often dominated by weak non-covalent interactions, making their in situ rearrangement under external stimuli challenging to understand. We investigate a pressure-induced single-crystal-to-single-crystal (SCSC) transformation between two polymorphs of 2,4,5-triiodo-1H-imidazole using machine learning potentials. This process involves the rearrangement of halogen and hydrogen bonds combined with proton transfer within a complex solid-state system. We developed a strategy to progressively approach the transition state along the phase transition path from both ends by using both the α and β crystal phases as initial structures for active learning. This method allowed us to develop a DFT-based machine learning potential that faithfully describes both of the stable phases and the transition processes. Our results demonstrate that these anisotropic interactions are represented accurately during molecular dynamic simulations. Bond breaking and reforming during proton transfer is observed and analysed in detail. This approach holds promise for simulating SCSC transitions in organic molecular crystals involving anisotropic interactions and chemical bond changes.

Toward On-Demand Polymorphic Transitions of Organic Crystals via Side Chain and Lattice Dynamics Engineering

Controlling polymorphism, namely, the occurrence of multiple crystal forms for a given compound, is still an open technological challenge that needs to be addressed for the reliable manufacturing of crystalline functional materials. Here, we devised a series of 13 organic crystals engineered to embody molecular fragments undergoing specific nanoscale motion anticipated to drive cooperative order-disorder phase transitions. By combining polarized optical microscopy coupled with a heating/cooling stage, differential scanning calorimetry, X-ray diffraction, low-frequency Raman spectroscopy, and calculations (density functional theory and molecular dynamics), we proved the occurrence of cooperative transitions in all the crystalline systems, and we demonstrated how both the molecular structure and lattice dynamics play crucial roles in these peculiar solid-to-solid transformations. These results introduce an efficient strategy to design polymorphic molecular crystalline materials endowed with specific molecular-scale lattice and macroscopic dynamics.

Porous isoreticular non-metal organic frameworks

Metal-organic frameworks (MOFs) are useful synthetic materials that are built by the programmed assembly of metal nodes and organic linkers1. The success of MOFs results from the isoreticular principle2, which allows families of structurally analogous frameworks to be built in a predictable way. This relies on directional coordinate covalent bonding to define the framework geometry. However, isoreticular strategies do not translate to other common crystalline solids, such as organic salts3-5, in which the intermolecular ionic bonding is less directional. Here we show that chemical knowledge can be combined with computational crystal-structure prediction6 (CSP) to design porous organic ammonium halide salts that contain no metals. The nodes in these salt frameworks are tightly packed ionic clusters that direct the materials to crystallize in specific ways, as demonstrated by the presence of well-defined spikes of low-energy, low-density isoreticular structures on the predicted lattice energy landscapes7,8. These energy landscapes allow us to select combinations of cations and anions that will form thermodynamically stable, porous salt frameworks with channel sizes, functionalities and geometries that can be predicted a priori. Some of these porous salts adsorb molecular guests such as iodine in quantities that exceed those of most MOFs, and this could be useful for applications such as radio-iodine capture9-12. More generally, the synthesis of these salts is scalable, involving simple acid-base neutralization, and the strategy makes it possible to create a family of non-metal organic frameworks that combine high ionic charge density with permanent porosity.

Machine-Learned Potentials by Active Learning from Organic Crystal Structure Prediction Landscapes

A primary challenge in organic molecular crystal structure prediction (CSP) is accurately ranking the energies of potential structures. While high-level solid-state density functional theory (DFT) methods allow for mostly reliable discrimination of the low-energy structures, their high computational cost is problematic because of the need to evaluate tens to hundreds of thousands of trial crystal structures to fully explore typical crystal energy landscapes. Consequently, lower-cost but less accurate empirical force fields are often used, sometimes as the first stage of a hierarchical scheme involving multiple stages of increasingly accurate energy calculations. Machine-learned interatomic potentials (MLIPs), trained to reproduce the results of ab initio methods with computational costs close to those of force fields, can improve the efficiency of the CSP by reducing or eliminating the need for costly DFT calculations. Here, we investigate active learning methods for training MLIPs with CSP datasets. The combination of active learning with the well-developed sampling methods from CSP yields potentials in a highly automated workflow that are relevant over a wide range of the crystal packing space. To demonstrate these potentials, we illustrate efficiently reranking large, diverse crystal structure landscapes to near-DFT accuracy from force field-based CSP, improving the reliability of the final energy ranking. Furthermore, we demonstrate how these potentials can be extended to more accurately model structures far from lattice energy minima through additional on-the-fly training within Monte Carlo simulations.

Frontiers of molecular crystal structure prediction for pharmaceuticals and functional organic materials

The reliability of organic molecular crystal structure prediction has improved tremendously in recent years. Crystal structure predictions for small, mostly rigid molecules are quickly becoming routine. Structure predictions for larger, highly flexible molecules are more challenging, but their crystal structures can also now be predicted with increasing rates of success. These advances are ushering in a new era where crystal structure prediction drives the experimental discovery of new solid forms. After briefly discussing the computational methods that enable successful crystal structure prediction, this perspective presents case studies from the literature that demonstrate how state-of-the-art crystal structure prediction can transform how scientists approach problems involving the organic solid state. Applications to pharmaceuticals, porous organic materials, photomechanical crystals, organic semi-conductors, and nuclear magnetic resonance crystallography are included. Finally, efforts to improve our understanding of which predicted crystal structures can actually be produced experimentally and other outstanding challenges are discussed.

Reducing overprediction of molecular crystal structures via threshold clustering

Crystal structure prediction is becoming an increasingly valuable tool for assessing polymorphism of crystalline molecular compounds, yet invariably, it overpredicts the number of polymorphs. One of the causes for this overprediction is in neglecting the coalescence of potential energy minima, separated by relatively small energy barriers, into a single basin at finite temperature. Considering this, we demonstrate a method underpinned by the threshold algorithm for clustering potential energy minima into basins, thereby identifying kinetically stable polymorphs and reducing overprediction.