Low-Scaling Algorithm for Nudged Elastic Band Calculations Using a Surrogate Machine Learning Model

The Importance of Reaction Energy in Predicting Chemical Reaction Barriers with Machine Learning Models

Improving our fundamental understanding of complex heterocatalytic processes increasingly relies on electronic structure simulations and microkinetic models based on calculated energy differences. In particular, calculation of activation barriers, usually achieved through compute-intensive saddle point search routines, remains a serious bottleneck in understanding trends in catalytic activity for highly branched reaction networks. Although the well-known Brønsted-Evans-Polyani (BEP) scaling - a one-feature linear regression model - has been widely applied in such microkinetic models, they still rely on calculated reaction energies and may not generalize beyond a single facet on a single class of materials, e. g., a terrace sites on transition metals. For highly branched and energetically shallow reaction networks, such as electrochemical CO2 reduction or wastewater remediation, calculating even reaction energies on many surfaces can become computationally intractable due to the combinatorial explosion of states that must be considered. Here, we investigate the feasibility of activation barrier prediction without knowledge of the reaction energy using linear and nonlinear machine learning (ML) models trained on a new database of over 500 dehydrogenation activation barriers. We also find that inclusion of the reaction energy significantly improves both classes of ML models, but complex nonlinear models can achieve performance similar to the simplest BEP scaling when predicting activation barriers on new systems. Additionally, inclusion of the reaction energy significantly improves generalizability to new systems beyond the training set. Our results suggest that the reaction energy is a critical feature to consider when building models to predict activation barriers, indicating that efforts to reliably predict reaction energies through, e. g., the Open Catalyst Project and others, will be an important route to effective model development for more complex systems.

Modelling the growth reaction pathways of zincone ALD/MLD hybrid thin films: a DFT study

ALD/MLD hybrid thin films can be fabricated by combining atomic layer deposition (ALD) and molecular layer deposition (MLD). Even though this deposition method has been extensively used experimentally, the computational work required to acquire the reaction paths during the thin film deposition process is still in dire demand. We investigated hybrid thin films consisting of diethyl zinc and either 4-aminophenol or hydroquinone using both gas-phase and surface reactions to gain extensive knowledge of the complex phenomena occurring during the process of hybrid thin film deposition. We used density functional theory (DFT) to obtain the activation energies of these kinetic-dependent deposition processes. Different processes of ethyl ligand removal as ethane were discovered, and we found that the hydroxyl group of 4-aminophenol was more reactive than the amino group in the migration of hydrogen to an ethyl ligand within a complicated branching reaction chain.

On the Capabilities of Transition Metal Carbides for Carbon Capture and Utilization Technologies

The search for cheap and active materials for the capture and activation of CO2 has led to many efforts aimed at developing new catalysts. In this context, earth-abundant transition metal carbides (TMCs) have emerged as promising candidates, garnering increased attention in recent decades due to their exceptional refractory properties and resistance to sintering, coking, and sulfur poisoning. In this work, we assess the use of Group 5 TMCs (VC, NbC, and TaC) as potential materials for carbon capture and sequestration/utilization technologies by combining experimental characterization techniques, first-principles-based multiscale modeling, vibrational analysis, and catalytic experiments. Our findings reveal that the stoichiometric phase of VC exhibits weak interactions with CO2, displaying an inability to adsorb or dissociate it. However, VC often exhibits the presence of surface carbon vacancies, leading to significant activation of CO2 at room temperature and facilitating its catalytic hydrogenation. In contrast, stoichiometric NbC and TaC phases exhibit stronger interactions with CO2, capable of adsorbing and even breaking of CO2 at low temperatures, particularly notable in the case of TaC. Nevertheless, NbC and TaC demonstrate poor catalytic performance for CO2 hydrogenation. This work suggests Group 5 TMCs as potential materials for CO2 abatement, emphasizes the importance of surface vacancies in enhancing catalytic activity and adsorption capability, and provides a reference for using the infrared spectra as a unique identifier to detect oxy-carbide phases or surface C vacancies within Group 5 TMCs.

Physical Prior Mean Function-Driven Gaussian Processes Search for Minimum-Energy Reaction Paths with a Climbing-Image Nudged Elastic Band: A General Method for Gas-Phase, Interfacial, and Bulk-Phase Reactions

The climbing-image nudged elastic band (CI-NEB) method serves as an indispensable tool for computational chemists, offering insight into minimum-energy reaction paths (MEPs) by delineating both transition states (TSs) and intermediate nonstationary structures along reaction coordinates. However, executing CI-NEB calculations for reactions with extensive reaction coordinate spans necessitates a large number of images to ensure a reliable convergence of the MEPs and TS structures, presenting a computationally demanding optimization challenge, even with mildly costly electronic-structure methods. In this study, we advocate for the utilization of physically inspired prior mean function-based Gaussian processes (GPs) to expedite MEP exploration and TS optimization via the CI-NEB method. By incorporating reliable prior physical approximations into potential energy surface (PES) modeling, we demonstrate enhanced efficiency in multidimensional CI-NEB optimization with surrogate-based optimizers. Our physically informed GP approach not only outperforms traditional nonsurrogate-based optimizers in optimization efficiency but also on-the-fly learns the reaction path valley during optimization, culminating in significant advancements. The surrogate PES derived from our optimization exhibits high accuracy compared to true PES references, aligning with our emphasis on leveraging reliable physical priors for robust and efficient posterior mean learning in GPs. Through a systematic benchmark study encompassing various reaction pathways, including gas-phase, bulk-phase, and interfacial/surface reactions, our physical GPs consistently demonstrate superior efficiency and reliability. For instance, they outperform the popular fast inertial relaxation engine optimizer by approximately a factor of 10, showcasing their versatility and efficacy in exploring reaction mechanisms and surface reaction PESs.

Locating Transition States by Variational Reaction Path Optimization with an Energy-Derivative-Free Objective Function

Locating transition states is essential for understanding molecular reactions. We propose a double-ended transition state search method by revisiting a variational reaction path optimization method known as the MaxFlux method. Although its original purpose is to add temperature effects to reaction paths, we conversely let the temperature approach zero to obtain an asymptotically exact minimum energy path and its corresponding transition state in variational formalism with an energy-derivative-free objective function. Using several numerical techniques to directly optimize the objective function, the present method reliably finds transition states with low computational cost. In particular, only three force evaluations per iteration are sufficient. This is confirmed on a variety of molecular reactions where the nudged elastic band method often fails. The present method is implemented in Python using the Atomic Simulation Environment and is available on GitHub.

A Q-learning method based on coarse-to-fine potential energy surface for locating transition state and reaction pathway

Transition state (TS) on the potential energy surface (PES) plays a key role in determining the kinetics and thermodynamics of chemical reactions. Inspired by the fact that the dynamics of complex systems are always driven by rare but significant transition events, we herein propose a TS search method in accordance with the Q-learning algorithm. Appropriate reward functions are set for a given PES to optimize the reaction pathway through continuous trial and error, and then the TS can be obtained from the optimized reaction pathway. The validity of this Q-learning method with reasonable settings of Q-value table including actions, states, learning rate, greedy rate, discount rate, and so on, is exemplified in 2 two-dimensional potential functions. In the applications of the Q-learning method to two chemical reactions, it is demonstrated that the Q-learning method can predict consistent TS and reaction pathway with those by ab initio calculations. Notably, the PES must be well prepared before using the Q-learning method, and a coarse-to-fine PES scanning scheme is thus introduced to save the computational time while maintaining the accuracy of the Q-learning prediction. This work offers a simple and reliable Q-learning method to search for all possible TS and reaction pathway of a chemical reaction, which may be a new option for effectively exploring the PES in an extensive search manner.

Area-selective atomic layer deposition on 2D monolayer lateral superlattices

The advanced patterning process is the basis of integration technology to realize the development of next-generation high-speed, low-power consumption devices. Recently, area-selective atomic layer deposition (AS-ALD), which allows the direct deposition of target materials on the desired area using a deposition barrier, has emerged as an alternative patterning process. However, the AS-ALD process remains challenging to use for the improvement of patterning resolution and selectivity. In this study, we report a superlattice-based AS-ALD (SAS-ALD) process using a two-dimensional (2D) MoS2-MoSe2 lateral superlattice as a pre-defining template. We achieved a minimum half pitch size of a sub-10 nm scale for the resulting AS-ALD on the 2D superlattice template by controlling the duration time of chemical vapor deposition (CVD) precursors. SAS-ALD introduces a mechanism that enables selectivity through the adsorption and diffusion processes of ALD precursors, distinctly different from conventional AS-ALD method. This technique facilitates selective deposition even on small pattern sizes and is compatible with the use of highly reactive precursors like trimethyl aluminum. Moreover, it allows for the selective deposition of a variety of materials, including Al2O3, HfO2, Ru, Te, and Sb2Se3.

GPAW: An open Python package for electronic structure calculations

We review the GPAW open-source Python package for electronic structure calculations. GPAW is based on the projector-augmented wave method and can solve the self-consistent density functional theory (DFT) equations using three different wave-function representations, namely real-space grids, plane waves, and numerical atomic orbitals. The three representations are complementary and mutually independent and can be connected by transformations via the real-space grid. This multi-basis feature renders GPAW highly versatile and unique among similar codes. By virtue of its modular structure, the GPAW code constitutes an ideal platform for the implementation of new features and methodologies. Moreover, it is well integrated with the Atomic Simulation Environment (ASE), providing a flexible and dynamic user interface. In addition to ground-state DFT calculations, GPAW supports many-body GW band structures, optical excitations from the Bethe-Salpeter Equation, variational calculations of excited states in molecules and solids via direct optimization, and real-time propagation of the Kohn-Sham equations within time-dependent DFT. A range of more advanced methods to describe magnetic excitations and non-collinear magnetism in solids are also now available. In addition, GPAW can calculate non-linear optical tensors of solids, charged crystal point defects, and much more. Recently, support for graphics processing unit (GPU) acceleration has been achieved with minor modifications to the GPAW code thanks to the CuPy library. We end the review with an outlook, describing some future plans for GPAW.

Graph Neural Network-Accelerated Multitasking Genetic Algorithm for Optimizing PdxTi1-xHy Surfaces under Various CO2 Reduction Reaction Conditions

Palladium (Pd) hydride-based catalysts have been reported to have excellent performance in the CO2 reduction reaction (CO2RR) and hydrogen evolution reaction (HER). Our previous work on doped PdH and Pd alloy hydrides showed that Ti-doped and Ti-alloyed Pd hydrides could improve the performance of the CO2 reduction reaction compared with pure Pd hydride. Compositions and chemical orderings of the surfaces with only one adsorbate under certain reaction conditions are linked to their stability, activity, and selectivity toward the CO2RR and HER, as shown in our previous work. In fact, various coverages, types, and mixtures of the adsorbates, as well as state variables such as temperature, pressure, applied potential, and chemical potential, could impact their stability, activity, and selectivity. However, these factors are usually fixed at common values to reduce the complexity of the structures and the complexity of the reaction conditions in most theoretical work. To address the complexities above and the huge search space, we apply a deep learning-assisted multitasking genetic algorithm to screen for PdxTi1-xHy surfaces containing multiple adsorbates for CO2RR under different reaction conditions. The ensemble deep learning model can greatly speed up the structure relaxations and retain a high accuracy and low uncertainty of the energy and forces. The multitasking genetic algorithm simultaneously finds globally stable surface structures under each reaction condition. Finally, 23 stable structures are screened out under different reaction conditions. Among these, Pd0.56Ti0.44H1.06 + 25%CO, Pd0.31Ti0.69H1.25 + 50%CO, Pd0.31Ti0.69H1.25 + 25%CO, and Pd0.88Ti0.12H1.06 + 25%CO are found to be very active for CO2RR and suitable to generate syngas consisting of CO and H2.

CoeffNet: predicting activation barriers through a chemically-interpretable, equivariant and physically constrained graph neural network

Activation barriers of elementary reactions are essential to predict molecular reaction mechanisms and kinetics. However, computing these energy barriers by identifying transition states with electronic structure methods (e.g., density functional theory) can be time-consuming and computationally expensive. In this work, we introduce CoeffNet, an equivariant graph neural network that predicts activation barriers using coefficients of any frontier molecular orbital (such as the highest occupied molecular orbital) of reactant and product complexes as graph node features. We show that using coefficients as features offer several advantages, such as chemical interpretability and physical constraints on the network's behaviour and numerical range. Model outputs are either activation barriers or coefficients of the chosen molecular orbital of the transition state; the latter quantity allows us to interpret the results of the neural network through chemical intuition. We test CoeffNet on a dataset of SN2 reactions as a proof-of-concept and show that the activation barriers are predicted with a mean absolute error of less than 0.025 eV. The highest occupied molecular orbital of the transition state is visualized and the distribution of the orbital densities of the transition states is described for a few prototype SN2 reactions.

A synergetic promotion of surface stability for high-voltage LiCoO2 by multi-element surface doping: a first-principles study

The utilization of high-voltage LiCoO2 is an effective approach to break through the bottleneck of practical energy density in lithium ion batteries. However, the structural and interfacial degradations at the deeply delithiated state as well as the associated safety concerns impede the application of high-voltage LiCoO2. Herein, we present a synergetic strategy for promoting the surface stability of LiCoO2 at high voltage by Ti-Mg-Al co-doping and systematically study the effects of the dopants on the surface stability, electronic structure and Li+ diffusion properties of the LiCoO2 (104) surface using first-principles calculations. It is found that Ti, Mg and Al dopants can be facilely introduced into the Co sites of the LiCoO2 (104) surface. Furthermore, the co-doping could significantly stabilize the surface oxygen of LiCoO2 at a high delithiation state. Particularly, by aggregating Ti-Mg-Al co-dopant distribution in the surface layer, surface oxygen loss is dramatically suppressed. In addition, analysis of the electronic structure indicates that Ti-Mg-Al co-doping can enhance the electronic conductivity of the LiCoO2 (104) surface and greatly inhibit the charge deficiency of the superficial lattice O atoms at a highly delithiated state. In spite of a negligible improvement in the surface Li+ diffusion kinetics, the Ti-Mg-Al surface-modified LiCoO2 is expected to exhibit improved electrochemical performance at high voltage due to its superior surface stability. Our results suggest that aggregating Ti, Mg and Al co-dopant distribution in the surface layer is a promising modulation strategy to synergistically promote the surface oxygen stability of LiCoO2 at high voltages.

Investigating Spillover Energy as a Descriptor for Single-Atom Alloy Catalyst Design

The identification of thermodynamic descriptors of catalytic performance is essential for the rational design of heterogeneous catalysts. Here, we investigate how spillover energy, a descriptor quantifying whether intermediates are more stable at the dopant or host metal sites, can be used to design single-atom alloys (SAAs) for formic acid dehydrogenation. Using theoretical calculations, we identify NiCu as a SAA with favorable spillover energy and demonstrate that formate intermediates produced after the initial O-H activation are more stable at Ni sites where rate-determining C-H activation occurs. Surface science experiments demonstrated that NiCu(111) SAAs are more reactive than Cu(111) while they still follow the formate reaction pathway. However, reactor studies of silica-supported NiCu SAA nanoparticles showed only a modest improvement over Cu resulting from surface coverage effects. Overall, this study demonstrates the potential of engineering SAAs using spillover energy as a design parameter and highlights the importance of adsorbate-adsorbate interactions under steady-state operation.

Computational Study of the Solid-State Incorporation of Sn(II) Acetate into Zeolite β

Sn-doped zeolites are potent Lewis acid catalysts for important reactions in the context of green and sustainable chemistry; however, their synthesis can have long reaction times and harsh chemical requirements, presenting an obstacle to scale-up and industrial application. To incorporate Sn into the β zeolite framework, solid-state incorporation (SSI) has recently been demonstrated as a fast and solvent-free synthetic method, with no impairment to the high activity and selectivity associated with Sn-β for its catalytic applications. Here, we report an ab initio computational study that combines periodic density functional theory with high-level embedded-cluster quantum/molecular mechanical (QM/MM) to elucidate the mechanistic steps in the synthetic process. Initially, once the Sn(II) acetate precursor coordinates to the β framework, acetic acid forms via a facile hydrogen transfer from the β framework onto the monodentate acetate ligand, with low kinetic barriers for subsequent dissociation of the ligand from the framework-bound Sn. Ketonization of the dissociated acetic acid can occur over the Lewis acidic Sn(II) site to produce CO2 and acetone with a low kinetic barrier (1.03 eV) compared to a gas-phase process (3.84 eV), helping to explain product distributions in good accordance with experimental analysis. Furthermore, we consider the oxidation of the Sn(II) species to form the Sn(IV) active site in the material by O2- and H2O-mediated mechanisms. The kinetic barrier for oxidation via H2 release is 3.26 eV, while the H2O-mediated dehydrogenation process has a minimum barrier of 1.38 eV, which indicates the possible role of residual H2O in the experimental observations of SSI synthesis. However, we find that dehydrogenation is facilitated more significantly by the presence of dioxygen (O2), introduced in the compressed air gas feed, via a two-step process oxidation process that forms H2O2 as an intermediate and has greatly reduced kinetic barriers of 0.25 and 0.26 eV. The results provide insight into how Sn insertion into β occurs during SSI and demonstrate the possible mechanism of top-down synthetic procedures for metal insertion into zeolites.

New Insight into the Pharmacological Importance of Atropine as the Potential Inhibitor of AKR1B1 via Detailed Computational Investigations: DFTs, ADMET, Molecular Docking, and Molecular Dynamics Studies

The aim of this research is to investigate the quantum geometric properties and chemical reactivity of atropine, a pharmaceutically active tropane alkaloid. Using density functional theory (DFT) computations with the B3LYP/SVP functional theory basis set, the most stable geometry of atropine was determined. Additionally, a variety of energetic molecular parameters were calculated, such as the optimized energy, atomic charges, dipole moment, frontier molecular orbital energies, HOMO-LUMO energy gap, molecular electrostatic potential, chemical reactivity descriptors, and molecular polarizability. To determine atropine's inhibitory potential, molecular docking was used to analyze ligand interactions within the active pockets of aldo-keto reductase (AKR1B1 and AKR1B10). The results of these studies showed that atropine has greater inhibitory action against AKR1B1 than AKR1B10, which was further validated through molecular dynamic simulations by analyzing root mean square deviation (RMSD) and root mean square fluctuations (RMSF). The results of the molecular docking simulation were supplemented with simulation data, and the ADMET characteristics were also determined to predict the drug likeness of a potential compound. In conclusion, the research suggests that atropine has potential as an inhibitor of AKR1B1 and could be used as a parent compound for the synthesis of more potent leads for the treatment of colon cancer associated with the sudden expression of AKR1B1.

Stability and reactivity of metal nanoclusters supported on transition metal carbides

Small particles of transition metals (TM) supported on transition metal carbides (TMC) - TM_n@TMC - provide a plethora of design opportunities for catalytic applications due to their highly exposed active centres, efficient atom utilisation and the physicochemical properties of the TMC support. To date, however, only a very small subset of TM_n@TMC catalysts have been tested experimentally and it is unclear which combinations may best catalyse which chemical reactions. Herein, we develop a high-throughput screening approach to catalyst design for supported nanoclusters based on density functional theory, and apply it to elucidate the stability and catalytic performance of all possible combinations between 7 monometallic nanoclusters (Rh, Pd, Pt, Au, Co, Ni and Cu) and 11 stable support surfaces of TMCs with 1 : 1 stoichiometry (TiC, ZrC, HfC, VC, NbC, TaC, MoC and WC) towards CH₄ and CO₂ conversion technologies. We analyse the generated database to unravel trends or simple descriptors in their resistance towards metal aggregate formation and sintering, oxidation, stability in the presence of adsorbate species, and study their adsorptive and catalytic properties, to facilitate the discovery of novel materials in the future. We identify 8 TM_n@TMC combinations as promising catalysts, all of them being new for experimental validation, thus expanding the chemical space for efficient conversion of CH₄ and CO₂.

How proton transfer impacts hachimoji DNA

Hachimoji DNA is a synthetic nucleic acid extension of DNA, formed by an additional four bases, Z, P, S, and B, that can encode information and sustain Darwinian evolution. In this paper, we aim to look into the properties of hachimoji DNA and investigate the probability of proton transfer between the bases, resulting in base mismatch under replication. First, we present a proton transfer mechanism for hachimoji DNA, analogous to the one presented by Löwdin years prior. Then, we use density functional theory to calculate proton transfer rates, tunnelling factors and the kinetic isotope effect in hachimoji DNA. We determined that the reaction barriers are sufficiently low that proton transfer is likely to occur even at biological temperatures. Furthermore, the rates of proton transfer of hachimoji DNA are much faster than in Watson-Crick DNA due to the barrier for Z-P and S-B being 30% lower than in G-C and A-T. Suggesting that proton transfer occurs more frequently in hachimoji DNA than canonical DNA, potentially leading to a higher mutation rate.

Tautomerisation Mechanisms in the Adenine-Thymine Nucleobase Pair during DNA Strand Separation

The adenine-thymine tautomer (A*-T*) has previously been discounted as a spontaneous mutagenesis mechanism due to the energetic instability of the tautomeric configuration. We study the stability of A*-T* while the nucleobases undergo DNA strand separation. Our calculations indicate an increase in the stability of A*-T* as the DNA strands unzip and the hydrogen bonds between the bases stretch. Molecular Dynamics simulations reveal the time scales and dynamics of DNA strand separation and the statistical ensemble of opening angles present in a biological environment. Our results demonstrate that the unwinding of DNA, an inherently out-of-equilibrium process facilitated by helicase, will change the energy landscape of the adenine-thymine tautomerization reaction. We propose that DNA strand separation allows the stable tautomerization of adenine-thymine, providing a feasible pathway for genetic point mutations via proton transfer between the A-T bases.

Revealing the Multifunctions of Li3N in the Suspension Electrolyte for Lithium Metal Batteries

Inorganic-rich solid-electrolyte interphases (SEIs) on Li metal anodes improve the electrochemical performance of Li metal batteries (LMBs). Therefore, a fundamental understanding of the roles played by essential inorganic compounds in SEIs is critical to realizing and developing high-performance LMBs. Among the prevalent SEI inorganic compounds observed for Li metal anodes, Li₃N is often found in the SEIs of high-performance LMBs. Herein, we elucidate new features of Li₃N by utilizing a suspension electrolyte design that contributes to the improved electrochemical performance of the Li metal anode. Through empirical and computational studies, we show that Li₃N guides Li electrodeposition along its surface, creates a weakly solvating environment by decreasing Li⁺-solvent coordination, induces organic-poor SEI on the Li metal anode, and facilitates Li⁺ transport in the electrolyte. Importantly, recognizing specific roles of SEI inorganics for Li metal anodes can serve as one of the rational guidelines to design and optimize SEIs through electrolyte engineering for LMBs.

Integrated carbon capture and conversion: A review on C2+ product mechanisms and mechanism-guided strategies

The need to reduce atmospheric CO₂ concentrations necessitates CO₂ capture technologies for conversion into stable products or long-term storage. A single pot solution that simultaneously captures and converts CO₂ could minimize additional costs and energy demands associated with CO₂ transport, compression, and transient storage. While a variety of reduction products exist, currently, only conversion to C₂₊ products including ethanol and ethylene are economically advantageous. Cu-based catalysts have the best-known performance for CO₂ electroreduction to C₂₊ products. Metal Organic Frameworks (MOFs) are touted for their carbon capture capacity. Thus, integrated Cu-based MOFs could be an ideal candidate for the one-pot capture and conversion. In this paper, we review Cu-based MOFs and MOF derivatives that have been used to synthesize C₂₊ products with the objective of understanding the mechanisms that enable synergistic capture and conversion. Furthermore, we discuss strategies based on the mechanistic insights that can be used to further enhance production. Finally, we discuss some of the challenges hindering widespread use of Cu-based MOFs and MOF derivatives along with possible solutions to overcome the challenges.

Chemical reaction networks and opportunities for machine learning

Chemical reaction networks (CRNs), defined by sets of species and possible reactions between them, are widely used to interrogate chemical systems. To capture increasingly complex phenomena, CRNs can be leveraged alongside data-driven methods and machine learning (ML). In this Perspective, we assess the diverse strategies available for CRN construction and analysis in pursuit of a wide range of scientific goals, discuss ML techniques currently being applied to CRNs and outline future CRN-ML approaches, presenting scientific and technical challenges to overcome.

Transition state search and geometry relaxation throughout chemical compound space with quantum machine learning

We use energies and forces predicted within response operator based quantum machine learning (OQML) to perform geometry optimization and transition state search calculations with legacy optimizers but without the need for subsequent re-optimization with quantum chemistry methods. For randomly sampled initial coordinates of small organic query molecules, we report systematic improvement of equilibrium and transition state geometry output as training set sizes increase. Out-of-sample S_N2 reactant complexes and transition state geometries have been predicted using the LBFGS and the QST2 algorithms with an root-mean-square deviation (RMSD) of 0.16 and 0.4 Å-after training on up to 200 reactant complex relaxations and transition state search trajectories from the QMrxn20 dataset, respectively. For geometry optimizations, we have also considered relaxation paths up to 5'595 constitutional isomers with sum formula C₇H₁₀O₂ from the QM9-database. Using the resulting OQML models with an LBFGS optimizer reproduces the minimum geometry with an RMSD of 0.14 Å, only using ∼6000 training points obtained from normal mode sampling along the optimization paths of the training compounds without the need for active learning. For converged equilibrium and transition state geometries, subsequent vibrational normal mode frequency analysis indicates deviation from MP2 reference results by on average 14 and 26 cm^-1, respectively. While the numerical cost for OQML predictions is negligible in comparison to density functional theory or MP2, the number of steps until convergence is typically larger in either case. The success rate for reaching convergence, however, improves systematically with training set size, underscoring OQML's potential for universal applicability.

Water adsorption and dynamics on graphene and other 2D materials: Computational and experimental advances

The interaction of water and surfaces, at molecular level, is of critical importance for understanding processes such as corrosion, friction, catalysis and mass transport. The significant literature on interactions with single crystal metal surfaces should not obscure unknowns in the unique behaviour of ice and the complex relationships between adsorption, diffusion and long-range inter-molecular interactions. Even less is known about the atomic-scale behaviour of water on novel, non-metallic interfaces, in particular on graphene and other 2D materials. In this manuscript, we review recent progress in the characterisation of water adsorption on 2D materials, with a focus on the nano-material graphene and graphitic nanostructures; materials which are of paramount importance for separation technologies, electrochemistry and catalysis, to name a few. The adsorption of water on graphene has also become one of the benchmark systems for modern computational methods, in particular dispersion-corrected density functional theory (DFT). We then review recent experimental and theoretical advances in studying the single-molecular motion of water at surfaces, with a special emphasis on scattering approaches as they allow an unparalleled window of observation to water surface motion, including diffusion, vibration and self-assembly.

Proton transfer during DNA strand separation as a source of mutagenic guanine-cytosine tautomers

Proton transfer between the DNA bases can lead to mutagenic Guanine-Cytosine tautomers. Over the past several decades, a heated debate has emerged over the biological impact of tautomeric forms. Here, we determine that the energy required for generating tautomers radically changes during the separation of double-stranded DNA. Density Functional Theory calculations indicate that the double proton transfer in Guanine-Cytosine follows a sequential, step-like mechanism where the reaction barrier increases quasi-linearly with strand separation. These results point to increased stability of the tautomer when the DNA strands unzip as they enter the helicase, effectively trapping the tautomer population. In addition, molecular dynamics simulations indicate that the relevant strand separation time is two orders of magnitude quicker than previously thought. Our results demonstrate that the unwinding of DNA by the helicase could simultaneously slow the formation but significantly enhance the stability of tautomeric base pairs and provide a feasible pathway for spontaneous DNA mutations.

Expanding the Material Search Space for Multivalent Cathodes

Multivalent batteries are an energy storage technology with the potential to surpass lithium-ion batteries; however, their performance have been limited by the low voltages and poor solid-state ionic mobility of available cathodes. A computational screening approach to identify high-performance multivalent intercalation cathodes among materials that do not contain the working ion of interest has been developed, which greatly expands the search space that can be considered for material discovery. This approach has been applied to magnesium cathodes as a proof of concept, and four resulting candidate materials [NASICON V₂(PO₄)₃, birnessite NaMn₄O₈, tavorite MnPO₄F, and spinel MnO₂] are discussed in further detail. In examining the ion migration environment and associated Mg²⁺ migration energy in these materials, local energy maxima are found to correspond with pathway positions where Mg²⁺ passes through a plane of anion atoms. While previous studies have established the influence of local coordination on multivalent ion mobility, these results suggest that considering both the type of the local bonding environment and available free volume for the mobile ion along its migration pathway can be significant for improving solid-state mobility.

FINETUNA: Fine-tuning Accelerated Molecular Simulations

Progress towards the energy breakthroughs needed to combat climate change can be significantly accelerated through the efficient simulation of atomistic systems. However, simulation techniques based on first principles, such as Density Functional Theory (DFT), are limited in their practical use due to their high computational expense. Machine learning approaches have the potential to approximate DFT in a computationally efficient manner, which could dramatically increase the impact of computational simulations on real-world problems. However, they are limited by their accuracy and the cost of generating labeled data. Here, we present an online active learning framework for accelerating the simulation of atomic systems efficiently and accurately by incorporating prior physical information learned by large-scale pre-trained graph neural network models from the Open Catalyst Project. Accelerating these simulations enables useful data to be generated more cheaply, allowing better models to be trained and more atomistic systems to be screened. We also present a method of comparing local optimization techniques on the basis of both their speed and accuracy. Experiments on 30 benchmark adsorbate-catalyst systems show that our method of transfer learning to incorporate prior information from pre-trained models accelerates simulations by reducing the number of DFT calculations by 91%, while meeting an accuracy threshold of 0.02 eV 93% of the time. Finally, we demonstrate a technique for leveraging the interactive functionality built in to VASP to efficiently compute single point calculations within our online active learning framework without the significant startup costs. This allows VASP to work in tandem with our framework while requiring 75% fewer self-consistent cycles than conventional single point calculations. The online active learning implementation, and examples using VASPInteractive, are available in the open source FINETUNA package on Github.

Computer-Aided Drug Design Boosts RAS Inhibitor Discovery

The Rat Sarcoma (RAS) family (NRAS, HRAS, and KRAS) is endowed with GTPase activity to regulate various signaling pathways in ubiquitous animal cells. As proto-oncogenes, RAS mutations can maintain activation, leading to the growth and proliferation of abnormal cells and the development of a variety of human cancers. For the fight against tumors, the discovery of RAS-targeted drugs is of high significance. On the one hand, the structural properties of the RAS protein make it difficult to find inhibitors specifically targeted to it. On the other hand, targeting other molecules in the RAS signaling pathway often leads to severe tissue toxicities due to the lack of disease specificity. However, computer-aided drug design (CADD) can help solve the above problems. As an interdisciplinary approach that combines computational biology with medicinal chemistry, CADD has brought a variety of advances and numerous benefits to drug design, such as the rapid identification of new targets and discovery of new drugs. Based on an overview of RAS features and the history of inhibitor discovery, this review provides insight into the application of mainstream CADD methods to RAS drug design.

Adsorption of Polycyclic Aromatic Hydrocarbons and C60 onto Forsterite: C-H Bond Activation by the Schottky Vacancy

Understanding how to catalytically break the C-H bond of aromatic molecules, such as polycyclic aromatic hydrocarbons (PAHs), is currently a big challenge and a subject of study in catalysis, astrochemistry, and planetary science. In the latter, the study of the breakdown reaction of PAHs on mineral surfaces is important to understand if PAHs are linked to prebiotic molecules in regions of star and planet formation. In this work, we employed a periodic density functional theory along with Grimme's D4 (DFT-D4) approach for studying the adsorption of a sample of PAHs (naphthalene, anthracene, fluoranthene, pyrene, coronene, and benzocoronene) and fullerene on the [010] forsterite surface and its defective surfaces (Fe-doped and Ni-doped surfaces and a MgO-Schottky vacancy) for their implications in catalysis and astrochemistry. On the basis of structural and binding energy analysis, large PAHs and fullerene present stronger adsorption on the pristine, Fe-doped, and Ni-doped forsterite surfaces than small PAHs. On a MgO-Schottky vacancy, parallel adsorption of the PAH leads to the chemisorption process (C-Si and/or C-O bonds), whereas perpendicular orientation of the PAH leads to the catalytic breaking of the aromatic C-H bond via a barrierless reaction. Spin density and charge analysis show that C-H dissociation is promoted by electron donation from the vacancy to the PAH. As a result of the undercoordinated Si and O atoms, the vacancy acts as a Frustrated Lewis Pair (FLP) catalyst. Therefore, a MgO-Schottky vacancy [010] forsterite surface proved to have potential catalytic activity for the activation of C-H bond in aromatic molecules.

Hydrogen Diffusion on, into and in Magnesium Probed by DFT: A Review

Hydrogen is an energy carrier that can be a sustainable solution for alternative energy with zero greenhouse gas emissions. Hydrogen storage is a key point for hydrogen energy. Metals provide an access for safe, controlled and reversible hydrogen storage and release. Magnesium, due to its outstanding hydrogen storage capacity, high natural abundance, low cost and non-toxicity is one of the most attractive materials for hydrogen storage. The economic efficiency of Mg as a hydrogen accumulator is limited by its sluggish hydrogen sorption kinetics and high stability of its hydride MgH2. Many attempts have been made to overcome these shortcomings. On a microscopic level, hydrogen absorption by metal is a complex multistep process that is impossible to survey experimentally. Theoretical studies help to elucidate this process and focus experimental efforts on the design of new effective Mg-based materials for hydrogen storage. This review reports on the results obtained within a density functional theory approach to studying hydrogen interactions with magnesium surfaces, diffusion on Mg surfaces, into and in bulk Mg, as well as hydrogen induced phase transformations in MgHx and hydrogen desorption from MgH2 surfaces.

Fast atomic structure optimization with on-the-fly sparse Gaussian process potentials. JOURNAL OF PHYSICS. CONDENSED MATTER : AN INSTITUTE OF PHYSICS JOURNAL 2022;34:344007. [PMID: 35675808 DOI: 10.1088/1361-648x/ac76ff]

We apply on-the-fly machine learning potentials (MLPs) using the sparse Gaussian process regression (SGPR) algorithm for fast optimization of atomic structures. Great acceleration is achieved even in the context of a single local optimization. Although for finding the exact local minimum, due to limited accuracy of MLPs, switching to another algorithm may be needed. For random gold clusters, the forces are reduced to ∼0.1 eV Å^-1within less than ten first-principles (FP) calculations. Because of highly transferable MLPs, this algorithm is specially suitable for global optimization methods such as random or evolutionary structure searching or basin hopping. This is demonstrated by sequential optimization of random gold clusters for which, after only a few optimizations, FP calculations were rarely needed.

Deep Learning-Assisted Investigation of Electric Field-Dipole Effects on Catalytic Ammonia Synthesis

External electric fields can modify binding energies of reactive surface species and enhance catalytic performance of heterogeneously catalyzed reactions. In this work, we used density functional theory (DFT) calculations-assisted and accelerated by a deep learning algorithm-to investigate the extent to which ruthenium-catalyzed ammonia synthesis would benefit from application of such external electric fields. This strategy allows us to determine which electronic properties control a molecule's degree of interaction with external electric fields. Our results show that (1) field-dependent adsorption/reaction energies are closely correlated to the dipole moments of intermediates over the surface, (2) a positive field promotes ammonia synthesis by lowering the overall energetics and decreasing the activation barriers of the potential rate-limiting steps (e.g., NH₂ hydrogenation) over Ru, (3) a positive field (>0.6 V/Å) favors the reaction mechanism by avoiding kinetically unfavorable N≡N bond dissociation over Ru(1013), and (4) local adsorption environments (i.e., dipole moments of the intermediates in the gas phase, surface defects, and surface coverage of intermediates) influence the resulting surface adsorbates' dipole moments and further modify field-dependent reaction energetics. The deep learning algorithm developed here accelerates field-dependent energy predictions with acceptable accuracies by five orders of magnitudes compared to DFT alone and has the capacity of transferability, which can predict field-dependent energetics of other catalytic surfaces with high-quality performance using little training data.

Theoretical Modeling of Electrochemical Proton-Coupled Electron Transfer

Proton-coupled electron transfer (PCET) plays an essential role in a wide range of electrocatalytic processes. A vast array of theoretical and computational methods have been developed to study electrochemical PCET. These methods can be used to calculate redox potentials and pK_a values for molecular electrocatalysts, proton-coupled redox potentials and bond dissociation free energies for PCET at metal and semiconductor interfaces, and reorganization energies associated with electrochemical PCET. Periodic density functional theory can also be used to compute PCET activation energies and perform molecular dynamics simulations of electrochemical interfaces. Various approaches for maintaining a constant electrode potential in electronic structure calculations and modeling complex interactions in the electric double layer (EDL) have been developed. Theoretical formulations for both homogeneous and heterogeneous electrochemical PCET spanning the adiabatic, nonadiabatic, and solvent-controlled regimes have been developed and provide analytical expressions for the rate constants and current densities as functions of applied potential. The quantum mechanical treatment of the proton and inclusion of excited vibronic states have been shown to be critical for describing experimental data, such as Tafel slopes and potential-dependent kinetic isotope effects. The calculated rate constants can be used as input to microkinetic models and voltammogram simulations to elucidate complex electrocatalytic processes.

The Energy Landscape Perspective: Encoding Structure and Function for Biomolecules

The energy landscape perspective is outlined with particular reference to biomolecules that perform multiple functions. We associate these multifunctional molecules with multifunnel energy landscapes, illustrated by some selected examples, where understanding the organisation of the landscape has provided new insight into function. Conformational selection and induced fit may provide alternative routes to realisation of multifunctionality, exploiting the possibility of environmental control and distinct binding modes.

Solid-State Structural Properties of Alloxazine Determined from Powder XRD Data in Conjunction with DFT-D Calculations and Solid-State NMR Spectroscopy: Unraveling the Tautomeric Identity and Pathways for Tautomeric Interconversion

We report the solid-state structural properties of alloxazine, a tricyclic ring system found in many biologically important molecules, with structure determination carried out directly from powder X-ray diffraction (XRD) data. As the crystal structures containing the alloxazine and isoalloxazine tautomers both give a high-quality fit to the powder XRD data in Rietveld refinement, other techniques are required to establish the tautomeric form in the solid state. In particular, high-resolution solid-state ¹⁵N NMR data support the presence of the alloxazine tautomer, based on comparison between isotropic chemical shifts in the experimental ¹⁵N NMR spectrum and the corresponding values calculated for the crystal structures containing the alloxazine and isoalloxazine tautomers. Furthermore, periodic DFT-D calculations at the PBE0-MBD level indicate that the crystal structure containing the alloxazine tautomer has significantly lower energy. We also report computational investigations of the interconversion between the tautomeric forms in the crystal structure via proton transfer along two intermolecular N-H···N hydrogen bonds; DFT-D calculations at the PBE0-MBD level indicate that the tautomeric interconversion is associated with a lower energy transition state for a mechanism involving concerted (rather than sequential) proton transfer along the two hydrogen bonds. However, based on the relative energies of the crystal structures containing the alloxazine and isoalloxazine tautomers, it is estimated that under conditions of thermal equilibrium at ambient temperature, more than 99.9% of the molecules in the crystal structure will exist as the alloxazine tautomer.

Hydrodeoxygenation of guaiacol over orthorhombic molybdenum carbide: a DFT and microkinetic study

The hydrodeoxygenation of guaiacol is modelled over a (100) β-Mo2C surface using density functional theory and microkinetic simulations.

Bayesian Optimization of High-Entropy Alloy Compositions for Electrocatalytic Oxygen Reduction*

Active, selective and stable catalysts are imperative for sustainable energy conversion, and engineering materials with such properties are highly desired. High‐entropy alloys (HEAs) offer a vast compositional space for tuning such properties. Too vast, however, to traverse without the proper tools. Here, we report the use of Bayesian optimization on a model based on density functional theory (DFT) to predict the most active compositions for the electrochemical oxygen reduction reaction (ORR) with the least possible number of sampled compositions for the two HEAs Ag‐Ir‐Pd‐Pt‐Ru and Ir‐Pd‐Pt‐Rh‐Ru. The discovered optima are then scrutinized with DFT and subjected to experimental validation where optimal catalytic activities are verified for Ag–Pd, Ir–Pt, and Pd–Ru binary alloys. This study offers insight into the number of experiments needed for optimizing the vast compositional space of multimetallic alloys which has been determined to be on the order of 50 for ORR on these HEAs.

Applying Active Learning to the Screening of Molecular Oxygen Evolution Catalysts

The oxygen evolution reaction (OER) can enable green hydrogen production; however, the state-of-the-art catalysts for this reaction are composed of prohibitively expensive materials. In addition, cheap catalysts have associated overpotentials that render the reaction inefficient. This impels the search to discover novel catalysts for this reaction computationally. In this communication, we present machine learning algorithms to enhance the hypothetical screening of molecular OER catalysts. By predicting calculated binding energies using Gaussian process regression (GPR) models and applying active learning schemes, we provide evidence that our algorithm can improve computational efficiency by guiding simulations towards candidates with promising OER descriptor values. Furthermore, we derive an acquisition function that, when maximized, can identify catalysts that can exhibit theoretical overpotentials that circumvent the constraints imposed by linear scaling relations by attempting to enforce a specific mechanism. Finally, we provide a brief perspective on the appropriate sets of molecules to consider when screening complexes that could be stable and active for this reaction.

Atomic Structure Optimization with Machine-Learning Enabled Interpolation between Chemical Elements

We introduce a computational method for global optimization of structure and ordering in atomic systems. The method relies on interpolation between chemical elements, which is incorporated in a machine-learning structural fingerprint. The method is based on Bayesian optimization with Gaussian processes and is applied to the global optimization of Au-Cu bulk systems, Cu-Ni surfaces with CO adsorption, and Cu-Ni clusters. The method consistently identifies low-energy structures, which are likely to be the global minima of the energy. For the investigated systems with 23-66 atoms, the number of required energy and force calculations is in the range 3-75.

First-principles materials simulation and design for alkali and alkaline metal ion batteries accelerated by machine learning

The challenge of regeneration of batteries requires a performance improvement in the alkali/alkaline metal ion battery (AMIB) materials, whereas the traditional research paradigm fully based on experiments and theoretical simulations needs massive research and development investment. During the last decade, machine learning (ML) has made breakthroughs in many complex disciplines, which testifies to their high processing speed and ability to capture relationships. Inspired by these achievements, ML has also been introduced to bring a new paradigm for shortening the development of AMIB materials. In this Perspective, the focus will be on how this new ML technology solves the key problems of redox potentials, ionic conductivity and stability parameters in first-principles materials' simulation and design for AMIBs. It is found that ML not only accelerates the property prediction, but also gives physicochemical insights into AMIB materials' design. In addition, the final part of this paper summarizes current achievements and looks forward to the progress of a novel paradigm in direct/inverse design with the increasing number of databases, skills, and ML technologies for AMIBs.

Infusing theory into deep learning for interpretable reactivity prediction

Despite recent advances of data acquisition and algorithms development, machine learning (ML) faces tremendous challenges to being adopted in practical catalyst design, largely due to its limited generalizability and poor explainability. Herein, we develop a theory-infused neural network (TinNet) approach that integrates deep learning algorithms with the well-established d-band theory of chemisorption for reactivity prediction of transition-metal surfaces. With simple adsorbates (e.g., *OH, *O, and *N) at active site ensembles as representative descriptor species, we demonstrate that the TinNet is on par with purely data-driven ML methods in prediction performance while being inherently interpretable. Incorporation of scientific knowledge of physical interactions into learning from data sheds further light on the nature of chemical bonding and opens up new avenues for ML discovery of novel motifs with desired catalytic properties.

Gaussian Process Regression for Materials and Molecules

We provide an introduction to Gaussian process regression (GPR) machine-learning methods in computational materials science and chemistry. The focus of the present review is on the regression of atomistic properties: in particular, on the construction of interatomic potentials, or force fields, in the Gaussian Approximation Potential (GAP) framework; beyond this, we also discuss the fitting of arbitrary scalar, vectorial, and tensorial quantities. Methodological aspects of reference data generation, representation, and regression, as well as the question of how a data-driven model may be validated, are reviewed and critically discussed. A survey of applications to a variety of research questions in chemistry and materials science illustrates the rapid growth in the field. A vision is outlined for the development of the methodology in the years to come.

Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems

Machine learning models are poised to make a transformative impact on chemical sciences by dramatically accelerating computational algorithms and amplifying insights available from computational chemistry methods. However, achieving this requires a confluence and coaction of expertise in computer science and physical sciences. This Review is written for new and experienced researchers working at the intersection of both fields. We first provide concise tutorials of computational chemistry and machine learning methods, showing how insights involving both can be achieved. We follow with a critical review of noteworthy applications that demonstrate how computational chemistry and machine learning can be used together to provide insightful (and useful) predictions in molecular and materials modeling, retrosyntheses, catalysis, and drug design.

Computational Discovery of Transition-metal Complexes: From High-throughput Screening to Machine Learning

Transition-metal complexes are attractive targets for the design of catalysts and functional materials. The behavior of the metal-organic bond, while very tunable for achieving target properties, is challenging to predict and necessitates searching a wide and complex space to identify needles in haystacks for target applications. This review will focus on the techniques that make high-throughput search of transition-metal chemical space feasible for the discovery of complexes with desirable properties. The review will cover the development, promise, and limitations of "traditional" computational chemistry (i.e., force field, semiempirical, and density functional theory methods) as it pertains to data generation for inorganic molecular discovery. The review will also discuss the opportunities and limitations in leveraging experimental data sources. We will focus on how advances in statistical modeling, artificial intelligence, multiobjective optimization, and automation accelerate discovery of lead compounds and design rules. The overall objective of this review is to showcase how bringing together advances from diverse areas of computational chemistry and computer science have enabled the rapid uncovering of structure-property relationships in transition-metal chemistry. We aim to highlight how unique considerations in motifs of metal-organic bonding (e.g., variable spin and oxidation state, and bonding strength/nature) set them and their discovery apart from more commonly considered organic molecules. We will also highlight how uncertainty and relative data scarcity in transition-metal chemistry motivate specific developments in machine learning representations, model training, and in computational chemistry. Finally, we will conclude with an outlook of areas of opportunity for the accelerated discovery of transition-metal complexes.

Differentiable sampling of molecular geometries with uncertainty-based adversarial attacks

Neural network (NN) interatomic potentials provide fast prediction of potential energy surfaces, closely matching the accuracy of the electronic structure methods used to produce the training data. However, NN predictions are only reliable within well-learned training domains, and show volatile behavior when extrapolating. Uncertainty quantification methods can flag atomic configurations for which prediction confidence is low, but arriving at such uncertain regions requires expensive sampling of the NN phase space, often using atomistic simulations. Here, we exploit automatic differentiation to drive atomistic systems towards high-likelihood, high-uncertainty configurations without the need for molecular dynamics simulations. By performing adversarial attacks on an uncertainty metric, informative geometries that expand the training domain of NNs are sampled. When combined with an active learning loop, this approach bootstraps and improves NN potentials while decreasing the number of calls to the ground truth method. This efficiency is demonstrated on sampling of kinetic barriers, collective variables in molecules, and supramolecular chemistry in zeolite-molecule interactions, and can be extended to any NN potential architecture and materials system.