Diffusion-based generative AI for exploring transition states from 2D molecular graphs

The exploration of transition state (TS) geometries is crucial for elucidating chemical reaction mechanisms and modeling their kinetics. Recently, machine learning (ML) models have shown remarkable performance for prediction of TS geometries. However, they require 3D conformations of reactants and products often with their appropriate orientations as input, which demands substantial efforts and computational cost. Here, we propose a generative approach based on the stochastic diffusion method, namely TSDiff, for prediction of TS geometries just from 2D molecular graphs. TSDiff outperforms the existing ML models with 3D geometries in terms of both accuracy and efficiency. Moreover, it enables to sample various TS conformations, because it learns the distribution of TS geometries for diverse reactions in training. Thus, TSDiff finds more favorable reaction pathways with lower barrier heights than those in the reference database. These results demonstrate that TSDiff shows promising potential for an efficient and reliable TS exploration.

Simultaneously improving reaction coverage and computational cost in automated reaction prediction tasks

Automated reaction prediction has the potential to elucidate complex reaction networks for applications ranging from combustion to materials degradation, but computational cost and inconsistent reaction coverage are still obstacles to exploring deep reaction networks. Here we show that cost can be reduced and reaction coverage can be increased simultaneously by relatively straightforward modifications of the reaction enumeration, geometry initialization and transition state convergence algorithms that are common to many prediction methodologies. These components are implemented in the context of yet another reaction program (YARP), our reaction prediction package with which we report reaction discovery benchmarks for organic single-step reactions, thermal degradation of a γ-ketohydroperoxide, and competing ring-closures in a large organic molecule. Compared with recent benchmarks, YARP (re)discovers both established and unreported reaction pathways and products while simultaneously reducing the cost of reaction characterization by nearly 100-fold and increasing convergence of transition states. This combination of ultra-low cost and high reaction coverage creates opportunities to explore the reactivity of larger systems and more complex reaction networks for applications such as chemical degradation, where computational cost is a bottleneck.

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

It is an ultimate goal in chemistry to predict reaction without recourse to experiment. Reaction prediction is not just the reaction rate determination of known reactions but, more broadly, the reaction exploration to identify new reaction routes. This review briefly overviews the theory on chemical reaction and the current methods for computing/estimating reaction rate and exploring reaction space. We particularly focus on the atomistic simulation methods for reaction exploration, which are benefited significantly by recently emerged machine learning potentials. We elaborate the stochastic surface walking global pathway sampling based on the global neural network (SSW-NN) potential, developed in our group since 2013, which can explore complex reactions systems unbiasedly and automatedly. Two examples, molecular reaction and heterogeneous catalytic reactions, are presented to illustrate the current status for reaction prediction using SSW-NN.

Water Radical Cations in the Gas Phase: Methods and Mechanisms of Formation, Structure and Chemical Properties

Water radical cations, (H₂O)_n^+•, are of great research interest in both fundamental and applied sciences. Fundamental studies of water radical reactions are important to better understand the mechanisms of natural processes, such as proton transfer in aqueous solutions, the formation of hydrogen bonds and DNA damage, as well as for the discovery of new gas-phase reactions and products. In applied science, the interest in water radicals is prompted by their potential in radiobiology and as a source of primary ions for selective and sensitive chemical ionization. However, in contrast to protonated water clusters, (H₂O)_nH⁺, which are relatively easy to generate and isolate in experiments, the generation and isolation of radical water clusters, (H₂O)_n^+•, is tremendously difficult due to their ultra-high reactivity. This review focuses on the current knowledge and unknowns regarding (H₂O)_n^+• species, including the methods and mechanisms of their formation, structure and chemical properties.

Solvation energies of the proton in methanol revisited and temperature effects

Various functionals assessing solvation free energies and enthalpies of the proton in methanol.

Ferrate(vi) initiated oxidative degradation mechanisms clarified by DFT calculations: a case for sulfamethoxazole

Ferrate(vi) is an efficient and environmentally friendly oxidant for the degradation of organic micropollutants. However, the related mechanism for the degradation is ambiguous and can hardly be elucidated empirically due to the rapid oxidation process and unstable intermediates for experimental trapping. Herein we performed density function theory (DFT) calculations to unveil the mechanism of ferrate(vi)-mediated degradation, taking sulfamethoxazole as a model compound. The results show that nucleophilic attack (rather than electrophilic attack) of HFeO₄^- on the isoxazole moiety of sulfamethoxazole initiates the subsequent degradations, and ferrate(vi) rather than the water molecule provides O atoms for the oxidation of the nitroso group and isoxazole moiety. Electron delocalization from the Fe atom to the isoxazole moiety is crucial for the ring-opening of isoxazole, and organometallic intermediates suggested previously are not the necessary ones in the oxidation of sulfamethoxazole by HFeO₄^-. Thus, this study has theoretically clarified the ferrate(vi) oxidation mechanisms for a representative sulfonamide, which were also partially corroborated by the intermediates and products observed in the previous experimental studies for phosphite and tryptophan. This study provides an exemplification on the application of quantum chemical calculations to clarify the degradation pathways of organic micropollutants, which is important for the prediction of degradation products needed in their engineering design.

Ab Initio Study of Ionized Water Radical Cation (H2O)8+ in Combination with the Particle Swarm Optimization Method

The structures of cationic water clusters (H₂O)₈⁺ have been globally explored by the particle swarm optimization method in combination with quantum chemical calculations. Geometry optimization and vibrational analysis for the 15 most interesting clusters were computed at the MP2/aug-cc-pVDZ level and infrared spectrum calculation at MPW1K/6-311++G** level. Special attention was paid to the relationships between their configurations and energies. Both MP2 and B3LYP-D3 calculations revealed that the cage-like structure is the most stable, which is different from a five-membered ring lowest energy structure but agrees well with a cage-like structure in the literature. Furthermore, our obtained cage-like structure is more stable by 0.87 and 1.23 kcal/mol than the previously reported structures at MP2 and B3LYP-D3 levels, respectively. Interestingly, on the basis of their relative Gibbs free energies and the temperature dependence of populations, the cage-like structure predominates only at very low temperatures, and the most dominating species transforms into a newfound four-membered ring structure from 100 to 400 K, which can contribute greatly to the experimental infrared spectrum. By topological analysis and reduced density gradient analysis, we also investigated the structural characteristics and bonding strengths of these water cluster radical cations.

Study of Potential Energy Surfaces towards Global Reaction Route Mapping

The potential energy surface (PES) is just a theoretical construct based on the Born-Oppenheimer approximation, but it underlies various phenomena, including molecular vibrations, collisional ionizations, and chemical reactions. This account describes how a new idea for global reaction route mapping (GRRM), which had seemed to be impossible for chemical systems with more than three atoms, was born and has been developed during the course of the study of the PES. GRRM has pioneered new fields of chemistry. Furthermore, techniques for GRRM are still developing, and GRRM is further extending its application to various areas of chemistry and chemical physics.

Stepwise formation of H3O(+)(H2O)n in an ion drift tube: Empirical effective temperature of association/dissociation reaction equilibrium in an electric field

We measured equilibrium constants for H3O(+)(H2O)n-1 + H2O↔H3O(+)(H2O)n (n = 4-9) reactions taking place in an ion drift tube with various applied electric fields at gas temperatures of 238-330 K. The zero-field reaction equilibrium constants were determined by extrapolation of those obtained at non-zero electric fields. From the zero-field reaction equilibrium constants, the standard enthalpy and entropy changes, ΔHn,n-1 (0) and ΔSn,n-1 (0), of stepwise association for n = 4-8 were derived and were in reasonable agreement with those measured in previous studies. We also examined the electric field dependence of the reaction equilibrium constants at non-zero electric fields for n = 4-8. An effective temperature for the reaction equilibrium constants at non-zero electric field was empirically obtained using a parameter describing the electric field dependence of the reaction equilibrium constants. Furthermore, the size dependence of the parameter was thought to reflect the evolution of the hydrogen-bond structure of H3O(+)(H2O)n with the cluster size. The reflection of structural information in the electric field dependence of the reaction equilibria is particularly noteworthy.

Structures and spectroscopy of protonated ammonia clusters at different temperatures

Protonated ammonia clusters are all Eigen structures and the first solvation shell of the related ammonium ion core is saturated by four ammonia molecules.

Disentangling the Contribution of Multiple Isomers to the Infrared Spectrum of the Protonated Water Heptamer

We use infrared/infrared double-resonance population labeling (IR(2)MS(2)) spectroscopy in the spectral region of the free and hydrogen-bonded OH stretching fundamentals (2880-3850 cm(-1)) to identify the number and to isolate the vibrational signatures of individual isomers contributing to the gas-phase IR spectra of the cryogenically cooled protonated water clusters H(+)(H2O)n·H2/D2 with n = 7-10. For n = 7, four isomers are identified and assigned. Surprisingly, the IR(2)MS(2) spectra of the protonated water octa-, nona-, and decamer show no evidence for multiple isomers. The present spectra support the prediction that the quasi-2D to 3D structural transition occurs in between n = 8 and 9 in the cold cluster regime. However, the same models have difficulty explaining the remarkable size dependence of the isomer population reported here.

Structures and relative stabilities of ammonia clusters at different temperatures: DFT vs. ab initio

The global minimum energy structures of (NH₃)_n=2–10are pointed out for the first time at a given temperature.

Reaction sampling and reactivity prediction using the stochastic surface walking method

The new theoretical method demonstrates the ability of automated reaction sampling and activity prediction for complex organic reactions.

Structural evolution and solvation of the OH radical in ionized water radical cations (H2O)n+, n = 5–8

Structural evolution of ionized water radical cations (H₂O)_n⁺, n = 5–8, is studied by ab intio methods.

Characterization of a solvent-separated ion-radical pair in cationized water networks: infrared photodissociation and Ar-attachment experiments for water cluster radical cations (H2O)n+(n = 3-8)

We present infrared spectra of nominal water cluster radical cations (H(2)O)(n)(+) (n = 3-8), or to be precise, ion-radical complexes H(+)(H(2)O)(n-1)(OH), with and without an Ar tag. These clusters are closely related to the ionizing radiation-induced processes in water and are a good model to characterize solvation structures of the ion-radical pair. The spectra of Ar-tagged species show narrower bandwidths relative to those of the bare clusters due to the reduced internal energy via an Ar-attachment. The observed spectra are analyzed by comparing with those of the similar system, H(+)(H(2)O)(n), and calculated ones. We find that the observed spectra are attributable to ion-radical-separated motifs in n ≥ 5, as reported in the previous study (Mizuse et al. Chem. Sci.2011, 2, 868-876). Beyond the structural trends found in the previous study, we characterize isomeric structures and determine the number of water molecules between the charged site and the OH radical in each cluster size. In all of the characterized cluster structures (n = 5-8), the most favorable position of OH radical is the next neighbor of the charged site (H(3)O(+) or H(5)O(2)(+)). The positions and cluster structures are governed by the balance among the hydrogen-bonding abilities of the charged site, H(2)O, and OH radical. These findings on the ionized water networks lead to understanding of the detailed processes of ionizing radiation-initiated reactions in liquid water and aqueous solutions.

Multilevel X-Pol: a fragment-based method with mixed quantum mechanical representations of different fragments

The explicit polarization (X-Pol) method is a fragment-based quantum mechanical model, in which a macromolecular system or other large or complex system in solution is partitioned into monomeric fragments. The present study extends the original X-Pol method, where all fragments are treated using the same electronic structure theory, to multilevel representations, called multilevel X-Pol, in which different electronic structure methods are used to describe different fragments. The multilevel X-Pol method has been implemented into a locally modified version of Gaussian 09. A key ingredient that is used to couple interfragment electrostatic interactions at different levels of theory is the use of the response density for the post-self-consistent-field energy. (The response density is also called the generalized density.) The method is useful for treating fragments in a small region of the system such as a solute molecule or the substrate and amino acids in the active site of an enzyme with a high-level theory, and the fragments in the rest of the system by a lower-level and computationally more efficient method. Multilevel X-Pol is illustrated here by applications to hydrogen bonding complexes in which one fragment is treated with the hybrid M06 density functional, Møller-Plesset perturbation theory, or coupled cluster theory, and the other fragments are treated by Hartree-Fock theory or the B3LYP or M06 hybrid density functionals.