Effects of Segregation on the Catalytic Properties of AgAuCuPdPt High-Entropy Alloy for CO Reduction Reaction.

Multicomponent alloys are promising catalysts for diverse chemical conversions, owing to the ability to tune their vast compositional space to maximize catalytic activity and product selectivity. However, elemental segregation, whereby the surface or grain boundaries of the material are enriched in a few elements, is a physically observed phenomenon in such alloys. Such segregation alters not only the composition but also the kinds of catalytically active sites present at the surface. Thus, elemental segregation, which can be achieved via various processing techniques, can be used as an additional knob in searching for alloy compositions that are both active and selective for a target chemical conversion. We demonstrate this using molecular simulations, machine learning, and Bayesian optimization to search for both random solid solution and "segregated" AgAuCuPdPt alloy compositions that are potentially active and selective for CO reduction reaction (CORR). Finally, we validate our findings by computing the reaction-free energy landscape for the CORR on the optimal alloy compositions via density functional theory calculations.

Systematic Identification of Atom-Centered Symmetry Functions for the Development of Neural Network Potentials.

Neural network potentials are emerging as promising classical force fields that can enable long-time and large-length scale simulations at close to ab initio accuracies. They learn the underlying potential energy surface by mapping the Cartesian coordinates of atoms to system energies using elemental neural networks. To ensure invariance with respect to system translation, rotation, and atom index permutations, in the Behler-Parrinnello type of neural network potential (BP-NNP), the Cartesian coordinates of atoms are transformed into "structural fingerprints" using atom-centered symmetry functions (ACSFs). Development of an accurate BP-NNP for any chemical system critically relies on the choice of these ACSFs. In this work, we have proposed a systematic framework for the identification of an optimal set of ACSFs for any target system, which not only considers the diverse atomic environments present in the training dataset but also inter-ACSF correlations. Our method is applicable to different kinds of ACSFs and across diverse chemical systems. We demonstrate this by building accurate BP-NNPs for water and Cu2S systems.

Applied machine learning for predicting the lanthanide-ligand binding affinities.

Binding affinities of metal-ligand complexes are central to a multitude of applications like drug design, chelation therapy, designing reagents for solvent extraction etc. While state-of-the-art molecular modelling approaches are usually employed to gather structural and chemical insights about the metal complexation with ligands, their computational cost and the limited ability to predict metal-ligand stability constants with reasonable accuracy, renders them impractical to screen large chemical spaces. In this context, leveraging vast amounts of experimental data to learn the metal-binding affinities of ligands becomes a promising alternative. Here, we develop a machine learning framework for predicting binding affinities (logK1) of lanthanide cations with several structurally diverse molecular ligands. Six supervised machine learning algorithms-Random Forest (RF), k-Nearest Neighbours (KNN), Support Vector Machines (SVM), Kernel Ridge Regression (KRR), Multi Layered Perceptrons (MLP) and Adaptive Boosting (AdaBoost)-were trained on a dataset comprising thousands of experimental values of logK1 and validated in an external 10-folds cross-validation procedure. This was followed by a thorough feature engineering and feature importance analysis to identify the molecular, metallic and solvent features most relevant to binding affinity prediction, along with an evaluation of performance metrics against the dimensionality of feature space. Having demonstrated the excellent predictive ability of our framework, we utilized the best performing AdaBoost model to predict the logK1 values of lanthanide cations with nearly 71 million compounds present in the PubChem database. Our methodology opens up an opportunity for significantly accelerating screening and design of ligands for various targeted applications, from vast chemical spaces.

Hydration structure and water exchange kinetics at xenotime-water interfaces: implications for rare earth minerals separation.

Hydration of surface ions gives rise to structural heterogeneity and variable exchange kinetics of water at complex mineral-water interfaces. Here, we employ ab initio molecular dynamics (AIMD) simulations and water adsorption calorimetry to examine the aqueous interfaces of xenotime, a phosphate mineral that contains predominantly Y3+ and heavy rare earth elements. Consistent with natural crystal morphology, xenotime is predicted to have a tetragonal prismatic shape, dominated by the {100} surface. Hydration of this surface induces multilayer interfacial water structures with distinct OH orientations, which agrees with recent crystal truncation rod measurements. The exchange kinetics between two adjacent water layers exhibits a wide range of underlying timescales (5-180 picoseconds), dictated by ion-water electrostatics. Adsorption of a bidentate hydroxamate ligand reveals that {100} xenotime surface can only accommodate monodentate coordination with water exchange kinetics strongly depending on specific ligand orientation, prompting us to reconsider traditional strategies for selective separation of rare-earth minerals.

Interfacial structure in the liquid-liquid extraction of rare earth elements by phosphoric acid ligands: a molecular dynamics study.

Solvent extraction (SX), wherein two immiscible liquids, one containing the extractant molecules and the other containing the solute to be extracted are brought in contact to effect the phase transfer of the solute, underpins metal extraction and recovery processes. The interfacial region is of utmost importance in the SX process, since besides thermodynamics, the physical and chemical heterogeneity at the interface governs the kinetics of the process. Yet, a fundamental understanding of this heterogeneity and its implications for the extraction mechanism are currently lacking. We use molecular dynamics (MD) simulations to study the liquid-liquid interface under conditions relevant to the SX of Rare Earth Elements (REEs) by a phosphoric acid ligand. Simulations revealed that the extractant molecules and varying amounts of acid and metal ions partitioned to the interface. The presence of these species had a significant effect on the interfacial thickness, hydrogen bond life times and orientations of the water molecules at the interface. Deprotonation of the ligands was essential for the adsorption of the metal ions at the interface, with these ions forming a number of different complexes at the interface involving one to three extractant molecules and four to eight water molecules. Although the interface itself was rough, no obvious 'finger-like' water protrusions penetrating the organic phase were seen in our simulations. While the results of our work help us gain fundamental insights into the sequence of events leading to the formation of a variety of interfacial complexes, they also emphasize the need to carry out a more detailed atomic level study to understand the full mechanism of extraction of REEs from the aqueous to organic phases by phosphoric acid ligands.

Bulk and surface DFT investigations of inorganic halide perovskites screened using machine learning and materials property databases.

In the recent past, there has been proliferation in high-throughput density functional theory and data-driven explorations of materials motivated by a need to reduce physical testing and costly computations for materials discovery. This has, in conjunction with the development of open-access materials property databases, encouraged accelerated and more streamlined discovery and screening of technologically relevant materials. In this work, we report our results on the screening and DFT studies of one such class of materials, i.e. ABX3 inorganic halide perovskites (A, B and X representing the monovalent, divalent and halide ions respectively) using a coupled machine-learning (ML) and density functional theory (DFT) approach. Utilizing the support vector machine algorithm, we predict the formability of 454 inorganic halide compounds in the perovskite phase. Compounds with a formation probability P≥ 0.8 are further checked for thermodynamic stability in at least one of these three open materials databases - Materials Project (MP), Automatic FLOW for Materials Discovery (AFLOW) and Open Quantum Materials Database (OQMD). The shortlisted candidate perovskites are then considered for DFT computations. Taking input geometries from MP's structure predictor, the optimized lattice parameters and computed band gaps (BG) for all screened compounds are compared with predictions across all databases. Subsequently, detailed studies on low index surfaces are presented for two halide perovksites - RbSnCl3 and RbSnBr3- having band-gaps in the favourable range for photovoltaics (PV). Different possible (100), (110) and (111) surface terminations are investigated for each of these compositions and the atomic relaxations, surface energies and electronic band structures are reported for each termination. To the best of our knowledge, no surface studies have been reported in the literature for any of the halide perovskites present in our database. These studies, therefore, are an important step towards gaining a fundamental understanding of the interfacial properties of perovskites, which can help facilitate further breakthroughs in the PV technology.

A comparative study of surface energies and water adsorption on Ce-bastnäsite, La-bastnäsite, and calcite via density functional theory and water adsorption calorimetry.

Bastnäsite, a fluoro-carbonate mineral, is the single largest mineral source of light rare earth elements (REE), La, Ce and Nd. Enhancing the efficiency of separation of the mineral from gangue through froth flotation is the first step towards meeting an ever increasing demand for REE. To design and evaluate collector molecules that selectively bind to bastnäsite, a fundamental understanding of the structure and surface properties of bastnäsite is essential. In our earlier work (J. Phys. Chem. C, 2016, 120, 16767), we carried out an extensive study of the structure, surface stability and water adsorption energies of La-bastnäsite. In this work, we make a comparative study of the surface properties of Ce-bastnäsite, La-bastnäsite, and calcite using a combination of density functional theory (DFT) and water adsorption calorimetry. Spin polarized DFT+U calculations show that the exchange interaction between the electrons in Ce 4f orbitals is negligible and that these orbitals do not participate in bonding with the oxygen atom of the adsorbed water molecule. In agreement with calorimetry, DFT calculations predict larger surface energies and stronger water adsorption energies on Ce-bastnäsite than on La-bastnäsite. The order of stabilities for stoichiometric surfaces is as follows: [101[combining macron]0] > [101[combining macron]1] > [101[combining macron]2] > [0001] > [112[combining macron]2] > [101[combining macron]4] and the most favorable adsorption sites for water molecules are the same as for La-bastnäsite. In agreement with water adsorption calorimetry, at low coverage water molecules are strongly stabilized via coordination to the surface Ce3+ ions, whereas at higher coverage they are adsorbed less strongly via hydrogen bonding interaction with the surface anions. Due to similar water adsorption energies on bastnäsite [101[combining macron]1] and calcite [101[combining macron]4] surfaces, the design of collector molecules that selectively bind to bastnäsite over calcite must exploit the structural differences in the predominantly exposed facets of these minerals.

Thermodynamic, Spectroscopic, and Computational Studies of f-Element Complexation by N-Hydroxyethyl-diethylenetriamine-N,N',N″,N″-tetraacetic Acid.

Potentiometric and spectroscopic techniques were combined with DFT calculations to probe the coordination environment and determine thermodynamic features of trivalent f-element complexation by N-hydroxyethyl-diethylenetriamine-N,N',N″,N″-tetraacetic acid, HEDTTA. Ligand protonation constants and lanthanide stability constants were determined using potentiometry. Five protonation constants were accessible in I = 2.0 M (H+/Na+)ClO4. UV-vis spectroscopy was used to determine stability constants for Nd3+ and Am3+ complexation with HEDTTA. Luminescence spectroscopy indicates two water molecules in the inner coordination sphere of the Eu/HEDTTA complex, suggesting HEDTTA is heptadentate. Luminescence data was supported by DFT calculations, which demonstrate that substitution of the acetate pendant arm by a N-hydroxyethyl group weakens the metal-nitrogen bond. This bond elongation is reflected in HEDTTA's ability to differentiate trivalent actinides from trivalent lanthanides. The trans-lanthanide Ln/HEDTTA complex stability trend is analogous to Ln/DTPA complexation; however, the loss of one chelate ring resulting from structural substitution weakens the complexation by ∼3 orders of magnitude. Successful separation of trivalent americium from trivalent lanthanides was demonstrated when HEDTTA was utilized as aqueous holdback complexant in a liquid-liquid system. Time-dependent extraction studies for HEDTTA were compared to diethylenetriamine-N,N,N',N″,N″-pentaacetic acid (DTPA) and N-hydroxyethyl-ethylenediamine-N,N',N'-triacetic acid (HEDTA). The results indicate substantially enhanced phase-transfer kinetic rates for mixtures containing HEDTTA.