Considering the environment, resource condition and cost-reducing issues at present, the great efforts to utilize the high-aluminum-content iron ore have been made in ironmaking production in China. The high content of Al2O3 in blast furnace (BF) slag with higher TLQ and higher viscosity has a great effect on BF operation, which makes the separation of slag iron difficult and causes the worst permeability of BF gas [1, 2, 3]. Increasing MgO content is a good way to ensure the stability and fluidity of molten slag in the actual BF production.
Recently, much of these attention in understanding the MgO effect on the viscosity of slag have been focused mostly on the final slag. Kim et al.  studied the effect of MgO on the viscosity of the CaO–SiO2–20 wt% Al2O3–MgO slag system with CaO/SiO2 from 1.0 to 1.2 and MgO content in the range of 5–13 wt% at 1,500℃. Wang et al.  revealed the influence of basicity and MgO/Al2O3 ratio on the viscous behavior of CaO–SiO2–Al2O3–MgO–CaCl2 slags under conditions of C/S=0.90–1.30 and MgO/Al2O3=0.40–0.67. It indicated that the MgO/Al2O3 ratio almost has no influence on the viscosity of the chlorine-containing slags at higher temperatures. Gao et al.  revealed the effect of basicity and MgO content on the viscosity of SiO2−CaO−MgO−Al2O3 slags, which indicated that increasing the basicity was found to be more effective than increasing the MgO content in decreasing the viscosity of the slag. X.F. Zhang et al.  focused on the effect of MgO/Al2O3 on the viscosity and break point temperature of the BF final slag system under the condition of 14.84–23.84 wt% Al2O3 content and relatively high MgO content of 8.87–17.6 wt% as well as a relatively high basicity of 1.14. However, in view of the final slag, the proper MgO/Al2O3 was recommended in a range of 0.6–0.7 by X.F. Zhang et al. while that was suggested in the range of 0.2–0.5 by F. M. Shen et al.  In the actual BF production, MgO/Al2O3 of final slag was in a wide range in China. More than 700 slag contents in BFs were investigated by the present authors. The results are shown in Figure 1. MgO/Al2O3 varies in a wide range from 0.35 to 1.0 and the average of MgO/Al2O3 is 0.6 in China.
Furthermore, as can be seen from the statistics data in Figure 1, the larger the volume of BFs, the lower the MgO/Al2O3 of the slags. So the BFs can have a smooth operation in a wide MgO/Al2O3 of final slag. Besides, the increase of the viscosity of BF final slag with the decrease of MgO content in the high temperature is negligible. The sulfur removal capacity and the fluidity of final slag have not significantly been affected by lowering the MgO content as binary basicity is the most important factor. Moreover, as well known, the performance of the primary slag is important for the BF operation. Thus, it is necessary to study the fluidity and the stability of BF primary slag with the change of MgO/Al2O3 for maintaining stable BF operations.
In terms of the reduction of iron ore in the BF, the viscosities of CaO–SiO2–Al2O3–MgO–FeO system slags were measured for slag compositions at C/S=1.15–1.6, 10–13 mass% Al2O3, 5–10 mass% MgO and 0–20 mass% FeO by Lee et al. . Kim et al.  studied the viscosities of CaO–SiO2–Al2O3–MgO–FeO slag which were measured under conditions of C/S1.35–1.45, 10–18 mass% Al2O3, 3.5–10 mass% MgO and 5 mass% FeO. However, few study on the influence of MgO/Al2O3 on the viscosity and TLQ of CaO–SiO2–Al2O3–MgO–FeO slag system. Thus, in this paper, CaO–SiO2–Al2O3–MgO–FeO slag systems were investigated in the conditions of C/S=1.25–1.40, 5–15 wt% FeO, 10–18 wt% Al2O3 and 4–10 wt% MgO to have a better understanding of MgO/Al2O3 on the BF slag and the preferential slag compositions for optimal BF operation.
The cell model was adopted and further developed by the CSIRO group in the past few decades. In addition, the available experimental data as well as the modeling studies results of the viscosity of iron-containing silicate slags in the CaO–MgO–FeO–Fe2O3–SiO2–Al2O3 system has been reviewed. Based on the analysis, a new model and database that includes a substantial list of oxide species commonly found in ferrous smelting was developed, which is incorporated in a computational package, multiphase equilibrium (MPE) software [11, 12].
The scope and capability of the MPE software covers a great number of elements in various condensed phases that are commonly found in processes such as ironmaking and steelmaking, copper smelting etc. The previous published work has provided some details of the assessment carried out as well as application of the model to predict the viscosities of the melts [13, 14]. Figure 2 shows a comparison of the modeling results with the measured viscosity of BF slags. Very close agreement is observed over several orders of magnitude between model predictions and experimental measurements. The MPE software has been applied by researchers and plant metallurgists for the prediction of the multiphase equilibria of the slag as well as the viscosity. In this paper, the effect of the C/S, MgO, Al2O3 and FeO on the viscosity and the liquidus temperature of the BF primary slags were analyzed by using MPE, and all of the data were from the MPE software.
Results and discussion
Effect of Al2O3 and MgO on viscosity
The effect of Al2O3 contents on the viscosity of the SiO2–CaO–MgO–Al2O3–FeO slag system at a fixed basicity of 1.3 as a function of temperature is shown in Figure 3. With the temperature increasing, the viscosity decreases smoothly in the fully liquid region. And, the slag viscosity increases rapidly as the temperature increases to a specific value. Thermodynamically, the liquid and solid phase can coexist in equilibrium at that temperature, which should be the liquidus temperature of the slag, TLQ.
Similar to formerly published work , the additions of Al2O3 in the molten slag can increase the slag viscosity, as shown in Figure 3(a). Regardless of the Al2O3 concentration in the slag, a lower temperature of the slag corresponds a higher viscosity, which clarifies that the effect of temperature is predominant compared to the Al2O3 additions and that the effect of temperature is much more influential at high Al2O3 content. With the increase of MgO (from Figure 3(a) to Figure 3(d)), the influence of Al2O3 on the slag viscosity becomes weaker as the temperature range becomes smaller.
The influence of Al2O3 on the slag viscosity in different MgO content is too complex to distinguish easily. For example, from the calculation curve which is on behalf of the result at 4 wt% MgO, the slag viscosity increases with the increase of Al2O3 both in the high slag temperature range and in the low-temperature range. The viscosity increases with the increase of Al2O3 in the high slag temperature range but varies differently in the low-temperature range for the result of 10 wt% MgO. Thus, for a clearer view of the viscosity stability, the liquidus temperatures of different MgO content and different Al2O3 content are calculated and shown in Figure 4.
In Figure 4, TLQ of slag with 10 wt% FeO and C/S=1.3 exhibits a minimum value at about 4 wt% MgO when Al2O3 is 10 wt% and TLQ increases with the increase of MgO. In the case of 12 wt% Al2O3, the TLQ of slag achieves a minimum value at approximately 8 wt% MgO followed by 6 wt% MgO. When Al2O3 is over 14 wt%, TLQ decreases with the increase of MgO. In addition, the TLQ of slag with 4 wt% MgO and 6 wt% MgO increases from about 1,320 to 1,440 ℃ by increasing Al2O3 content from 10 wt% to 18 wt%, while that of slag with 8 wt% MgO and 10 wt% MgO first decreases and then increases with the increase of Al2O3. The TLQ of slag has a minimum value at about 12 wt% Al2O3 and 14 wt% Al2O3 when MgO contents are 8 wt% and 10 wt%, respectively. Thus, it can be indicated in terms of TLQ of slag that MgO content in primary slag is prone to be maintained at a different value when Al2O3 is different for the stable BF operation. Furthermore, the melting temperature and the viscosity of primary slag play a significant role for BF operation because the temperature of cohesive region, at which slag formation proceeds, is less than 1,500 ℃. In addition, compared to the final slag, Al2O3 content in primary slag is not assimilated sufficiently with ash as part of SiO2 and Al2O3 are from coke and/or coal. So Al2O3 in the primary slag is much lower than that of the final slag.
The isoviscosity curves of the slag are shown in Figure 5 at different MgO/Al2O3 and the fixed Al2O3 content as a function of temperature. The contents of Al2O3 in Figure 5(a)–5(e) are 10 wt%, 12 wt%, 14 wt%, 1 6 wt%, and 18 wt%, respectively. With the temperature decreasing and the Al2O3 content increasing, the isoviscosity variations become closer and closer, which clarifies that the thermal stability of primary slag starts to deteriorate. Al2O3 plays much more influence on the slag viscosity at high Al2O3 concentration with the Al2O3 content increasing and the temperature decreasing. It can be seen that the viscosities at constant MgO/Al2O3 ratio increase with the decrease of temperature in particular at a lower MgO/Al2O3 ratio and the value rises abruptly as the decrease of temperature.
In addition, it is found from Figure 5 that the MgO/Al2O3 ratio of the slag presents obvious effect on the slag viscosity. For example, in the area where the MgO/Al2O3 ratio is low in Figure 5(b), with the increase of the MgO/Al2O3 ratio, the isoviscosity curves become scarce gradually, which declares that the MgO/Al2O3 ratio can reduce the effect of the temperature on the slag viscosity gradually and that the MgO/Al2O3 ratio can improve the slag stability. However, it has an opposite effect in the area where the MgO/Al2O3 ratio of the slag is high. Therefore, the MgO/Al2O3 ratio of the slag cannot remain too much deviated, and the stability and the fluidity of the slag can be obviously improved by a suitable MgO/Al2O3 ratio. In addition, comparing Figure 5(a)–5(e), it indicates that the proper value of the MgO/Al2O3 ratio increases with the increase of Al2O3 content from 10 wt% to 18 wt% in the slag and the proper MgO/Al2O3 ratios are 0.4, 0.5, 0.55, 0.55 and more than 0.55, respectively.
The dependence of slag viscosity on the binary basicity (C/S) of the slag from 1.25 to 1.4 is shown in Figure 6. Al2O3 in Figure 6(a) is 10 wt% while that in Figure 6(b) is 16 wt%. As shown in Figure 6, the slag viscosity increases with C/S in the lower temperature when Al2O3 is 10 wt% while the slag viscosity decreases with C/S when Al2O3 is 16 wt%. Therefore, C/S has an important effect on the slag viscosity. The isoviscosity curves of 10 wt% Al2O3 and 16 wt% Al2O3 are indicated in Figure 7. It can be seen that when Al2O3 is in a low level, increasing C/S can reduce the slag viscosity. It is a good way to reduce the slag viscosity by decreasing C/S if Al2O3 is in a high level. Figure 8 exhibits the dependence of slag TLQ on the slag C/S from 1.25 to 1.4. The slag TLQ decreases with the increase of C/S and then increases at higher C/S. Besides, the TLQ at C/S=1.25 and 1.3 exhibits minimum values at about 12 wt% Al2O3, while the TLQ at C/S=1.35 and 1.4 exhibits minimum values at about 14 wt% Al2O3. The TLQ of the slag increases with the increase of the C/S when Al2O3 is lower than 14 wt%, while the TLQ of the slag decreases with the increase of the C/S when Al2O3 is higher than 14 wt%. In general, the viscosity of slags decreases with the increase of C/S because the structure of the silicate network in the melt will be destroyed, producing a simple structure with the increase in basic oxides [14, 16]. Theoretical analysis of silicate polymerization degree can be used to explain the result of slag viscosity in the composition of Al2O3 lower than 14 wt%. It is well known that Al2O3 is amphoteric oxide and its behavior depends on the composition of the slag . In the condition of high Al2O3 content slag in the study, Al2O3 would behave as acid oxide and the increasing of viscosity with the increase of Al2O3 can be understood based on the increase of the polymerization degree by the incorporation of the [AlO4]-tetrahedral into the [SiO4]-tetrahedral units, as explained in the reference .
Figure 9 exhibits the dependence of primary slag viscosity on the FeO content from 5 wt% to 15 wt%. It indicates that the slag viscosity decreases with the increase of FeO. It is also demonstrated by the isoviscosity curves in Figure 10. It can be seen that the isoviscosity curves become closer and closer when FeO is low. Dependence of TLQ on FeO content and Al2O3 content at C/S=1.3 and 8 wt% MgO is shown in Figure 11. The TLQ of the slags increase first and then decrease as Al2O3 content increases from 10 wt% to 18 wt%. It can be seen from Figure 11 that the typical slags containing Al2O3 content which is more than 14 wt% exhibit relatively high TLQ. The TLQ decreases with the increase of FeO content in the slag. Thus, the slag fluidity in BF cohesive zone could be affected by the reduction rate of iron ore and the content of the slag. Toop and Samis found that in CaO–FeO–SiO2 melts, FeO does not associate with the silicate groups in the low basicity region but supplies free Fe2+ and O2– ions to the silicate melts and causes the mixing entropy of silicate melt to increase . FeO in the slag can lower the melting temperature of slags . It indicates that improving the reduction rate of iron ore can achieve a lower TLQ and have a smooth operation in high Al2O3 content slag.
The TLQ of slag exhibits a minimum value at about 12 wt% Al2O3 and 14 wt% Al2O3 when MgO contents are 8 wt% and 10 wt%, respectively. The MgO content in primary slag is prone to be maintained at different values when Al2O3 is different for the stable BF operation.
The suitable value of the MgO/Al2O3 ratio increases with the increase of Al2O3 content from 10 wt% to 18 wt% in the slag and the proper MgO/Al2O3 ratios are 0.4–0.55.
The TLQ of the slag increases with the increase of C/S when Al2O3 is lower than 14 wt% while that decreases with the increase of the C/S when Al2O3 is higher than 14 wt%.
The slag viscosity decreases with the increase of FeO. The isoviscosity curves become closer and closer when FeO is low. Improving the reduction rate of iron ore can achieve a lower TLQ and have a smooth operation in the high Al2O3 content.
This work was financially supported by the Fundamental Research Funds for the Central Universities (FRF-BD-17-010A) and (FRF-TP-17-040A1), The National Science Foundation for Young Scientists of China (51704019) and Major Science and Technology Program for Water Pollution Control and Treatment (2017ZX07402001).
A. Shankar, M. Gornerup, A.K. Lahiri and S. Seetharaman, Metall. Trans. B, 37B (2006) 941.Google Scholar
K.X. Jiao, J.L. Zhang, Z.Y. Wang and C.L. Chen, Steel Res. Int., 88 (2017) 1.Google Scholar
C. Wang, J.L. Zhang, K.X. Jiao and Z.J. Liu, Metall. Res. Technol., 114 (2017) 205.Google Scholar
F.M. Shen, H.Y. Zheng, X. Jiang and G. Wei, Iron and Steel, 491 (2014) 1. Google Scholar
L. Zhang, S. Jahanshahi, S. Sun, M. Lim, B. Bourke, S. Wright and M. Somerville, Mater. Forum, 25 (2001) 136–153. Google Scholar
L. Zhang, S. Jahanshahi, S. Sun, C. Chen, B. Bourke, S. Wright and M. Somerville, JOM, 5411 (2002) 51. Google Scholar
L. Zhang and J. Sharif, Metall. Trans. B., 29B (1998) 187. Google Scholar
L. Zhang and S. Jahanshashi, Metall. Mater. Trans. B, 29B (1998) 177. Google Scholar
K. Hunk, H.O. Matsuura, F.M. Tsukihashi, W.L. Wang, D.J. Min and I. Sohn, Metall. Trans. B, 44B (2013) 5. Google Scholar
J.O.M. Bockris and D.C. Lowe, Proc. Roy. Soc. (London), 226A (1954) 423. Google Scholar
J.H. Park, D.J. Min and H.S. Song, Metall. Trans. B, 35B (2004) 269. Google Scholar
G.H. Zhang and K.C. Chou, Metall. Trans. B, 43B (2012) 849. Google Scholar
Q.W. Toop and C.S. Samis, Trans. Metall. Soc. AIME, 224 (1962) 878. Google Scholar
P.A. Tanskanen, S.M. Huttunen, P.H. Mannila and J.J. Harkki, Proc. 6th Int. Conf. Molten Slags, Fluxes, and Salts, KTH, Stockholm (2000), pp. 445. Google Scholar
About the article
Published Online: 2018-11-16
Published in Print: 2019-02-25
Citation Information: High Temperature Materials and Processes, Volume 38, Issue 2019, Pages 354–361, ISSN (Online) 2191-0324, ISSN (Print) 0334-6455, DOI: https://doi.org/10.1515/htmp-2018-0019.
© 2019 Walter de Gruyter GmbH, Berlin/Boston. This work is licensed under the Creative Commons Attribution 4.0 Public License. BY 4.0