Nowadays, digitalization and automation in both industrial and research activities are driving forces of innovations. In recent years, machine learning (ML) techniques have been widely applied in these areas. A paramount direction in the application of ML models is the prediction of the material service time in heating devices. The results of ML algorithms are easy to interpret and can significantly shorten the time required for research and decision-making, substituting the trial-and-error approach and allowing for more sustainable processes. This work presents the state of the art in the application of machine learning for the investigation of MgO-C refractories, which are materials mainly consumed by the steel industry. Firstly, ML algorithms are presented, with an emphasis on the most commonly used ones in refractories engineering. Then, we reveal the application of ML in laboratory and industrial-scale investigations of MgO-C refractories. The first group reveals the implementation of ML techniques in the prediction of the most critical properties of MgO-C, including oxidation resistance, optimization of the C content, corrosion resistance, and thermomechanical properties. For the second group, ML was shown to be mostly utilized for the prediction of the service time of refractories. The work is summarized by indicating the opportunities and limitations of ML in the refractories engineering field. Above all, reliable models require an appropriate amount of high-quality data, which is the greatest current challenge and a call to the industry for data sharing, which will be reimbursed over the longer lifetimes of devices.

Image analysis is a powerful tool that can be applied in scientific research, industry, and everyday life, but still, there is more room to use it in materials science. The interdisciplinary cooperation between materials scientists and computer scientists can unlock the potential of digital image analysis. Traditional image analysis used in materials science, manual or computer-aided, permits for the quantitative assessment of the coexisting components at the cross-sections, based on stereological law. However, currently used cutting-edge tools for computer image analysis can greatly speed up the process of microstructure analysis, e.g., via simultaneous extraction of quantitative data of all phases in an SEM image. The dedicated digital image processing software Aphelion was applied to develop an algorithm for the automated image analysis of multi-phase high-temperature ceramic material. The algorithm recognizes each phase and simultaneously calculates its quantity. In this work, we compare the traditional stereology-based methods of image analysis (linear and planimetry) to the automated method using a developed algorithm. The analysis was performed on a digital SEM microstructural image of high-temperature ceramic material from the Cu-Al-Fe-O system, containing four different phase components. The results show the good agreement of data obtained by classical stereology-based methods and the developed automated method. This presents an opportunity for the fast extraction of both qualitative and quantitative from the SEM images.

In this paper, the corrosion mechanism of commercial alumina-spinel refractory was investigated at 1350 and 1450 °C. Disc samples were coated with shells of two different slags containing 4 and 10 wt.% SiO₂. The after-corrosion refractory was investigated in view of changes in its microstructure and phase composition by SEM/EDS and XRD techniques, respectively. At 1350 °C slags slightly infiltrated the microstructure, whereas at 1450 °C slags infiltrated the alumina-spinel refractory causing its significant corrosion. As a result of corrosion, new phases were formed, including calcium dialuminate (Ca₂Al₄O₇), calcium hexaluminate (CaAl₁₂O₁₉), and gehlenite (Ca₂AlSi₂O₇). Formation of calcium aluminate layers in the microstructure of the refractory inhibited further dissolution of alumina aggregates; however, expansive behavior of CaAl₁₂O₁₉ raised the microstructure porosity. The additional SiO₂ in the slag doubled the amount of low melting gehlenite in the matrix, accelerating the corrosion process of alumina-spinel brick at high temperatures.

Chemical resistance of commercial refractory raw materials against Cu slag is critical to consider them as candidates for the production of refractories used in Cu metallurgy. In this study, we show the comparative results for the corrosion resistance of four commercial refractory raw materials—magnesia chromite co-clinkers FMC 45 and FMC 57, PAK, and fused spinel SP AM 70—against aggressive, low-melting PbO-rich Cu slag (Z1) determined by hot-stage microscopy (up to 1450 °C) and pellet test (1100 and 1400 °C). Samples were characterized after the pellet test by XRD, SEM/EDS, and examination of their physicochemical properties to explore the corrosion reactions and then assess comparatively their chemical resistance. Since many works have focused on corrosion resistance of refractory products, the individual refractory raw materials have not been investigated so far. In this work, we show that magnesia chromite co-clinker FMC 45 exhibits the most beneficial properties considering its application in the production of refractories for the Cu industry. Forsterite (Mg₂SiO₄) and güggenite (Cu₂MgO₃) solid solutions constitute corrosion products in FMC 45, and its mixture with slag shows moderate dimensional stability at high temperatures. On the other hand, the fused spinel SP AM 70 is the least resistant to PbO-rich Cu slag (Z1); it starts to sinter at 970 °C, followed by a fast 8%-shrinkage caused by the formation of güggenite solid solution in significant amounts.