Oleg N. Dubinin

This work couples statistical synthesis optimization with AI-guided electrode-design to boost the electrochemical performance of additively manufactured rGO-Cobalt oxide supercapacitor electrodes. The optimal synthesis conditions found using a full-factorial design (DoE) of experiment (yielded a nanocomposite with well-dispersed Co3O4 and CoO domains embedded on a conductive rGO matrix to be further applied in inks. To decouple geometry-dependent effects from material contributions, 30 printed electrode architectures were systematically evaluated, producing specific capacitances of 59.7-96.2 F/g and establishing electrode topology as a first-order performance determinant. A machine learning model was subsequently trained to extract nonlinear structure-property relationships from geometric descriptors, including connectivity, curvature, and ion transport-path metrics and generated a computationally derived electrode pattern. This AI-optimised architecture achieved a specific capacitance of 104.8 F/g, measured by cyclic voltammetry, surpassing the tested pattern library by approximately 9% under identical experimental conditions. The optimised electrode further delivered 210.5 F/g at 0.5 A/g with a rate-capability retention of ~92%, corresponding to an estimated gravimetric energy density of ~7.3 Wh/kg. These findings demonstrate that coupling DoE-validated synthesis control with AI-generated electrode topology enables simultaneous improvement of capacitance, rate performance, and energy/power metrics without chemical reformulation, establishing a scalable and transferable digital-design framework for next-generation high-performance supercapacitor electrodes.

Dental implants are thought to be implanted for life, but throughout their lifespan, they function in aggressive oral environment, resulting in corrosion of the material itself as well as possible inflammation of adjacent tissues. Therefore, materials and oral products for people with metallic intraoral appliances must be chosen carefully. The purpose of this study was to investigate the corrosion behavior of common titanium and cobalt–chromium alloys in interaction with various dry mouth products using electrochemical impedance spectroscopy (EIS). The study showed that different dry mouth products lead to different open circuit potentials, corrosion voltages, and currents. The corrosion potentials of Ti64 and CoCr ranged from −0.3 to 0 V and −0.67 to 0.7 V, respectively. In contrast to titanium, pitting corrosion was observed for the cobalt–chromium alloy, leading to the release of Co and Cr ions. Based on the results, it can be argued that the commercially available dry mouth remedies are more favorable for dental alloys in terms of corrosion compared to Fusayama Meyer’s artificial saliva. Thus, to prevent undesirable interactions, the individual characteristics of not only the composition of each patient’s tooth and jaw structure, but also the materials already used in their oral cavity and oral hygiene products, must be taken into account.

The efficiency of electronic microchip-based devices increases with advancements in technology, while their size decreases. This miniaturization leads to significant overheating of various electronic components, such as power transistors, processors, and power diodes, leading to a reduction in their lifespan and reliability. To address this issue, researchers are exploring the use of materials that offer efficient heat dissipation. One promising material is a polymer–boron nitride composite. This paper focuses on 3D printing using digital light processing of a model of a composite radiator with different boron nitride fillings. The measured absolute values of the thermal conductivity of such a composite in the temperature range of 3–300 K strongly depend on the concentration of boron nitride. Filling the photopolymer with boron nitride leads to a change in the behavior of the volt–current curves, which may be associated with the occurrence of percolation currents during the deposition of boron nitride. The ab initio calculations show the behavior and spatial orientation of BN flakes under the influence of an external electric field at the atomic level. These results demonstrate the potential use of photopolymer-based composite materials filled with boron nitride, which are manufactured using additive techniques, in modern electronics.

The coefficient of linear thermal expansion and the specific heat capacity of laser-deposited Cu-Fe alloys fabricated from tin, aluminum, chromium bronze (89–99 wt.% Cu), and SS 316L were studied. The investigated alloys had a 1:1 and a 3:1 bronze–steel ratio. The Al–bronze-based alloy showed the lowest value of linear thermal expansion coefficient: (1.212 ± 0.095)∙10⁻⁵ K⁻¹. Contrarily, this value was the highest {[(1.878–1.959) ± 0.095]∙10⁻⁵ K⁻¹} in the case of functionally graded parts created from alternating layers of bronze and steel. Differential scanning calorimetry provided experimental results about the specific heat capacity of the materials. In the case of Al–bronze-based specimens, it demonstrated a decrease in the specific heat capacity until ~260 °C and its further increase during a heating cycle. Exothermic peaks related to polymorphic transformations were observed in the Al–bronze-based specimens. Cooling cycles showed monotonous behavior for specific heat capacities. It had exothermic peaks in the case of Cr–bronze-based alloys. A Lennard-Jones potential equation was used for testing the relation between heat capacity and thermal expansion. A three-way interaction regression model validated the results and provided the relative thermal expansion of commercially pure DED-fabricated SS 316L. Its specific heat capacity was also studied experimentally and was 15–20% higher in comparison to the traditional method of production.

Additive manufacturing is a modern technique to produce parts with a complex geometry. However, the choice of the printing parameters is a time-consuming and costly process. In this study, the parameter optimization for the laser powder bed fusion process was investigated. Using state-of-the art multi-objective Bayesian optimization, the set of the most-promising process parameters (laser power, scanning speed, hatch distance, etc.), which would yield parts with the desired hardness and porosity, was established. The Gaussian process surrogate model was built on 57 empirical data points, and through efficient sampling in the design space, we were able to obtain three points in the Pareto front in just over six iterations. The produced parts had a hardness ranging from 224–235 HV and a porosity in the range of 0.2–0.37%. The trained model recommended using the following parameters for high-quality parts: 58 W, 257 mm/s, 45 µm, with a scan rotation angle of 131 degrees. The proposed methodology greatly reduces the number of experiments, thus saving time and resources. The candidate process parameters prescribed by the model were experimentally validated and tested.