Giant In-Plane Flexoelectricity and Radial Polarization in Janus IV-VI Monolayers and Nanotubes.

Nanotubes have established a new paradigm in nanoscience because of their atomically thin geometries and intriguing properties. However, because of their typical metastability compared to their 2D and 3D counterparts, it is still fundamentally challenging to synthesize nanotubes with controlled size. New strategies have been suggested for synthesizing nanotubes with a controlled geometry. One of these is considering Janus 2D layers, which can self-roll to form a nanotube. Herein, we study 412 nanotubes (along the armchair and zigzag directions) based on 36 Janus IV-VI compounds using density functional theory (DFT) calculations. By investigating the energy-radius relationship using structural models and Bayesian predictions, the most stable nanotubes show negative strain energies and radii below 20 Å, where curvature effects can play a significant role. The band structures show that the selected nanotubes exhibit sizable band gaps and size-dependent electronic properties. More strikingly, the flexoelectricity along the in-plane directions and radial directions in these nanotubes is significantly larger than that in other nanotubes and their 2D counterparts. This work opens up an avenue of structure-property relationships of Janus IV-VI nanotubes and demonstrates giant flexoelectricity in these nanotubes for future electronic and energy applications.

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.

Understanding the electronic and structural effects in ORR intermediate binding on anion-substituted zirconia surfaces.

For oxygen reduction reaction (ORR), the surface adsorption energies of O* and OH* intermediates are key descriptors for catalytic activity. In this work, we investigate anion-substituted zirconia catalyst surfaces and determine that adsorption energies of O* and OH* intermediates is governed by both structural and electronic effects. When the adsorption energies are not influenced by the structural effects of the catalyst surface, they exhibit a linear correlation with integrated crystal orbital Hamiltonian population (ICOHP) of the adsorbate-surface bond. The influence of structural effects, due to re-optimisation slab geometry after adsorption of intermediate species, leads to stronger adsorption of intermediates. Our calculations show that there is a change in the bond order to accommodate the incoming adsorbate species which leads to stronger adsorption when both structural and electronic effects influence the adsorption phenomena. The insights into the catalyst-adsorbate interactions can guide the design of future ORR catalysts.

Accelerated chemical science with AI.

In light of the pressing need for practical materials and molecular solutions to renewable energy and health problems, to name just two examples, one wonders how to accelerate research and development in the chemical sciences, so as to address the time it takes to bring materials from initial discovery to commercialization. Artificial intelligence (AI)-based techniques, in particular, are having a transformative and accelerating impact on many if not most, technological domains. To shed light on these questions, the authors and participants gathered in person for the ASLLA Symposium on the theme of 'Accelerated Chemical Science with AI' at Gangneung, Republic of Korea. We present the findings, ideas, comments, and often contentious opinions expressed during four panel discussions related to the respective general topics: 'Data', 'New applications', 'Machine learning algorithms', and 'Education'. All discussions were recorded, transcribed into text using Open AI's Whisper, and summarized using LG AI Research's EXAONE LLM, followed by revision by all authors. For the broader benefit of current researchers, educators in higher education, and academic bodies such as associations, publishers, librarians, and companies, we provide chemistry-specific recommendations and summarize the resulting conclusions.

High-throughput Compositional Screening of PdxTi1-xHy and PdxNb1-xHy Hydrides for CO2 Reduction.

Electrochemical experiments and theoretical calculations have shown that Pd-based metal hydrides can perform well for the CO2 reduction reaction (CO2RR). Our previous work on doped-PdH showed that doping Ti and Nb into PdH can improve the CO2RR activity, suggesting that the Pd alloy hydrides with better performance are likely to be found in the PdxTi1-xHy and PdxNb1-xHy phase space. However, the vast compositional and structural space with different alloy hydride compositions and surface adsorbates, makes it intractable to screen out the stable and active PdxM1-xHy catalysts using density functional theory calculations. Herein, an active learning cluster expansion (ALCE) surrogate model equipped with Monte Carlo simulated annealing (MCSA), a CO* binding energy filter and a kinetic model are used to identify promising PdxTi1-xHy and PdxNb1-xHy catalysts with high stability and superior activity. Using our approach, we identify 24 stable and active candidates of PdxTi1-xHy and 5 active candidates of PdxNb1-xHy. Among these candidates, the Pd0.23Ti0.77H, Pd0.19Ti0.81H0.94, and Pd0.17Nb0.83H0.25 are predicted to display current densities of approximately 5.1, 5.1 and 4.6 µA cm-2 at -0.5 V overpotential, respectively, which are significantly higher than that of PdH at 3.7 µA cm-2.

MultiXC-QM9: Large dataset of molecular and reaction energies from multi-level quantum chemical methods.

Well curated extensive datasets have helped spur intense molecular machine learning (ML) method development activities over the last few years, encouraging nonchemists to be part of the effort as well. QM9 dataset is one of the benchmark databases for small molecules with molecular energies based on B3LYP functional. G4MP2 based energies of these molecules were published later. To enable a wide variety of ML tasks like transfer learning, delta learning, multitask learning, etc. with QM9 molecules, in this article, we introduce a new dataset with QM9 molecule energies estimated with 76 different DFT functionals and three different basis sets (228 energy numbers for each molecule). We additionally enumerated all possible A ↔ B monomolecular interconversions within the QM9 dataset and provided the reaction energies based on these 76 functionals, and basis sets. Lastly, we also provide the bond changes for all the 162 million reactions with the dataset to enable structure- and bond-based reaction energy prediction tools based on ML.

Graph neural network interatomic potential ensembles with calibrated aleatoric and epistemic uncertainty on energy and forces.

Inexpensive machine learning (ML) potentials are increasingly being used to speed up structural optimization and molecular dynamics simulations of materials by iteratively predicting and applying interatomic forces. In these settings, it is crucial to detect when predictions are unreliable to avoid wrong or misleading results. Here, we present a complete framework for training and recalibrating graph neural network ensemble models to produce accurate predictions of energy and forces with calibrated uncertainty estimates. The proposed method considers both epistemic and aleatoric uncertainty and the total uncertainties are recalibrated post hoc using a nonlinear scaling function to achieve good calibration on previously unseen data, without loss of predictive accuracy. The method is demonstrated and evaluated on two challenging, publicly available datasets, ANI-1x (Smith et al. J. Chem. Phys., 2018, 148, 241733.) and Transition1x (Schreiner et al. Sci. Data, 2022, 9, 779.), both containing diverse conformations far from equilibrium. A detailed analysis of the predictive performance and uncertainty calibration is provided. In all experiments, the proposed method achieved low prediction error and good uncertainty calibration, with predicted uncertainty correlating with expected error, on energy and forces. To the best of our knowledge, the method presented in this paper is the first to consider a complete framework for obtaining calibrated epistemic and aleatoric uncertainty predictions on both energy and forces in ML potentials.

Structural and electronic properties of double wall MoSTe nanotubes.

Janus nanotubes originating from rolling up asymmetric dichalcogenide monolayers have shown unique properties compared to their 2D and 3D counterparts. Most of the work on Janus nanotubes is focused on single-wall (SW) tubes. In this work, we have investigated the structural and electronic properties of double wall (DW) MoSTe nanotubes using Density Functional Theory (DFT). The most stable DW, corresponding to a minimum of the strain energy, is formed by combining 16- and 24-unit cells for the inner and outer tubes. This DW configuration shows a slightly smaller inner diameter than the SW tube, which was formed by 18-unit cells due to the intra-wall interaction. The investigation of the band gaps of 2D structures under strain and SW/DW nanotubes revealed that the curvature of the nanotube and the strain induced when forming the tube are the two primary factors enabling the band gap tuning. Moreover, we found that the band gaps of the DW MoSTe tubes close, compared to the SWs, generating tubes with a metallic-like behavior. This property makes DW MoSTe nanotubes promising for electrochemical applications.

Neural network potentials for accelerated metadynamics of oxygen reduction kinetics at Au-water interfaces.

The application of ab initio molecular dynamics (AIMD) for the explicit modeling of reactions at solid-liquid interfaces in electrochemical energy conversion systems like batteries and fuel cells can provide new understandings towards reaction mechanisms. However, its prohibitive computational cost severely restricts the time- and length-scales of AIMD. Equivariant graph neural network (GNN) based accurate surrogate potentials can accelerate the speed of performing molecular dynamics after learning on representative structures in a data efficient manner. In this study, we combined uncertainty-aware GNN potentials and enhanced sampling to investigate the reactive process of the oxygen reduction reaction (ORR) at an Au(100)-water interface. By using a well-established active learning framework based on CUR matrix decomposition, we can evenly sample equilibrium structures from MD simulations and non-equilibrium reaction intermediates that are rarely visited during the reaction. The trained GNNs have shown exceptional performance in terms of force prediction accuracy, the ability to reproduce structural properties, and low uncertainties when performing MD and metadynamics simulations. Furthermore, the collective variables employed in this work enabled the automatic search of reaction pathways and provide a detailed understanding towards the ORR reaction mechanism on Au(100). Our simulations identified the associative reaction mechanism without the presence of *O and a low reaction barrier of 0.3 eV, which is in agreement with experimental findings. The methodology employed in this study can pave the way for modeling complex chemical reactions at electrochemical interfaces with an explicit solvent under ambient conditions.

Cation-Coordinated Inner-Sphere CO2 Electroreduction at Au-Water Interfaces.

Electrochemical CO2 reduction reaction (CO2RR) is a promising technology for the clean energy economy. Numerous efforts have been devoted to enhancing the mechanistic understanding of CO2RR from both experimental and theoretical studies. Electrolyte ions are critical for the CO2RR; however, the role of alkali metal cations is highly controversial, and a complete free energy diagram of CO2RR at Au-water interfaces is still missing. Here, we provide a systematic mechanism study toward CO2RR via ab initio molecular dynamics simulations integrated with the slow-growth sampling (SG-AIMD) method. By using the SG-AIMD approach, we demonstrate that CO2RR is facile at the inner-sphere interface in the presence of K cations, which promote the CO2 activation with the free energy barrier of only 0.66 eV. Furthermore, the competitive hydrogen evolution reaction (HER) is inhibited by the interfacial cations with the induced kinetic blockage effect, where the rate-limiting Volmer step shows a much higher energy barrier (1.27 eV). Eventually, a comprehensive free energy diagram including both kinetics and thermodynamics of the CO2RR to CO and the HER at the electrochemical interface is derived, which illustrates the critical role of cations on the overall performance of CO2 electroreduction by facilitating CO2 adsorption while suppressing the hydrogen evolution at the same time.

Oxidation and de-alloying of PtMn particle models: a computational investigation.

We present a computational study of the energetics and mechanisms of oxidation of Pt-Mn systems. We use slab models and simulate the oxidation process over the most stable (111) facet at a given Pt2Mn composition to make the problem computationally affordable, and combine Density-Functional Theory (DFT) with neural network potentials and metadynamics simulations to accelerate the mechanistic search. We find, first, that Mn has a strong tendency to alloy with Pt. This tendency is optimally realized when Pt and Mn are mixed in the bulk, but, at a composition in which the Mn content is high enough such as for Pt2Mn, Mn atoms will also be found in the surface outmost layer. These surface Mn atoms can dissociate O2 and generate MnOx species, transforming the surface-alloyed Mn atoms into MnOx surface oxide structures supported on a metallic framework in which one or more vacancy sites are simultaneously created. The thus-formed vacancies promote the successive steps of the oxidation process: the vacancy sites can be filled by surface oxygen atoms, which can then interact with Mn atoms in deeper layers, or subsurface Mn atoms can intercalate into interstitial sites. Both these steps facilitate the extraction of further bulk Mn atoms into MnOx oxide surface structures, and thus the progress of the oxidation process, with typical rate-determining energy barriers in the range 0.9-1.0 eV.

Transition1x - a dataset for building generalizable reactive machine learning potentials.

Machine Learning (ML) models have, in contrast to their usefulness in molecular dynamics studies, had limited success as surrogate potentials for reaction barrier search. This is primarily because available datasets for training ML models on small molecular systems almost exclusively contain configurations at or near equilibrium. In this work, we present the dataset Transition1x containing 9.6 million Density Functional Theory (DFT) calculations of forces and energies of molecular configurations on and around reaction pathways at the ωB97x/6-31 G(d) level of theory. The data was generated by running Nudged Elastic Band (NEB) with DFT on 10k organic reactions of various types while saving intermediate calculations. We train equivariant graph message-passing neural network models on Transition1x and cross-validate on the popular ANI1x and QM9 datasets. We show that ML models cannot learn features in transition state regions solely by training on hitherto popular benchmark datasets. Transition1x is a new challenging benchmark that will provide an important step towards developing next-generation ML force fields that also work far away from equilibrium configurations and reactive systems.

Cheap Turns Superior: A Linear Regression-Based Correction Method to Reaction Energy from the DFT.

Workflows to predict chemical reaction networks based on density functional theory (DFT) are prone to systematic errors in reaction energy due to the extensive use of cheap DFT exchange-correlation functionals to limit computational cost. Recently, machine learning-based models are increasingly applied to mitigate this problem. However, machine learning models require systems similar to trained data, and the models often perform poorly for out-of-distribution systems. Here, we present a simple bond-based correction method that improves the accuracy of DFT-derived reaction energies. It is based on linear regression, and the correction terms for each bond are derived from reactions among the QM9 data set. We demonstrate the effectiveness of this method with three DFT functionals in three different rungs of Jacob's ladder. The simple correction method is effective for all rungs but especially so for the cheapest PBE functional. Finally, we applied the correction method to a few reactions with molecules significantly different from those in the QM9 data set that was used to fit the linear regression model. Once corrected by this method, we found that the DFT reaction energies for such out-of-distribution reactions are within 0.05 eV of the G4MP2 method.

Transition-Metal-Free Barium Hydride Mediates Dinitrogen Fixation and Ammonia Synthesis.

Transition-metal-mediated dinitrogen fixation has been intensively investigated. The employment of main group elements for this vital reaction has recently sparked interest because of new dinitrogen reaction chemistry. We report ammonia synthesis via a chemical looping process mediated by a transition-metal-free barium hydride (BaH2 ). Experimental and computational studies reveal that the introduction of hydrogen vacancies is essential for creating multiple coordinatively unsaturated Ba sites for N2 activation. The adjacent lattice hydridic hydrogen (H- ) then undergoes both reductive elimination and reductive protonation to convert N2 to NHx . The ammonia production rate supports this hydride-vacancy mechanism via a chemical looping route that far exceeds that of a catalytic process. The BaH2 -mediated chemical looping process has prospects in future technologies for ammonia synthesis using transition-metal-free materials.

Structure and energetics of liquid water-hydroxyl layers on Pt(111).

The interactions between liquid water and hydroxyl species on Pt(111) surfaces have been intensely investigated due to their importance to fuel cell electrocatalysis. Here we present a molecular dynamics study of their structure and energetics using an ensemble of neural network potentials, which allow us to obtain unprecedented statistical sampling. We first study the energetics of hydroxyl formation, where we find a near-linear adsorption energy profile, which exhibits a soft and gradual increase in the differential adsorption energy at high hydroxyl coverages. This is strikingly different from the predictions of the conventional bilayer model, which displays a kink at 1/3ML OH coverage indicating a sizeable jump in differential adsorption energy, but within the statistical uncertainty of previously reported ab initio molecular dynamics studies. We then analyze the structure of the interface, where we provide evidence for the water-OH/Pt(111) interface being hydrophobic at high hydroxyl coverages. We furthermore explain the observed adsorption energetics by analyzing the hydrogen bonding in the water-hydroxyl adlayers, where we argue that the increase in differential adsorption energy at high OH coverage can be explained by a reduction in the number of hydrogen bonds from the adsorbed water molecules to the hydroxyls.

Phase-Field Investigation of Lithium Electrodeposition at Different Applied Overpotentials and Operating Temperatures.

Li metal is an exciting anode for high-energy Li-ion batteries and other future battery technologies due to its high energy density and low redox potential. Despite their high promise, the commercialization of Li-metal-based batteries has been hampered due to the formation of dendrites that lead to mechanical instability, energy loss, and eventual internal short circuits. In recent years, the mechanism of dendrite formation and the strategies to suppress their growth have been studied intensely. However, the effect of applied overpotential and operating temperature on dendrite formation and their growth rate remains to be fully understood. Here, we elucidate the correlation between the applied overpotential and operating temperature to the dendrite height and tortuosity of the Li-metal surface during electrodeposition using phase-field model simulations. We identify an optimal operating temperature of a half-cell consisting of a Li metal anode and 1 M LiPF6 in EC/DMC (1/1), which increases gradually as the magnitude of the overpotential increases. The investigation reveals that the temperature dependence identified in the simulations and experiments often disagree because they are primarily conducted under galvanostatic and potentiostatic conditions, respectively. The temperature increase under potentiostatic conditions increases the induced current while it decreases the induced overpotential under galvanostatic conditions. Therefore, the analysis and comparison of temperature-dependent characteristics must be carried out with care.

Metal-Doped PdH(111) Catalysts for CO2 Reduction.

PdH-based catalysts hold promise for both CO2 reduction to CO and the hydrogen evolution reaction. Density functional theory is used to systematically screen for stability, activity, and selectivity of transition metal dopants in PdH. The transition metal elements Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Ru, Rh, Ag, Cd, Hf, Ta, W, and Re are doped into PdH(111) surface with six different doping configurations: single, dimer, triangle, parallelogram, island, and overlayer. We find that several dopants, such as Ti and Nb, have excellent predicted catalytic activity and CO2 selectivity compared to the pure PdH hydride. In addition, they display good stability due to their negative doping formation energy. The improved performance can be assigned to reaction intermediates forming two bonds consisting of one C-Metal and one O-Metal bond on the PdH surface, which break the scaling relations of intermediates, and thus have stronger HOCO* binding facilitating CO2 activation.

Dual Role of Mo6 S8 in Polysulfide Conversion and Shuttle for Mg-S Batteries.

Magnesium-Sulfur batteries are one of most appealing options among the post-lithium battery systems due to its potentially high energy density, safe and sustainable electrode materials. The major practical challenges are originated from the soluble magnesium polysulfide intermediates and their shuttling between the electrodes, which cause high overpotentials, low sulfur utilization, and poor Coulombic efficiency. Herein, a functional Mo6 S8 modified separator is designed to effectively address these issues. Both the experimental results and density functional theory calculations show that the electrochemically active Mo6 S8 layer has a superior adsorption capability of polysulfides and simultaneously acts as a mediator to accelerate the polysulfide conversion kinetics. Remarkably, the magnesium-sulfur cell assembled with the functional separator delivers a high specific energy density (942.9 mA h g-1 in the 1st cycle) and can be cycled at 0.2 C for 200 cycles with a Coulombic efficiency of 96%. This work demonstrates a new design concept toward high-performance metal-sulfur batteries.

Artificial Intelligence Applied to Battery Research: Hype or Reality?

This is a critical review of artificial intelligence/machine learning (AI/ML) methods applied to battery research. It aims at providing a comprehensive, authoritative, and critical, yet easily understandable, review of general interest to the battery community. It addresses the concepts, approaches, tools, outcomes, and challenges of using AI/ML as an accelerator for the design and optimization of the next generation of batteries─a current hot topic. It intends to create both accessibility of these tools to the chemistry and electrochemical energy sciences communities and completeness in terms of the different battery R&D aspects covered.

Is the water/Pt(111) interface ordered at room temperature?

The structure of the water/Pt(111) interface has been a subject of debate over the past decades. Here, we report the results of a room temperature molecular dynamics study based on neural network potentials, which allow us to access long time scale simulations while retaining ab initio accuracy. We find that the water/Pt(111) interface is characterized by a double layer composed of a primary, strongly bound adsorption layer with a coverage of ∼0.15 ML, which is coupled to a secondary, weakly bound adsorption layer with a coverage of ∼0.58 ML. By studying the order of the primary adsorption layer, we find that there is an effective repulsion between the adsorbed water molecules, which gives rise to a dynamically changing, semi-ordered interfacial structure, where the water molecules in the primary adsorption layer are distributed homogeneously across the interface, forming frequent hydrogen bonds to water molecules in the secondary adsorption layer. We further show that these conclusions are beyond the time scales accessible to ab initio molecular dynamics.