IntroductionMgO-CaO refractories exhibit superior characteristics such as high refractoriness, good thermodynamic stability, and superior slag resistance. They perform particularly well in steel purification and improving steel quality, playing an important role in high-temperature industries. However, utilizing used MgO-CaO refractories becomes a challenge as the use of these refractories expands. Effective recycling of used MgO-CaO refractories can mitigate environmental pollution, conserve valuable land resources, and lower energy costs and carbon emissions during raw material production. The high-value utilization of used MgO-CaO refractories is thus critical for sustainable development.The recycling of used MgO-CaO refractories has attracted recent attention. Some researchers regenerated MgO-CaO refractories using used MgO-CaO bricks, reporting that the regenerated bricks exhibited comparable physical properties to the original ones and showed an enhanced resistance to low-basicity slag. However, the effect of used brick incorporation on the slag resistance of MgO-CaO refractories are still lack, and the underlying mechanism remains unclear.The existing research on the slag-erosion mechanism of MgO-CaO refractories mainly involves static and dynamic slag resistance experiments or on-site industrial tests to analyze the original bricks. However, the random distribution of aggregates complicates the analysis of the effect of used refractory content on slag resistance, making it difficult to elucidate its impact on the refractory. In this study, six groups of MgO-CaO refractory matrix samples with different used brick contents (i.e., 0%-50%, in mass fraction) were prepared via mixing MgO-CaO aggregate, MgO-CaO powder, MgO powder, and used MgO-CaO bricks, and crushed into fine powders. This approach eliminated the influence of aggregate distribution and focused on the matrix portion of MgO-CaO refractories. The slag-erosion behavior of the matrix with different used brick contents was analyzed to clarify the impact mechanism of used brick incorporation on the slag resistance of MgO-CaO refractories.MethodsMgO-CaO aggregate (bulk density: 3.18 g/cm³, apparent porosity: 4.96%-6.62%), used brick (bulk density: 3.04 g/cm³, apparent porosity: 7.34%-7.84%), MgO-CaO powder, MgO powder, and high-silica slag, which were provided by ShanXi Luweibao Taigang Refractories Co., Ltd., China, were used as raw materials. After batching, each aggregate was separately crushed in a vibratory crusher for 1 min to obtain a mixed powder with a particle size of 200 mesh (75 μm). The mixed powder was then placed in a mixing barrel and homogenized with 200 mesh MgO-CaO powder and MgO powder for 30 min. Subsequently, the homogenized powder was placed into a mold and pressed into 𽔶 mm cylindrical samples under 100 MPa. The green samples were sintered at 1600 ℃ for 3 h, followed by furnace-cooling. Finally, the high-silica slag-erosion test was conducted by a sessile drop method. High-silica slag was pressed into cubic blocks (2 g) with the sizes of 10 mm×10 mm×10 mm and placed at the center of the MgO-CaO matrix samples. The samples and slag were heated to 1600 ℃ for 3 h at a rate of 5 ℃/min, and then cooled within the furnace.Results and discussionAt the addition of 10% used bricks, F10 exhibits the deepest slag penetration depth (i.e., 1059 μm), the largest spread area (i.e., 95%), and the smallest contact angle (i.e., 19.5°). The slag penetration depth gradually decreases as the used brick content increases. Microscopically, the interface between each sample and high-silica slag is comprised of a slag layer, a slag-erosion layer, and an original layer. F0 has the thinnest slag-erosion layer (i.e., 214 μm), while F10 displays the thickest layer (i.e., 488 μm). The slag-erosion layer thickness decreases when the used brick content exceeds 30%. The pore analysis shows that the average pore diameter is the largest (i.e., 31.2 μm) in F10 and smaller in F50 (i.e., 24.6 μm). The pore diameter is smaller than that of F0 when the used brick content exceeds 40%.According to the thermodynamic calculation and the MgO-CaO-SiO₂ ternary phase diagram, the primary reaction products between high-silica slag and MgO-CaO refractories are Ca₂SiO₅ and Ca₂SiO₄. The slag-erosion model simulations reveal that the maximum difference in liquid phase content is only 5 g, indicating that minor variations in phase composition are not a main cause of the different erosion behaviors. However, liquid phase content significantly affects the sintering behavior. An appropriate liquid phase content enhances a wettability between solid particles, promoting particle rearrangement and filling voids, forming fine closed pores. Liquid phases encapsulate pores, preventing them from being trapped within crystals and preserving their shape and position, thus leading to a uniform pore structure. F10 has larger and more pores, while F50 exhibits finer pores. As the temperature increases, high-silica slag melts, and smaller pores absorb slag more readily due to the more intense capillary action. In contrast, larger pores with a weaker capillary action accommodate more slag and have a greater surface area, thus increasing slag penetration. The interfacial reaction in F10 is more intense, decreasing the contact angle and promoting further reactions. Compared with MgO, CaO phase exhibits a higher reactivity with SiO₂, forming Ca₃SiO₅ or Ca₂SiO₄, which are gradually eroded by slag.ConclusionsHigh-silica slag exhibited an erosive effect primarily on CaO in the MgO-CaO refractory matrix. In the slag-erosion model simulations, the introduction of used bricks slightly increased the liquid phase content at 1600 ℃, while the composition and variation of the phases during the slag-erosion process showed minimal differences. This was not main factor contributing to the different erosion behaviors. The liquid phase content significantly affected the sintering behavior of the raw materials. The pore diameter reached its maximum when adding 10% used bricks, and as the content of used bricks increased, the pore sizes in the matrix became more uniform and smaller, enhancing the slag-erosion resistance of the regenerated MgO-CaO refractory matrix. Utilizing MgO-CaO aggregate, MgO-CaO powder, MgO powder, and used bricks in appropriate proportions could effectively enhance the slag-erosion resistance of the regenerated MgO-CaO refractory matrix. This improvement could be further optimized via strictly controlling pore size according to the adjustments of material composition, particle size distribution, forming pressure, and sintering conditions, thereby significantly enhancing the slag-erosion resistance of MgO-CaO refractories.