Advanced Theory and Simulations最新文献

Graphene oxide (GO) is a promising carbon-based adsorbent due to its active oxygen functional groups and tunable electronic properties. This study employs density functional theory (DFT) at the B3LYP/6-31G* level to investigate the adsorption of H₂S, NO₂, and SO₂ on four functionalized GO models: surface hydroxyl-functionalized GO (SOH-GO), surface epoxy-functionalized GO (SO-GO), edge hydroxyl-functionalized GO (EOH-GO), and edge epoxy-functionalized GO (EO-GO). The strongest adsorption energies were obtained for SOH-GO-H₂S (−2.68 kcal/mol), EOH-GO-NO₂ (−5.08 kcal/mol), and EO-GO-SO₂ (−4.87 kcal/mol). Frontier Molecular Orbital (FMO) analysis showed notable band-gap reduction, with EO-GO-SO₂ decreasing to 1.08 eV, indicating enhanced charge transfer. Density of States (DOS) profiles revealed increased peak intensities, such as SOH-GO-H₂S rising from 2, 1, and 5 eV⁻¹ to 3, 2, and 6 eV⁻¹. Chemical hardness values were the highest for SOH-GO-H₂S (η = 1.04 eV) and EOH-GO-NO₂ (η = 0.83 eV), while EO-GO-SO₂ exhibited maximum softness (β = 5.71 eV). The UV–visible bathochromic shifts up to 651 nm validated the adsorption mechanism. These findings provide strong computational insight for designing selective GO-based gas sensors.

The numerical simulations of diarrhea disease with treatment and vaccination system (DDTVS) by applying the artificial computational procedure are presented in this study. These stochastic computing neural network performances have been applied along with the Bayesian regularization computing scheme (BRCS) for the nonlinear DDTVS. The mathematical form of the diarrhea disease presents a nonlinear system based on the susceptible, infected, and recovered model, along with the treatment and vaccination factors. The neural network procedures are applied by applying twenty neurons, a log-sigmoid activation function, and the data distribution is provided as 78% for training, while the other two statistics, validation and testing, have been taken as 12%. The designed neural network approach's correctness is observed by the overlapping of source and obtained solutions, as well as the reducible absolute error. Additionally, the dependability and trustworthiness of the proposed neural network structure is examined by using function fitness, histogram measures, and correlation/regression.

The focus of the current work investigation was structural isomerism, caused by variations in atomic connectivity and positional isomerism, resulting in unique molecular geometries and electronic structures that significantly influence the photoelectronic behavior of isomers. This study examines three sets of structural isomers, based on tetracene, pyrene, and chrysene cores, namely PTA-1 and PTA-2, PTA-3 and PTA-4, PTA-5 and PTA-6, respectively. Among all the isomers, PTA-5 has the best aromatic character, with average HOMA score of 0.749, underscoring that the substitution position holds the key to π-electron delocalization. The charge transfer character of excited states is examined through Hole–Electron analysis. Among all the investigated systems, PTA-5 exhibits the highest D-index with a value of 0.529 Å, clearly indicating pronounced charge transfer in excited states. A quantum chemical approach utilizing the M06-2X functional and 6–311G** basis set is employed to examine the optical and nonlinear optical properties. PTA-1 stands out with a higher α_iso value (110.75 × 10⁻³⁶ esu), whereas PTA-3 exhibits the largest <γ> with the value of 486.23 × 10⁻³⁶ esu, with the lowest transition energy 3.54 eV. Additionally, a red-shift absorption peak at 312.7 nm in the UV-visible spectra of PTA-5 indicates potential of these compounds in optoelectronic applications.

This study examines the electrical, structural, and thermoelectric properties of newly constructed germanide halides of the rare earth metal Scandium, with chemical formula Sc₂GeX₂ (where X = Cl, Br, or I). These materials exhibit semiconductor behavior with an indirect narrow bandgap estimated to be 0.14 eV for Sc₂GeBr₂, and 0.24 eV for Sc₂GeI₂. The thermoelectric properties are analyzed using the first principles method in conjunction with the Boltzmann transport equations (BTE). Two out of three materials of a Scandium germanide halogen series are mechanically and dynamically stable. The Figure of Merit (ZT) is determined by evaluating and integrating thermoelectric coefficients such as thermal conductivity, electrical conductivity, and Seebeck coefficient. The lattice thermal conductivity values for these materials are computed for Sc₂GeI₂, exhibiting the lowest value of 1.32 Wm⁻¹ K⁻¹ and Sc₂GeBr₂ with lowest value of 2.96 Wm⁻¹ K⁻¹ at 300 K. The highest figure of merit (ZT) between these novel materials is 0.54 for Sc₂GeI₂ with Seebeck coefficient of 349.98 µV K⁻¹, electrical conductivity 13.97 × 10⁵ S m⁻¹ and electronic thermal conductivity 9.66 Wm⁻¹ K⁻¹ whereas Sc₂GeBr₂ shows lower value of figure of merit, i.e., 0.40. These results suggest that these materials can be considered as promising candidates for energy harvesting in thermoelectric applications.

Graph Neural Networks (GNNs) models are intensively used for predicting the properties of molecular materials. To improve the extraction of structural features from molecular representations, the line graph representation introduced in Atomistic Line Graph Neural Network (ALIGNN) can explicitly incorporate bond angle information. However, in terms of the convolution process, ALIGNN still employs the edge-gated graph convolution from previous GNNs, making it challenging to extract more effective information. In this work, combined with the line graph representation, we introduce multiple attention mechanisms into GNNs and propose two GNN models, N-ALIGNN and T-ALIGNN, which include the Node-Attention Layer (NAL) and Self-Attention Layer (SAL) respectively. These models can extract more structural features from both bond graph and line graph, thereby improving prediction accuracy. We conducted tests on three public datasets: JARVIS-DFT, Materials Project, and QM9. The results demonstrate that the performance of the two models is improved in different prediction tasks, proving that the introduction of appropriate attention mechanisms can enhance the prediction accuracy of the models.

Silicene, once constrained by stringent synthesis conditions, is recently reported to be synthesizable using state-of-the-art techniques such as VANS (Vacuum–Nitrogen Assisted protocol) and S-SEDNE (Seamless Substrate-Engineered Delamination and Encapsulation). This renewed feasibility drives our exploration of silicene-based spin-FETs considering the possible compatibility with existing electronic devices, which are predominantly constructed from silicon materials. Utilizing a silicene nanoribbon with a mirror-version Z-shaped zigzag structure as the channel material, a novel two-gate spin-field effect transistor (spin-FET) is proposed and the impact of channel geometry is studied. The gate voltage-driven current–voltage characteristics reveal oscillations in the electron current, which is a hallmark of the Datta–Das spin-FET. Our results show that the proposed mirror-version Z-shaped geometry yields a nearly 100-fold enhancement in drain current compared to its straight-channel counterpart, highlighting the critical role of channel design in optimizing spintronic device performance.

Van der Waals homobilayers, comprising two identical layers of a 2D material, are significant due to their tunable electronic properties and potential applications in advanced electronic and optoelectronic devices, such as transistors, photodetectors, and solar cells. This study utilizes machine learning (ML) and optimization methods to predict the bandgaps of these homobilayers. The Extra Tree Regressor (ETR), which features 48 attributes, provides reasonable accuracy. To enhance predictive performance, we integrate three optimization approaches evaluated across 1000 experiments. The ETR model achieves precise results while necessitating a reduced number of density functional theory (DFT) calculations. The Gaussian Process (GP) model, in conjunction with Bayesian optimization, improves prediction by utilizing probabilistic outputs and uncertainty assessments. We subsequently employed a hybrid methodology that combines ETR and GP models: initially utilizing ETR for fast exploration and then shifting to GP for enhanced optimization. The suggested hybrid method exhibits efficiency and accuracy, attaining robust prediction ability while utilizing fewer computer resources - showing its applicability in future research.

Platinum (Pt) alloys are essential for extremely high-temperature applications owing to their superior strength and oxidation resistance. Rhodium (Rh) is a critical alloying element, yet the atomistic mechanisms by which Rh content governs Pt-based solid solution performance remain elusive. In particular, the effects of Rh on lattice distortion and electronic structure require clarification. Here, we systematically investigate Pt-xRh (x = 0–40 wt.%) using density-functional theory (DFT) to reveal the influence of Rh content on structural, electronic, elastic, and thermal properties. The results show that increasing Rh enhances Young's and shear moduli while simultaneously intensifying mechanical anisotropy. Ideal tensile simulations indicate that higher Rh raises the elastic stress, reflecting strengthening associated with Rh-induced lattice and bonding modifications. Electronic structure analysis confirms excellent metallic conductivity, with localized charge accumulation around Rh atoms leading to notable lattice distortions. Thermal analyses demonstrate that increasing Rh elevates the Debye temperature, reduces thermal expansion, and accelerates heat capacity saturation toward the Dulong–Petit limit. These findings provide fundamental insights into the role of Rh in Pt-based solid solutions and offer valuable guidance for the rational design of high-performance noble metal solid solutions for demanding high-temperature applications.