Computational Materials Science最新文献

Metallic glass matrix composites (MGMCs) represent a promising avenue for enhancing the ductility of monolithic metallic glass. These composites utilize a secondary crystalline phase to aid in the delocalization of strain. This work seeks to understand the mechanisms underlying strain delocalization in MGMCs to guide further advancements in this class of material. Employing a mesoscale shear transformation zone (STZ) dynamics model, we investigate how variation in dendritic microstructural sizes and spacings impact the shear banding behaviors of MGMCs subjected to uniaxial tensile loading. Statistical analysis of shear banding characteristics reveals that the competition of shear band nucleation and propagation rates can encourage strain delocalization in MGMCs. The introduction of a crystalline dendritic structure into the amorphous matrix increases the number of shear band nucleation events while reducing shear band propagation rates. Furthermore, reducing dendrite sizes leads to greater strain delocalization among more shear bands and delays the onset of run-away shear bands, resulting in lower overall shear band growth rates. Therefore, this study sheds light on the crucial role of dendritic microstructural sizes in influencing shear banding characteristics and strain delocalization in MGMCs, offering valuable insights to inform the design and development of advanced materials with superior mechanical properties.

In this study, machine learning (ML) models were successfully employed to predict the short-term electrochemical corrosion behavior of high-entropy alloys (HEAs) based on their chemical compositions. Considering the vast compositional space of HEAs, which restricts the development of corrosion-resistant HEAs, and the lack of non-destructive methods to qualitatively assess their corrosion resistance, this work represents a significant advancement in the field. The “three-step” method was applied to select the optimal feature set from 38 features, and six ML regression models were trained and compared. The eXtreme Gradient Boosting (XGBoost) and Gradient Boosting Decision Tree (GBDT) algorithms demonstrated the highest predictive accuracy (R² = 81.02 % and 84.64 %, respectively) among the six algorithms. The model’s robust generalization capabilities were confirmed through validation on an additional dataset. Moreover, the interpretability of the model was enhanced by employing two analysis methods, which revealed that pH as an environmental factor, electronegativity difference and average electronegativity as empirical parameters, and the concentrations of Cr and Cu as compositional parameters have the most significant impact on the corrosion resistance of HEAs. The proposed methodology and framework have the potential to optimize alloy composition, facilitating the design and development of new corrosion-resistant HEAs.

Atomistic modeling of carbon nanosprings under tension is performed. Two types of nanosprings are considered, the $l$-helicene (C${}_{l^2}$H${}_l$)${}_{\infty }$ in the form of a helicoid and the $l$-kekulene (C${}_{l^2-1}$H${}_{l+1}$)${}_{\infty }$ in the form of a spiral graphene nanoribbon. The molecules exhibit large elastic (reversible) deformations of 200%–500%. Surprisingly, $l$-helicene with $l>3$ and $l$-kekulene with $l>2$ are stretched inhomogeneously, so that the domains with small and large tensile deformation are observed within certain range of relative elongation. Moreover, within this range of elongation, the stretching occurs at constant tensile force. This behavior is explained by calculating the potential energy of homogeneously stretched nanosprings as a function of elongation. These curves have a non-convex shape within the range of relative elongation where the inhomogeneous deformation occurs. When the constraint of homogeneous tension is not applied, the system does not follow the non-convex dependence of energy on elongation, but rather the tangent line to this curve. Since energy is a linear function of elongation, the force is constant in the region of inhomogeneous deformation. The results presented demonstrate the possibility of creating graphene nanosprings that deform over a wide range of strain with a constant tensile force.

The dynamic response of shock wave impact on single crystal aluminium and lightweight multicomponent alloy Al-Cu-Li-Mg is simulated by using the combination of Ab initio Molecular Dynamics (AIMD) and Multi-Scale Shock Technique (MSST), with the analysis carried out at the atomic/electronic levels. The simulation is verified by comparing the particle velocity of single crystal obtained in this work with the data in literature. The shock compression process not only involves the migration of atoms, but also is related to electronic transition. Two stages could be found in the shock compression process: oscillatory compression of the crystal cell and oscillatory migration of the atoms. The crystal structure of the multicomponent alloy could be disordered even at low shock speed, due to the difference in the ability to migrate between different kinds of atoms. As the sample is shock-compressed, the contribution proportion of crystal orbitals shows a sharp decrease for D orbital, while it increases significantly for S orbital and P orbital. The electron structure shows a quicker response to the shock wave compression process than the crystal structure. The orbital contribution from P orbital of the crystal is mainly due to the P orbital of Al atoms, while the orbital contribution from D orbital of the crystal is mainly due to the D orbital of Cu atoms. Total Density of States (TDOS) is mainly contributed by the Projected Density of State (PDOS) of Cu atoms in the occupied state of energy levels, while it is close to the PDOS of Al atoms in the non-occupied state of energy levels.

Semiconducting half-Heusler alloys are potential candidates for thermoelectric generators operational at high temperatures. In this work, the stability, electronic, and thermoelectric properties of 18 valence electron TiXPb (X$=$Ni, Pd, Pt) compounds are investigated using density functional theory and semi-classical Boltzmann transport theory. The compounds are both thermodynamically and dynamically stable. We find them to be semiconductors with indirect band gaps lying between 0.32−0.64 eV. Our calculations show that from thermoelectric performance perspective electrons exhibit better transport properties than holes. A combination of large power factor and low lattice thermal conductivity results in $zT>1$ in all the materials. Our calculations predict that amongst the three compounds, TiPtPb have a maximum value of $zT$ for both electrons and holes. In this material our calculation yields a maximum $zT$ of 2.22 at 900 K for n-type doping at a doping concentration of 9.46 × $10^{20}$ ${\mathrm{cm}}^{-3}$ and 1.80 at 900 K for p-type doping at a doping concentration of 4.51 × $10^{20}$ ${\mathrm{cm}}^{-3}$.

In this study, an innovative approach is explored that combines Density Functional Tight Binding (DFTB) with Computer Vision (CV) techniques to analyze the electronic structure and enhance the photocatalytic capabilities of carbon-doped titanium oxide nanoparticles (C-doped TiO${}_2$ NPs). The findings reveal that C doping, in levels ranging from 0.1% to 0.6% progressively alters the material’s electronic structure and photocatalytic activity. Specifically, the energy gap decreases significantly from 3.160 eV for undoped TiO${}_2$ to 0.565 eV at 0.6% doping, with no substantial changes observed beyond 0.6% doping. A notable correlation between increased C doping and a rise in total energy suggests a complex interaction between C incorporation and the energetic as well as structural dynamics of TiO${}_2$ NPs. This interaction could enhance photocatalytic efficiency, especially under visible light, by reducing the band gap through C doping. The use of CV methodologies improves computational efficiency and predictive accuracy. These techniques validate the DFTB results and accelerate the material discovery process via machine learning models. The ability of CV methods to accurately predict the properties of C-doped TiO${}_2$ NPs across various doping levels, combined with their computational advantages, represents a significant advancement in materials science.

In recent years, there has been increasing interest in two-dimensional (2D) thermoelectric materials owing to their potential to achieve enhanced thermoelectric conversion efficiency, driven by quantum size effects. Here, we employed density functional theory (DFT) to investigate the thermoelectric and mechanisms properties of a novel monolayer NbOX₂ (X=Cl, Br, I). We identified a broad and flat band below the Fermi level, mainly contributed by the d_z² orbitals of the niobium atoms. This flat band will result in a relatively large effective mass (m*) and manifests as a peak in the density of states (DOS), known as a Van Hove singularity. Notably, the monolayer NbOX₂ exhibits outstanding electronic transport properties along the x-direction and demonstrates reduced lattice thermal conductivity (κ_l) along the y-direction. The anisotropic κ_l of monolayer NbOI₂ is measured at only 0.76 (0.45) W/m/K along the x (y) direction at room temperature, potentially attributed to strong nonharmonic interaction. Moreover, as the atomic number of the halogen elements increases, it leads to an enhancement of the power factor and a reduction in κ_l. At 300 K, the maximum anisotropic ZT values with n-type and p-type for NbOI₂ in the x (y) direction are recorded at 2.91 (0.87) and 2.12 (0.97), respectively. When the temperature rises to 500 K, its p(n)-type ZT in the x direction can attain values as high as 3.96 (3.20), indicating that the NbOI₂ has good application potential in the realm of thermoelectrics.

The nanostructure and tribological properties of molybdenum disulfide (MoS${}_2$) doped with nickel (Ni) were investigated using reactive molecular dynamics simulations. Sputtering deposition simulations captured the formation of MoS${}_2$ films with different Ni concentrations (0%, 2%, 10%, and 15% by weight) and temperatures (300 K and 670 K). The morphology of the deposited films was characterized in terms of density, crystallinity, and Ni clustering. The deposited films were then compressed and sheared in simulations designed to mimic the function of the material as a dry film lubricant. Results showed that, in these simulations, the 2% and 10% Ni-doped films exhibited lower shear stress than the 0% and 15% Ni-doped films. This non-monotonic trend was analyzed in terms of the evolution of the film nanostructure during shear. It was found that the films that exhibited low shear stress formed a lubricious particle, characterized by a crystalline core and amorphous, Ni-containing surface. The lubricious particle was formed through a combination of density, crystallinity, and Ni clustering conditions only possible with the intermediate amount of Ni. The findings suggest that optimizing the Ni concentration during sputtering may be a promising approach to improve the tribological performance of MoS${}_2$ dry film lubricants.

High-throughput $abinitio$ calculations are the indispensable parts of data-driven discovery of new materials with desirable properties, as reflected in the establishment of several online material databases. The accumulation of extensive theoretical data through computations enables data-driven discovery by constructing machine learning and artificial intelligence models to predict novel compounds and forecast their properties. Efficient usage and extraction of data from these existing online material databases can accelerate the next stage materials discovery that targets different and more advanced properties, such as electron–phonon coupling for phonon-mediated superconductivity. However, extracting data from these databases, generating tailored input files for different $abinitio$ calculations, performing such calculations, and analyzing new results can be demanding tasks. Here, we introduce a software package named “HTESP” (High-Throughput Electronic Structure Package) written in Python and Bash languages, which automates the entire workflow including data extraction, input file generation, calculation submission, result collection and plotting. Our HTESP will help speed up future computational materials discovery processes.

Graphene is a promising material with high strength that can be reduced by the presence of defects. Defect engineering can be an effective way of property control for such two-dimensional structures like graphene. In the present work, the mechanical properties of graphene with dislocation dipoles under uniaxial tension have been studied using molecular dynamics. A dislocation dipole consists of two heptagon–pentagon pairs (dislocations) separated by a dipole arm with length from 0 to 30 Å. Graphene wrinkling is allowed to reveal the underlying deformation mechanisms. Tensile deformation was applied at temperatures ranging from 0 to 3000 K. The tensile strength of defect-free graphene and graphene with Stone–Wales defect is more sensitive to temperature and loading direction than that of graphene with dislocation dipoles. The value of the dipole arm has no significant effect on the fracture strain and stress, but the presence of any dipole significantly reduces the fracture strain. With increasing temperature, the tensile strength and the anisotropy of the mechanical properties decrease. The present study provides insight into the behavior of defective graphene under uniaxial tension, which will help in its application in the design of next-generation flexible devices.