Integrated workflows and interfaces for data-driven semi-empirical electronic structure calculations.

Modern software engineering of electronic structure codes has seen a paradigm shift from monolithic workflows toward object-based modularity. Software objectivity allows for greater flexibility in the application of electronic structure calculations, with particular benefits when integrated with approaches for data-driven analysis. Here, we discuss different approaches to create deep modular interfaces that connect big-data workflows and electronic structure codes and explore the diversity of use cases that they can enable. We present two such interface approaches for the semi-empirical electronic structure package, DFTB+. In one case, DFTB+ is applied as a library and provides data to an external workflow; in another, DFTB+receives data via external bindings and processes the information subsequently within an internal workflow. We provide a general framework to enable data exchange workflows for embedding new machine-learning-based Hamiltonians within DFTB+ or enabling deep integration of DFTB+ in multiscale embedding workflows. These modular interfaces demonstrate opportunities in emergent software and workflows to accelerate scientific discovery by harnessing existing software capabilities.

The application of QM/MM simulations in heterogeneous catalysis.

The QM/MM simulation method is provenly efficient for the simulation of biological systems, where an interplay of extensive environment and delicate local interactions drives a process of interest through a funnel on a complex energy landscape. Recent advances in quantum chemistry and force-field methods present opportunities for the adoption of QM/MM to simulate heterogeneous catalytic processes, and their related systems, where similar intricacies exist on the energy landscape. Herein, the fundamental theoretical considerations for performing QM/MM simulations, and the practical considerations for setting up QM/MM simulations of catalytic systems, are introduced; then, areas of heterogeneous catalysis are explored where QM/MM methods have been most fruitfully applied. The discussion includes simulations performed for adsorption processes in solvent at metallic interfaces, reaction mechanisms within zeolitic systems, nanoparticles, and defect chemistry within ionic solids. We conclude with a perspective on the current state of the field and areas where future opportunities for development and application exist.

SuSMoST: Surface Science Modeling and Simulation Toolkit.

We present to the scientific community the Surface Science Modeling and Simulation Toolkit (SuSMoST), which includes a number of utilities and implementations of statistical physics algorithms and models. With SuSMoST it is possible to predict or explain the structure and thermodynamic properties of adsorption layers. SuSMoST automatically builds formal graph and tensor-network models based on atomic description of adsorption complexes and helps to do ab initio calculations of interactions between adsorbed species. Using methods of various nature SuSMoST generates representative samples of adsorption layers and computes its thermodynamic quantities such as mean energy, coverage, density, and heat capacity. From these data one can plot phase diagrams of adsorption systems, assess thermal stability of self-assembled structures, simulate thermal desorption spectra, and so forth.

A systematic computational study of the structure crossover and coordination number distribution of metallic nanoparticles.

In this study, we identified stable configurations for three nanoparticle structure motifs (icosahedral, decahedral and cuboctahedral) of eight transition metals (Cu, Ag, Au, Pd, Ni, Rh, Ir, and Pt) ranging in size from 140 to 3000 atoms. We made simple yet precise analytical approximations of the energy of the stable configurations as a function of nanoparticle size and calculated the structure crossover sizes from these approximations. We then analyzed the surface structure of the nanoparticles in terms of the distribution of the coordination numbers and active sites. We found that low-coordinated atoms are most preferable for cuboctahedral forms and for lighter metals - Cu, Ni and Rh. Compared to other considered metals, gold nanoparticles exhibited unique features as follows: the least amount of low-coordinated atoms, the largest fraction of (111) faces on its surface and a concave reconstruction of five-fold vertices.

2-Carboxyethylgermanium Sesquioxide as A Promising Anode Material for Li-Ion Batteries.

Various forms of germanium and germanium-containing compounds and materials are actively investigated as energy-intensive alternatives to graphite as the anode of lithium-ion batteries. The most accessible form-germanium dioxide-has the structure of a 3D polymer, which accounts for its rapid destruction during cycling, and requires the development of further approaches to the production of nanomaterials and various composites based on it. For the first time, we propose here the strategy of using 2-carboxyethylgermanium sesquioxide ([O1.5 GeCH2 CH2 CO2 H]n , 2-CEGS), in lieu of GeO2 , as a promising, energy-intensive, and stable new source system for building lithium-ion anodes. Due to the presence of the organic substituent, the formed polymer has a 1D or a 2D space organization, which facilitates the reversible penetration of lithium into its structure. 2-CEGS is common and commercially available, completely safe and non-toxic, insoluble in organic solvents (which is important for battery use) but soluble in water (which is convenient for manufacturing diverse materials from it). This paper reports the preparation of micro- (flower-shaped agglomerates of ≈1 µm thick plates) and nanoformed (needle-shaped nanoparticles of ≈500×(50-80) nm) 2-CEGS using methods commonly available in laboratories and industry such as vacuum and freeze-drying of aqueous solutions of 2-CEGS. Lithium half-cell anodes based on 2-CEGS show a capacity of ≈400 mAh g-1 for microforms and up to ≈700 mAh g-1 for nanoforms, which is almost two times higher than the maximal theoretical capacity of graphite. These anodes are stable during the cycling at various rates. The results of DFT simulations suggest that Li atoms form the stable Li2 O with the oxygen atoms of 2-CEGS, and actual charge-discharge cycles involve deoxygenated GeC3 H5 molecules. Thus, C3 chains loosen the anode structure compared to pure Ge, improving its ability to accommodate Li ions.