Thermodynamic analysis of strontium deoxidizing ability in liquid iron at presence of aluminum

Phase diagram of the ternary oxide system FeO - SrO -Al2O3 was constructed for the first time. In this system, the following compounds can be formed: hercynite FeAl2O4 and five strontium aluminates - Sr4Al2O7 , Sr3Al2O6 , SrAl2O4 , SrAl4O7 , SrAl12O19 . According to the calculations performed, solid solutions of oxides are not formed in the system, as it is confirmed by the literature data. In the course of modeling, the optimal energy parameters of the theory of subregular ionic solutions were selected for the components of the oxide melt (FeO, SrO, Al2O3 ). Thermodynamic analysis of strontium deoxidizing ability in liquid iron at presence of aluminum was carried out using the technique for constructing the surface of solubility of strontium and aluminum in metal for steelmaking temperatures (1550 and 1600 °C) and carbon concentrations of 0.1 and 0.4 %. The equilibrium constants of the reactions of formation of strontium aluminates Sr3Al2O6 and SrAl2O4 from the components of the metal melt were calculated for the temperature range of 1550 - 1650 °C. It was found that the rest of strontium aluminates can be formed in liquid metal only at temperatures above 1750 °C. The base of thermodynamic data for the studied systems is given: temperature dependences of equilibrium constants for reactions occurring between components; values of interaction parameters of the first order (according to Wagner) for elements in liquid iron; values of energy parameters of the theory of subregular ionic solutions (for oxide melt). It follows from the calculations that the formation of strontium monoaluminate SrAl2O4 and corundum Al2O3 is most probable as the interaction products in Fe -Al - Sr - O and Fe -Al - Sr - C - O systems.

Synthesis, Optical Properties and Applications of Ternary Oxide Nanoparticles by a Microwave Technique

Nanocrystal preparation is in high demand due to the development of nanodevices. Nanostructures or microstructures of binary oxides such as TiO2, ZnO, and Al2O3 have been extensively prepared and studied. Synthesis of ternary oxide nanomaterials with controlled structures, morphologies, and sizes are of interest due to their potential applications in nanodevices caused by tunable energy level, bandgap, and structure. In this work, a solution-based method is used to prepare CaWO4 and LaPO4 ternary oxide nanomaterials by microwave technique. Controlled sizes and morphologies including nanorods and microspheres are synthesized by the microwave method, which is believed to be a facile and low-cost technique for the synthesis of ternary oxide nanomaterials. Furthermore, the (relatively) quick process enables high efficiency of the production. The structural and optical properties of the prepared nanomaterials are also investigated in this work. This work benefits nanomaterial synthesis for nanomanufacturing and applications in lighting and photodynamic activations as well as optical storage.

Scalable Gaussian Processes for Data-Driven Design using Big Data with Categorical Factors (IDETC2021-71570)

Scientific and engineering problems often require the use of artificial intelligence to aid understanding and the search for promising designs. While Gaussian processes (GP) stand out as easy-to-use and interpretable learners, they have difficulties in accommodating big datasets, qualitative inputs, and multi-type responses obtained from different simulators, which has become a common challenge for data-driven design applications. In this paper, we propose a GP model that utilizes latent variables and functions obtained through variational inference to address the aforementioned challenges simultaneously. The method is built upon the latent variable Gaussian process (LVGP) model where qualitative factors are mapped into a continuous latent space to enable GP modeling of mixed-variable datasets. By extending variational inference to LVGP models, the large training dataset is replaced by a small set of inducing points to address the scalability issue. Output response vectors are represented by a linear combination of independent latent functions, forming a flexible kernel structure to handle multi-type responses. Comparative studies demonstrate that the proposed method scales well for large datasets, while outperforming state-of-the-art machine learning methods without requiring much hyperparameter tuning. In addition, an interpretable latent space is obtained to draw insights into the effect of qualitative factors, such as those associated with “building blocks” of architectures and element choices in metamaterial and materials design. Our approach is demonstrated for machine learning of ternary oxide materials and topology optimization of a multiscale compliant mechanism with aperiodic microstructures and multiple materials.

Data-Driven Design via Scalable Gaussian Processes for Multi-Response Big Data With Qualitative Factors

Scientific and engineering problems often require an inexpensive surrogate model to aid understanding and the search for promising designs. While Gaussian processes (GP) stand out as easy-to-use and interpretable learners in surrogate modeling, they have difficulties in accommodating big datasets, qualitative inputs, and multi-type responses obtained from different simulators, which has become a common challenge for a growing number of data-driven design applications. In this paper, we propose a GP model that utilizes latent variables and functions obtained through variational inference to address the aforementioned challenges simultaneously. The method is built upon the latent variable Gaussian process (LVGP) model where qualitative factors are mapped into a continuous latent space to enable GP modeling of mixed-variable datasets. By extending variational inference to LVGP models, the large training dataset is replaced by a small set of inducing points to address the scalability issue. Output response vectors are represented by a linear combination of independent latent functions, forming a flexible kernel structure to handle multi-type responses. Comparative studies demonstrate that the proposed method scales well for large datasets with over 104 data points, while outperforming state-of-the-art machine learning methods without requiring much hyperparameter tuning. In addition, an interpretable latent space is obtained to draw insights into the effect of qualitative factors, such as those associated with “building blocks” of architectures and element choices in metamaterial and materials design. Our approach is demonstrated for machine learning of ternary oxide materials and topology optimization of a multiscale compliant mechanism with aperiodic microstructures and multiple materials.

Magnetic and structural properties of the solid solution CuAl2(1−x)Ga2xO4

CuAl2O4 is a ternary oxide spinel with Cu2+ ions ($$s=1/2$$ s = 1 / 2 ) primarily populating the A-site diamond sublattice. The compound is reported to display evidence of spin glass behavior but possess a non-frozen magnetic ground state below the transition temperature. On the other hand, the spinel CuGa2O4 displays spin glass behavior at ~ 2.5 K with Cu2+ ions more readily tending to the B-site pyrochlore sublattice. Therefore, we investigate the magnetic and structural properties of the solid solution CuAl2(1-x)Ga2xO4 examining the evolution of the magnetic behavior as Al3+ is replaced with a much larger Ga3+ ion. Our results show that the Cu2+ ions tend to migrate from tetrahedral to octahedral sites as the Ga3+ ion concentration increases, resulting in a concomitant change in the glassy magnetic properties of the solution. Results indicate glassy behavior for much of the solution with a general trend towards decreasing magnetic frustration as the Cu2+ ion shifts to the B-site. However, the $$x=0.1$$ x = 0.1 and 0.2 members of the system do not show glassy behavior down to our measurement limit (1.9 K) suggesting a delayed spin glass transition. We suggest that these two members are additional candidates for investigation to access highly frustrated exotic quantum states.