Reformulating Reactivity Design for Data-Efficient Machine Learning.

Machine learning (ML) can deliver rapid and accurate reaction barrier predictions for use in rational reactivity design. However, model training requires large data sets of typically thousands or tens of thousands of barriers that are very expensive to obtain computationally or experimentally. Furthermore, bespoke data sets are required for each region of interest in reaction space as models typically struggle to generalize. We have therefore reformulated the ML barrier prediction problem toward a much more data-efficient process: finding a reaction from a prespecified set with a desired target value. Our reformulation enables the rapid selection of reactions with purpose-specific activation barriers, for example, in the design of reactivity and selectivity in synthesis, catalyst design, toxicology, and covalent drug discovery, requiring just tens of accurately measured barriers. Importantly, our reformulation does not require generalization beyond the domain of the data set at hand, and we show excellent results for the highly toxicologically and synthetically relevant data sets of aza-Michael addition and transition-metal-catalyzed dihydrogen activation, typically requiring less than 20 accurately measured density functional theory (DFT) barriers. Even for incomplete data sets of E2 and SN2 reactions, with high numbers of missing barriers (74% and 56% respectively), our chosen ML search method still requires significantly fewer data points than the hundreds or thousands needed for more conventional uses of ML to predict activation barriers. Finally, we include a case study in which we use our process to guide the optimization of the dihydrogen activation catalyst. Our approach was able to identify a reaction within 1 kcal mol-1 of the target barrier by only having to run 12 DFT reaction barrier calculations, which illustrates the usage and real-world applicability of this reformulation for systems of high synthetic importance.

Photoredox-HAT Catalysis for Primary Amine α-C-H Alkylation: Mechanistic Insight with Transient Absorption Spectroscopy.

The synergistic use of (organo)photoredox catalysts with hydrogen-atom transfer (HAT) cocatalysts has emerged as a powerful strategy for innate C(sp3)-H bond functionalization, particularly for C-H bonds α- to nitrogen. Azide ion (N3-) was recently identified as an effective HAT catalyst for the challenging α-C-H alkylation of unprotected, primary alkylamines, in combination with dicyanoarene photocatalysts such as 1,2,3,5-tetrakis(carbazol-9-yl)-4,6-dicyanobenzene (4CzIPN). Here, time-resolved transient absorption spectroscopy over sub-picosecond to microsecond timescales provides kinetic and mechanistic details of the photoredox catalytic cycle in acetonitrile solution. Direct observation of the electron transfer from N3- to photoexcited 4CzIPN reveals the participation of the S1 excited electronic state of the organic photocatalyst as an electron acceptor, but the N3• radical product of this reaction is not observed. Instead, both time-resolved infrared and UV-visible spectroscopic measurements implicate rapid association of N3• with N3- (a favorable process in acetonitrile) to form the N6•- radical anion. Electronic structure calculations indicate that N3• is the active participant in the HAT reaction, suggesting a role for N6•- as a reservoir that regulates the concentration of N3•.

Photochemical Fingerprinting Is a Sensitive Probe for the Detection of Synthetic Cannabinoid Receptor Agonists; toward Robust Point-of-Care Detection.

With synthetic cannabinoid receptor agonist (SCRA) use still prevalent across Europe and structurally advanced generations emerging, it is imperative that drug detection methods advance in parallel. SCRAs are a chemically diverse and evolving group, which makes rapid detection challenging. We have previously shown that fluorescence spectral fingerprinting (FSF) has the potential to provide rapid assessment of SCRA presence directly from street material with minimal processing and in saliva. Enhancing the sensitivity and discriminatory ability of this approach has high potential to accelerate the delivery of a point-of-care technology that can be used confidently by a range of stakeholders, from medical to prison staff. We demonstrate that a range of structurally distinct SCRAs are photochemically active and give rise to distinct FSFs after irradiation. To explore this in detail, we have synthesized a model series of compounds which mimic specific structural features of AM-694. Our data show that FSFs are sensitive to chemically conservative changes, with evidence that this relates to shifts in the electronic structure and cross-conjugation. Crucially, we find that the photochemical degradation rate is sensitive to individual structures and gives rise to a specific major product, the mechanism and identification of which we elucidate through density-functional theory (DFT) and time-dependent DFT. We test the potential of our hybrid "photochemical fingerprinting" approach to discriminate SCRAs by demonstrating SCRA detection from a simulated smoking apparatus in saliva. Our study shows the potential of tracking photochemical reactivity via FSFs for enhanced discrimination of SCRAs, with successful integration into a portable device.

Atom and step economical synthesis of acyclic quaternary centers via iridium-catalyzed hydroarylative cross-coupling of 1,1-disubstituted alkenes.

Quaternary benzylic centers are accessed with high atom and step economy by Ir-catalyzed alkene hydroarylation. These studies provide unique examples of the use of non-polarized 1,1-disubstituted alkenes in branch selective Murai-type hydro(hetero)arylations. Detailed mechanistic studies have been undertaken, and these indicate that the first irreversible step is the demanding alkene carbometallation process. Structure-reactivity studies show that the efficiency of this is critically dependent on key structural features of the ligand. Computational studies have been undertaken to rationalize this experimental data, showing how more sterically demanding ligands reduce the reaction barrier via predistortion of the reacting intermediate. The key insight disclosed here will underpin the ongoing development of increasingly sophisticated branch selective Murai hydroarylations.

Eliminating Transition State Calculations for Faster and More Accurate Reactivity Prediction in Sulfa-Michael Additions Relevant to Human Health and the Environment.

Fast and accurate computational approaches to predicting reactivity in sulfa-Michael additions are required for high-throughput screening in toxicology (e.g., predicting excess aquatic toxicity and skin sensitization), chemical synthesis, covalent drug design (e.g., targeting cysteine), and data set generation for machine learning. The kinetic glutathione chemoassay is a time-consuming in chemico method used to extract kinetic data in the form of log(k GSH) for organic electrophiles. In this work, we use density functional theory to compare the use of transition states (TSs) and enolate intermediate structures following C-S bond formation in the prediction of log(k GSH) for a diverse group of 1,4 Michael acceptors. Despite the widespread use of transition state calculations in the literature to predict sulfa-Michael reactivity, we observe that intermediate structures show much better performance for the prediction of log(k GSH), are faster to calculate, and easier to obtain than TSs. Furthermore, we show how linear combinations of atomic charges from the isolated Michael acceptors can further improve predictions, even when using inexpensive semiempirical quantum chemistry methods. Our models can be used widely in the chemical sciences (e.g., in the prediction of toxicity relevant to the environment and human health, synthesis planning, and the design of cysteine-targeting covalent inhibitors), and represent a low-cost, sustainable approach to reactivity assessment.

B(C6 F5 )3 -Catalyzed E-Selective Isomerization of Alkenes.

Herein, we report the B(C6 F5 )3 -catalyzed E-selective isomerization of alkenes. The transition-metal-free method is applicable across a diverse array of readily accessible substrates, giving access to a broad range of synthetically useful products containing versatile stereodefined internal alkenes. The reaction mechanism was investigated by using synthetic and computational methods.

Machine learning and semi-empirical calculations: a synergistic approach to rapid, accurate, and mechanism-based reaction barrier prediction.

Modern QM modelling methods, such as DFT, have provided detailed mechanistic insights into countless reactions. However, their computational cost inhibits their ability to rapidly screen large numbers of substrates and catalysts in reaction discovery. For a C-C bond forming nitro-Michael addition, we introduce a synergistic semi-empirical quantum mechanical (SQM) and machine learning (ML) approach that allows the prediction of DFT-quality reaction barriers in minutes, even on a standard laptop using widely available modelling software. Mean absolute errors (MAEs) are obtained that are below the accepted chemical accuracy threshold of 1 kcal mol-1 and substantially better than SQM methods without ML correction (5.71 kcal mol-1). Predictive power is shown to hold when the ML models are applied to an unseen set of compounds from the toxicology literature. Mechanistic insight is also achieved via the generation of full SQM transition state (TS) structures which are found to be very good approximations for the DFT-level geometries, revealing important steric interactions in some TSs. This combination of speed, accuracy, and mechanistic insight is unprecedented; current ML barrier models compromise on at least one of these important criteria.

Noncovalent Interactions in the Oxazaborolidine-Catalyzed Enantioselective Mukaiyama Aldol.

Current models for oxazaborolidine-catalyzed transition-state structures are determined by C-H···O-B and C-H···O═S formyl hydrogen bonding between the electrophile and catalyst. However, selectivity in the oxazaborolidine-catalyzed Mukaiyama aldol cannot be fully rationalized using these models. Combined density functional theory and noncovalent interaction analyses reveal a new reaction model relying on C-H···O, C-H···π, and π-π interactions between the nucleophile, electrophile, and catalyst to induce selectivity.

Iron-Catalyzed H/D Exchange of Primary Silanes, Secondary Silanes, and Tertiary Siloxanes.

A synthetic study into the catalytic hydrogen/deuterium (H/D) exchange of 1° silanes, 2° silanes, and 3° siloxanes is presented, facilitated by iron-ß-diketiminato complexes (1a and 1b). Near-complete H/D exchange is observed for a variety of aryl- and alkyl-containing hydrosilanes and hydrosiloxanes. The reaction tolerates alternative hydride source pinacolborane (HBpin), with quantitative H/D exchange. A synthetic and density functional theory (DFT) investigation suggests that a monomeric iron-deuteride is responsible for the H/D exchange.

Comparing the Performances of Force Fields in Conformational Searching of Hydrogen-Bond-Donating Catalysts.

Here, we compare the relative performances of different force fields for conformational searching of hydrogen-bond-donating catalyst-like molecules. We assess the force fields by their predictions of conformer energies, geometries, low-energy, nonredundant conformers, and the maximum numbers of possible conformers. Overall, MM3, MMFFs, and OPLS3e had consistently strong performances and are recommended for conformationally searching molecules structurally similar to those in this study.

Organocatalytic Enantioselective Sulfa-Michael Additions to α,ß-Unsaturated Diazoketones.

Enantioselective sulfa-Michael additions to α,ß unsaturated diazocarbonyl compounds have been developed. Quinine-derived squaramide was found to be the best catalyst to promote C-S bond formation in a highly stereoselective fashion for alkyl and aryl thiols. The easy-to-follow protocol allowed the preparation of 22 examples in enantiomeric ratios up to 97:3 and reaction yields up to 94%. The mechanism and origins of enantioselectivity were determined through density functional theory (DFT) calculations.

Chiral Phosphoric Acid Catalysis: The Terada Model Revisited.

Since Akiyama and Terada independently reported the introduction of chiral phosphoric acids (CPAs) as effective catalysts for Mannich-type reactions in 2004, the field of CPA catalysis has grown immensely. Terada reported in 2008 the first example of the activation of aldehydes by a CPA. Based on density functional theory (DFT) calculations, Terada proposed a dual activation mode for this enantioselective aza-ene-type reaction between an aldehyde and an enecarbamate. In this model, hydrogen bonds between the catalyst's hydroxyl group and the carbonyl oxygen and the catalyst's P═O and the formyl proton were observed; the nucleophile then attacks without coordination to the catalyst. This reaction model provided the mechanistic basis for understanding Terada's reaction and many other asymmetric transformations. In the present study, DFT calculations are reported that identify a lower-energy mechanism for this landmark reaction. In this new model, hydrogen bonds between the catalyst's hydroxyl group and the aldehyde oxygen and the catalyst's P═O and the NH group of the enecarbamate are seen. The new model rationalizes the stereoselective outcome of Terada's reaction and offers insight into why a more sterically demanding catalyst gives lower levels of enantioselectivity.

A molecular communication channel consisting of a single reversible chain of hydrogen bonds in a conformationally flexible oligomer.

Communication of information through the global switching of conformation in synthetic molecules has hitherto entailed the inversion of chirality. Here, we report a class of oligomer through which information may be communicated through a global reversal of polarity. Ethylene-bridged oligoureas are constitutionally symmetrical, conformationally flexible molecules organized by a single chain of hydrogen bonds running the full length of the oligomer. NMR reveals that this hydrogen-bonded chain may undergo a coherent reversal of directionality. The directional uniformity of the hydrogen-bond chain allows it to act as a channel for the spatial communication of information on a molecular scale. A binding site at the terminus of an oligomer detects local information about changes in pH or anion concentration and transmits that information-in the form of a directionality switch in the hydrogen-bond chain-to a remote polarity-sensitive fluorophore. This propagation of polarity-encoded information provides a new mechanism for molecular communication.

Understanding the mechanism of the chiral phosphoric acid-catalyzed aza-Cope rearrangement.

The first catalytic enantioselective aza-Cope rearrangement was reported in 2008 by Rueping et al. The reaction is catalyzed by a 1,1'-bi-2-naphthol-derived (BINOL-derived) phosphoric acid and achieved high yields and enantioselectivities (up to 97 : 3 er with 75% yield). This work utilizes Density Functional Theory to understand the mechanism of the reaction and explain the origins of the enantioselectivity. An extensive conformational search was carried out to explore the different activation modes by the catalyst and, the Transition State (TS) leading to the major product was found to be 1.3 kcal mol-1 lower in energy than the TS leading to the minor product. The origin of this stabilization was rationalized with NBO and NCI analysis: it was found that the major TS has a greater number of non-bonding interactions between the substrate and the catalyst, and shows stronger H-bond interactions between H atoms in the substrate and the O atoms in the phosphate group of the catalyst.

Density Functional Theory in the Prediction of Mutagenicity: A Perspective.

As a field, computational toxicology is concerned with using in silico models to predict and understand the origins of toxicity. It is fast, relatively inexpensive, and avoids the ethical conundrum of using animals in scientific experimentation. In this perspective, we discuss the importance of computational models in toxicology, with a specific focus on the different model types that can be used in predictive toxicological approaches toward mutagenicity (SARs and QSARs). We then focus on how quantum chemical methods, such as density functional theory (DFT), have previously been used in the prediction of mutagenicity. It is then discussed how DFT allows for the development of new chemical descriptors that focus on capturing the steric and energetic effects that influence toxicological reactions. We hope to demonstrate the role that DFT plays in understanding the fundamental, intrinsic chemistry of toxicological reactions in predictive toxicology.

Control of Crystallinity and Stereocomplexation of Synthetic Carbohydrate Polymers from d- and l-Xylose.

Manipulating the stereochemistry of polymers is a powerful method to alter their physical properties. Despite the chirality of monosaccharides, reports on the impact of stereochemistry in natural polysaccharides and synthetic carbohydrate polymers remain absent. Herein, we report the cocrystallisation of regio- and stereoregular polyethers derived from d- and l-xylose, leading to enhanced thermal properties compared to the enantiopure polymers. To the best of our knowledge, this is the first example of a stereocomplex between carbohydrate polymers of opposite chirality. In contrast, atactic polymers obtained from a racemic mixture of monomers are amorphous. We also show that the polymer hydroxyl groups are amenable to post-polymerisation functionalization. These strategies afford a family of carbohydrate polyethers, the physical and chemical properties of which can both be controlled, and which opens new possibilities for polysaccharide mimics in biomedical applications or as advanced materials.

Computational Studies of Chiral Hydroxyl Carboxylic Acids: The Allylboration of Aldehydes.

The mechanism of the asymmetric BINOL-derived hydroxyl carboxylic acid catalyzed allylboration of benzaldehyde was investigated using density functional theory calculations. A new reaction model is proposed, and the roles of the two Brønsted acidic sites of the catalyst elucidated. Catalyst distortion was found to be a key factor in determining stereoselectivity. The flexibility of the hydroxyl carboxylic acid catalyst leads to significant differences in the mechanism and origins of selectivity compared to the equivalent phosphoric acid catalyzed reaction.

Reactivity prediction in aza-Michael additions without transition state calculations: the Ames test for mutagenicity.

Animal testing remains a contentious ethical issue in predictive toxicology. Thus, a fast, versatile, low-cost quantum chemical model is presented for predicting the risk of Ames mutagenicity in a series of 1,4 Michael acceptor type compounds. This framework eliminates the need for transition state calculations, and uses an intermediate structure to probe the reactivity of aza-Michael acceptors. This model can be used in a variety of settings e.g., the design of targeted covalent inhibitors and polyketide biosyntheses.

Formicamycin biosynthesis involves a unique reductive ring contraction.

Fasamycin natural products are biosynthetic precursors of the formicamycins. Both groups of compounds are polyketide natural products that exhibit potent antibacterial activity despite displaying different three-dimensional topologies. We show here that transformation of fasamycin into formicamycin metabolites requires two gene products and occurs via a novel two-step ring expansion-ring contraction pathway. Deletion of forX, encoding a flavin dependent monooxygenase, abolished formicamycin production and leads to accumulation of fasamycin E. Deletion of the adjacent gene forY, encoding a flavin dependent oxidoreductase, also abolished formicamycin biosynthesis and led to the accumulation of new lactone metabolites that represent Baeyer-Villiger oxidation products of the fasamycins. These results identify ForX as a Baeyer-Villiger monooxygenase capable of dearomatizing ring C of the fasamycins. Through in vivo cross feeding and biomimetic semi-synthesis experiments we showed that these lactone products represent biosynthetic intermediates that are reduced to formicamycins in a unique reductive ring contraction reaction catalyzed by ForY.

Enantioselective Total Synthesis of (-)-Finerenone Using Asymmetric Transfer Hydrogenation.

(-)-Finerenone is a nonsteroidal mineralocorticoid receptor antagonist currently in phase III clinical trials for the treatment of chronic kidney disease in type 2 diabetes. It contains an unusual dihydronaphthyridine core. We report a 6-step synthesis of (-)-finerenone, which features an enantioselective partial transfer hydrogenation of a naphthyridine using a chiral phosphoric acid catalyst with a Hantzsch ester. The process is complicated by the fact that the naphthyridine exists as a mixture of two atropisomers that react at different rates and with different selectivities. The intrinsic kinetic resolution was converted into a kinetic dynamic resolution at elevated temperature, which enabled us to obtain (-)-finerenone in both high yield and high enantioselectivity. DFT calculations have revealed the origin of selectivity.