Abstract: The surface of the sliding nozzle slide plate after ladle use is analyzed to study the reasons for the loosening of the working surface structure. The slide plate is made of Al2O3-zrO2-C material fired at high temperature. When heated under negative pressure, an obvious loosening layer can be formed, which is consistent with the thermodynamic calculation results. When heated under Ar atmosphere or N2 atmosphere, a slight loosening phenomenon can be observed. Compared with heating under negative pressure conditions, the degree of loosening is lighter; when heated under a buried carbon atmosphere, there is almost no loosening phenomenon. Through the heating test under negative pressure conditions, the loosening reaction of the slide gate plate working surface tissue can be reproduced, which is also an effective method to evaluate the surface damage of the slide gate plate.
Keywords: slide gate plate; working surface; negative pressure; loosening; thermodynamic calculation
1 Introduction
Sliding nozzle slides for ladles and tundishes focus on thermal peeling resistance and oxidation resistance. Most use Al₂O₃-C materials. The typical damage to the sliding nozzle slide is to the working surface, often called roughening. Many studies have examined the morphology of hair pulling. Most studies show that decarburization and oxidation cause loosening and peeling of the molten layer. This occurs due to reactions between O₂, Mn, and inclusions in molten steel at the sliding nozzle and slide plate interface during casting.
Mizobuchi et al. studied increased damage to the sliding plate’s working surface when casting ultra-low carbon aluminum-killed steel (Al-K steel). They observed reactions between various steel types and Al₂O₃-C materials while controlling the electric furnace atmosphere. They found that reactions between Al₂O₃ particles in the Al₂O₃-C sliding plate and C generated and escaped gas phases like Al(g) and CO(g), forming a loose layer.
Chifeng et al. studied the sliding nozzle slide’s damage state when casting steel types with different dissolved oxygen concentrations. They found that casting steel with low dissolved oxygen caused Al₂O₃ particles and C in the slide’s working surface to disappear, forming a loose layer. They observed penetration of slag components like Fe and CaO in the loose layer. Using a high-frequency induction furnace, Chifeng et al. studied reactions between molten steel with almost no dissolved oxygen and a sliding nozzle slide. The results showed that Al₂O₃ particles and C disappeared, forming a loose layer.
In this study, we further observed and analyzed the working surfaces of several used slide gate plates. We found that use under negative pressure conditions aggravated the loosening of the slide gate plate working surfaces. This article reports on this research.
2 Analysis of slide gate plate after use
We used the sliding nozzle slide A from a steel plant’s ladle in this study. Table 1 shows its original physical and chemical performance indicators. The slide plate, made of Al₂O₃-ZrO₂-C material fired at high temperature, is widely used in tundish and ladle applications. Figure 1 shows the appearance of the upper and lower slide plates after 6 heats (a total of 764 minutes). Steel slag is clearly attached to the upper slide plate.
Figure 2 shows the microstructure photos near the nozzle hole of the working surface. Both the upper and lower sliding plates show steel slag penetration into the working surface to a depth of 1500 μm. The lower sliding plate has a metamorphic layer about 300 μm thick from the working surface. However, the working surface remains slidable. The Al₂O₃ particles and C near the working surface of the upper sliding plate have disappeared, causing the structure to become loose. A metamorphic layer of ZrO₂-mullite aggregate can be observed in the loose layer.
We inferred that the generation of the loose layer leads to tissue peeling, causing stroke surface roughening damage. After long-term casting use, the upper slide plate is more likely to experience structural loosening than the lower slide plate.
Table 1 Performance of slide gate plate materials
| project | index |
| chemical components/% | |
| Al2O3 | 79 |
| zO2 | 10 |
| siO2 | 4 |
| F.C. | 5 |
| Bulk density/(g .cm-3) | 3 . 35 |
| Apparent porosity/% | 6.3 |
| Room temperature flexural strength/Mpa | 29 |
| High temperature flexural strength (1400℃)/Mpa | 23 |
| Linear expansion rate (1500℃)/% | 1 . 00 |

Figure 1 Photo of the appearance of slide A after using 6 heats

Figure 2 Microstructure photos of slide plate A after using 6 heats
3. Mechanism analysis of loosening of working surface structure
The reason for the loosening of the working surface structure: the oxygen in the air that invades between the upper and lower sliding plate mating surfaces oxidizes the carbon in the sliding plate material. The dissolved oxygen and inclusions in the molten steel react with the slide gate plate material, causing the slide gate plate to oxidize, decarburize and loosen its structure; in addition, there are also reactions inside the skateboard material that cause its structure to become loose.
The oxygen in the air reacts with the carbon in the refractory material, causing the carbon to vaporize and escape, causing the structure of the refractory material to become loose.
C( S ) +O2 ( g) = CO( g)↑ (1)
Similarly, the dissolved oxygen in the molten steel reacts with the carbon in the refractory material, causing the carbon to vaporize and escape, and the carbon in the refractory material dissolves into the molten steel, which causes the structure of the refractory material to become loose.
C( S ) + [O] = CO( g)↑ (2)
C( S) =[C] (3)
The reaction inside the refractory material should be the reaction between components such as Al2 O3, ZrO2 and mullite in the refractory material and the carbon component. For example, when Al2O3 reacts with carbon, Al(g), Al2O(g), and CO(g) are formed and released. The reaction formula is as follows:
Al2 O3 ( S ) + 3C(S) = 2Al(g)↑+ 3CO(g)↑ (4)
Al2 O3 ( S ) + 2C(S) = Al2O(g)↑ + 2CO(g)↑ (5)
zrO2 ( S ) + 2C(S) = ZrC(S ) + 2CO(g)↑ (6)
Al6Si2O13(S)+6C(S)=3Al2O3(S)+2SiC(S)+4CO(g) (7)
For slide A after use, the disappearance of Al2O3 and carbon components and the loosening of the organizational structure observed under the microscope can be explained by reaction equations (4) and (5). The metamorphic phenomenon of ZrO2-mullite particles can be explained by reaction equations (6) and (7).
The schematic diagram of the loosening of the slide gate plate structure is shown in Figure 3. When casting when the slide plate is not fully open, a non-filled area of molten steel will appear near the nozzle hole of the upper slide plate, which will generate negative pressure. Fluent software was used to simulate the thermal fluid, and the results showed that the non-filled area was almost a vacuum. For the skateboard in the negative pressure zone, the Al(g), Al2O(g) and CO(g) produced by the above reactions (4)~(7) accelerate the dissipation, thereby promoting the progress of these reactions. In particular, the long casting time and the small opening of the slide plate (the working surface area in the negative pressure zone is large) are the main factors that aggravate the damage to the upper slide plate.

Figure 3 Schematic diagram of the formation of negative pressure area and loose layer of slide gate plate
4 Thermodynamic calculation verification
4.1 Al2 03 and C coexistence conditions
We used the thermodynamic calculation software Factsage to calculate and verify the above reaction under negative pressure conditions. Calculation conditions assume the temperature near the working surface is 1550°C. We calculated the amount of various thermodynamically stable phases present when the Al₂O₃ and C coexistence system (Al₂O₃-C system) reaches equilibrium under different system pressures at 1550°C. Taking reaction (4) as an example, we assumed that the starting material contains 1 mol of C and 0.33 mol of Al₂O₃. Figure 4 shows the relationship between the amount of each component in the Al₂O₃-C system and the system pressure.
The figure shows that both Al₂O₃ and C exist stably in the pressure range of 0.01 to 1 atm. However, when the pressure is less than 0.01 atm, they decompose to generate CO (g), Al (g), and Al₂O (g). The lower the system pressure, the greater the reaction extent. This indicates that the lower the system pressure, the easier it is to form a loose layer. In other words, the reaction between Al₂O₃ and C is more likely to occur on the upper sliding plate working surface in the negative pressure zone, exacerbating structural loosening.
Figure 5 shows the relationship between the amount of CO (g) generated in the system and the system pressure at different temperatures. The figure indicates that the higher the temperature, the smaller the degree of negative pressure required to generate CO (g). This means high temperature promotes the decomposition reaction of Al₂O₃.

Figure 4 The relationship between the amount of each component in the Al2O3-C system and the system pressure at 1550°C

Figure 5 The relationship between the production amount of CO (g) and the system pressure in the Al2 O3 -C system at different temperatures.
4.2 Zr02 and C coexist
We calculated the amount of various thermodynamically stable phases present when the ZrO₂ and C coexistence system (ZrO₂-C system) reaches equilibrium under different system pressures at 1550°C. Taking reaction (6) as an example, we assumed that the starting material contains 1 mol of C and 0.33 mol of ZrO₂. Figure 6 shows the relationship between the amount of each component in the ZrO₂-C system and the system pressure.
The figure shows that both ZrO₂ and C exist stably when the system pressure is 1 atm. However, when the system pressure is less than 0.5 atm, they react to form ZrC. This indicates that the reaction between ZrO₂ and C under negative pressure causes C to disappear, leading to the loosening of the material structure.
Figure 7 shows the relationship between the amount of CO (g) generated in the ZrO₂-C system and the system pressure at different temperatures. The figure indicates that the higher the temperature, the smaller the degree of negative pressure required to generate CO (g). This means high temperature promotes the reaction between ZrO₂ and C.

Figure 6 The relationship between the amount of each component and the system pressure in the ZrO2-C system at 1550℃

Figure 7 The relationship between the production amount of CO(g) and system pressure in ZrO2-C system at different temperatures.
4.3 Coexistence of mullite and C
We calculated the amount of thermodynamically stable phases present when the coexistence system of mullite and C (mullite-C system) reaches equilibrium under different system pressures at 1550°C. Taking reaction (7) as an example, we assumed the starting material contains 1 mol of C and 0.46 mol of mullite. Figure 8 shows the relationship between the amount of each component in the mullite-C system and the system pressure.
The figure shows that mullite undergoes a decomposition-reduction reaction and converts into Al₂O₃ and SiC at a pressure of 1 atm. This decomposition-reduction reaction intensifies when the pressure drops below 1 atm. When the pressure is less than 0.01 atm, SiO(g), Al(g), and CO(g) are generated, further aggravating the loosening of the material. This means mullite undergoes a decomposition-reduction reaction with C under negative pressure conditions. It converts into gas phase components of C, Si, and Al, which then escape, resulting in the loosening of the material structure.
Figure 9 shows the relationship between the production amount of CO(g) in the mullite-C system and the system pressure at different temperatures. The figure indicates that the higher the temperature, the smaller the degree of negative pressure required to generate CO(g). This means high temperature promotes the reduction reaction of Al₂O₃.
For Al2O3 -SiO2 -ZrO2 -C materials, the reactions of Al2O3, mullite, ZrO2 and C are as follows: the lower the system pressure and the higher the temperature, the easier it is to react and the easier it is to form a loose layer; In addition, the order of reduction and decomposition reactions from easy to difficult is mullite, ZrO2, and Al2O3.

Figure 8 The relationship between the amount of each component and the system pressure in the mullite-C system at 1550℃

Figure 9 The relationship between the production amount of CO (g) and system pressure in the mullite-C system at different temperatures.
5 Loose layer formation test
To verify that negative pressure promotes the reduction reaction of Al₂O₃, mullite, and ZrO₂, we conducted a heating test. We placed slide gate plate A in a horizontal tube furnace and reduced the pressure to 400 Pa (0.004 atm). The sample was heated to 1500°C at 300°C per hour. After 5 hours at 1500°C, heating was stopped, but 400 Pa pressure was maintained until cooling.
We also conducted heating tests at 1500°C for 5 hours in Ar, N₂, and buried carbon atmospheres. For Ar and N₂ tests, we alternated decompression and inflation three times before heating. This involved decompression and inflation with Ar or N₂.
Figure 10 shows the sample’s appearance before and after heating. The sample surface was black before heating. After heating under negative pressure, Ar, or N₂, the sample surface turned gray, indicating C disappearance. After heating under carbon-embedded conditions, the sample surface remained black, indicating C still existed.
Table 2 shows the mass change rate, linear change rate, volume density, and apparent porosity before and after heating.

Figure 10 Appearance photos of the sample before heating and after heating under different atmosphere conditions
Table 2 Properties of samples before heating and after heating under different atmosphere conditions
| project | Negative pressure heating | Ar medium heating | N2 medium heating | buried carbon heating | before heating |
| Quality change rate/% | -2 . 86 | -0 . 75 | -1 . 19 | 0 . 65 | |
| Line change rate/% | 0 . 22 | 0 . 48 | 0 . 10 | 0 . 58 | |
| Bulk density/(g .cm-3) | 3 . 26 | 3 . 30 | 3 . 31 | 3 . 34 | 3 . 36 |
| Apparent porosity/% | 13.6 | 11.1 | 10.8 | 9.2 | 6.7 |
Table 2 data show that samples lost mass after heating under negative pressure, Ar atmosphere, and N₂ atmosphere. This mass reduction is due to material components converting to gas and escaping. The sample’s mass increased after heating with buried carbon, likely due to reactions involving elemental Al and Si. Compared to Ar and N₂ atmospheres, samples heated under negative pressure showed smaller mass reduction and volume density. However, their apparent porosity was significantly larger. This indicates significant dissipation of material components. Figure 11 shows microstructural photos of samples heated under different atmospheres.

Figure 11 Microstructure photos of samples heated under different atmosphere conditions
The figure shows that heating under negative pressure creates a loose layer about 300 μm thick from the surface inward. Voids form in this layer due to the disappearance of Al₂O₃ particles and C, and ZrO₂-mullite metamorphoses. This verifies the previous thermodynamic conclusion that “the reaction between oxide aggregate and carbon under negative pressure can loosen the material structure.” It also confirms the structural loosening of the sliding plate’s working surface after use. Heating in an Ar atmosphere modified some ZrO₂-mullite inside the sample, but Al₂O₃ particles did not disappear. A 50 μm thick oxide layer formed on the sample’s surface due to a small reaction between Al(g) and surface oxygen. The sample’s condition after heating in N₂ atmosphere mirrored that in Ar atmosphere. Heating in a CO atmosphere did not deteriorate the ZrO₂-mullite, and the material structure remained intact.
Using FactSage, we calculated the stable phases when the material with the initial composition in Table 1 reaches equilibrium at 1500°C under different atmospheres. Under negative pressure, we assumed a system pressure of 0.001 atm. For Ar and N₂ atmospheres, we set the total initial components to 100% (w), assuming sufficient Ar and N₂, calculated as more than 1% (w). For CO atmosphere, we also set the initial components to 100% (w) and assumed sufficient CO(g), calculated as more than 1% (w). Figure 12 shows the results.
Under negative pressure, stable phases include Al₂O₃(s), SiC(s), ZrC(s), CO(g), SiO(g), Al(g), and Al₂O(g). The reduction reactions of C, mullite, ZrO₂, and Al₂O₃ produce gas phases like CO(g), SiO(g), and Al(g). These reactions reduce the solid phase to 86.8% (w) and increase the gas phase to 13.2% (w). These escaping gases cause a greater mass reduction after heating. The disappearance of Al₂O₃ particles and C in the loose layer and the metamorphism of ZrO₂-mullite observed in microstructures match this result.
Heating in Ar atmosphere calculated stable phases as Al₂O₃(s), mullite(s), C(s), SiC(s), ZrC(s), Ar(g), CO(g), and SiO(g). The reactions of C, some mullite, and ZrO₂ reduce the solid phase to 99.3% (w) and the gas phase to 1.7% (w). This confirms the slight mass reduction after heating.
In N₂ atmosphere, stable phases include Al₂O₃(s), ZrO₂(s), C(s), and Si₂Al₈O₉N₈(s). The reactions of C and mullite reduce the solid phase to 98.7% (w) and the gas phase to 2.3% (w). This confirms the slight mass reduction after heating.
In CO atmosphere, stable phases include Al₂O₃(s), mullite(s), ZrO₂(s), C(s), and CO(g). Since C hardly reacts, the solid phase remains 100% (w) and the gas phase is 1.0% (w). This matches the test results showing no mass reduction and no internal tissue deterioration.
These results show that heating under negative pressure reproduces the structural loosening of the sliding nozzle slide’s working surface. Thus, it effectively evaluates the loosening mechanism of the sliding nozzle slide.

Figure 12 Relative content of each stable phase in the sample before heating and when heating reaches equilibrium at 1500°C under different atmospheres
6 Conclusion
Analysis of the slide gate plate’s working surface after use revealed that Al₂O₃ particles and C disappeared near the working surface. The ZrO₂-mullite aggregate deteriorated and became loose. A negative pressure zone formed in the non-filled part near the nozzle hole of the upper slide plate. During casting, the reaction between Al₂O₃ particles, ZrO₂-mullite aggregate, and carbon intensified near the negative pressure zone. The generated gas phase escaped outward, causing the slide gate plate’s working surface to become loose.
Thermodynamically, lower pressure and higher temperature make the aggregate react with carbon more easily, leading to loosening. The reduction and decomposition reactions, from easiest to hardest, are mullite, ZrO₂, and Al₂O₃.
The test on the formation of surface loosening layer under various heating conditions shows that heating under negative pressure can form an obvious loosening layer, which is consistent with the thermodynamic calculation results. When heated under Ar atmosphere or N₂ atmosphere, a slight loosening phenomenon can be observed. Compared with heating under negative pressure conditions, the degree of loosening is lighter; when heated under a buried carbon atmosphere, there is almost no loosening phenomenon. Through the heating test under negative pressure conditions, the loosening reaction of the slide gate plate working surface tissue can be reproduced, which is also an effective method to evaluate the surface damage of the slide gate plates.


