Studies have shown that replacing the aluminium oxide-carbon (Al₂0₃-C) nozzle material with zirconium oxide-carbon (Zr02-C) can reduce corrosion, and substituting a portion of the Zr0₂ with zirconium boride (ZrB₂) can provide even greater corrosion resistance.

In the continuous casting of steel, corrosion in the slag zone often severely limits the life of the submerged nozzles. Studies have shown that replacing the aluminium oxide-carbon (Al₂0₃-C) nozzle material with zirconium oxide-carbon (Zr02-C) can reduce this corrosion, and substituting a portion of the Zr0₂ with zirconium boride (ZrB₂) can provide even greater corrosion resistance.^1,2

Tests performed by Ogata et al.¹ have confirmed that an increased ZrB₂ content reduces corrosion. However, the results vary considerably for different slag compositions due to the strong assault of boric acid (B203), which builds during the oxidation of the ZrB₂ on the stabilized Zr0₂. This reaction releases a stabilizer additive and modifies the Zr0₂ into an unstabilized material, which is attacked more vigorously by the slag.

In accordance with the experiments carried out by Ogata et al.¹, tests were carried out with comparable slag types. A commercial composition of Zr0₂-C was chosen as the matrix material. The Zr0₂ was exchanged with ZrB₂ in 2:1 and 1:2 ratios. Additionally, a metallic component, alumina (Al), was added to enable the B₂0₃ to react to a boride and therefore prevent the attack on the Zr0₂ stabilizer. The Al works as an antioxidant and has already proven itself in combination with boron compounds in magnesium oxide-carbon (Mg0-C) refractory material.

Experiments

Five test samples were produced in conditions simulating actual production at Didier Werke AG in Germany. The composition of the samples is listed in Table 1. The Zr0₂, partly stabilized with calcium oxide (Ca0), was used in two grain fractures (<500 µm and <100 µm). The <100 µm proportion was substituted with <100 µm of ZrB₂ (Samples 2-5). Three percent Al was added as additional metallic component in Samples 3 and 5.

The tests were carried out under argon in an induction unit. An ultra-low carbon (ULC) steel was melted in a crucible of corundum refractory concrete. After the steel was melted, a casting powder slag was added that built up a molten slag layer about 10 mm thick on the steel melt. The composition of both slags is listed in Table 2.

The test samples, with a cross sectional area of 20 by 20 mm, were fixed in a lid of refractory concrete and suspended for 135 minutes in the 1600°C melt. They were submerged at a depth of 50 mm.

To evaluate the results, the corroded surface area of the test piece was compared before and after the experiments through digital picture analysis. The test pieces were also metallographically examined.

The tests and analyses of the results were carried out by DIFK in Bonn, Germany.

Results and Discussion

The rate of corrosion in the various tests is shown in Figure 1. For both slags, the corrosion reduces as the test number increases. The best results were delivered by the compositions with a Zr0₂:ZrB₂ ratio of 1:2 and an addition of 3% Al. Compared to the matrix test (Sample 1), the corrosion rate is almost halved in tests with both slags (Sample 5), and Slag 2 demonstrates stronger aggression than Slag 1.

The photographs of the microstructures show that the slag penetrates much deeper into the Zr0₂-C material than the Zr0₂-ZrB₂-C material. This is exemplified in Figures 2-6 for Samples 1-5 with Slag 1. The ZrB₂ addition reduces the wettability of the refractory material and, thus, prevents the penetration of the molten slag. Furthermore, the microstructure analysis of Samples 2 and 3 shows that the addition of Al limits the oxidation of ZrB₂.

While in Figure 3 (Sample 2) no ZrB₂ particles are found in the area of the phase boundary between slag and refractory material, they still exist in Sample 3 (Figure 4) in direct contact with the slag. Both Samples 4 and 5 (Figures 5 and 6) show ZrB2 particles in contact with the slag because of the high ZrB2 content. But the penetration of the slag into the microstructure is less for Sample 5 with the Al addition.

Work by Rigaud et al.⁴ has shown that B₂0₃ acts as a protection against oxidation for graphite. For refractory products containing C, corrosion progresses most rapidly through the graphite phase. The remaining refractory components are thereby released from the matrix and carried off by the slag. This can be prevented by antioxidants such as boron compounds, which attract oxygen and stop the oxidation of graphite. Additionally, the build-up of B₂0₃ protects the graphite from further oxidation, slowing the penetration of the slag into the layers and reducing its damaging effects. The Al also attracts oxygen, preventing the early oxidation of ZrB₂ to a certain degree and, with it, the build-up of the oxidation product, B₂0₃, which, according to Ogata et al.¹, has a negative effect on the stability of Zr0₂ in contact with the slag.

The idea, which lay at the root of adding Al, is that the build-up of B₂0₃ is bonded as Al-boride, and so its attack on the Zr0₂ stabilizer is reduced. Far more important, though, is that the oxidation of ZrB₂ is greatly reduced through the antioxidation effect of Al because much less B₂0₃ exists. This is sufficient to protect the graphite from oxidation. However, there is also less B₂0₃ available to destabilize the Zr0₂. Whether the resultant Al₂0₃ creates a boride with the B₂0₃ remains to be clarified.

The differences in the results of Ogata et al.¹ with similar slags and test conditions could lie in the basic matrix of the Zr0₂-C material. In our tests, a nearly commercial composition from the company Didier was employed (see Table 1, Sample 1). This composition differs from the material used in the Ogata et al. tests, which consists of exclusively Zr0₂ and flaked graphite. In addition, a finer ZrB₂ was used. No comment was made concerning the particle size of the Zr0₂. However, the particle size may also have a large influence on the corrosive behavior. Research into this aspect was not part of this study.

Improving Corrosion Resistance

The fact that adding ZrB₂ to Zr0₂-C material prevents corrosion in the slag region of submerged nozzles has been described on many occasions.^1,2 Previous tests have shown, however, that these improvements are only minor, especially with a slag composition like that of Slag 1. For this reason, ZrB₂, which is relatively expensive compared to other materials, is only being used for submerged entry nozzles in special cases within industry, such as the strongly corrosive Slag 2.

However, as the results in Figure 1 demonstrate, a clear increase in corrosion resistance is provided for both slags used. The higher the proportion of ZrB₂ in the Zr0₂-C material, the less rapid the corrosion, and the greater the positive effect of the metallic antioxidant.

Adding ZrB₂ to the Zr0₂-C material provides a reduction of the wettability through the slag, thereby preventing its penetration into the microstructure. Additionally, it provides a moderate build-up of B₂0₃ through oxidation, which slows the attack on the graphite. Adding Al further reduces the oxidation of ZrB₂ and, thus, its transformation into the thermo-dynamically unstable Zr0₂.

Substituting Zr0₂ with ZrB₂ in a ratio of 1:2 and simultaneously adding Al provides an almost 100% improvement in the corrosion resistance against the slag, making the use of ZrB₂ a sensible cost-effective solution.

Acknowledgements

The authors would like to thank the Didier Research Institute Wiesbaden for manufacturing the test pieces. We would also like to thank Dr. K. Mavrommatis (Inst. für Eisenhüttenkunde, RWTH Aachen), Dr. M. Ollig and Dipl.-Ing. G. Routschka (both from DIFK GmbH, Bonn) for their technical support throughout the study.

For More Information

For more information about the ZrB2 additives, contact Dr. Ing. Siegfried Prietzel, Elektroschmelzwerk Kempten GmbH, P.O. Box 1526, D-87405, Kempten, Germany; (49) 831-5618-506; fax: (49) 831-5618-495; or e-mail siegfried.prietzel@wacker.com.

*Dr.-Ing. K. Hunold and Dr.-Ing. S. Prietzel; Elektroschmelzwerk Kempten GmbH
**Dipl.-Ing. U. Kross, Prof. Dr.-Ing. habil. J. Pötschke and Dipl.-Ing. M. Thiesen, DIFK (Deutsches Institut für Feuerfest und Keramik) GmbH