Search

Ni‐CeO2 nanoparticles (NPs) are promising nanocatalysts for water splitting and water gas shift reactions due to the ability of ceria to temporarily donate oxygen to the catalytic reaction and accept oxygen after the reaction is completed. Therefore, elucidating how different properties of the Ni‐Ceria NPs relate to the activity and selectivity of the catalytic reaction, is of crucial importance for the development of novel catalysts. In this work the active learning (AL) method based on machine learning regression and its uncertainty is used for the global optimization of Ce(4‐x)NixO(8‐x) (x = 1, 2, 3) nanoparticles, employing density functional theory calculations. Additionally, further investigation of the NPs by mass‐scaled parallel‐tempering Born‐Oppenheimer molecular dynamics resulted in the same putative global minimum structures found by AL, demonstrating the robustness of our AL search to learn from small datasets and assist in the global optimization of complex electronic structure systems. [ABSTRACT FROM AUTHOR]

In this work, we explore the possibility of using computationally inexpensive electronic structure methods, such as semiempirical and DFTB calculations, for the search of the global minimum (GM) structure of chemical systems. The basic prerequisite that these inexpensive methods will need to fulfill is that their lowest energy structures can be used as starting point for a subsequent local optimization at a benchmark level that will yield its GM. If this is possible, one could bypass the global optimization at the expensive method, which is currently impossible except for very small molecules. Specifically, we test our methods with clusters of second row elements including systems of several bonding types, such as alkali, metal, and covalent clusters. The results reveal that the DFTB3 method yields reasonable results and is a potential candidate for this type of applications. Even though the DFTB2 approach using standard parameters is proven to yield poor results, we show that a re-parametrization of only its repulsive part is enough to achieve excellent results, even when applied to larger systems outside the training set. [ABSTRACT FROM AUTHOR]

Genetic algorithms (GAs) are stochastic global search methods inspired by biological evolution. They have been used extensively in chemistry and materials science coupled with theoretical methods, ranging from force‐fields to high‐throughput first‐principles methods. The methodology allows an accurate and automated structural determination for molecules, atomic clusters, nanoparticles, and solid surfaces, fundamental to understanding chemical processes in catalysis and environmental sciences, for instance. In this work, we propose a new genetic algorithm software, GAMaterial, implemented in Python3.x, that performs global searches to elucidate the structures of atomic clusters, doped clusters or materials and atomic clusters on surfaces. For all these applications, it is possible to accelerate the GA search by using machine learning (ML), the ML@GA method, to build subsequent populations. Results for ML@GA applied for the dopant distributions in atomic clusters are presented. The GAMaterial software was applied for the automatic structural search for the Ti6O12 cluster, doping Al in Si11 (4Al@Si11) and Na10 supported on graphene (Na10@graphene), where DFTB calculations were used to sample the complex search surfaces with reasonably low computational cost. Finally, the global search by GA of the Mo8C4 cluster was considered, where DFT calculations were made with the deMon2k code, which is interfaced with GAMaterial. [ABSTRACT FROM AUTHOR]

Active learning (AL) has been successfully applied in materials science for the global optimization of clusters and defects in materials. Many important chemistry problems require the structural elucidation of molecules as a first step to the mechanistic elucidation of complex heterogeneous catalysis phenomena. Theoretical methods coupled with global optimization algorithms are successfully used for this purpose. However, it is challenging to find the global minimum structure together with the proper electronic spin multiplicity. In this work, we present an AL implementation for global optimization of atomic clusters where the spin multiplicity is considered in the search loop (SM@AL). The method was implemented in the QMLMaterial software, interfaced with the deMon2k program to perform local structure optimizations. In this work, we present applications of SM@AL for the global optimization, in terms of molecular structure and electronic spin of 3Al@Si11, where Si11 is doped by 3 Al, and Mo4C2 with spin multiplicities 2, 4 and 6 and 1, 3, 5, 7, 9 and 11, respectively. [ABSTRACT FROM AUTHOR]

In catalysis, an accurate structural elucidation of molecules, atomic clusters, nanoparticles and solid surfaces is required to understand chemical processes. Therefore, an efficient and automatic structure determination for these systems is of great benefit since it requires a global search within huge chemical spaces. In this work, we propose a new active learning (AL) method intended for atomic clusters that uses different supervised machine learning techniques and their uncertainties to decide the promising non-observed (virtual) structures to be evaluated from quantum calculations. The method was developed for structural elucidation of (I) clusters where all atomic coordinates are allowed to change in a continuous chemical space and (II) doped ones where the atoms exchange in a sufficiently rigid structure, thus, a discrete search space where all cluster descriptors are known. Particularly for case I, a genetic algorithm operator was used to create the unknown virtual structures (and then their descriptors) from the observed ones to improve the performance of the AL search. The proposed AL was applied to the global optimization of heteronuclear (Al4Si7 and 4Al@Si11) and homonuclear (Na20) clusters using self-consistent charge density-functional tight-binding (SCC-DFTB), where a new repulsion parameter was developed to reproduce isomers evaluated from high-level calculations. The performance of the Gaussian process and artificial neural network algorithms was evaluated together with several uncertainty quantification methods: from Gaussian process, K-fold cross-validation and nonparametric bootstrap (BS) resampling. The efficiency of the AL was compared to conventional global optimization methods, such as random search and a genetic algorithm. The results show that the AL can find efficiently the global minimum of atomic clusters. [ABSTRACT FROM AUTHOR]

The study of materials in the nanoscale regime has important applications for catalytic reactions, the energy industry and medicine. We performed exploratory density functional theory calculations for molybdenum disulphide (MoS2) and calcium carbonate (CaCO3) nanoparticles (NPs), the former being developed as a hydrogenation and coke-prevention catalyst and the latter representing the catalytic support in some reservoirs. Utilizing Born–Oppenheimer Molecular Dynamics with reduced masses as a global optimization method, we found Mo8S16 and Mo16S32 NPs with lower-lying energies than those of locally optimized crystal geometries. Our results suggest that MoS2 NPs prefer a tetragonal lattice arrangement which is in agreement with previous studies of smaller MoS2 NPs. It remains to be seen how prenucleation MoS2 clusters of the hexagonal phase are formed. The CaCO3 NPs showed small energy differences between vastly different conformations. Therefore, in order to capture only their supporting effects in reservoirs, one should consider keeping a substantial part of the models fixed. [ABSTRACT FROM AUTHOR]