Quantum chemical modeling of the structure and properties of SnO2 nanoclusters

By the method of density functional theory with exchange-correlation functional B3LYP and basis set 3‑21G (d), the structural and energy characteristics have been considered of the molecular models of SnO2 nanoclusters of different size and composition with the number of Sn atoms from 1 to 10. Incompletely coordinated surface tin atoms were terminated by hydroxyl groups. It has been shown that the Sn–O bond length in nanoclusters does not depend on the cluster size and on the coordination number of Sn atoms, but is determined by the coordination type of neighboring oxygen atoms. Namely, the bond length Sn–O(3) (@ 2.10 Å) is greater than that of Sn–O (2) (@ 1.98 Å). The calculated values of Sn–O (3) bond lengths agree well with the experimental ones for crystalline SnO 2 samples (2.05 Å). The theoretically calculated width of the energy gap decreases naturally with increasing cluster size (from 6.14 to 3.46 eV) and approaches the experimental value of the band gap of the SnO 2 crystal (3.6 eV). The principle of additivity was used to analyze the energy characteristics of the considered models and to estimate the corresponding values for a cassiterite crystal. According to this principle, a molecular model can be represented as a set of atoms or atomic groups of several types that differ in the coordination environment and, therefore, make different contributions to the total energy of the system. The calculated value of the atomization energy for SnO2 is 1661 kJ/mol and corresponds satisfactorily to the experimentally measured specific atomization energy of crystalline SnO2 (1381 kJ/mol). It has been shown that a satisfactory reproduction of the experimental characteristics of crystalline tin dioxide is possible when using clusters containing at least 10 state atoms, for example, (SnO2)10×14H2O.

The prediction of molecule atomization energy using neural network and extreme gradient boosting

Machine Learning is an artificial intelligence system, where the system has the ability to learn automatically from experience without being explicitly programmed. The learning process from Machine Learning starts from observing the data and then looking at the pattern of the data. The main purpose of this process is to make computers learn automatically. In this study, we will use Machine Learning to predict molecular atomization energy. From various methods in Machine Learning, we use two methods namely Neural Network and Extreme Gradient Boosting. Both methods have several parameters that must be adjusted so that the predicted value of the atomization energy of the molecule has the lowest possible error. We are trying to find the right parameter values for both methods. For the neural network method, it is quite difficult to find the right parameter value because it takes a long time to train the model of the neural network to find out whether the model is good or bad, while for the Extreme Gradient Boosting method the time needed to train the model is shorter, so it is quite easy to find the right parameter values for the model. This study also looked at the effects of the modification on the dataset with the output transformation of normalization and standardization then removing molecules containing Br atoms and changing the entry in the Coulomb matrix to 0 if the distance between atoms in the molecule exceeds 2 angstrom.

Applying simulation results of high-boron compounds of structure at the atomic level to estimate their chemical hardness

Determining the macrocharacteristics of materials based on the results of ab initio calculations is one of the most relevant and promising areas of research. One of the most important performance characteristics of the material is its hardness. The presented approach to determining the chemical Vickers hardness of substances based on using ab initio calculated values of atomization energy and molar volume atomic clusters, which are elements of the structure of the studied compounds. Clusters of boron, aluminum and magnesium borides of different atomic structure, which are obtained using simulation modeling of their evolution, are considered. The results of quantum chemical calculations of the values of atomization energy and molar volume of the considered fragments, obtained using the Gaussian'03 software package in the framework of the theory of electron density functional in the B3LYP / STO-3G approximation, are presented. The hardness of materials, structural elements of which are tested atomic clusters, obtained by the developed approach are presented. The calculated hardness is compared with its values determined by both experimental and other theoretical methods. The comparison showed a high correlation of the obtained results with the experimental data already at the cluster size equal to 12—25 atoms. Analysis of the results of applying the proposed approach to various modifications of boron and some boron-containing compounds showed that quantum-chemical calculations of atomic energy and molar volume values within the cluster model provide the ability to establish reliable estimates of the hardness of existing compounds of this class. The developed approach, together with simulation modeling of the evolution of hypothetical phases, can also be applied to predict their hardness. Keywords: boron, borides, cluster model, Vickers hardness.