Modular, multi-robot integration of laboratories: an autonomous workflow for solid-state chemistry.

Automation can transform productivity in research activities that use liquid handling, such as organic synthesis, but it has made less impact in materials laboratories, which require sample preparation steps and a range of solid-state characterization techniques. For example, powder X-ray diffraction (PXRD) is a key method in materials and pharmaceutical chemistry, but its end-to-end automation is challenging because it involves solid powder handling and sample processing. Here we present a fully autonomous solid-state workflow for PXRD experiments that can match or even surpass manual data quality, encompassing crystal growth, sample preparation, and automated data capture. The workflow involves 12 steps performed by a team of three multipurpose robots, illustrating the power of flexible, modular automation to integrate complex, multitask laboratories.

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.

Pushing Technique Boundaries to Probe Conformational Polymorphism.

We present an extensive exploration of the solid-form landscape of chlorpropamide (CPA) using a combined experimental-computational approach at the frontiers of both fields. We have obtained new conformational polymorphs of CPA, placing them into context with known forms using flexible-molecule crystal structure prediction. We highlight the formation of a new polymorph (ζ-CPA) via spray-drying experiments despite its notable metastability (14 kJ/mol) relative to the thermodynamic α-form, and we identify and resolve the ball-milled η-form isolated in 2019. Additionally, we employ impurity- and gel-assisted crystallization to control polymorphism and the formation of novel multicomponent forms. We, thus, demonstrate the power of this collaborative screening approach to observe, rationalize, and control the formation of new metastable forms.

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.

Experimental Confirmation of a Predicted Porous Hydrogen-Bonded Organic Framework.

Hydrogen-bonded organic frameworks (HOFs) with low densities and high porosities are rare and challenging to design because most molecules have a strong energetic preference for close packing. Crystal structure prediction (CSP) can rank the crystal packings available to an organic molecule based on their relative lattice energies. This has become a powerful tool for the a priori design of porous molecular crystals. Previously, we combined CSP with structure-property predictions to generate energy-structure-function (ESF) maps for a series of triptycene-based molecules with quinoxaline groups. From these ESF maps, triptycene trisquinoxalinedione (TH5) was predicted to form a previously unknown low-energy HOF (TH5-A) with a remarkably low density of 0.374 g cm-3 and three-dimensional (3D) pores. Here, we demonstrate the reliability of those ESF maps by discovering this TH5-A polymorph experimentally. This material has a high accessible surface area of 3,284 m2 g-1 , as measured by nitrogen adsorption, making it one of the most porous HOFs reported to date.

Seeking Rules Governing Mixed Molecular Crystallization.

Mixed crystals result when components of the structure are randomly replaced by analogues in ratios that can be varied continuously over certain ranges. Mixed crystals are useful because their properties can be adjusted by increments, simply by altering the ratio of components. Unfortunately, no clear rules exist to predict when two compounds are similar enough to form mixed crystals containing substantial amounts of both. To gain further understanding, we have used single-crystal X-ray diffraction, computational methods, and other tools to study mixed crystallizations within a selected set of structurally related compounds. This work has allowed us to begin to clarify the rules governing the phenomenon by showing that mixed crystals can have compositions and properties that vary continuously over wide ranges, even when the individual components do not normally crystallize in the same way. Moreover, close agreement of the results of our experiments and computational modeling demonstrates that reliable predictions about mixed crystallization can be made, despite the complexity of the phenomenon.

Densest plane group packings of regular polygons.

Packings of regular convex polygons (n-gons) that are sufficiently dense have been studied extensively in the context of modeling physical and biological systems as well as discrete and computational geometry. Former results were mainly regarding densest lattice or double-lattice configurations. Here we consider all two-dimensional crystallographic symmetry groups (plane groups) by restricting the configuration space of the general packing problem of congruent copies of a compact subset of the two-dimensional Euclidean space to particular isomorphism classes of the discrete group of isometries. We formulate the plane group packing problem as a nonlinear constrained optimization problem. By means of the Entropic Trust Region Packing Algorithm that approximately solves this problem, we examine some known and unknown densest packings of various n-gons in all 17 plane groups and state conjectures about common symmetries of the densest plane group packings for every n-gon.

Targeted design of porous materials without strong, directional interactions.

A porous molecular crystal (TSCl) was found to crystallise from dichloromethane and water during the synthesis of tetrakis(4-sulfophenylmethane). Crystal structure prediction (CSP) rationalises the driving force behind the formation of this porous TSCl phase and the intermolecular interactions that direct its formation. Gas sorption analysis showed that TSCl is permanently porous with selective adsorption of CO2 over N2, H2 and CH4 and a maximum CO2 uptake of 74 cm3 g-1 at 195 K. Calculations revealed that TSCl assembles via a combination of weak hydrogen bonds and strong dispersion interactions. This illustrates that CSP can underpin approaches to crystal engineering that do not involve more intuitive directional interactions, such as hydrogen bonding.

Analogy Powered by Prediction and Structural Invariants: Computationally Led Discovery of a Mesoporous Hydrogen-Bonded Organic Cage Crystal.

Mesoporous molecular crystals have potential applications in separation and catalysis, but they are rare and hard to design because many weak interactions compete during crystallization, and most molecules have an energetic preference for close packing. Here, we combine crystal structure prediction (CSP) with structural invariants to continuously qualify the similarity between predicted crystal structures for related molecules. This allows isomorphous substitution strategies, which can be unreliable for molecular crystals, to be augmented by a priori prediction, thus leveraging the power of both approaches. We used this combined approach to discover a rare example of a low-density (0.54 g cm-3) mesoporous hydrogen-bonded framework (HOF), 3D-CageHOF-1. This structure comprises an organic cage (Cage-3-NH2) that was predicted to form kinetically trapped, low-density polymorphs via CSP. Pointwise distance distribution structural invariants revealed five predicted forms of Cage-3-NH2 that are analogous to experimentally realized porous crystals of a chemically different but geometrically similar molecule, T2. More broadly, this approach overcomes the difficulties in comparing predicted molecular crystals with varying lattice parameters, thus allowing for the systematic comparison of energy-structure landscapes for chemically dissimilar molecules.

Surprising Chemistry of 6-Azidotetrazolo[5,1-a]phthalazine: What a Purported Natural Product Reveals about the Polymorphism of Explosives.

6-Azidotetrazolo[5,1-a]phthalazine (ATPH) is a nitrogen-rich compound of surprisingly broad interest. It is purported to be a natural product, yet it is closely related to substances developed as explosives and is highly polymorphic despite having a nearly planar structure with little flexibility. Seven solid forms of ATPH have been characterized by single-crystal X-ray diffraction. The structures show diverse patterns of molecular organization, including both stacked sheets and herringbone packing. In all cases, N···N and C-H···N interactions play key roles in ensuring molecular cohesion. The high polymorphism of ATPH appears to arise in part from the ability of virtually every atom of nitrogen and hydrogen in the molecule to take part in close N···N and C-H···N contacts. As a result, adjacent molecules can adopt many different relative orientations that are energetically similar, thereby generating a polymorphic landscape with an unusually high density of potential structures. This landscape has been explored in detail by the computational prediction of crystal structures. Studying ATPH has provided insights into the field of energetic materials, where access to multiple polymorphs can be used to improve performance and clarify how it depends on molecular packing. In addition, our work with ATPH shows how valuable insights into molecular crystallization, often gleaned from statistical analyses of structural databases, can also come from in-depth empirical and theoretical studies of single compounds that show distinctive behavior.

De Novo Crystal Structure Determination from Machine Learned Chemical Shifts.

Determination of the three-dimensional atomic-level structure of powdered solids is one of the key goals in current chemistry. Solid-state NMR chemical shifts can be used to solve this problem, but they are limited by the high computational cost associated with crystal structure prediction methods and density functional theory chemical shift calculations. Here, we successfully determine the crystal structures of ampicillin, piroxicam, cocaine, and two polymorphs of the drug molecule AZD8329 using on-the-fly generated machine-learned isotropic chemical shifts to directly guide a Monte Carlo-based structure determination process starting from a random gas-phase conformation.

Global analysis of the energy landscapes of molecular crystal structures by applying the threshold algorithm.

Polymorphism in molecular crystals has important consequences for the control of materials properties and our understanding of crystallization. Computational methods, including crystal structure prediction, have provided important insight into polymorphism, but have usually been limited to assessing the relative energies of structures. We describe the implementation of the Monte Carlo threshold algorithm as a method to provide an estimate of the energy barriers separating crystal structures. By sampling the local energy minima accessible from multiple starting structures, the simulations yield a global picture of the crystal energy landscapes and provide valuable information on the depth of the energy minima associated with crystal structures. We present results from applying the threshold algorithm to four polymorphic organic molecular crystals, examine the influence of applying space group symmetry constraints during the simulations, and discuss the relationship between the structure of the energy landscape and the intermolecular interactions present in the crystals.

Accelerating computational discovery of porous solids through improved navigation of energy-structure-function maps.

While energy-structure-function (ESF) maps are a powerful new tool for in silico materials design, the cost of acquiring an ESF map for many properties is too high for routine integration into high-throughput virtual screening workflows. Here, we propose the next evolution of the ESF map. This uses parallel Bayesian optimization to selectively acquire energy and property data, generating the same levels of insight at a fraction of the computational cost. We use this approach to obtain a two orders of magnitude speedup on an ESF study that focused on the discovery of molecular crystals for methane capture, saving more than 500,000 central processing unit hours from the original protocol. By accelerating the acquisition of insight from ESF maps, we pave the way for the use of these maps in automated ultrahigh-throughput screening pipelines by greatly reducing the opportunity risk associated with the choice of system to calculate.

Inherent Ethyl Acetate Selectivity in a Trianglimine Molecular Solid.

Ethyl acetate is an important chemical raw material and solvent. It is also a key volatile organic compound in the brewing industry and a marker for lung cancer. Materials that are highly selective toward ethyl acetate are needed for its separation and detection. Here, we report a trianglimine macrocycle (TAMC) that selectively adsorbs ethyl acetate by forming a solvate. Crystal structure prediction showed this to be the lowest energy solvate structure available. This solvate leaves a metastable, "templated" cavity after solvent removal. Adsorption and breakthrough experiments confirmed that TAMC has adequate adsorption kinetics to separate ethyl acetate from azeotropic mixtures with ethanol, which is a challenging and energy-intensive industrial separation.

Exploration and Optimization in Crystal Structure Prediction: Combining Basin Hopping with Quasi-Random Sampling.

We describe the implementation of a Monte Carlo basin hopping (BH) global optimization procedure for the prediction of molecular crystal structures. The BH method is combined with quasi-random (QR) structure generation in a hybrid method for crystal structure prediction, QR-BH, which combines the low-discrepancy sampling provided by QR sequences with BH efficiency at locating low energy structures. Through tests on a set of single-component molecular crystals and co-crystals, we demonstrate that QR-BH provides faster location of low energy structures than pure QR sampling, while maintaining the efficient location of higher energy structures that are important for identifying important polymorphs.

Digital navigation of energy-structure-function maps for hydrogen-bonded porous molecular crystals.

Energy-structure-function (ESF) maps can aid the targeted discovery of porous molecular crystals by predicting the stable crystalline arrangements along with their functions of interest. Here, we compute ESF maps for a series of rigid molecules that comprise either a triptycene or a spiro-biphenyl core, functionalized with six different hydrogen-bonding moieties. We show that the positioning of the hydrogen-bonding sites, as well as their number, has a profound influence on the shape of the resulting ESF maps, revealing promising structure-function spaces for future experiments. We also demonstrate a simple and general approach to representing and inspecting the high-dimensional data of an ESF map, enabling an efficient navigation of the ESF data to identify 'landmark' structures that are energetically favourable or functionally interesting. This is a step toward the automated analysis of ESF maps, an important goal for closed-loop, autonomous searches for molecular crystals with useful functions.

Structure prediction of crystals, surfaces and nanoparticles.

We review the current techniques used in the prediction of crystal structures and their surfaces and of the structures of nanoparticles. The main classes of search algorithm and energy function are summarized, and we discuss the growing role of methods based on machine learning. We illustrate the current status of the field with examples taken from metallic, inorganic and organic systems. This article is part of a discussion meeting issue 'Dynamic in situ microscopy relating structure and function'.

Multifidelity Statistical Machine Learning for Molecular Crystal Structure Prediction.

The prediction of crystal structures from first-principles requires highly accurate energies for large numbers of putative crystal structures. High accuracy of solid state density functional theory (DFT) calculations is often required, but hundreds or more structures can be present in the low energy region of interest, so that the associated computational costs are prohibitive. Here, we apply statistical machine learning to predict expensive hybrid functional DFT (PBE0) calculations using a multifidelity approach to re-evaluate the energies of crystal structures predicted with an inexpensive force field. The method uses an autoregressive Gaussian process, making use of less expensive GGA DFT (PBE) calculations to bridge the gap between the force field and PBE0 energies. The method is benchmarked on the crystal structure landscapes of three small, hydrogen-bonded organic molecules and shown to produce accurate predictions of energies and crystal structure ranking using small numbers of the most expensive calculations; the PBE0 energies can be predicted with errors of less than 1 kJ mol-1 with between 4.2 and 6.8% of the cost of the full calculations. As the model that we have developed is probabilistic, we discuss how the uncertainties in predicted energies impact the assessment of the energetic ranking of crystal structures.

Minimizing Polymorphic Risk through Cooperative Computational and Experimental Exploration.

We combine state-of-the-art computational crystal structure prediction (CSP) techniques with a wide range of experimental crystallization methods to understand and explore crystal structure in pharmaceuticals and minimize the risk of unanticipated late-appearing polymorphs. Initially, we demonstrate the power of CSP to rationalize the difficulty in obtaining polymorphs of the well-known pharmaceutical isoniazid and show that CSP provides the structure of the recently obtained, but unsolved, Form III of this drug despite there being only a single resolved form for almost 70 years. More dramatically, our blind CSP study predicts a significant risk of polymorphism for the related iproniazid. Employing a wide variety of experimental techniques, including high-pressure experiments, we experimentally obtained the first three known nonsolvated crystal forms of iproniazid, all of which were successfully predicted in the CSP procedure. We demonstrate the power of CSP methods and free energy calculations to rationalize the observed elusiveness of the third form of iproniazid, the success of high-pressure experiments in obtaining it, and the ability of our synergistic computational-experimental approach to "de-risk" solid form landscapes.