An adaptive design approach for defects distribution modeling in materials from first-principle calculations

Lourenço, Maicon Pierre; Anastácio, Alexandre dos Santos; Rosa, A. L.; Frauenheim, Thomas; Silva, Mauricio Chagas da

doi:10.1007/s00894-020-04438-w

Cited by 15 publications

(24 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Active learning can also be applied in a problem specific context, where further savings in training data are possible, because accurate predictions are especially relevant only for subsets of the full configurational space. This includes configurational problems that are local in nature, such as structure relaxation 42,43 and transition state determination [44][45][46] as well as more ambitious tasks such as molecular dynamics [47][48][49][50][51][52] , chemical reaction networks 53,54 and finally global structure search 3,[55][56][57][58][59][60][61] , including our recently proposed Global Optimization with First-principles Energy Expressions (GOFEE) structure search method 62 , which will be further detailed in this work, along with some additional improvements to the population and local convergence of structures. For global structure search an active learning approach can utilize the fact that accuracy is increasingly important for lower energy structures, such that higher energy structures can be screened based on only rough energy predictions.…”

Section: Introductionmentioning

confidence: 99%

Global optimization of atomistic structure enhanced by machine learning

Bisbo,

Hammer

2020

Preprint

View full text Add to dashboard Cite

Global Optimization with First-principles Energy Expressions (GOFEE) is an efficient method for identifying low energy structures in computationally expensive energy landscapes such as the ones described by density functional theory (DFT), van der Waals-enabled DFT, or even methods beyond DFT. GOFEE relies on a machine learned surrogate model of energies and forces, trained on-thefly, to explore configuration space, eliminating the need for expensive relaxations of all candidate structures using first-principles methods. In this paper we elaborate on the importance of the use of a Gaussian kernel with two length scales in the Gaussian Process Regression (GPR) surrogate model. We further explore the role of the use in GOFEE of the lower confidence bound for relaxation and selection of candidate structures. In addition, we present two improvements to the method: 1) the population generation now relies on a clustering of all low-energy structures evaluated with DFT, with the lowest energy member of each cluster making up the population. 2) the very final relaxations in well-sampled basins of the energy landscape, "the final exploitation steps", are now performed as continued relaxation paths within the first-principles method, to allow for arbitrarily fine relaxations of the best structures, independently of the predictive resolution of the surrogate model. The versatility of the GOFEE method is demonstrated by applying it to identify the lowenergy structures of gas-phase fullerene-type 24-atom carbon clusters and of dome-shaped 18-atom carbon clusters supported on Ir(111).

show abstract

Section: Introductionmentioning

confidence: 99%

Global optimization of atomistic structure enhanced by machine learning

Bisbo,

Hammer

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The classical Warburg semi-infinite diffusion depends upon the surface coverage, as shown in equations 18 and 19. 67 Z diff = σt -1/2 (1 -j) (18) In equation 18, Z diff is the impedance of the Warburg element, σ is the Warburg coefficient t is the applied frequency and j is equal to -1 1/2 . The Warburg coefficient for a particular situation which the diffusion of reduced In equation 19, R is the universal gas constant, T the absolute temperature, A s the active electrode area, F the Faraday's constant, C the concentration of the electrochemical species, D the diffusion coefficient and θ the surface coverage of the electrode.…”

Section: Electrochemical Studiesmentioning

confidence: 99%

“…16 In this context, the understanding of the interactions between the corrosion inhibitor and the metallic surface at a microscopic level has great importance for corrosion inhibition researchers, since one knows which molecular or electronic properties could be related to corrosion inhibition, making possible to correlate the calculated molecular properties with experimental data, such as ε inh , for proposing chemical modifications of the molecule to synthesize new molecules with better corrosion inhibition performance. 17 In recent years, the so-called computational chemistry has been present in research not only on the subject of corrosion inhibitors, but also in the chemistry of materials such as the use of machine learning tools to study failures in metallic structures 18 or even using the ab initio method based on density functional theory (DFT) methodology, which uses quantum mechanics to determine molecular properties to evaluate macroscopic phenomena such as adsorption of inhibitor at the electrode/solution interface. 19 Thus, the correlation between experimental results and theoretical calculations allows to deepen the physical and chemical understandings of the phenomenon of corrosion inhibition.…”

Section: Introductionmentioning

confidence: 99%

Inhibition of Copper Corrosion in Acid Medium by Imidazole-Based Compounds: Electrochemical and Molecular Approaches

Costa¹,

Almeida-Neto²,

Marinho³

et al. 2023

J. Braz. Chem. Soc.

View full text Add to dashboard Cite

Chemically modified imidazole molecules: 4-(1H-imidazol-1-yl)aniline, 4-(1H-imidazol1-yl)benzaldehyde, 4-(1H-imidazol-1-yl)phenol and (4-(1H-imidazol-1-yl)phenyl)methanol were investigated as inhibitors of the copper (Cu0) corrosion in 0.5 mol L-1 H2SO4 medium. The electrochemical corrosion data were obtained by monitoring open circuit potential, linear potentiodynamic polarization and electrochemical impedance spectroscopy techniques, while the computational density functional theory (DFT) method was applied to correlate the electronic properties of the molecules with corrosion inhibition efficiencies. All molecules had inhibited the Cu corrosion, and the inhibition values lied between 80 and 94%. A good correlation between the inhibition efficiencies values and Gibbs adsorption energy was found, showing that the more negative Gibbs energy, better interaction between the corrosion inhibitor with the Cu0 surface, diminishing its corrosion in 0.5 mol L-1 H2SO4 medium. The DFT calculations showed significative differences in electronic and reactivity properties of imidazole and other molecules. The higher corrosion inhibition of imidazole derivates could be explained by electrophilic characteristic of these molecules, since there are empty molecular orbitals spread over mainly in benzene rings that make a metal-ligand charge transfer, receiving electronic density from the copper surface by backbonding, according to the electronic Fukui functions and the potential charge distribution considering the map of electrostatic potential.

show abstract

“…While the underlying problem is similar to before, i.e., finding suitable materials from large configuration spaces, here the constraints on suitability are based on electrical properties such as the band gap or the photoelectric conversion coefficient (PCE). ML has been employed in a number of studies [240], [241], [242], [243], [244], [245], [246], [247], [248], [249], [250], [251], [252], [253], [254], [236], [255], [256], [257], [258], [259], [260], [261], [262], [263], [264], [265], [266], [267].…”

Section: Discovery Of Novel Stable Materialsmentioning

confidence: 99%

A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry

Fiedler¹,

Shah²,

Bußmann³

et al. 2021

Preprint

View full text Add to dashboard Cite

With the growth of computational resources, the scope of electronic structure simulations has increased greatly. Artificial intelligence and robust data analysis hold the promise to accelerate large-scale simulations and their analysis to hitherto unattainable scales. Machine learning is a rapidly growing field for the processing of such complex datasets. It has recently gained traction in the domain of electronic structure simulations, where density functional theory takes the prominent role of the most widely used electronic structure method. Thus, DFT calculations represent one of the largest loads on academic high-performance computing systems across the world. Accelerating these with machine learning can reduce the resources required and enables simulations of larger systems. Hence, the combination of density functional theory and machine learning has the potential to rapidly advance electronic structure applications such as in-silico materials discovery and the search for new chemical reaction pathways. We provide the theoretical background of both density functional theory and machine learning on a generally accessible level. This serves as the basis of our comprehensive review including research articles up to December 2020 in chemistry and materials science that employ machine-learning techniques. In our analysis, we categorize the body of research into main threads and extract impactful results. We conclude our review with an outlook on exciting research directions in terms of a citation analysis.

show abstract

An adaptive design approach for defects distribution modeling in materials from first-principle calculations

Cited by 15 publications

References 53 publications

Global optimization of atomistic structure enhanced by machine learning

Global optimization of atomistic structure enhanced by machine learning

Inhibition of Copper Corrosion in Acid Medium by Imidazole-Based Compounds: Electrochemical and Molecular Approaches

A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry

Contact Info

Product

Resources

About