Predicting crystal form stability under real-world conditions.

The physicochemical properties of molecular crystals, such as solubility, stability, compactability, melting behaviour and bioavailability, depend on their crystal form1. In silico crystal form selection has recently come much closer to realization because of the development of accurate and affordable free-energy calculations2-4. Here we redefine the state of the art, primarily by improving the accuracy of free-energy calculations, constructing a reliable experimental benchmark for solid-solid free-energy differences, quantifying statistical errors for the computed free energies and placing both hydrate crystal structures of different stoichiometries and anhydrate crystal structures on the same energy landscape, with defined error bars, as a function of temperature and relative humidity. The calculated free energies have standard errors of 1-2 kJ mol-1 for industrially relevant compounds, and the method to place crystal structures with different hydrate stoichiometries on the same energy landscape can be extended to other multi-component systems, including solvates. These contributions reduce the gap between the needs of the experimentalist and the capabilities of modern computational tools, transforming crystal structure prediction into a more reliable and actionable procedure that can be used in combination with experimental evidence to direct crystal form selection and establish control5.

Assessing the Accuracy and Efficiency of Free Energy Differences Obtained from Reweighted Flow-Based Probabilistic Generative Models.

Computing free energy differences between metastable states characterized by nonoverlapping Boltzmann distributions is often a computationally intensive endeavor, usually requiring chains of intermediate states to connect them. Targeted free energy perturbation (TFEP) can significantly lower the computational cost of FEP calculations by choosing a set of invertible maps used to directly connect the distributions of interest, achieving the necessary statistically significant overlaps without sampling any intermediate states. Probabilistic generative models (PGMs) based on normalizing flow architectures can make it much easier via machine learning to train invertible maps needed for TFEP. However, the accuracy and applicability of approaches based on empirically learned maps depend crucially on the choice of reweighting method adopted to estimate the free energy differences. In this work, we assess the accuracy, rate of convergence, and data efficiency of different free energy estimators, including exponential averaging, Bennett acceptance ratio (BAR), and multistate Bennett acceptance ratio (MBAR), in reweighting PGMs trained by maximum likelihood on limited amounts of molecular dynamics data sampled only from end-states of interest. We carry out the comparisons on a set of simple but representative case studies, including conformational ensembles of alanine dipeptide and ibuprofen. Our results indicate that BAR and MBAR are both data efficient and robust, even in the presence of significant model overfitting in the generation of invertible maps. This analysis can serve as a stepping stone for the deployment of efficient and quantitatively accurate ML-based free energy calculation methods in complex systems.

Efficient Crystal Structure Prediction for Structurally Related Molecules with Accurate and Transferable Tailor-Made Force Fields.

Crystal structure prediction (CSP) is generally used to complement experimental solid form screening and applied to individual molecules in drug development. The fast development of algorithms and computing resources offers the opportunity to use CSP earlier and for a broader range of applications in the drug design cycle. This study presents a novel paradigm of CSP specifically designed for structurally related molecules, referred to as Quick-CSP. The approach prioritizes more accurate physics through robust and transferable tailor-made force fields (TMFFs), such that significant efficiency gains are achieved through the reduction of expensive ab initio calculations. The accuracy of the TMFF is increased by the introduction of electrostatic multipoles, and the fragment-based force field parameterization scheme is demonstrated to be transferable for a family of chemically related molecules. The protocol is benchmarked with structurally related compounds from the Bromodomain and Extraterminal (BET) domain inhibitors series. A new convergence criterion is introduced that aims at performing only as many ab initio optimizations of crystal structures as required to locate the bottom of the crystal energy landscape within a user-defined accuracy. The overall approach provides significant cost savings ranging from three- to eight-fold less than the full-CSP workflow. The reported advancements expand the scope and utility of the underlying CSP building blocks as well as their novel reassembly to other applications earlier in the drug design cycle to guide molecule design and selection.

Carbons with Regular Pore Geometry Yield Fundamental Insights into Supercapacitor Charge Storage.

We conduct molecular dynamics simulations of electrical double-layer capacitors (EDLCs) using a library of ordered, porous carbon electrode materials called zeolite templated carbons (ZTCs). The well-defined pore shapes of the ZTCs enable us to determine the influence of pore geometry on both charging dynamics and charge storage mechanisms in EDLCs, also referred to as supercapacitors. We show that charging dynamics are negatively correlated with the pore-limiting diameter of the electrode material and display signatures of both progressive charging and ion trapping. However, the equilibrium capacitance, unlike charging dynamics, is not strongly correlated to commonly used, purely geometric descriptors such as pore size. Instead, we find a strong correlation of capacitance to the charge compensation per carbon (CCpC), a descriptor we define in this work as the average charge of the electrode atoms within the coordination shell of a counterion. A high CCpC indicates efficient charge storage, as the strong partial charges of the electrode are able to screen counterion charge, enabling higher ion loading and thus more charge storage within the electrode at a fixed applied voltage. We determine that adsorption sites with a high CCpC tend to be found within pockets with a smaller radius of curvature, where the counterions are able to minimize their distance with multiple points on the electrode surface, and therefore induce stronger local partial charges.

Can Metal-Organic Frameworks Be Used for Cannabis Breathalyzers?

Δ9-Tetrahydrocannabinol (THC) is the principal psychoactive component of cannabis, and there is an urgent need to build low-cost and portable devices that can detect its presence from breath. Similarly to alcohol detectors, these tools can be used by law enforcement to determine driver intoxication and enforce safer and more regulated use of cannabis. In this work, we propose to use a class of microporous crystals, metal-organic frameworks (MOFs), to selectively adsorb THC that can be later detected using optical, electrochemical, or fluorescence-based sensing methods. We computationally screened more than 5000 MOFs, highlighting the materials that have the largest affinity with THC, as well as the highest selectivity against water, showing that it is thermodynamically feasible for MOFs to adsorb THC from humid breath. We propose and compare different models for THC and different computational protocols to rank the promising materials, also presenting a novel approach to assess the permeability of a porous framework to nonspherical molecules. We identified three adsorption motifs in MOFs with high affinity to THC, which we refer to as "narrow channels", "thick walls", and "parking spots". Therefore, we expect our protocols and our findings to be generalizable for different classes of microporous materials and also for investigating the adsorption properties of other large molecules that, like THC, have a nonspherical shape.

Amine Dynamics in Diamine-Appended Mg2(dobpdc) Metal-Organic Frameworks.

Variable-temperature 15N solid-state NMR spectroscopy is used to uncover the dynamics of three diamines appended to the metal-organic framework Mg2(dobpdc) (dobpdc4- = 4,4'-dioxidobiphenyl-3,3'-dicarboxylate), an important family of CO2 capture materials. The results imply both bound and free amine nitrogen environments exist when diamines are coordinated to the framework open Mg2+ sites. There are rapid exchanges between two nitrogen environments for all three diamines, the rates and energetics of which are quantified by 15N solid-state NMR data and corroborated by density functional theory calculations and molecular dynamics simulations. The activation energy for the exchange provides a measure of the metal-amine bond strength. The unexpected negative correlation between the metal-amine bond strength and CO2 adsorption step pressure reveals that metal-amine bond strength is not the only important factor in determining the CO2 adsorption properties of diamine-appended Mg2(dobpdc) metal-organic frameworks.

Predicting Product Distribution of Propene Dimerization in Nanoporous Materials.

In this work, a theoretical framework is developed to explain and predict changes in the product distribution of the propene dimerization reaction, which yields a mixture of C6 olefin isomers, resulting from the use of different porous materials as catalysts. The MOF-74 class of materials has shown promise in catalyzing the dimerization of propene with high selectivity for valuable linear olefin products. We show that experimentally observed changes in the product distribution can be explained in terms of the contribution of the pores to the free energy of formation, which are directly computed using molecular simulation. Our model is used to screen a library of 118 existing and hypothetical MOF and zeolite structures to study how product distribution can be tuned by changing pore size, shape, and composition of porous materials. Using these molecular descriptors, catalyst properties are identified that increase the selective reaction of linear olefin isomers, which are valued as industrial feedstocks. A pore size commensurate with the size of the desired linear products enhances linear conversion by sterically hindering the branched isomers. Another promising feature is the presence of open metal sites, which interact with the olefin π-bond to provide favorable binding sites for the linear isomers.