Machine Learning for Electronically Excited States of Molecules

Electronically excited states of molecules are at the heart of photochemistry, photophysics, as well as photobiology and also play a role in material science. Their theoretical description requires highly accurate quantum chemical calculations, which are computationally expensive. In this review, we focus on not only how machine learning is employed to speed up such excited-state simulations but also how this branch of artificial intelligence can be used to advance this exciting research field in all its aspects. Discussed applications of machine learning for excited states include excited-state dynamics simulations, static calculations of absorption spectra, as well as many others. In order to put these studies into context, we discuss the promises and pitfalls of the involved machine learning techniques. Since the latter are mostly based on quantum chemistry calculations, we also provide a short introduction into excited-state electronic structure methods and approaches for nonadiabatic dynamics simulations and describe tricks and problems when using them in machine learning for excited states of molecules.

Machine Learning for Electronically Excited States of Molecules

Electronically excited states of molecules are at the heart of photochemistry, photophysics, as well as photobiology and also play a role in material science. Their theoretical description requires highly accurate quantum chemical calculations, which are computationally expensive. In this review, we focus on not only how machine learning is employed to speed up such excited-state simulations but also how this branch of artificial intelligence can be used to advance this exciting research field in all its aspects. Discussed applications of machine learning for excited states include excited-state dynamics simulations, static calculations of absorption spectra, as well as many others. In order to put these studies into context, we discuss the promises and pitfalls of the involved machine learning techniques. Since the latter are mostly based on quantum chemistry calculations, we also provide a short introduction into excited-state electronic structure methods and approaches for nonadiabatic dynamics simulations and describe tricks and problems when using them in machine learning for excited states of molecules.

Perspective on integrating machine learning into computational chemistry and materials science

Machine learning (ML) methods are being used in almost every conceivable area of electronic structure theory and molecular simulation. In particular, ML has become firmly established in the construction of high-dimensional interatomic potentials. Not a day goes by without another proof of principle being published on how ML methods can represent and predict quantum mechanical properties-be they observable, such as molecular polarizabilities, or not, such as atomic charges. As ML is becoming pervasive in electronic structure theory and molecular simulation, we provide an overview of how atomistic computational modeling is being transformed by the incorporation of ML approaches. From the perspective of the practitioner in the field, we assess how common workflows to predict structure, dynamics, and spectroscopy are affected by ML. Finally, we discuss how a tighter and lasting integration of ML methods with computational chemistry and materials science can be achieved and what it will mean for research practice, software development, and postgraduate training.

Machine Learning for Chemical Reactions

Machine learning (ML) techniques applied to chemical reactions have a long history. The present contribution discusses applications ranging from small molecule reaction dynamics to computational platforms for reaction planning. ML-based techniques can be particularly relevant for problems involving both computation and experiments. For one, Bayesian inference is a powerful approach to develop models consistent with knowledge from experiments. Second, ML-based methods can also be used to handle problems that are formally intractable using conventional approaches, such as exhaustive characterization of state-to-state information in reactive collisions. Finally, the explicit simulation of reactive networks as they occur in combustion has become possible using machine-learned neural network potentials. This review provides an overview of the questions that can and have been addressed using machine learning techniques, and an outlook discusses challenges in this diverse and stimulating field. It is concluded that ML applied to chemistry problems as practiced and conceived today has the potential to transform the way with which the field approaches problems involving chemical reactions, in both research and academic teaching.

Breaking the Coupled Cluster Barrier for Machine-Learned Potentials of Large Molecules: The Case of 15-Atom Acetylacetone

Machine-learned potential energy surfaces (PESs) for molecules with more than 10 atoms are typically forced to use lower-level electronic structure methods such as density functional theory (DFT) and second-order Møller-Plesset perturbation theory (MP2). While these are efficient and realistic, they fall short of the accuracy of the "gold standard" coupled-cluster method, especially with respect to reaction and isomerization barriers. We report a major step forward in applying a Δ-machine learning method to the challenging case of acetylacetone, whose MP2 barrier height for H-atom transfer is low by roughly 1.1 kcal/mol relative to the benchmark CCSD(T) barrier of 3.2 kcal/mol. From a database of 2151 local CCSD(T) energies and training with as few as 430 energies, we obtain a new PES with a barrier of 3.5 kcal/mol in agreement with the LCCSD(T) barrier of 3.5 kcal/mol and close to the benchmark value. Tunneling splittings due to H-atom transfer are calculated using this new PES, providing improved estimates over previous ones obtained using an MP2-based PES.

Computational approaches to dissociative chemisorption on metals: towards chemical accuracy

We review the state-of-the-art in the theory of dissociative chemisorption (DC) of small gas phase molecules on metal surfaces, which is important to modeling heterogeneous catalysis for practical reasons, and for achieving an understanding of the wealth of experimental information that exists for this topic, for fundamental reasons. We first give a quick overview of the experimental state of the field. Turning to the theory, we address the challenge that barrier heights (E_b, which are not observables) for DC on metals cannot yet be calculated with chemical accuracy, although embedded correlated wave function theory and diffusion Monte-Carlo are moving in this direction. For benchmarking, at present chemically accurate E_b can only be derived from dynamics calculations based on a semi-empirically derived density functional (DF), by computing a sticking curve and demonstrating that it is shifted from the curve measured in a supersonic beam experiment by no more than 1 kcal mol^-1. The approach capable of delivering this accuracy is called the specific reaction parameter (SRP) approach to density functional theory (DFT). SRP-DFT relies on DFT and on dynamics calculations, which are most efficiently performed if a potential energy surface (PES) is available. We therefore present a brief review of the DFs that now exist, also considering their performance on databases for E_b for gas phase reactions and DC on metals, and for adsorption to metals. We also consider expressions for SRP-DFs and briefly discuss other electronic structure methods that have addressed the interaction of molecules with metal surfaces. An overview is presented of dynamical models, which make a distinction as to whether or not, and which dissipative channels are modeled, the dissipative channels being surface phonons and electronically non-adiabatic channels such as electron-hole pair excitation. We also discuss the dynamical methods that have been used, such as the quasi-classical trajectory method and quantum dynamical methods like the time-dependent wave packet method and the reaction path Hamiltonian method. Limits on the accuracy of these methods are discussed for DC of diatomic and polyatomic molecules on metal surfaces, paying particular attention to reduced dimensionality approximations that still have to be invoked in wave packet calculations on polyatomic molecules like CH₄. We also address the accuracy of fitting methods, such as recent machine learning methods (like neural network methods) and the corrugation reducing procedure. In discussing the calculation of observables we emphasize the importance of modeling the properties of the supersonic beams in simulating the sticking probability curves measured in the associated experiments. We show that chemically accurate barrier heights have now been extracted for DC in 11 molecule-metal surface systems, some of which form the most accurate core of the only existing database of E_b for DC reactions on metal surfaces (SBH10). The SRP-DFs (or candidate SRP-DFs) that have been derived show transferability in many cases, i.e., they have been shown also to yield chemically accurate E_b for chemically related systems. This can in principle be exploited in simulating rates of catalyzed reactions on nano-particles containing facets and edges, as SRP-DFs may be transferable among systems in which a molecule dissociates on low index and stepped surfaces of the same metal. In many instances SRP-DFs have allowed important conclusions regarding the mechanisms underlying observed experimental trends. An important recent observation is that SRP-DFT based on semi-local exchange DFs has so far only been successful for systems for which the difference of the metal work function and the molecule's electron affinity exceeds 7 eV. A main challenge to SRP-DFT is to extend its applicability to the other systems, which involve a range of important DC reactions of e.g. O₂, H₂O, NH₃, CO₂, and CH₃OH. Recent calculations employing a PES based on a screened hybrid exchange functional suggest that the road to success may be based on using exchange functionals of this category.

Facilitated inversion complicates the stereodynamics of an SN2 reaction at nitrogen center

Bimolecular nucleophilic substitution (S_N2) reactions at carbon center are well known to proceed with the stereospecific Walden-inversion mechanism. Reaction dynamics simulations on a newly developed high-level ab initio analytical potential energy surface for the F⁻ + NH₂Cl nitrogen-centered S_N2 and proton-transfer reactions reveal a hydrogen-bond-formation-induced multiple-inversion mechanism undermining the stereospecificity of the N-centered S_N2 channel. Unlike the analogous F⁻ + CH₃Cl S_N2 reaction, F⁻ + NH₂Cl → Cl⁻ + NH₂F is indirect, producing a significant amount of NH₂F with retention, as well as inverted NH₂Cl during the timescale within the unperturbed NH₂Cl molecule gets inverted with only low probability, showing the important role of facilitated inversions via an FH…NHCl⁻-like transition state. Proton transfer leading to HF + NHCl⁻ is more direct and becomes the dominant product channel at higher collision energies.

Multiple-inversion, the analogue of the double-inversion pathway recently revealed for S_N2@C, is the key mechanism in S_N2 at N center undermining stereospecificity.

Dynamics of the isotope exchange reaction of D with H3 +, H2D+, and D2H. J Chem Phys 2021;154:084307. [PMID: 33639774 DOI: 10.1063/5.0038434]

We have measured the merged-beams rate coefficient for the titular isotope exchange reactions as a function of the relative collision energy in the range of ∼3 meV-10 eV. The results appear to scale with the number of available sites for deuteration. We have performed extensive theoretical calculations to characterize the zero-point energy corrected reaction path. Vibrationally adiabatic minimum energy paths were obtained using a combination of unrestricted quadratic configuration interaction of single and double excitations and internally contracted multireference configuration interaction calculations. The resulting barrier height, ranging from 68 meV to 89 meV, together with the various asymptotes that may be reached in the collision, was used in a classical over-the-barrier model. All competing endoergic reaction channels were taken into account using a flux reduction factor. This model reproduces all three experimental sets quite satisfactorily. In order to generate thermal rate coefficients down to 10 K, the internal excitation energy distribution of each H₃ ⁺ isotopologue is evaluated level by level using available line lists and accurate spectroscopic parameters. Tunneling is accounted for by a direct inclusion of the exact quantum tunneling probability in the evaluation of the cross section. We derive a thermal rate coefficient of <1×10^-12 cm³ s^-1 for temperatures below 44 K, 86 K, and 139 K for the reaction of D with H₃ ⁺, H₂D⁺, and D₂H⁺, respectively, with tunneling effects included. The derived thermal rate coefficients exceed the ring polymer molecular dynamics prediction of Bulut et al. [J. Phys. Chem. A 123, 8766 (2019)] at all temperatures.

Supercollisions of fast H-atom with ethylene on an accurate full-dimensional potential energy surface

The collisions transferring large portions of energy are often called supercollisions. In the H + C₂H₂ reactive system, the rovibrationally cold C₂H₂ molecule can be activated with substantial internal excitations by its collision with a translationally hot H atom. It is interesting to investigate the mechanisms of collisional energy transfer in other important reactions of H with hydrocarbons. Here, an accurate, global, full-dimensional potential energy surface (PES) of H + C₂H₄ was constructed by the fundamental invariant neural network fitting based on roughly 100 000 UCCSD(T)-F12a/aug-cc-pVTZ data points. Extensive quasi-classical trajectory calculations were carried out on the full-dimensional PES to investigate the energy transfer process in collisions of the translationally hot H atoms with C₂H₄ in a wide range of collision energies. The computed function of the energy-transfer probability is not a simple exponential decay function but exhibits large magnitudes in the region of a large amount of energy transfer, indicating the signature of supercollisions. The supercollisions among non-complex-forming nonreactive (prompt) trajectories are frustrated complex-forming processes in which the incoming H atom penetrates into C₂H₄ with a small C-H distance but promptly and directly leaves C₂H₄. The complex-forming supercollisions, in which either the attacking H atom leaves (complex-forming nonreactive collisions) or one of the original H atoms of C₂H₄ leaves (complex-forming reactive trajectories), dominate large energy transfer from the translational energy to internal excitation of molecule. The current work sheds valuable light on the energy transfer of this important reaction in the combustion and may motivate related experimental investigations.

Full-dimensional potential energy surface for acetylacetone and tunneling splittings

New, full-dimensional potential energy surface for acetylacetone allows for description of H-tunneling dynamics and characterization of stationary points.

Constructing feed-forward artificial neural networks to fit potential energy surfaces for molecular simulation of high-temperature gas flows

Kinetic rates for thermochemical nonequilibrium models are generally computed from quasiclassical trajectory (QCT) calculations on accurate ab initio potential energy surfaces (PES). In this article, we use a feed-forward artificial neural network (ANN) to fit existing single-point energies for N_{2}+N_{2} interactions [Bender et al., J. Chem. Phys. 143, 054304 (2015)JCPSA60021-960610.1063/1.4927571] to construct a PES suitable for molecular simulation of high-temperature gas flows. We then perform detailed comparisons with a widely used N_{4} PES that was built using the permutation invariant polynomials (PIP) method. Specific physical considerations in the construction of the ANN for this application are detailed. Translation, rotation, and permutation invariance are precisely satisfied by mapping the interatomic distances onto a set of permutation invariant inputs, known as fundamental invariants (FI) that generate the permutation invariant polynomial ring. The diatomic energy is imposed by decomposing the total potential energy into a sum of a two-body and a many-body energy contribution. To obtain the correct dynamical behavior with the most basic, yet computationally efficient ANN, spurious long-distance interactions must be removed to avoid incorrect physical behavior at the dissociation threshold. We use a simple apodization function to smoothly taper off to zero any residual many-body interaction at large separations. Both accuracy and performance of the FI-ANN PES are assessed. QCT calculations are used to compute dissociation probabilities and vibrational energy distributions at various equilibrium temperatures. Excellent agreement with the results obtained from the PIP PES is found. For our test case, the ANN PES is also significantly more computationally efficient than the PIP PES at comparable root-mean-square error levels.

Experimental study of the proton-transfer reaction C + H2+ → CH+ + H and its isotopic variant (D2+)

We report absolute integral cross section (ICS) measurements using a dual-source merged-fast-beams apparatus to study the titular reactions over the relative translational energy range of Er ∼ 0.01-10 eV. We used photodetachment of C- to produce a pure beam of atomic C in the ground electronic 3P term, with statistically populated fine-structure levels. The H2+ and D2+ were formed in an electron impact ionization source, with well known vibrational and rotational distributions. The experimental work is complemented by a theoretical study of the CH2+ electronic system in the reactant and product channels, which helps to clarify the possible reaction mechanisms underlying the ICS measurements. Our measurements provide evidence that the reactions are barrierless and exoergic. They also indicate the apparent absence of an intermolecular isotope effect, to within the total experimental uncertainties. Capture models, taking into account either the charge-induced dipole interaction potential or the combined charge-quadrupole and charge-induced dipole interaction potentials, produce reaction cross sections that lie a factor of ∼4 above the experimental results. Based on our theoretical study, we hypothesize that the reaction is most likely to proceed adiabatically through the 14A' and 14A'' states of CH2+via the reaction C(3P) + H2+(2Σ+g) → CH+(3Π) + H(2S). We also hypothesize that at low collision energies only H2+(v ≤ 2) and D2+(v ≤ 3) contribute to the titular reactions, due to the onset of dissociative charge transfer for higher vibrational v levels. Incorporating these assumptions into the capture models brings them into better agreement with the experimental results. Still, for energies ⪅0.1 eV where capture models are most relevant, the modified charge-induced dipole model yields reaction cross sections with an incorrect energy dependence and lying ∼10% below the experimental results. The capture cross section obtained from the combined charge-quadrupole and charge-induced dipole model better matches the measured energy dependence but lies ∼30-50% above the experimental results. These findings provide important guidance for future quasiclassical trajectory and quantum mechanical treatments of this reaction.

Neural Network Potential Energy Surfaces for Small Molecules and Reactions

We review progress in neural network (NN)-based methods for the construction of interatomic potentials from discrete samples (such as ab initio energies) for applications in classical and quantum dynamics including reaction dynamics and computational spectroscopy. The main focus is on methods for building molecular potential energy surfaces (PES) in internal coordinates that explicitly include all many-body contributions, even though some of the methods we review limit the degree of coupling, due either to a desire to limit computational cost or to limited data. Explicit and direct treatment of all many-body contributions is only practical for sufficiently small molecules, which are therefore our primary focus. This includes small molecules on surfaces. We consider direct, single NN PES fitting as well as more complex methods that impose structure (such as a multibody representation) on the PES function, either through the architecture of one NN or by using multiple NNs. We show how NNs are effective in building representations with low-dimensional functions including dimensionality reduction. We consider NN-based approaches to build PESs in the sums-of-product form important for quantum dynamics, ways to treat symmetry, and issues related to sampling data distributions and the relation between PES errors and errors in observables. We highlight combinations of NNs with other ideas such as permutationally invariant polynomials or sums of environment-dependent atomic contributions, which have recently emerged as powerful tools for building highly accurate PESs for relatively large molecular and reactive systems.

On the separability of large-amplitude motions in anharmonic frequency calculations

Nuclear vibrational theories based upon the Watson Hamiltonian are ubiquitous in quantum chemistry, but are generally unable to model systems in which the wavefunction can delocalise over multiple energy minima, i.e. molecules that have low-energy torsion and inversion barriers. In a 2019 Chemical Reviews article, Puzzarini et al. note that a common workaround is to simply decouple these problematic modes from all other vibrations in the system during anharmonic frequency calculations. They also point out that this approximation can be "ill-suited", but do not quantify the errors introduced. In this work, we present the first systematic investigation into how separating out or constraining torsion and inversion vibrations within potential energy surface (PES) expansions affects the accuracy of computed fundamental wavenumbers for the remaining vibrational modes, using a test set of 19 tetratomic molecules for which high quality analytic potential energy surfaces and fully-coupled anharmonic reference fundamental frequencies are available. We find that the most effective and efficient strategy is to remove the mode in question from the PES expansion entirely. This introduces errors of up to +10 cm-1 in stretching fundamentals that would otherwise couple to the dropped mode, and ±5 cm-1 in all other fundamentals. These errors are approximately commensurate with, but not necessarily additional to, errors due to the choice of electronic structure model used in constructing spectroscopically accurate PES.

Quantitative Dynamics of the N2O + C2H2 → Oxadiazole Reaction: A Model for 1,3-Dipolar Cycloadditions

The reaction N₂O + C₂H₂ → oxadiazole has been considered as a prototype for 1,3-dipolar cycloadditions. Here, we report a comprehensive dynamical study of this important reaction on a full-dimensional potential energy surface, which is fitted to about 64 000 high-level ab initio data by a machine learning approach. Comprehensive dynamical simulations are carried out to provide quantitative chemical insight into its reaction dynamics. In addition to confirming the enhancement effect of the N₂O bending mode on the reactivity, intricate mode specificity effects of other vibrational modes in reactants are revealed for the first time. The asymmetric stretching mode of N₂O and the C-C-H bending mode of C₂H₂ show no effect. All remaining modes can enhance the reactivity. In particular, the vibrational excitation of the N₂O symmetric stretching mode shows similar enhancement effect on the title reaction, compared to its bending mode excitation. Detailed analysis reveals that the concerted mechanism dominates with the reactants propelled sufficiently close to each other to yield product. This study advances our understanding of the chemical dynamics of the title reaction.

Exploring the Mechanism of Catalysis with the Unified Reaction Valley Approach (URVA)—A Review

The unified reaction valley approach (URVA) differs from mainstream mechanistic studies, as it describes a chemical reaction via the reaction path and the surrounding reaction valley on the potential energy surface from the van der Waals region to the transition state and far out into the exit channel, where the products are located. The key feature of URVA is the focus on the curving of the reaction path. Moving along the reaction path, any electronic structure change of the reacting molecules is registered by a change in their normal vibrational modes and their coupling with the path, which recovers the curvature of the reaction path. This leads to a unique curvature profile for each chemical reaction with curvature minima reflecting minimal change and curvature maxima, the location of important chemical events such as bond breaking/forming, charge polarization and transfer, rehybridization, etc. A unique decomposition of the path curvature into internal coordinate components provides comprehensive insights into the origins of the chemical changes taking place. After presenting the theoretical background of URVA, we discuss its application to four diverse catalytic processes: (i) the Rh catalyzed methanol carbonylation—the Monsanto process; (ii) the Sharpless epoxidation of allylic alcohols—transition to heterogenous catalysis; (iii) Au(I) assisted [3,3]-sigmatropic rearrangement of allyl acetate; and (iv) the Bacillus subtilis chorismate mutase catalyzed Claisen rearrangement—and show how URVA leads to a new protocol for fine-tuning of existing catalysts and the design of new efficient and eco-friendly catalysts. At the end of this article the pURVA software is introduced. The overall goal of this article is to introduce to the chemical community a new protocol for fine-tuning existing catalytic reactions while aiding in the design of modern and environmentally friendly catalysts.

Fitting potential energy surfaces with fundamental invariant neural network. II. Generating fundamental invariants for molecular systems with up to ten atoms

Symmetry adaptation is crucial in representing a permutationally invariant potential energy surface (PES). Due to the rapid increase in computational time with respect to the molecular size, as well as the reliance on the algebra software, the previous neural network (NN) fitting with inputs of fundamental invariants (FIs) has practical limits. Here, we report an improved and efficient generation scheme of FIs based on the computational invariant theory and parallel program, which can be readily used as the input vector of NNs in fitting high-dimensional PESs with permutation symmetry. The newly developed method significantly reduces the evaluation time of FIs, thereby extending the FI-NN method for constructing highly accurate PESs to larger systems beyond five atoms. Because of the minimum size of invariants used in the inputs of the NN, the NN structure can be very flexible for FI-NN, which leads to small fitting errors. The resulting FI-NN PES is much faster on evaluating than the corresponding permutationally invariant polynomial-NN PES.

Efficient Generation of Permutationally Invariant Potential Energy Surfaces for Large Molecules

An efficient method is described for generating a fragmented, permutationally invariant polynomial basis to fit electronic energies and, if available, gradients for large molecules. The method presented rests on the fragmentation of a large molecule into any number of fragments while maintaining the permutational invariance and uniqueness of the polynomials. The new approach improves on a previous one reported by Qu and Bowman by avoiding repetition of polynomials in the fitting basis set and speeding up gradient evaluations while keeping the accuracy of the PES. The method is demonstrated for CH₃–NH–CO–CH₃ (N-methylacetamide) and NH₂–CH₂–COOH (glycine).

Exact quantum dynamics background of dispersion interactions: case study for CH4·Ar in full (12) dimensions

A full-dimensional ab initio potential energy surface of spectroscopic quality is developed for the van-der-Waals complex of a methane molecule and an argon atom. Variational vibrational states are computed on this surface including all twelve (12) vibrational degrees of freedom of the methane-argon complex using the GENIUSH computer program and the Smolyak sparse grid method. The full-dimensional computations make it possible to study the fine details of the interaction and distortion effects and to make a direct assessment of the reduced-dimensionality models often used in the quantum dynamics study of weakly-bound complexes. A 12-dimensional (12D) vibrational computation including only a single harmonic oscillator basis function (9D) to describe the methane fragment (for which we use the ground-state effective structure as the reference structure) has a 0.40 cm^-1 root-mean-square error (rms) with respect to the converged 12D bound-state excitation energies, which is less than half of the rms of the 3D model set up with the 〈r〉₀ methane structure. Allowing 10 basis functions for the methane fragment in a 12D computation performs much better than the 3D models by reducing the rms of the bound state vibrational energies to 0.07 cm^-1. The full-dimensional potential energy surface correctly describes the dissociation of the system, which together with further development of the variational (ro)vibrational methodology opens a route to the study of the role of dispersion forces in the excited methane vibrations and the energy transfer from the intra- to the intermolecular vibrational modes.

On the development of a gold-standard potential energy surface for the OH− + CH3I reaction

We develop the first accurate full-dimensional ab initio PES for the OH⁻ + CH₃I S_N2 and proton-transfer reactions treating the failure of CCSD(T) at certain geometries.

Benchmark ab initio and dynamical characterization of the stationary points of reactive atom + alkane and SN2 potential energy surfaces

We review composite ab initio and dynamical methods and their applications to characterize stationary points of atom/ion + molecule reactions.

Intramolecular vibrational energy redistribution and the quantum ergodicity transition: a phase space perspective

The onset of facile intramolecular vibrational energy flow can be related to features in the connected network of anharmonic resonances in the classical phase space.

Kinetics and dynamics study of the OH + C2H6 → H2O + C2H5 reaction based on an analytical global potential energy surface

To describe the gas-phase hydrogen abstraction reaction between the hydroxyl radical and the ethane molecule, an analytical full-dimensional potential energy surface was developed within the Born-Oppenheimer approximation. This reactive process is a ten-body system with 24 degrees of freedom, which represents a theoretical challenge. The new surface, named PES-2020, presents low barrier, 3.76 kcal mol-1, high exothermicity, -16.20 kcal mol-1, and intermediate complexes in the entrance and exit channels. To test the quality and accuracy of the analytical surface several stringent tests were performed and, in general, PES-2020 reasonably simulates the theoretical information used as input, it is a continuous and smooth potential, without spurious minima, it presents great versatility and a reasonable description of this ten-body reaction. Based on this surface, an exhaustive kinetics and dynamics study was performed with a double objective: to analyze the capacity of the new surface to simulate the experimental evidence, and to help understand the mechanism of reaction and the role of the ethyl group in the reaction. In the kinetics study, three approaches were used: variational transition-state theory with multidimensional tunnelling (VTST/MT), ring polymer molecular dynamics (RPMD) and quasi-classical trajectory (QCT) results, in the temperature range 200-2000 K. There is general agreement between the three approaches and they reasonably simulate the experimental behaviour, which gives confidence to the fitness of the new surface to describe the system. In the dynamics study, QCT calculations were performed at 298 K for a direct comparison with the only experimental result reported. We found that ethyl fragment presents a noticeable internal energy (∼20%) and so cannot be considered as a spectator. The water product vibrational energy is reasonably reproduced, though when a level-by-level distribution is analyzed the agreement is only qualitative.

A global ab initio potential energy surface and dynamics of the proton-transfer reaction: OH− + D2 → HOD + D−

The reaction mechanisms of OH⁻ + D₂ → HOD + D⁻ were first revealed by theory, based on an accurate full-dimensional PES.

Approximating Periodic Potential Energy Surfaces with Sparse Trigonometric Interpolation

The potential energy surface (PES) describes the energy of a chemical system as a function of its geometry and is a fundamental concept in computational chemistry. A PES provides much useful information about the system, including the structures and energies of various stationary points, such as local minima, maxima, and transition states. Construction of full-dimensional PESs for molecules with more than 10 atoms is computationally expensive and often not feasible. Previous work in our group used sparse interpolation with polynomial basis functions to construct a surrogate reduced-dimensional PESs along chemically significant reaction coordinates, such as bond lengths, bond angles, and torsion angles. However, polynomial interpolation does not preserve the periodicity of the PES gradient with respect to angular components of geometry, such as torsion angles, which can lead to nonphysical phenomena. In this work, we construct a surrogate PES using trigonometric basis functions, for a system where the selected reaction coordinates all correspond to the torsion angles, resulting in a periodically repeating PES. We find that a trigonometric interpolation basis not only guarantees periodicity of the gradient but also results in slightly lower approximation error than polynomial interpolation.

Imaging pair-correlated reaction cross sections in F + CH3D(νb = 0, 1) → CH2D(ν4 = 1) + HF(ν)

The title reactions were studied in a crossed-beam experiment at collisional energies (Ec) from 0.5 to 4.7 kcal mol-1. The νb (ν4) vibrational mode denotes the bending (umbrella) motion of the CH3D reactant (CH2D product). Using a time-sliced, velocity-map imaging technique, we extracted the state-specific, pair-correlated integral and differential cross sections. As with other isotopically analogous ground-state reactions, an inverted vibrational population of the HF coproduct was observed. Both the step-like excitation function near the threshold and the oscillatory forward-backward peakings in the Ec-evolution of the two dominant pair-correlated angular distributions at lower Ec suggest a resonance-mediated, time-delay mechanism. As Ec increases, the angular distribution of the HF(ν = 2) product evolves into a smooth and broad swath in the backward hemisphere, indicative of a direct rebound mechanism. One quantum excitation of the bending modes of CH3D(νb = 1) promotes the reaction rate by two- to three-fold up to Ec = 2.1 kcal mol-1. Broadly speaking, all major findings are qualitatively in line with previous results in the reactions of the F atom with other isotopologues. However, the rainbow feature recently observed in the CH2D(00) + HF(ν = 3) product channel is entirely absent. A possible rationale is put forward, which reinforces the previous reactive rainbow conjecture and calls for future theoretical investigations.

Molecular Dynamics Simulations on Relaxed Reduced-Dimensional Potential Energy Surfaces

Molecular dynamics (MD) simulations with full-dimensional potential energy surfaces (PESs) obtained from high-level ab initio calculations are frequently used to model reaction dynamics of small molecules (i.e., molecules with up to 10 atoms). Construction of full-dimensional PESs for larger molecules is, however, not feasible since the number of ab initio calculations required grows rapidly with the increase of dimension. Only a small number of coordinates are often essential for describing the reactivity of even very large systems, and reduced-dimensional PESs with these coordinates can be built for reaction dynamics studies. While analytical methods based on transition-state theory framework are well established for analyzing the reduced-dimensional PESs, MD simulation algorithms capable of generating trajectories on such surfaces are more rare. In this work, we present a new MD implementation that utilizes the relaxed reduced-dimensional PES for standard microcanonical (NVE) and canonical (NVT) MD simulations. The method is applied to the pyramidal inversion of a NH₃ molecule. The results from the MD simulations on a reduced, three-dimensional PES are validated against the ab initio MD simulations, as well as MD simulations on full-dimensional PES and experimental data.

A fragmented, permutationally invariant polynomial approach for potential energy surfaces of large molecules: Application to N-methyl acetamide

We describe and apply a method to extend permutationally invariant polynomial (PIP) potential energy surface (PES) fitting to molecules with more than 10 atoms. The method creates a compact basis of PIPs as the union of PIPs obtained from fragments of the molecule. An application is reported for trans-N-methyl acetamide, where B3LYP/cc-pVDZ electronic energies and gradients are used to develop a full-dimensional potential for this prototype peptide molecule. The performance of several fragmented bases is verified against a benchmark PES using all (66) Morse variables. The method appears feasible for much larger molecules.

A critical comparison of neural network potentials for molecular reaction dynamics with exact permutation symmetry

Several symmetry strategies have been compared in fitting full dimensional accurate potentials for reactive systems based on a neural network approach.

Modelling potential energy surfaces for small clusters using Shepard interpolation with Gaussian-form nodal functions

A new interpolation method based on Gaussian functions to reliably generate potential energy surfaces.

Anomalous kinetics of the reaction between OH and HO2on an accurate triplet state potential energy surface

The quasi-classical trajectory predicts the rate coefficient of the OH + HO₂→ H₂O + O₂reaction based on a full dimensional accurate PIP-NN PES, which is fit to 108 000 points calculated at the CCSD(T)-F12a/AVTZ level.