Slag Resistance of Products Made from Alumina-Enriched Clays
Abstract Generated abstract
This study examines the slag resistance of alumina-enriched clay refractories in comparison with fireclay, fused mullite, and fused corundum under attack by basic slags. It combines phase-diagram based estimates of solubility and slag viscosity changes in the alumina-silica-calcia system with dissolution tests at 1450°C and crucible corrosion experiments using clay bonded compositions containing chamotte, corundum, or finely dispersed alumina. The results indicate that ceramic high-alumina bodies on a clay bond may be less resistant than ordinary fireclay because their porosity permits slag penetration and bond dissolution, despite lower calculated solubility. Fused mullite shows no clear advantage over fireclay in slags containing more than 40 percent lime, while dense fused corundum displays markedly superior resistance to all slags tested.
Full Text
Slag Resistance of Products Made from Clays Enriched with Alumina
P. G. Usov
Introduction
The enrichment of refractory clays with alumina in the manufacture of products is intended to improve their physicochemical properties. Theoretically, this aim is fully justified.
Alumina, being itself a highly refractory material, when reacting with silica forms a chemical compound that melts at a high temperature, thereby contributing to an increase in the refractoriness of the body.
At the same time, by introducing alumina into clay it is in principle possible to obtain a constant qualitative composition of the crystalline phases and to reduce the amount of glass in the body. The formation of such a structure is the principal condition in the manufacture of products with constant physicochemical properties. Free silica, liberated as a result of the mullitization of clays during firing, will also react with the introduced alumina with the formation of mullite. Ultimately, the body will consist of a crystalline phase—mullite—and glass, the composition and amount of which will depend on the composition and amount of impurities in the starting materials and on the firing temperature.
In products containing more alumina than is present in mullite, its excess forms a second crystalline phase, and the properties of the body will be determined by the differences in the chemical and thermal properties of the crystalline phases and the glass.
High-alumina refractory materials, owing to their high refractoriness, attract exceptional attention from workers in various branches of industry and from scientific institutions. It may be said that, at present, sufficient laboratory material has been accumulated on the manufacture and on certain properties of high-alumina products, but their slag resistance has been almost completely unstudied. The available literature on this question contains contradictory material and can hardly be used without additional investigation of slag attack.
In the present work, material is presented on the investigation of the behavior of high-alumina compositions toward basic slags containing small amounts of the oxides of iron, manganese, and magnesium. Data on the slag resistance of high-alumina compositions are given in comparison with the better-known refractory material—fireclay; for this purpose, tests of specimens of all compositions were accompanied by tests, under the same conditions, of specimens made of fireclay.
Study of Slag Corrosion
a) Preliminary study
A preliminary study of the interaction of molten slags with refractory materials of the \( \mathrm{Al_2O_3—SiO_2} \) system was carried out on the ternary diagram \( \mathrm{CaO—Al_2O_3—SiO_2} \), assuming thereby that neither the slag nor the refractory material contains impurities of iron, manganese, and magnesium oxides. If we take as representative points the points of intersection of the straight line connecting the composition of the slag with the refractory material with the isotherms at \(1400^\circ\) and \(1500^\circ\mathrm{C}\), then, in the melt of the composition of the representative points at the temperatures of the corresponding isotherms, there will be in equilibrium
\[ \frac{Y}{x+y} \]
—the amount of refractory material with
\[ \frac{X}{x+y} \]
—the amount of slag,
where \(X\) is the segment of the straight line—refractory material–isotherm,
\(Y\) is the segment isotherm–slag.
For calculating quantitative ratios, the liquid slag may be regarded as the solvent, and the refractory material as the soluble substance.
Then the solubility of the refractory material in the slag for the adopted compositions of the representative points will be expressed by the ratio of the amount of refractory material to slag that are in equilibrium, i.e.,
\[ \left(\frac{Y}{x+y} : \frac{X}{x+y}\right)\cdot 100. \]
Or
\[ \frac{x}{y}\cdot 100 \]
—in grams per 100 g of pure slag at the corresponding temperatures.
The calculated solubility of refractory materials of the \( \mathrm{Al_2O_3—SiO_2} \) system in slags of the following composition:
\[ \begin{aligned} &\text{No. 1 } \mathrm{SiO_2}=40\%; \qquad \mathrm{Al_2O_3}=20\%; \qquad \mathrm{CaO}=40\%;\\ &\text{No. 2 } \mathrm{SiO_2}=40\%; \qquad \mathrm{Al_2O_3}=10\%; \qquad \mathrm{CaO}=50\%;\\ &\text{No. 3 } \mathrm{SiO_2}=50\%; \qquad \mathrm{Al_2O_3}=10\%; \qquad \mathrm{CaO}=40\% \end{aligned} \]
at temperatures of \(1400^\circ\) and \(1500^\circ\mathrm{C}\) is shown on the curves in Figs. 1, 2, and 3.
The actual process of corrosion of refractory materials by slag in furnace installations differs from that considered in that the refractory material–slag system has a definite interface, at which the refractory material, at a temperature below its melting point, as a result of the action of molten slag, passes from the solid state into the melt, changing its composition and viscosity.
The actual process of corrosion of a refractory material will be determined not only by the chemical incompatibility of the reacting substances, but will also depend on the structure of the body being corroded and, chiefly, on the viscosity of the slag and its change, since the corrosion process is connected with the spatial displacement of the reacting particles.
Most silicate systems in the molten state are viscous media, and the spatial displacement of reacting particles in them proceeds considerably more slowly in comparison with the chemical reactions of silicate formation, as a result of which the total rate
Fig. 1. Solubility of refractory materials of the \(Al_2O_3—SiO_2\) system in slag No. 1.
Fig. 2. Solubility of refractory materials of the \(Al_2O_3—SiO_2\) system in slag No. 2.
the process of corroding the refractory material will be determined by the diffusion rate of the reacting particles in the slag, as the slowest process. Consequently, the calculated data obtained for solubility would be practically applicable if the refractory-material—slag system reached complete equilibrium, which in practice almost never occurs.
Thus, in order to assess the activity of a slag acting on a refractory material, the solubility data must be supplemented with data characterizing the change in slag viscosity as the refractory material dissolves in it at the interaction temperature.
Fig. 3. Solubility of refractory materials of the
\( \mathrm{Al_2O_3—SiO_2} \) system in slag No. 3
Using the same method as in calculating solubility, for the slags given above we calculated the change in viscosity upon dissolution in them of equal amounts of refractory materials of the \( \mathrm{Al_2O_3—SiO_2} \) system at temperatures of 1400°C and 1500°C.
The calculation was made from the viscosity diagrams of MacCaffery. The data obtained on the change in viscosity of slags upon dissolution in them of 10, 20, 30, etc., g of refractory per 100 g of slag are presented in Figs. 4, 5, 6, 7, 8, and 9.
From a preliminary examination of the dissolving action of the three slags given above on refractory materials of the \( \mathrm{Al_2O_3—SiO} \) system, it follows that the greatest solubility in these slags is exhibited by compositions containing about 20% \( \mathrm{Al_2O_3} \) and 80% \( \mathrm{SiO_2} \). With an increase in the content of \( \mathrm{Al_2O_3} \) or \( \mathrm{SiO_2} \) in the refractory material, their solubility decreases; moreover, the lowering effect of the oxides \( \mathrm{Al_2O_3} \) and \( \mathrm{SiO_2} \) on solubility is manifested almost equally at temperatures up to 1500°.
Figure 4
Graph labels:
- Slag
- \(T = 1400^\circ\)
- \(\mathrm{SiO_2} - 40\%\)
- \(\mathrm{Al_2O_3} - 20\%\)
- \(\mathrm{CaO} - 40\%\)
- Vertical axis: Viscosity in poises
- Horizontal axis: Composition of the refractory
- Composition scale:
- \(\mathrm{Al_2O_3}\), %: 100, 80, 70, 50, 40, 30, 20, 10, 0
- \(\mathrm{SiO_2}\), %: 0, 20, 30, 50, 60, 70, 80, 90, 100
- Curves labeled: 10 g, 20 g, 30 g, 40 g, 50 g
Figure 5
Graph labels:
- Slag
- \(T = 1500^\circ\)
- \(\mathrm{SiO_2} - 40\%\)
- \(\mathrm{Al_2O_3} - 20\%\)
- \(\mathrm{CaO} - 40\%\)
- Vertical axis: Viscosity in poises
- Horizontal axis: Composition of the refractory
- Composition scale:
- \(\mathrm{Al_2O_3}\), %: 100, 80, 70, 50, 40, 30, 20, 10, 0
- \(\mathrm{SiO_2}\), %: 0, 20, 30, 50, 60, 70, 80, 90, 100
- Curves labeled: 10 g, 20 g, 30 g, 40 g, 50 g
Figs. 4 and 5. Change in the viscosity of slag No. 1 upon dissolution in it of refractory materials of the \(\mathrm{Al_2O_3}-\mathrm{SiO_2}\) system at temperatures of 1400° and 1500°C.
Figs. 6 and 7. Change in the viscosity of slag No. 2 during the dissolution in it of refractory materials of the \(\mathrm{Al_2O_3}—\mathrm{SiO_2}\) system at temperatures of 1400° and 1500°C.
Fig. 8 and 9. Change in the viscosity of slag No. 3 during dissolution in it of refractory materials of the \(Al_2O_3—SiO_2\) system at temperatures of 1400° and 1500°C.
Silica-brick compositions, like high-alumina ones, dissolve only slightly in the slags cited. Corundum compositions have still lower solubility.
With the dissolution of refractory materials in slags, their viscosity increases either directly or after passing through a certain minimum; in this case, the increase in slag viscosity occurs most sharply when siliceous compositions and pure silica dissolve. The dissolution of high-alumina and corundum compositions, however, changes the viscosity of slags only slightly.
The difference in the effect on the change in slag viscosity upon dissolution of compositions with different contents of $\mathrm{Al_2O_3}$ and $\mathrm{SiO_2}$ is especially pronounced at temperatures above $1400^\circ\mathrm{C}$ (at $1500^\circ$ and higher).
Acid slags, owing to the great tendency of $\mathrm{SiO_2}$ toward polymerization, remain viscous even at considerable superheating. For example, the eutectic of the $\mathrm{CaO—Al_2O_3—SiO_2}$ system, of composition 62% $\mathrm{SiO_2}$, 14.75% $\mathrm{Al_2O_3}$, and 23.25% $\mathrm{CaO}$, melts at a temperature of $1170^\circ\mathrm{C}$, while the melt obtained is so thick that it lacks fluidity even at a temperature of $1500^\circ\mathrm{C}$*), which is not observed in slags with a high alumina content.
Porous high-alumina products, despite their lower absolute solubility, should be corroded more rapidly by basic slags than chamotte and acid compositions, since the latter are not impregnated by slag and are protected by a thicker diffusion layer.
Experimental Investigation
The practical investigation of slag corrosion was carried out by two methods.
First method. Slag weighing about 50 g (calculated as 10 g of slag per 1 g of the specimen tested) was melted in a graphite crucible in an electric furnace and heated to a temperature of $1450^\circ$. After the specified temperature had been established, a preheated specimen, in the form of a cube weighing about 5 g, was immersed in the slag. The moment of immersion of the specimen into the slag was taken as the beginning of the experiment. During dissolution, the molten slag was stirred slightly with a carbon rod every 5 minutes. The moment of complete dissolution of the specimen was taken as the end of the experiment. The time for dissolution of 1 g of specimen in the slag was taken as the measure of the slag resistance of the material. The temperature in the furnace during the experiment was monitored continuously with a total-radiation radiation pyrometer aimed at the mirror of the molten slag.
Second—crucible method. Crucibles were prepared from the masses subject to study. After firing and determination of water absorption and porosity, the cells of the crucibles were filled with finely ground slag (6 g per cell). The crucibles were then placed in an electric furnace and heated to a temperature of $1450^\circ$, with a holding time at this temperature of 1.5 hours, after which heating was stopped. The cooled crucibles were sawed along the diameter for qualitative study of the nature of the corrosion.
Initial Materials
The specimens were prepared by semi-dry pressing of mixes with clay from the Arinichevskoye deposit, of the following composition: $\mathrm{SiO_2}=53.38\%$; $\mathrm{Al_2O_3}=27.00\%$; $\mathrm{Fe_2O_3}=1.91\%$; $\mathrm{CaO}=0.20\%$; $\mathrm{MgO}=0.30\%$, $\mathrm{TiO_2}=1.40\%$ and loss on ignition—9.24%, and with fillers of: a) chamotte from the same clay; b) corundum of granulometric composition (Table 1); c) $\mathrm{Al_2O_3}$ from hydroxide, obtained by calcination.
*) Pavlov M. A.—Recent investigations of the physical properties of blast-furnace slags and the application of their results to the calculation of blast furnaces.
Table 1
| Sieve with number of openings per cm² | 900 | 1600 | 2500 | 4900 | 6400 | 10000 | Passed through sieve 10000 |
|---|---|---|---|---|---|---|---|
| Residue on sieve, % | 0.16 | 1.12 | 2.40 | 24.85 | 13.60 | 14.14 | 43.75 |
Compositions enriched with \(\mathrm{Al_2O_3}\) in various forms, in comparison with chamotte compositions, much more slowly develop the dense structure of the products that is most advantageous for slag resistance. Tables 2 and 3 give the porosity of specimens with chamotte and \(\mathrm{Al_2O_3}\) fillers, prepared by the same method and fired at different temperatures without holding and with holding.
Table 2
| No. | Composition, mass % | Porosity of specimens, %, fired at temperatures, °C (without holding): 1100 | 1200 | 1300 | 1400 | 1500 | 1600 |
|---|---|---|---|---|---|---|---|
| 1 | Clay 50 + chamotte 50 | 31.8 | 13.1 | 9.55 | 9.0 | 5.9 | 0.0 |
| 2 | Clay 50 + corundum 50 | 30.5 | 21.0 | 19.2 | 16.5 | 16.3 | 16.5 |
| 3 | Clay 50 + \(\mathrm{Al_2O_3}\) from hydroxide 50 | 59.7 | 44.5 | 49.2 | 41.0 | 36.7 | 32.5 |
Table 3
| No. | Composition, mass | Porosity of specimens, %, fired at 1400°C with holding, h: 0.0 | 4.0 | 8.0 | 12.0 | 20.0 | Porosity of specimens, %, fired at 1450°C with holding, h: 0.0 | 12.0 | 24.0 | 36.0 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Clay 50 + chamotte 50 | 11.0 | 10.7 | 4.08 | 4.0 | 4.2 | 8.0 | 0.8 | 0.0 | 3.18 |
| 2 | Clay 50 + corundum 50 | 17.0 | 13.8 | 13.1 | 10.8 | 9.5 | 17.0 | 10.8 | 7.9 | 7.3 |
| 3 | Clay 50 + \(\mathrm{Al_2O_3}\) from hydroxide—50 | 42.0 | 31.0 | 26.1 | 22.7 | 13.0 | 26.6 | 4.5 | 2.18 | 2.18 |
| 4 | Clay 50 + \(\mathrm{Al_2O_3}\) from aluminum chloride 50 | — | — | — | — | — | 33.8 | 18.4 | 6.0 | 5.12 |
Note: Aluminum chloride was mixed with the clay in the form of an aqueous solution of known concentration.
Finely dispersed alumina enters into reaction with the binder more readily than corundum; this difference is especially noticeable at firing temperatures of 1450°. Compositions with a higher corundum content sinter considerably more difficulty. The prepared specimens with different \(\mathrm{Al_2O_3}\) contents were treated with slags by the first method.
Table 4
| Slag compositions, % | \(\mathrm{SiO_2}\) | \(\mathrm{Al_2O_3}\) | \(\mathrm{Fe_2O_3}\) | \(\mathrm{CaO}\) | \(\mathrm{MgO}\) | \(\mathrm{MnO}\) |
|---|---|---|---|---|---|---|
| Slag No. 1 | 38.22 | 21.00 | 0.53 | 36.40 | 2.48 | 0.60 |
| Slag No. 2 | 38.60 | 12.80 | 0.30 | 46.50 | 1.51 | 0.30 |
| Slag No. 3 | 46.50 | 18.30 | 0.46 | 32.00 | 2.18 | 0.52 |
The results of the action of slag No. 1 on the specimens are given in Tables 5, 6, and 7. By the same method, products with different \(\mathrm{Al_2O_3}\) contents, factory-made and specially selected at the Kuznetsk Metallurgical Plant, were investigated. To preserve the structure, the test specimens were cut out of bricks. The characteristics of the specimens and the results of the action of slag on them are given in Table No. 7.
Table 5
| No. | Composition of the investigated specimens, % | Water absorption, % | Porosity, % | Time for dissolution of 1 g of specimen, min. (slag resistance) |
|---|---|---|---|---|
| 1 | Clay 50 + chamotte 50 | 11.47 | 22.0 | 14.8 |
| 2 | Same | 10.82 | 21.27 | 14.0 |
| 3 | Same | 9.60 | 18.97 | 15.0 |
| 4 | Same | 10.27 | 20.87 | 14.7 |
| 5 | Clay 50 + corundum 50 | 5.48 | 12.71 | 7.7 |
| 6 | Same | 4.85 | 11.21 | 8.2 |
| 7 | Clay 50 + Al₂O₃ from hydroxide 50 | 2.33 | 5.60 | 12.2 |
| 8 | Same | 3.55 | 8.76 | 11.6 |
| 9 | Clay 50 + Al₂O₃ from aluminum chloride 50 | 2.35 | 5.42 | 11.2 |
| 10 | Same | 1.94 | 4.67 | 12.6 |
Table 6
| No. | Composition of the investigated specimens, % | Water absorption, % | Porosity, % | Time for dissolution of 1 g of specimen, min. (slag resistance) |
|---|---|---|---|---|
| 1 | Clay 50 + chamotte 50 | 2.8 | 5.0 | 16.0 |
| 2 | Same | 2.7 | 5.1 | 14.0 |
| 3 | Clay 85 + corundum 15 | 3.4 | 6.2 | 15.0 |
| 4 | Same | 6.95 | 11.9 | 8.3 |
| 5 | Clay 70 + corundum 30 | 4.3 | 8.0 | 6.5 |
| 6 | Same | 3.35 | 7.5 | 7.8 |
| 7 | Clay 55 + corundum 45 | 6.5 | 12 | 8.8 |
| 8 | Same | 5.5 | 10.6 | 7.0 |
| 9 | Clay 40 + corundum 60 | 9.0 | 16.5 | 11.0 |
| 10 | Same | 12.40 | 24.0 | 8.9 |
| 11 | Clay 25 + corundum 75 | 10.3 | 19.5 | 7.9 |
| 12 | Same | 16.05 | 29.0 | 7.0 |
| 13 | Clay 10 + corundum 90 | 16.2 | 33.2 | 5.8 |
| 14 | Same | 16.37 | 37.0 | 5.4 |
Table 7
| No. | Name of specimen | SiO₂ | Al₂O₃ | Fe₂O₃ | CaO | MgO | TiO₂ | Water absorption, % | Porosity, % | Time for dissolution of 1 g of specimen, min. (slag resistance) |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Semi-acid | 82.16 | 13.76 | 2.74 | 0.20 | 0.51 | — | 11.8 | 22.17 | 12.8 |
| 2 | Chamotte | 62.70 | 31.90 | 2.44 | 0.25 | 0.24 | 2.72 | 9.15 | 17.85 | 14.0 |
| 3 | Chamotte-corundum with 5% MgO, fired at 1350° | 27.2 | 65.6 | 1.33 | 0.11 | 5.06 | 1.24 | 15.5 | 29.2 | 5.5 |
| 4 | Same, fired at 1450° | — | — | — | — | — | — | 1.33 | 3.36 | 8.8 |
The results of the action of slag No. 2 are given in Table 8.
Table 8
| No. | Composition of specimens, % | Water absorption, % | Porosity, % | Time for dissolution of 1 g of specimen, min. (slag resistance) |
|---|---|---|---|---|
| 1 | Clay 50 + chamotte 50 | 11.36 | 20.0 | 6.7 |
| 2 | Clay 70 + corundum 30 | 3.75 | 8.8 | 5.0 |
| 3 | Clay 55 + corundum 45 | 10.15 | 21.7 | 3.0 |
| 4 | Clay 40 + corundum 60 | 10.4 | 19.0 | 3.6 |
| 5 | Clay 25 + corundum 75 | 13.0 | 27.2 | 2.0 |
| 6 | Clay 10 + corundum 90 | 12.5 | 27.2 | 2.0 |
The results of the action of slag No. 3 are given in Table 9.
Table 9
| No. | Composition of specimens, % | Water absorption, % | Porosity, % | Time for dissolution of 1 g of specimen, min. (slag resistance) |
|---|---|---|---|---|
| 1 | Clays 50 + chamotte 50 | 11.4 | 20.0 | 25.0 |
| 2 | Clays 70 + corundum 30 | 4.35 | 10.5 | 14.0 |
| 3 | Clays 55 + corundum 45 | 7.0 | 17.6 | 17.0 |
| 4 | Clays 40 + corundum 60 | 10.6 | 24.0 | 20.0 |
| 5 | Clays 25 + corundum 75 | 13.4 | 30.0 | 10.0 |
| 6 | Clays 10 + corundum 90 | 15.7 | 37.0 | 6.0 |
The action of the three slags considered was checked on compositions with different contents of \(\mathrm{Al_2O_3}\) and with a dense structure of the products. The specimens were cut out in the form of monoliths and subjected to treatment with slag.
The results of the treatment are given in Table 10.
Table 10
| No. | Name of specimen | \(\mathrm{SiO_2}\), % | \(\mathrm{Al_2O_3}\), % | \(\mathrm{Fe_2O_3}\), % | \(\mathrm{TiO_2}\), % | \(\mathrm{CaO}+\mathrm{MgO}+\), % | Water absorption, % | Porosity, % | Slag No. 1: dissolution of 1 g of specimen, min. | Slag No. 2: dissolution of 1 g of specimen, min. | Slag No. 3: dissolution of 1 g of specimen, min. |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Quartz (monolith) | 99.9 | — | — | — | — | 1.01 | 2.07 | 24.0 | 10.3 | 35 |
| 2 | Chamotte glass bar (monolith) | 60.0 | 35.2 | 1.1 | 1.41; alk.— | 1.0; 1.4 | 1.5 | 2.8 | 15.0 | 7.2 | 32 |
| 3 | Electrofused mullite from the Yerevan plant (monolith) | 22.20 | 72.62 | 1.80 | 4.7 | 0.1 | 0.5 | 1.2 | 14.7 | 12.0 | 50 |
| 4 | Electrofused corundum, gray (monolith) | 2.0 | 96.0 | 1.7 | — | 0.3 | 0.0 | 0.0 | 300 | 400 | 600 |
Note: Electrocorrundum specimens were not dissolved by the slag to the end. The loss resulting from the action of the slag over the course of 150 minutes was determined by removing the surface of the specimen attacked by the slag by treatment in hydrochloric acid and by mechanical cleaning.
The study of slag corrosion by the second method was carried out with crucibles using a bond of the same clay from the Arinich deposit. The composition and characteristics of the crucibles are given in Table 11.
Table 11
| No. | Composition of crucibles, % | Water absorption, % |
|---|---|---|
| 1 | Clays 50 + chamotte 50 | 12.4 |
| 2 | Clays 70 + corundum 30 | 7.0 |
| 3 | Clays 55 + corundum 45 | 9.5 |
| 4 | Clays 40 + corundum 60 | 12.2 |
| 5 | Clays 25 + corundum 75 | 16.0 |
| 6 | Clays 10 + corundum 90 | 17.5 |
The crucibles were treated with slag No. 1.
After treatment, the cells in the first three crucibles were filled with slag. The boundary between the slag and the body of the crucible is sharply defined. In crucible No. 4, considerable penetration of the slag into the structure was observed, while in crucibles No. 5 and No. 6 the slag was completely absorbed by the structure and in some places penetrated to the outer surface. In cross-section it is quite clearly visible: crucibles with
with a high corundum content are corroded more strongly by slag. Fireclay compositions are corroded mainly from the surface, and no appreciable penetration of slag into the structure of the body is observed in them.
Into compositions with corundum, even when their initial porosity is low, slag penetrates more readily. In specimens containing up to 45% corundum, migration of iron oxides from the slag into the crucible body is observed; as a result, the slag in the cell becomes decolorized, while the crucible body, starting from the contact zone with the slag, is colored black.
In crucibles with a high corundum content—more porous ones—at high temperatures the pores are filled with molten slag to a considerable depth.
Slag absorbed by the pores dissolves the bond, and the article disintegrates. Therefore specimens of this composition are destroyed almost equally rapidly in slags of different chemical activity.
When finely dispersed \( \mathrm{Al_2O_3} \) (calcined hydroxide) is used for preparing high-alumina bodies, firing the articles to a temperature of \(1450^\circ\) does not produce in them a developed crystalline structure, which is more desirable for increasing slag resistance.
Conclusions
-
High-alumina articles manufactured by the ceramic method on a clay bond with a filler of \( \mathrm{Al_2O_3} \) are unstable against the action of basic molten slags. They are corroded more rapidly than compositions of ordinary fireclay.
-
In slag resistance, fused mullite has no advantage over articles of ordinary fireclay composition when acted upon by slags with more than 40% \( \mathrm{CaO} \).
Against the corrosive action of acid slags, fused mullite is more slag-resistant than articles of fireclay composition.
- Dense fused corundum has exceptionally high slag resistance, in comparison with fused mullite and fireclay, against the action of the slags investigated.