Reaction prediction via atomistic simulation: from quantum mechanics to machine learning

Nanoscale chemical reaction exploration with a quantum magnifying glass

Nanoscopic systems exhibit diverse molecular substructures by which they facilitate specific functions. Theoretical models of them, which aim at describing, understanding, and predicting these capabilities, are difficult to build. Viable quantum-classical hybrid models come with specific challenges regarding atomistic structure construction and quantum region selection. Moreover, if their dynamics are mapped onto a state-to-state mechanism such as a chemical reaction network, its exhaustive exploration will be impossible due to the combinatorial explosion of the reaction space. Here, we introduce a "quantum magnifying glass" that allows one to interactively manipulate nanoscale structures at the quantum level. The quantum magnifying glass seamlessly combines autonomous model parametrization, ultra-fast quantum mechanical calculations, and automated reaction exploration. It represents an approach to investigate complex reaction sequences in a physically consistent manner with unprecedented effortlessness in real time. We demonstrate these features for reactions in bio-macromolecules and metal-organic frameworks, diverse systems that highlight general applicability.

A human-machine interface for automatic exploration of chemical reaction networks

Autonomous reaction network exploration algorithms offer a systematic approach to explore mechanisms of complex chemical processes. However, the resulting reaction networks are so vast that an exploration of all potentially accessible intermediates is computationally too demanding. This renders brute-force explorations unfeasible, while explorations with completely pre-defined intermediates or hard-wired chemical constraints, such as element-specific coordination numbers, are not flexible enough for complex chemical systems. Here, we introduce a STEERING WHEEL to guide an otherwise unbiased automated exploration. The STEERING WHEEL algorithm is intuitive, generally applicable, and enables one to focus on specific regions of an emerging network. It also allows for guiding automated data generation in the context of mechanism exploration, catalyst design, and other chemical optimization challenges. The algorithm is demonstrated for reaction mechanism elucidation of transition metal catalysts. We highlight how to explore catalytic cycles in a systematic and reproducible way. The exploration objectives are fully adjustable, allowing one to harness the STEERING WHEEL for both structure-specific (accurate) calculations as well as for broad high-throughput screening of possible reaction intermediates.

Exploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential

Atomistic simulation has a broad range of applications from drug design to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive ab initio simulations. For this reason, chemistry and materials science would greatly benefit from a general reactive MLIP, that is, an MLIP that is applicable to a broad range of reactive chemistry without the need for refitting. Here we develop a general reactive MLIP (ANI-1xnr) through automated sampling of condensed-phase reactions. ANI-1xnr is then applied to study five distinct systems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early earth small molecules. In all studies, ANI-1xnr closely matches experiment (when available) and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N and O elements in the condensed phase, enabling high-throughput in silico reactive chemistry experimentation.

Conjugated quantitative structure-property relationship models: Prediction of kinetic characteristics linked by the Arrhenius equation

Conjugated QSPR models for reactions integrate fundamental chemical laws expressed by mathematical equations with machine learning algorithms. Herein we present a methodology for building conjugated QSPR models integrated with the Arrhenius equation. Conjugated QSPR models were used to predict kinetic characteristics of cycloaddition reactions related by the Arrhenius equation: rate constantl o g k ${{\rm l}{\rm o}{\rm g}k}$ , pre-exponential factorl o g A ${{\rm l}{\rm o}{\rm g}A}$ , and activation energyE a ${{E}_{{\rm a}}}$ . They were benchmarked against single-task (individual and equation-based models) and multi-task models. In individual models, all characteristics were modeled separately, while in multi-task modelsl o g k ${{\rm l}{\rm o}{\rm g}k}$ ,l o g A ${{\rm l}{\rm o}{\rm g}A}$ andE a ${{E}_{{\rm a}}}$ were treated cooperatively. An equation-based model assessedl o g k ${{\rm l}{\rm o}{\rm g}k}$ using the Arrhenius equation andl o g A ${{\rm l}{\rm o}{\rm g}A}$ andE a ${{E}_{{\rm a}}}$ values predicted by individual models. It has been demonstrated that the conjugated QSPR models can accurately predict the reaction rate constants at extreme temperatures, at which reaction rate constants hardly can be measured experimentally. Also, in the case of small training sets conjugated models are more robust than related single-task approaches.

Making the Case for Quantum Mechanics in Predictive Toxicology─Nearly 100 Years Too Late?

The use of quantum mechanics (QM) has long been the norm to study covalent-binding phenomena in chemistry and biochemistry. The pharmaceutical industry leverages QM models explicitly in covalent drug discovery and implicitly to characterize short-range interactions in noncovalent binding. Predictive toxicology has resisted widespread adoption of QM, including in the pharmaceutical industry, despite its obvious relevance to the metabolic processes in the upstream of adverse outcome pathways and advances in both QM methods and computational resources, which support fit-for-purpose applications in reasonable timeframes. Here, we make the case for embracing QM as an indispensable part of a toxicologist's toolkit. We argue that QM provides the necessary orthogonality to alert-based expert systems and traditional QSARs, consistent with calls for animal-free integrated testing strategies for safety assessments of commercial chemicals. We outline existing roadblocks to this transition, including the need to train model developers in QM and the shift toward service-based toxicity models that utilize high-performance computing clusters. Lastly, we describe recent examples of successful implementations of QM in hazard assessments and propose how in silico toxicology can be further advanced by integrating QM with artificial intelligence.

Synthesis of Disubstituted Carboxonium Derivatives of Closo-Decaborate Anion [2,6-B10H8O2CC6H5]-: Theoretical and Experimental Study

A comprehensive study focused on the preparation of disubstituted carboxonium derivatives of closo-decaborate anion [2,6-B10H8O2CC6H5]- was carried out. The proposed synthesis of the target product was based on the interaction between the anion [B10H11]- and benzoic acid C6H5COOH. It was shown that the formation of this product proceeds stepwise through the formation of a mono-substituted product [B10H9OC(OH)C6H5]-. In addition, an alternative one-step approach for obtaining the target derivative is postulated. The structure of tetrabutylammonium salts of carboxonium derivative ((C4H9)4N)[2,6-B10H8O2CC6H5] was established with the help of X-ray structure analysis. The reaction pathway for the formation of [2,6-B10H8O2CC6H5]- was investigated with the help of density functional theory (DFT) calculations. This process has an electrophile induced nucleophilic substitution (EINS) mechanism, and intermediate anionic species play a key role. Such intermediates have a structure in which one boron atom coordinates two hydrogen atoms. The regioselectivity for the process of formation for the 2,6-isomer was also proved by theoretical calculations. Generally, in the experimental part, the simple and available approach for producing disubstituted carboxonium derivative was introduced, and the mechanism of this process was investigated with the help of theoretical calculations. The proposed approach can be applicable for the preparation of a wide range of disubstituted derivatives of closo-borate anions.

Experimental and Computational Approaches to Sulfonated Poly(arylene ether sulfone) Synthesis Using Different Halogen Atoms at the Reactive Site

Engineering thermoplastics, such as poly(arylene ether sulfone), are more often synthesized using F-containing monomers rather than Cl-containing monomers because the F atom is considered more electronegative than Cl, leading to a better condensation polymerization reaction. In this study, the reaction's spontaneity improved when Cl atoms were used compared to the case using F atoms. Specifically, sulfonated poly(arylene ether sulfone) was synthesized by reacting 4,4'-dihydroxybiphenyl with two types of biphenyl sulfone monomers containing Cl and F atoms. No significant difference was observed in the structural, elemental, and chemical properties of the two copolymers based on nuclear magnetic resonance spectroscopy, Fourier transform infrared spectroscopy, thermogravimetric analysis, X-ray diffraction, transmission electron microscopy, and electrochemical impedance spectroscopy. However, the solution viscosity and mechanical strength of the copolymer synthesized with the Cl-terminal monomers were slightly higher than those of the copolymer synthesized with the F-terminal monomers due to higher reaction spontaneity. The first-principle study was employed to elucidate the underlying mechanisms of these reactions.

An automated method for graph-based chemical space exploration and transition state finding

Algorithms that automatically explore the chemical space have been limited to chemical systems with a low number of atoms due to expensive involved quantum calculations and the large amount of possible reaction pathways. The method described here presents a novel solution to the problem of chemical exploration by generating reaction networks with heuristics based on chemical theory. First, a second version of the reaction network is determined through molecular graph transformations acting upon functional groups of the reacting. Only transformations that break two chemical bonds and form two new ones are considered, leading to a significant performance enhancement compared to previously presented algorithm. Second, energy barriers for this reaction network are estimated through quantum chemical calculations by a growing string method, which can also identify non-octet species missed during the previous step and further define the reaction network. The proposed algorithm has been successfully applied to five different chemical reactions, in all cases identifying the most important reaction pathways.

High-Throughput Screening of Bicationic Redox Materials for Chemical Looping Ammonia Synthesis

Ammonia recently has gained increasing attention as a carrier for the efficient and safe usage of hydrogen to further advance the hydrogen economy. However, there is a pressing need to develop new ammonia synthesis techniques to overcome the problem of intense energy consumption associated with the widely used Haber-Bosch process. Chemical looping ammonia synthesis (CLAS) is a promising approach to tackle this problem, but the ideal redox materials to drive these chemical looping processes are yet to be discovered. Here, by mining the well-established MP database, the reaction free energies for CLAS involving 1699 bicationic inorganic redox pairs are screened to comprehensively investigate their potentials as efficient redox materials in four different CLAS schemes. A state-of-the-art machine learning strategy is further deployed to significantly widen the chemical space for discovering the promising redox materials from more than half a million candidates. Most importantly, using the three-step H2 O-CL as an example, a new metric is introduced to determine bicationic redox pairs that are "cooperatively enhanced" compared to their corresponding monocationic counterparts. It is found that bicationic compounds containing a combination of alkali/alkaline-earth metals and transition metal (TM)/post-TM/metalloid elements are compounds that are particularly promising in this respect.

Methanol Synthesis from CO2/CO Mixture on Cu-Zn Catalysts from Microkinetics-Guided Machine Learning Pathway Search

Methanol synthesis on industrial Cu/ZnO/Al₂O₃ catalysts via the hydrogenation of CO and CO₂ mixture, despite several decades of research, is still puzzling due to the nature of the active site and the role of CO₂ in the feed gas. Herein, with the large-scale machine learning atomic simulation, we develop a microkinetics-guided machine learning pathway search to explore thousands of reaction pathways for CO₂ and CO hydrogenations on thermodynamically favorable Cu-Zn surface structures, including Cu(111), Cu(211), and Zn-alloyed Cu(211) surfaces, from which the lowest energy pathways are identified. We find that Zn decorates at the step-edge at Cu(211) up to 0.22 ML under reaction conditions with the Zn-Zn dimeric sites being avoided. CO₂ and CO hydrogenations occur exclusively at the step-edge of the (211) surface with up to 0.11 ML Zn coverage, where the low coverage of Zn (0.11 ML) does not much affect the reaction kinetics, but the higher coverages of Zn (0.22 ML) poison the catalyst. It is CO₂ hydrogenation instead of CO hydrogenation that dominates methanol synthesis, agreeing with previous isotope experiments. While metallic steps are identified as the major active site, we show that the [-Zn-OH-Zn-] chains (cationic Zn) can grow on Cu(111) surfaces under reaction conditions, which suggests the critical role of CO in the mixed gas for reducing the cationic Zn and exposing metal sites for methanol synthesis. Our results provide a comprehensive picture on the dynamic coupling of the feed gas composition, the catalyst active site, and the reaction activity in this complex heterogeneous catalytic system.

Artificial intelligence pathway search to resolve catalytic glycerol hydrogenolysis selectivity

The complex interaction between molecules and catalyst surfaces leads to great difficulties in understanding and predicting the activity and selectivity in heterogeneous catalysis. Here we develop an end-to-end artificial intelligence framework for the activity prediction of heterogeneous catalytic systems (AI-Cat method), which takes simple inputs from names of molecules and metal catalysts and outputs the reaction energy profile from the input molecule to low energy pathway products. The AI-Cat method combines two neural network models, one for predicting reaction patterns and the other for providing the reaction barrier and energy, with a Monte Carlo tree search to resolve the low energy pathways in a reaction network. We then apply AI-Cat to resolve the reaction network of glycerol hydrogenolysis on Cu surfaces, which is a typical selective C-O bond activation system and of key significance for biomass-derived polyol utilization. We show that glycerol hydrogenolysis features a huge reaction network of relevant candidates, containing 420 reaction intermediates and 2467 elementary reactions. Among them, the surface-mediated enol-keto tautomeric resonance is a key step to facilitate the primary C-OH bond breaking and thus selects 1,2-propanediol as the major product on Cu catalysts. 1,3-Propanediol can only be produced under strong acidic conditions and high surface H coverage by following a hydrogenation-dehydration pathway. AI-Cat further discovers six low-energy reaction patterns for C-O bond activation on metals that is of general significance to polyol catalysis. Our results demonstrate that the reaction prediction for complex heterogeneous catalysis is now feasible with AI-based atomic simulation and a Monte Carlo tree search.

Inferring potential landscapes from noisy trajectories of particles within an optical feedback trap

While particle trajectories encode information on their governing potentials, potentials can be challenging to robustly extract from trajectories. Measurement errors may corrupt a particle’s position, and sparse sampling of the potential limits data in higher energy regions such as barriers. We develop a Bayesian method to infer potentials from trajectories corrupted by Markovian measurement noise without assuming prior functional form on the potentials. As an alternative to Gaussian process priors over potentials, we introduce structured kernel interpolation to the Natural Sciences which allows us to extend our analysis to large datasets. Structured-Kernel-Interpolation Priors for Potential Energy Reconstruction (SKIPPER) is validated on 1D and 2D experimental trajectories for particles in a feedback trap.

•

A feedback trap was used to generate noisy Langevin microbead trajectories

•

The potential energy surface is recovered using a Bayesian formulation

•

The formulation uses a structured-kernel-interpolation Gaussian process (SKI-GP) to tractably approximate Gaussian process regression for larger datasets

•

Thanks to our adaptation of SKI-GP, we have broadened the use of Gaussian processes for natural science applications

An Ensemble Structure and Physiochemical (SPOC) Descriptor for Machine-Learning Prediction of Chemical Reaction and Molecular Properties

Feature representations, or descriptors, are machines' chemical language that largely shapes the prediction capability, generalizability and interpretability of machine learning models. To develop a generally applicable descriptor is highly warranted for chemists to deal with conventional prediction tasks in the context of sparsely distributed and small datasets. Inspired by the chemist's vision on molecules, we presented herein an ensemble descriptor, SPOC, curated on the principles of physical organic chemistry that integrates Structure and Physicochemical property (SPOC) of a molecule. SPOC could be readily constructed by combining molecular fingerprints, representing the structure of a given molecule, and molecular physicochemical properties extracted from RDKit or Mordred molecular descriptors. The applicability of SPOC was fully surveyed in a range of well-structured chemical databases with machine learning tasks varying from regression to classifications.

Autonomous design of new chemical reactions using a variational autoencoder

Artificial intelligence based chemistry models are a promising method of exploring chemical reaction design spaces. However, training datasets based on experimental synthesis are typically reported only for the optimal synthesis reactions. This leads to an inherited bias in the model predictions. Therefore, robust datasets that span the entirety of the solution space are necessary to remove inherited bias and permit complete training of the space. In this study, an artificial intelligence model based on a Variational AutoEncoder (VAE) has been developed and investigated to synthetically generate continuous datasets. The approach involves sampling the latent space to generate new chemical reactions. This developed technique is demonstrated by generating over 7,000,000 new reactions from a training dataset containing only 7,000 reactions. The generated reactions include molecular species that are larger and more diverse than the training set.

Dynamics of Heterogeneous Catalytic Processes at Operando Conditions

The rational design of high-performance catalysts is hindered by the lack of knowledge of the structures of active sites and the reaction pathways under reaction conditions, which can be ideally addressed by an in situ/operando characterization. Besides the experimental insights, a theoretical investigation that simulates reaction conditions-so-called operando modeling-is necessary for a plausible understanding of a working catalyst system at the atomic scale. However, there is still a huge gap between the current widely used computational model and the concept of operando modeling, which should be achieved through multiscale computational modeling. This Perspective describes various modeling approaches and machine learning techniques that step toward operando modeling, followed by selected experimental examples that present an operando understanding in the thermo- and electrocatalytic processes. At last, the remaining challenges in this area are outlined.

An automatized workflow from molecular dynamic simulation to quantum chemical methods to identify elementary reactions and compute reaction constants

We present an automatized workflow which, starting from molecular dynamics simulations, identifies reaction events, filters them, and prepares them for accurate quantum chemical calculations using, for example, Density Functional Theory (DFT) or Coupled Cluster methods. The capabilities of the automatized workflow are demonstrated by the example of simulations for the combustion of some polycyclic aromatic hydrocarbons (PAHs). It is shown how key elementary reaction candidates are filtered out of a much larger set of redundant reactions and refined further. The molecular species in question are optimized using DFT and reaction energies, barrier heights, and reaction rates are calculated. The setup is general enough to include at this stage configurational sampling, which can be exploited in the future. Using the introduced machinery, we investigate how the observed reaction types depend on the gas atmosphere used in the molecular dynamics simulation. For the re-optimization on the DFT level, we show how the additional information needed to switch from reactive force-field to electronic structure calculations can be filled in and study how well ReaxFF and DFT agree with each other and shine light on the perspective of using more accurate semi-empirical methods in the MD simulation.

Application of computational approach in plastic pyrolysis kinetic modelling: a review

During the past decade, pyrolysis routes have been identified as one of the most promising solutions for plastic waste management. However, the industrial adoption of such technologies has been limited and several unresolved blind spots hamper the commercial application of pyrolysis. Despite many years and efforts to explain pyrolysis models based on global kinetic approaches, recent advances in computational modelling such as machine learning and quantum mechanics offer new insights. For example, the kinetic and mechanistic information about plastic pyrolysis reactions necessary for scaling up processes is unravelling. This selective literature review reveals some of the foundational knowledge and accurate views on the reaction pathways, product yields, and other features of pyrolysis created by these new tools. Pyrolysis routes mapped by machine learning and quantum mechanics will gain more relevance in the coming years, especially studies that combine computational models with different time and scale resolutions governed by “first principles.” Existing research suggests that, as machine learning is further coupled to quantum mechanics, scientists and engineers will better predict products, yields, and compositions, as well as more complicated features such as ideal reactor design.

Comprehensive strategies of machine-learning-based quantitative structure-activity relationship models

Early quantitative structure-activity relationship (QSAR) technologies have unsatisfactory versatility and accuracy in fields such as drug discovery because they are based on traditional machine learning and interpretive expert features. The development of Big Data and deep learning technologies significantly improve the processing of unstructured data and unleash the great potential of QSAR. Here we discuss the integration of wet experiments (which provide experimental data and reliable verification), molecular dynamics simulation (which provides mechanistic interpretation at the atomic/molecular levels), and machine learning (including deep learning) techniques to improve QSAR models. We first review the history of traditional QSAR and point out its problems. We then propose a better QSAR model characterized by a new iterative framework to integrate machine learning with disparate data input. Finally, we discuss the application of QSAR and machine learning to many practical research fields, including drug development and clinical trials.

Predicting Rate Constants of Hydroxyl Radical Reactions with Alkanes Using Machine Learning

The hydrogen abstraction reactions of the hydroxyl radical with alkanes play an important role in combustion chemistry and atmospheric chemistry. However, site-specific reaction constants are difficult to obtain experimentally and theoretically. Recently, machine learning has proved its ability to predict chemical properties. In this work, a machine learning approach is developed to predict the temperature-dependent site-specific rate constants of the title reactions. Multilayered neural network (NN) models are developed by training the site-specific rate constants of 11 reactions, and several schemes are designed to improve the prediction accuracy. The results show that the proposed NN models are robust in predicting the site-specific and overall rate constants.

Perspective on computational reaction prediction using machine learning methods in heterogeneous catalysis

Heterogeneous catalysis plays a significant role in the modern chemical industry. Towards the rational design of novel catalysts, understanding reactions over surfaces is the most essential aspect. Typical industrial catalytic processes such as syngas conversion and methane utilisation can generate a large reaction network comprising thousands of intermediates and reaction pairs. This complexity not only arises from the permutation of transformations between species but also from the extra reaction channels offered by distinct surface sites. Despite the success in investigating surface reactions at the atomic scale, the huge computational expense of ab initio methods hinders the exploration of such complicated reaction networks. With the proliferation of catalysis studies, machine learning as an emerging tool can take advantage of the accumulated reaction data to emulate the output of ab initio methods towards swift reaction prediction. Here, we briefly summarise the conventional workflow of reaction prediction, including reaction network generation, ab initio thermodynamics and microkinetic modelling. An overview of the frequently used regression models in machine learning is presented. As a promising alternative to full ab initio calculations, machine learning interatomic potentials are highlighted. Furthermore, we survey applications assisted by these methods for accelerating reaction prediction, exploring reaction networks, and computational catalyst design. Finally, we envisage future directions in computationally investigating reactions and implementing machine learning algorithms in heterogeneous catalysis.

In Situ Surface Structures of PdAg Catalyst and Their Influence on Acetylene Semihydrogenation Revealed by Machine Learning and Experiment

PdAg alloy is an industrial catalyst for acetylene-selective hydrogenation in excess ethene. While significant efforts have been devoted to increase the selectivity, there has been little progress in the catalyst performance at low temperatures. Here by combining a machine-learning atomic simulation and catalysis experiment, we clarify the surface status of PdAg alloy catalyst under the reaction conditions and screen out a rutile-TiO₂ supported Pd₁Ag₃ catalyst with high performance: i.e., 85% selectivity at >96% acetylene conversion over a 100 h period in an experiment. The machine-learning global potential energy surface exploration determines the Pd-Ag-H bulk and surface phase diagrams under the reaction conditions, which reveals two key bulk compositions, Pd₁Ag₁ (R3̅m) and Pd₁Ag₃ (Pm3̅m), and quantifies the surface structures with varied Pd:Ag ratios under the reaction conditions. We show that the catalyst activity is controlled by the PdAg patterns on the (111) surface that are variable under reaction conditions, but the selectivity is largely determined by the amount of Pd exposure on the (100) surface. These insights provide the fundamental basis for the rational design of a better catalyst via three measures: (i) controlling the Pd:Ag ratio at 1:3, (ii) reducing the nanoparticle size to limit PdAg local patterns, (iii) searching for active supports to terminate the (100) facets.